1600 John F. Kennedy Blvd. Ste 1800 Philadelphia, PA 19103-2899
Otologic Surgery ISBN: 978-1-4160-4665-3 Copyright © 2010, 2001, 1994 by Saunders, an imprint of Elsevier Inc. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permissions may be sought directly from Elsevier’s Health Sciences Rights Department in Philadelphia, PA, USA: phone: (+1) 215 239 3804, fax: (+1) 215 239 3805, e-mail:
[email protected]. You may also complete your request on-line via the Elsevier homepage (http://www.elsevier.com), by selecting “Customer Support” and then “Obtaining Permissions”.
Notice Knowledge and best practice in this field are constantly changing. As new research and experience broaden our knowledge, changes in practice, treatment and drug therapy may become necessary or appropriate. Readers are advised to check the most current information provided (i) on procedures featured or (ii) by the manufacturer of each product to be administered, to verify the recommended dose or formula, the method and duration of administration, and contraindications. It is the responsibility of the practitioner, relying on his or her own experience and knowledge of the patient, to make diagnoses, to determine dosages and the best treatment for each individual patient, and to take all appropriate safety precautions. To the fullest extent of the law, neither the Publisher nor the Editors assume any liability for any injury and/or damage to persons or property arising out of or related to any use of the material contained in this book. The Publisher
Library of Congress Cataloging-in-Publication Data Otologic surgery / [edited by] Derald E. Brackmann, Clough Shelton, Moisés A. Arriaga. — 3rd ed. p. ; cm. Includes bibliographical references and index. ISBN 978-1-4160-4665-3 1. Ear—Surgery. I. Brackmann, Derald E. II. Shelton, Clough. III. Arriaga, Moises A. [DNLM: 1. Ear—surgery. 2. Otologic Surgical Procedures. WV 200 O878 2010] RF126.O87 2010 617.8’059—dc22 ���������������������������������������������������������������������� 2009032119
Acquisitions Editor: Stefanie Jewell-Thomas Developmental Editor: Rachel Yard Project Manager: Jagannathan Varadarajan Design Direction: Ellen Zanolle Publishing Services Manager: Hemamalini Rajendrababu
Printed in the United States of America Last digit is the print number: 9 8 7 6 5 4 3 2 1
Dedication This book is dedicated to our mentors and teachers, Drs. Howard P. House, William F. House, and James L. Sheehy. Each of these outstanding physicians has special talents and characteristics that, when melded together, resulted in an outstanding clinical, research, and educational facility, The House Ear Clinic and Institute. Regrettably, Drs. Howard House and James Sheehy have passed away since the publication of the previous edition of this book. Howard House, the founder of our institution, was among the first to concentrate his activities in the field of otology. He devoted his career to the treatment of otosclerosis. In addition to his surgical genius, Howard was recognized as an outstanding statesman and fundraiser. Without him the House Ear Institute, which has provided so many opportunities for all of us, would not exist. He died in 2003 at the age of 95. At the time of his death, he was still coming to the office regularly and was active in development work for the Institute.
William F. House joined his brother in practice after completing his residency. A creative genius, Bill recognized that the future of otology lay in the diagnosis and treatment of diseases of the inner ear. He introduced the operating microscope and microsurgical techniques to the field of neurosurgery and revolutionized the treatment of acoustic tumors and other neurotologic problems. Bill is also recognized as instrumental in bringing the cochlear implant to the state of a practical clinical device that is now widely applied. Bill is now retired from clinical practice but, at the age of 86, remains extremely creative and is currently pursuing a number of new innovations in otology and audiology.
The final link in the chain that resulted in the success of the House Ear Clinic and Institute was Dr. James L. Sheehy. His special interest was in the field of chronic otitis media. In addition to his outstanding surgical ability, Jim possessed exceptional talent in organizational ability and teaching. Jim was responsible for developing all the patient educational materials as well as serving as the editor for all of the many publications produced by members of the House Ear Clinic. His course development, panel discussions, and slide preparation techniques became standards for our specialty. Jim had been a member of the House Ear Clinic for 48 years and died in 2006. It was our great privilege to be under the personal tutelage of each of these outstanding men. In addition to all the attributes enumerated above, first and foremost each was an outstanding physician. They practiced the art and science of surgery in the finest fashion, making it most appropriate that this book on surgical technique be dedicated to them. Derald E. Brackmann, MD Clough Shelton, MD Moisés A. Arriaga, MD
In Memoriam
On October 19, 1996, the field of otology lost one of its most influential leaders of modern times. Harold Frederick Schuknecht, MD, Professor Emeritus of the Department of Otology and Laryngology at the Harvard Medical School and Chief Emeritus of the Department of Otolarynology at the Massachusetts Eye and Ear Infirmary, was a world-renowned clinical otologist, otopathologist, teacher, and scholar. His contribution to human otopathology is unparalleled. His book, Pathology of the Ear, which he solely authored, is without question the most complete and comprehensive thesis on the subject. His clinical approach and technical innovations were based on scientific principle, and he unabashedly held others to the same standard. His influence as a teacher and role model is evidenced by the unprecedented number of his students who have followed in his footsteps and have risen as leaders in our specialty. Through his life’s work and through the lives of those he has touched, his influence lives on. Harold Frederick Schuknecht
Mendell Robinson, MD, known for his eponymous stapes prosthesis, passed away on September 29, 2007. A sketch on a napkin during an air flight in 1960 led to the development of this popular and successful prosthesis. Dr. Robinson was an internationally renowned otosclerosis surgeon and had a successful otologic practice in Providence, Rhode Island, for almost 50 years. He was so appreciated that the mayor of Providence officially declared “Mendell Robinson Day” on two separate occasions. We have chosen to leave his chapter unchanged from the previous edition.
Mendell Robinson
vii
viii
In Memoriam
As this edition of Otologic Surgery was going to press, we were saddened by the sudden death of our dear colleague Antonio De la Cruz. He succumbed to a malignant lymphoma after a very brief illness. Antonio was a member of the House Ear Clinic and Institute for 34 years and director of the Institute’s Department of Education. He directed hundreds of temporal bone dissection courses at the Institute and was responsible for teaching otologic surgery to thousands of physicians. His colleagues recognized him by election to the presidency of the American Academy of Otolaryngology– Head and Neck Surgery and the American Otologic Society. Antonio participated in more national and international courses than any physician in the history of our specialty. All of us marveled at his tireless energy, which allowed him to travel at least on a monthly basis to courses around the world. In addition to his teaching activities, Antonio maintained an active otologic and neurotologic practice, benefiting many patients with his expertise. He contributed greatly in many areas, particularly in the surgical correction of congenital atresia of the external auditory canal. His techniques are described in the chapter that he contributed to this volume. A former House Fellow wrote the following: “I am saddened to hear of Antonio’s passing. He had a unique ability to encourage others to perceive the skills of the expert to be achievable by them. His humble style, though, belied a high level of skill and savvy. His focused energy, his keen intellect, and his eagerness to teach all made him a great mentor and colleague, roles that touched so many of us over the last 30+ years. I am sure many, many will miss him but will forever cherish the perspective, skills, and tips he gave so freely. His contributions will live on.”
Antonio De la Cruz
Contributors Oliver F. Adunka, MD
Derald E. Brackmann, MD, FACS
Assistant Professor, Department of Otolaryngology−Head and Neck Surgery, University of North Carolina School of Medicine, Chapel Hill, North Carolina
Clinical Professor, Otolaryngology-Head and Neck Surgery and Neurological Surgery, University of Southern California School of Medicine; President, House Ear Clinic, and Board of Directors, House Ear Institute, Los Angeles, ���������� California
Translabyrinthine Vestibular Neurectomy
Moisés A. Arriaga, MD, MBA, FACS Clinical Professor of Otolaryngology and Neurosurgery, and Director of Otology and Neurotology, Department of Otorhinolaryngology−Head and Neck Surgery, Louisiana State University Health Science Center, New Orleans; Medical Director, Hearing and Balance Center, Our Lady of the Lake Regional Medical Center, Baton Rouge, Louisiana
Malignancies of the Temporal Bone—Limited Temporal Bone Resection; Mastoidectomy—Canal W all Down Procedure; Overview of Transtemporal Skull Base Surgery; Surgery for Glomus Tumors and Other Lesions of the Jugular Foramen; Anterior and Subtemporal Approaches to the Infratemporal Fossa
Gregory A. Ator, MD Associate Professor, Department of Otolaryngology−Head and Neck Surgery; Director, Division of Otology/Neurotology, University of Kansas Medical Center, Kansas City, Kansas
Traumatic Facial Paralysis
James E. Benecke, Jr., MD Clinical Professor, Otolaryngology−Head and Neck Surgery, St. Louis University School of Medicine; Section Chief, Otolaryngology, Missouri Baptist Medical Center; Director, Otology Associates, Inc. St. Louis, Missouri
Otologic Instrumentation
Leonard P. Berenholz, MD Trumbull Memorial Hospital, Department of Otolaryngology, Warren, Ohio
Special Problems of Otosclerosis Surgery
K. Paul Boyev, MD Assistant Professor, �������������������������������������� Department of Otolaryngology-Head and Neck Surgery������������������������������������������������ ; Director, Division of Otology/Neurotology, and Director, Hearing and Balance Center, University of South Florida College of Medicine, Tampa, Florida
Drainage Procedures for Petrous Apex Lesions; Surgery for Glomus Tumors and Other Lesions of the Jugular Foramen; Middle Fossa Approach; Auditory Implants for the Central Nervous System; Management of Postoperative Cerebrospinal Fluid Leaks
Craig A. Buchman, MD, FACS Professor and Chief, Division of Otology, Neurotology, and Skull Base Surgery, Department of Otolaryngology-Head and Neck Surgery, University of North Carolina School of Medicine, Chapel Hill, North Carolina
Translabyrinthine Vestibular Neurectomy
John P. Carey, MD Associate Professor, Otolaryngology-Head and Neck Surgery, Johns Hopkins University School of Medicine; Attending, The Johns Hopkins Hospital, Baltimore, �������� Maryland
Superior Semicircular Canal Dehiscence Syndrome
Ricardo L. Carrau, MD Professor, Departments of Neurological Surgery and Otolaryngology, University of Pittsburgh School of Medicine; Minimally Invasive EndoNeurosurgery Center and Center for Skull Base Surgery, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania
Anterior and Subtemporal Approaches to the Infratemporal Fossa; Endoscopic Endonasal Approaches to the Skull Base and Paranasal Sinuses
Stephen P. Cass, MD Associate Professor, Department of Otolaryngology, University of Colorado, Denver, Colorado
Chemical Treatment of the Labyrinth
Ray C. Chang, MD Department of ������������������������������������� Otolaryngology, University of Miami; Department of ������������������������������������������ Otolaryngology, Jackson Memorial Hospital/ University of Miami, Miami, Florida
Intraoperative Neurophysiologic Monitoring
Cochleosacculotomy ix
Contributors
Douglas A. Chen, MD
Shervin R. Dashti, MD, PhD
Adjunct Associate Professor of Surgery, Allegheny University of the Health Sciences; Director, Division of Neurology, Hearing and Balance Center, Allegheny General Hospital, Pittsburgh, ������������ Pennsylvania
Neurosurgical Institute of Kentucky, Norton Neuroscience Institute, Louisville, Kentucky
Dural Herniation and Cerebrospinal Fluid Leak
Henry H. Chen, MD, MBA Resident Physician, ����������������������������������������� Department of ��������������������������� Otolaryngology, University of Colorado Health Sciences Center, Denver, Colorado
Traumatic Facial Paralysis
Vascular Considerations in Neurotologic Surgery
M. Jennifer Derebery, MD Clinical Professor of Otolaryngology, University of Southern California School of Medicine; Staff, St. Vincent’s Medical Center, and Associate, House Ear Clinic, Los Angeles, California
Surgery of Ventilation and Mucosal Disease
Joseph M. Chen, MD, FRCS(C)
Shaun C. Desai, MD
Associate Professor, Department of Otolaryngology, University of Toronto; Staff Surgeon, Sunnybrook and Women’s College Health Science Center, Toronto, Ontario, Canada
The George Washington University Medical Center, Washington, DC
Middle Cranial Fossa—Vestibular Neurectomy; Transotic Approach
Vivek R. Deshmukh, MD
Sarah S. Connell, MD Neuro-otology Fellow, ������������������������������ Department of ���������������� Otolaryngology, University of Miami Ear Institute, Miami, Florida; Physician, Otolaryngology, Kaiser Permanente, Walnut Creek, California
Intraoperative Neurophysiologic Monitoring
Benjamin T. Crane, MD, PhD Assistant Professor, ����������������������������������������� Department of ��������������������������� Otolaryngology, University of Rochester; Assistant Professor, Otolaryngology, Strong Memorial Hospital, Rochester, New York
Superior Semicircular Canal Dehiscence Syndrome
Antonio De la Cruz, MD† Formerly Clinical Professor, Otolaryngology-Head and Neck Surgery, University of Southern California, Los Angeles, California
Congenital Malformation of the External Auditory Canal and Middle Ear; Transcochlear Approach to Cerebellopontine Angle Lesions
Robert D. Cullen, MD Otologic Center, Inc., and Midwest Ear Institute, Kansas City, �������� Missouri
Surgery for Cochlear Implantation; Translabyrinthine Approach
Calhoun D. Cunningham III, MD Consultant Clinical Professor of Otolaryngology−Head and Neck Surgery, Duke University School of Medicine; VicePresident of Carolina Ear and Hearing Clinic, and Co-Director of Carolina Research Institute, Raleigh, North Carolina
Ossicular Reconstruction †Deceased
Vascular Considerations in Neurotologic Surgery
Director, Cerebrovascular and Endovascular Neurosurgery, Department of Neurosurgery, George Washington University, Washington, DC
Vascular Considerations in Neurotologic Surgery
John L. Dornhoffer, MD, FACS Professor and Director, Division of Neurotology, Department of Otolaryngology−Head and Neck Surgery; Professor, Department of Neurobiology and Developmental Sciences; and ���������������������������������������������������� Medical Director, ENT Clinic and Audiology Services, University of Arkansas for Medical Sciences, Little Rock, Arkansas
Cartilage Tympanoplasty
Adrien A. Eshraghi, MD Associate Professor, Department of Otolaryngology, University of Miami, Miller School of Medicine, Miami, Florida
Intraoperative Neurophysiologic Monitoring
Jose N. Fayad, MD Associate Professor, Clinical Otolaryngology, University of Southern California; Associate, Otology/Neurotology, House Ear Clinic, Los Angeles, California
Otologic Instrumentation; Tympanoplasty—Outer Surface Grafting Technique; Auditory Implants for the Central Nervous System
Ugo Fisch, MD Professor Emeritus of Otolaryngology, University of Zurich; Head of ENT and Skull Base Surgery, University Hospital, Zurich, Switzerland
Middle Cranial Fossa—Vestibular Neurectomy; Transotic Approach
Contributors
xi
David R. Friedland, MD, PhD
Neil A. Giddings, MD
Associate Professor and Chief, Division of Otology and Neuro-otologic Skull Base Surgery, Department of Otolaryngology and Communication Sciences, Medical College of Wisconsin; Attending Physician, Division of Pediatric Otolaryngology, Children’s Hospital of Wisconsin; Attending ��������������������������������������������� Physician,����������������������������������� Department of Otolaryngology-Head and Neck Surgery, Froedtert and The Medical College of Wisconsin Hospital, Milwaukee, Wisconsin
Sacred Heart Medical Center and Deaconess Medical Center, Spokane, ���������� Washington
Stereotactic Radiosurgery of Skull Base Tumors
Rick A. Friedman, MD, PhD Associate Professor, Otolaryngology, University of Southern California; Chief, Division of Skull Base Surgery, Cedars-Sinai Medical Center; Chief, Section on Hereditary Disorders, Cell and Biology Genetics Division, House Ear Clinic/House Ear Institute, Los Angeles, ���������� California
Surgery of V entilation and Mucosal Disease; Translabyrinthine Approach; Extended Middle Cranial Fossa Approach
Takanori Fukushima, MD, MMSc Consulting Professor, Department of Neurosurgery, Duke University Medical Center, Durham, North Carolina; Professor, Department of Neurosurgery, West Virigina University, Morgantown, West Virginia
Extreme Lateral Infrajugular Transcondylar Approach for Resection of Skull Base Tumors
Bruce J. Gantz, MD, FACS Professor and Department Head, Otolaryngology-Head and Neck Surgery, University of Iowa; Professor and Department Head, Otolaryngology-Head and Neck Surgery, University of Iowa Hospitals and Clinics, Iowa City, Iowa
Canal W all Reconstruction Tympanomastoidectomy; Management of Bell’s Palsy and Ramsay Hunt Syndrome
Gale Gardner, MD Clinical Professor, Otolaryngology/Head and Neck Surgery, Louisiana State University Health Science Center-Shreveport; Active Staff, Otolaryngology-Head and Neck Surgery, Louisiana State University ��������������� Medical Center, Shreveport, ��������� Louisiana
Tympanoplasty—Undersurface Graft Technique: Transcanal Approach
Paul A. Gardner, MD Department of Neurological Surgery, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania
Tympanoplasty—Undersurface Graft Technique: Transcanal Approach; Endoscopic Endonasal Approaches to the Skull Base and Paranasal Sinuses
Drainage Procedures for Petrous Apex Lesions
Michael E. Glasscock III, MD, FACS Adjunct Professor, Otolaryngology−Otology, Vanderbilt University Medical Center, Nashville, Tennessee
Tympanoplasty—Undersurface Graft Technique: Postauricular Approach
Samuel P. Gubbels, MD Assistant Professor, Surgery, Division of Otolaryngology, University of Wisconsin–Madison, Madison, ��������� Wisconsin
Canal Wall Reconstruction Tympanomastoidectomy; Management of Bell’s Palsy and Ramsay Hunt Syndrome
Ophir Handzel, MD, LLB Department of Otology and Laryngology, Harvard Medical School; Fellow, Otology/Neurotology, Otolaryngology, Massachusetts Eye and Ear Infirmary, Boston, ������������� Massachusetts
Diagnosis and Management of the Patulous Eustachian Tube
Steven A. Harvey, MD Clinical Assistant Professor, Department of Otolaryngology, Medical College of Wisconsin, Milwaukee, Wisconsin
Complications of Surgery for Chronic Otitis Media
Todd A. Hillman, MD Adjunct Clinical Faculty, Otolaryngology, Louisiana State University, New Orleans, Louisiana; Associate, Pittsburgh Ear Associates, and Staff, Division of Neurotology, Allegheny General Hospital, Pittsburgh, Pennsylvania
Petrosal Approach
William E. Hitselberger, MD Neurosurgery, St. Vincent’s Hospital, Los Angeles, California
Auditory Implants for the Central Nervous System
Barbara Stahl Hoskins, RN, BSN Allied Health Practitioner, Doheny Surgery, St. Vincent’s Medical Center, Los Angeles; Allied Health Practitioner, Surgery, Kaiser Sunset, Hollywood; Private Scrub Nurse, Neurosurgical Department, St. Vincent’s Medical Center, Los Angeles, ���������� California
Otologic Instrumentation
xii
Contributors
Howard P. House, MD†
Herman A. Jenkins, MD
Formerly Professor Emeritus, University of Southern California; Founder and Chairman Emeritus, House Ear Institute, St. Vincent’s Medical Center, Los Angeles, California
Professor and Chairman, Department of Otolaryngology, University of Colorado Health Sciences Center; Chief, Otolaryngology Service, University of Colorado Hospital, Denver, Colorado
Total Stapedectomy
Traumatic Facial Paralysis
John W. House, MD
Amin B. Kassam, MD, FRCS(C)
Clinical Professor, Department of Otolaryngology, University of Southern California School of Medicine; President, House Ear Institute, Los Angeles, California
Associate Professor, Department of Neurological Surgery, University of Pittsburgh School of Medicine; Director, Minimally Invasive EndoNeurosurgery Center, and Co-Director, Center for Skull Base Surgery, University of Pittsburgh Medical Center, Pittsburgh, ������������ Pennsylvania
Translabyrinthine Approach
William F. House, MD Formerly at Hoog Hospital, Newport Beach, California
Middle Fossa Approach
Brandon B. Isaacson, MD Assistant Professor, Otolaryngology-Head and Neck Surgery, University of Texas Southwestern Medical Center at Dallas; Attending, Otolaryngology-Head and Neck Surgery, Zale Lipshy University Hospital, Dallas, Texas
Office Management of Tympanic Membrane Perforation and the Draining Ear
Robert K. Jackler, MD Sewall Professor and Chair and Associate Dean (CME), Otolaryngology-Head and Neck Surgery, Stanford University School of Medicine, Stanford, ���������� California
Retrosigmoid Approach to Tumors of the Cerebellopontine Angle
C. Gary ���������������������� Jackson, MD, FACS Clinical Professor, Department of Surgery, University of North Carolina School of Medicine, Chapel Hill, North Carolina
Tympanoplasty—Undersurface Graft Technique: Postauricular Approach
Ivo P. Janecka, MD, FACS Professor, Harvard Medical School; Director, Skull Base International; Staff, Children’s Hospital, Brigham and Women’s Hospital, Boston, Massachusetts
Anterior and Subtemporal Approaches to the Infratemporal Fossa; Endoscopic Endonasal Approaches to the Skull Base and Paranasal Sinuses
David M. Kaylie, MD Associate Professor, Department of Surgery, Division of Otolaryngology, Duke University Medical Center, Durham, North Carolina
Tympanoplasty—Undersurface Graft Technique: Postauricular Approach
Bradley W. Kesser, MD Assistant Professor, Department of Otolaryngology-Head and Neck Surgery, University of Virginia Health System, Charlottesville, Virginia
Surgery of Ventilation and Mucosal Disease
Joe W. Kutz, MD Assistant Professor, Otolaryngology-Head and Neck Surgery, University of Texas Southwestern Medical Center at Dallas; Attending, Otolaryngology-Head and Neck Surgery, Zale Lipshy University Hospital, Dallas, Texas
Office Management of Tympanic Membrane Perforation and the Draining Ear
Jed A. Kwartler, MD, MBA Clinical Associate Professor, Division of Otolaryngology, University of Medicine and Dentistry of New Jersey, Newark, New Jersey
Total Stapedectomy
Malignancies of the Temporal Bone-Radical Temporal Bone Resection
John P. Leonetti, MD
Peter J. Jannetta, MD
Otolaryngology-Head and Neck Surgery, ������������������ Loyola University Medical Center,���������� Maywood, Illinois ���������
Professor of Neurosurgery, Department of Neurosurgery, Drexel University College of Medicine, Philadelphia; Vice Chairman, Department of Neurosurgery, Allegheny General Hospital, Pittsburgh, Pennslyvania
Operations for V ascular Compressive Syndromes †Deceased
Malignancies of the Temporal Bone—Limited Temporal Bone Resection
Contributors
xiii
Robert E. Levine, MD
Lloyd B. Minor, MD
Clinical Professor, Ophthalmology, University of Southern California Keck School of Medicine; Co-Founder and CoDirector, Facial Nerve Clinic, House Ear Clinic, Los Angeles, California,
Andelot Professor and Director, Department of Otolaryngology-Head and Neck Surgery, The Johns Hopkins University School of Medicine, Baltimore, Maryland
Care of the Eye in Facial Paralysis
James Lin, MD Assistant Professor, Department of Otolaryngology, Louisiana State University Health Sciences Center, New Orleans, Louisiana
Translabyrinthine Approach
William H. Lippy, MD, FACS Chief Physician/Surgeon, Otolaryngology, The Lippy Group for Ear, Nose, and Throat; Physician, Otolaryngology, Trumbell Memorial Hospital, Warren, Ohio
Special Problems of Otosclerosis Surgery
Philip D. Littlefield, MD Otology and Neurotology, Department of Otolaryngology, Walter Reed Army Medical Center, Washington, DC
Complications of Surgery for Chronic Otitis Media
Teresa M. Lo, MD Department of Otolaryngology, The Southeast Permanente Medical Group, Atlanta, Georgia
Perilymphatic Fistula
Larry B. Lundy, MD Associate Professor, Otolaryngology-Head and Neck Surgery, Mayo Clinic, Jacksonville, Florida,
Laser Revision Stapedectomy
William M. Luxford, MD Clinical Professor, Otolaryngology, University of Southern California Keck School of Medicine; Associate, House Clinic, Los Angeles, ���������� California
Hypoglossal Facial Anastomosis Surgery for Cochlear Implantation
John T. McElveen, Jr., MD Consultant Clinical Professor of Otolaryngology–Head and Neck Surgery, Duke University School of Medicine; President of Carolina Ear and Hearing Clinic, and Director of Carolina Research Institute, Raleigh, North Carolina
Ossicular Reconstruction
Michael J. McKenna, MD Professor of Otology and Laryngology, Department of Otology and Laryngology, Harvard Medical School; Surgeon in Otolaryngology, Department of Otolaryngology, Massachusetts Eye and Ear Infirmary, Boston, ������������� Massachusetts
Cochleosacculotomy; Transcanal Labyrinthectomy
Superior Semicircular Canal Dehiscence Syndrome
Edwin M. Monsell, MD, PhD Professor, Otolaryngology-Head and Neck Surgery, Wayne State University, Detroit, Michigan
Chemical Treatment of the Labyrinth
Joseph B. Nadol, Jr., MD Walter Augustus Lecompte Professor and Chair, Department of Otology and Laryngology, Harvard Medical School; Chief of Otolaryngology, Department of Otolaryngology, Massachusetts Eye and Ear Infirmary, Boston, ������������� Massachusetts
Transcanal Labyrinthectomy
Julian M. Nedzelski, MD, FRCS Professor Emeritus, Department of Otolaryngology-Head and Neck Surgery, University of Toronto; Consultant, Department of Otolaryngology-Head and Neck Surgery, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada
Chemical Treatment of the Labyrinth
J. Gail �������������������� Neely, MD, FACS Professor and Director, Otology/Neurotology/Base of Skull Surgery; Director of Research, Department of OtolaryngologyHead and Neck Surgery, Washington University School of Medicine; Attending, Otolaryngology-Head and Neck Surgery, BarnesJewish Hospital and St. Louis Children’s Hospital, St. Louis, Missouri
Surgery of Acute Infections and Their Complications
James L. Netterville, MD The Mark C. Smith Professor, and Director of Head and Neck Surgery, Vanderbilt Department of OtolaryngologyHead and Neck Surgery, Vanderbilt Medical Center, Nashville, Tennessee
Rehabilitation of Lower Cranial Nerve Deficits after Neurotologic Skull Base Surgery
Steven R. Otto, MA Chief Audiologist and Coordinator, ABI Program, Department of Auditory Implants and Perception, House Ear Institute, Los Angeles, California
Auditory Implants for the Central Nervous System
Mark D. Packer, MD Neurotology Fellow, Otolaryngology, The Ohio State University; Lieutenant Colonel, United States Air Force, Columbus, Ohio
Surgery of the Endolymphatic Sac
xiv
Contributors
Lorne S. Parnes, MD, FRCS(C)
Miriam I. Redleaf, MD
Professor, Otolaryngology and Clinical Neurological Sciences, University of Western Ontario; Site Chief, Otolaryngology, University Hospital–London Health Sciences Centre, London, Ontario, Canada
Assistant Professor, Otolaryngology, University of Chicago and University of Chicago Medical Center, Chicago, Illinois
Posterior Semicircular Canal Occlusion for Benign Paroxysmal Positional V ertigo
Joseph B. Roberson, Jr., MD
Rodney Perkins, MD Clinical Professor of Surgery, Stanford University, Palo Alto, California
Canalplasty for Exostoses of the External Auditory Canal and Miscellaneous Auditory Canal Problems
Brian P. Perry, MD, FACS Clinical Associate Professor, The University of Texas, Health Science Center–San Antonio, San Antonio, Texas
Management of Bell’s Palsy and Ramsay Hunt Syndrome
Thomas M. Pilkington, MD Resident Physician, Division of Otolaryngology-Head and Neck Surgery, Department of Surgery, Duke University Medical Center, Durham, North Carolina
Extreme Lateral Infrajugular Transcondylar Approach for Resection of Skull Base Tumors
Dennis S. Poe, MD Associate Professor, Otology and Laryngology, Harvard Medical School; Associate Surgeon, Otolaryngology, Children’s Hospital, Boston, ������������� Massachusetts
Diagnosis and Management of the Patulous Eustachian Tube
Sanjay Prasad, MD Clinical Assistant Professor, Department of OtolaryngologyHead and Neck Surgery, Georgetown University Medical Center, Washington, DC; President and Founder, Metropolitan Ear Group, Besthesda, Mayland
Malignancies of the Temporal Bone—Radical Temporal Bone Resection
Daniel M. Prevedello, MD Department of Neurological Surgery, University of Pittsburgh School of Medicine; Minimally Invasive EndoNeurosurgery Center and Center for Skull Base Surgery, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania
Endoscopic Endonasal Approaches to the Skull Base and Paranasal Sinuses
Management of Bell’s Palsy and Ramsay Hunt Syndrome
Stanford University Hospital, Palo Alto, and San Ramon Regional Medical Center, San Ramon, California
Canalplasty for Exostoses of the External Auditory Canal and Miscellaneous Auditory Canal Problems; Avoidance and Management of Complications of Otosclerosis Surgery
Michael A. Roberts, MD Comprehensive Ophthalmologist, St. Vincent’s Medical Center, Los Angeles, California
Care of the Eye in Facial Paralysis
Mendell Robinson, MD† Formerly Clinical Associate Professor, Brown University School of Medicine; Senior Surgeon, Miriam Hospital, Rhode Island Hospital, Providence, Rhode Island
Partial Stapedectomy
Grayson K. Rodgers, MD Birmingham Hearing and Balance Center, Birmingham, Alabama
Management of Postoperative Cerebrospinal Fluid Leaks
Peter S. Roland, MD Professor and Chairman, Otolaryngology-Head and Neck Surgery, University of Texas Southwestern Medical Center at Dallas; Chief of Service, Otolaryngology-Head and Neck Surgery, Zale Lipshy University Hospital, Dallas, Texas
Office Management of Tympanic Membrane Perforation and the Draining Ear
Christina L. Runge-Samuelson, PhD Associate Professor and Co-Director, Koss Cochlear Implant Program, Department of Otolaryngology and Communication Sciences, Medical College of Wisconsin; Division of Pediatric Otolaryngology, Children’s Hospital of Wisconsin, and Department of Otolaryngology-Head and Neck Surgery, Froedtert and The Medical College of Wisconsin Hospital, Milwaukee, Wisconsin
Stereotactic Radiosurgery of Skull Base Tumors
Leonard P. Rybak, MD, PhD Professor of Surgery, Southern Illinois University; Memorial Medical Center and St. John’s Hospital, Springfield, Illinois
Chemical Treatment of the Labyrinth
Contributors
xv
Barry M. Schaitkin, MD
David W. Sim, FRCS Ed (URL)
Professor, Department of Otolaryngology, University of Pittsburgh School of Medicine, University of Pittsburgh Medical Center Shadyside, Pittsburgh, Pennsylvania
Clinical Tutor, University of Edinburgh, Edinburgh, Scotland
Facial Reanimation Techniques
Harlod F. Schuknecht, MD† Formerly Professor and Chairman Emeritus, Department of Otology and Laryngology, Harvard Medical School; Emeritus Chief of Otolaryngology, Department of Otolaryngology, Massachusetts General Hospital, Boston, Massachusetts
Cochleosacculotomy
Raymond F. Sekula, Jr., MD Assistant Professor Neurological Surgery, Department of Neurological Surgery, Allegheny General Hospital, Pittsburgh, Pennsylvania
Operations for V ascular Compressive Syndromes
Robert V. Shannon, PhD Adjunct Professor, Biomedical Engineering, University of Southern California; �������������������������� Department Head, Auditory Implants ������������� and Perception, House Ear Institute, Los Angeles, California
Auditory Implants for the Central Nervous System
M. Coyle Shea, Jr., MD Associate Clinical Professor, Department of Otolaryngology-Head and Neck Surgery, University of Tennessee Center for the Health Sciences; Active Staff, Baptist Memorial Hospitals; Courtesy Staff, Methodist Hospitals, Memphis, ��������� Tennessee
Tympanoplasty—Undersurface Graft Technique: Transcanal Approach
James L. Sheehy, MD† Formerly Clinical Professor of Surgery and Otolaryngology, University of Southern California School of Medicine, Los Angeles, California
Tympanoplasty—Outer Surface Grafting Technique; Ossicular Reconstruction; Mastoidectomy—Intact Canal W all Procedure; Tympanoplasty—Staging and Use of Plastic
Retrosigmoid Approach to Tumors of the Cerebellopontine Angle
George T. Singleton, MD Professor Emeritus, Otolaryngology-Head and Neck Surgery, University of Florida College of Medicine; OtolaryngologyHead and Neck Surgery, Shands Teaching Hospital; Otolaryngology-Head and Neck Surgery, Malcolm Randall VA Medical Center, Gainesville, ������� Florida
Perilymphatic Fistula
William H. Slattery III, MD Clinical Professor, University of Southern California; ���������� Director, Clinical Studies, House Ear Institute; Associate, House Ear Clinic, Los Angeles, California
Implantable Hearing Devices; Neurofibromatosis
Carl H. Snyderman, MD Departments of Neurological Surgery and Otolaryngology, University of Pittsburgh School of Medicine; Minimally Invasive EndoNeurosurgery Center and Center for Skull Base Surgery, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania
Endoscopic Endonasal Approaches to the Skull Base and Paranasal Sinuses
Robert F. Spetzler, MD Professor, Department of Surgery, Section of Neurosurgery, University of Arizona College of Medicine, Tuscon; ��������� Director and J. N. Harber Chairman of Neurological Surgery, Barrow Neurological Institute, Phoenix, Arizona
Vascular Consideration in Neurotologic Surgery
Barry Strasnick, MD, FACS Professor and Chairman, Department of Otolaryngology-Head and Neck Surgery, Eastern Virginia Medical School, Norfolk, Virginia
Tympanoplasty—Undersurface Graft Technique: Postauricular Approach
Clough Shelton, MD, FACS
Christopher A. Sullivan, MD
C. Charles Hetzel, Jr., MD and Alice Barker Hetzel Presidential Endowed Chair in Otolaryngology; Professor, Otology, Neuro-Otology and Skull Base Surgery; Chief, Otolaryngology-Head and Neck Surgery, University of Utah; Medical Director, Otolaryngology-Head and Neck Surgery and Otology/Neuro-Otology Surgeon, University of Utah Hospital and Clinics, Salt Lake City, Utah
Assistant Professor, Otolaryngology-Head and Neck Surgery, Wake Forest University School of Medicine; Staff Surgeon, Otolaryngology-Head and Neck Surgery, North Carolina Baptist Hospital; Assistant Professor, Regenerative Medicine, Wake Forest Institute for Regenerative Medicine, Winston Salem, �������������� North Carolina
Tympanoplasty—Staging and Use of Plastic; Laser Stapedotomy, Facial Nerve Tumors; Middle Fossa Approach
†Deceased
Rehabilitation of Lower Cranial Nerve Deficits after Neurotologic Skull Base Surgery
†Deceased
xvi
Contributors
Mark J. Syms, MD
P. Ashley Wackym, MD, FACS, FAAP
Neurologist, Section of Neurotology, Department of Neurosurgery, Barrow Neurological Institute, Phoenix, Arizona
John C. Koss Professor and Chairman, Department of Otolaryngology and Communication Sciences, Medical College of Wisconsin; Chief, Section of Otology, Division of Pediatric Otolaryngology, Children’s Hospital of Wisconsin; Chief, Department of Otolaryngology-Head and Neck Surgery, Froedtert and The Medical College of Wisconsin Hospital, Milwaukee, Wisconsin; ������������������� Vice President for Research, Legacy Health System, Portland, Oregon
Mastoidectomy—Intact Canal W all Procedure
Charles A. Syms III, MD Clinical Professor, Department of Otolaryngology-Head and Neck Surgery, University of Texas Health Science Center, San Antonio; President, Ear Medical Group, San Antonio, Texas
Mastoidectomy—Intact Canal Wall Procedure
Steven A. Telian, MD John L. Kemink Professor of Otorhinolaryngology; Director, Division of Otology, Neurology, and Skull Base Surgery; and Medical Director, Cochlear Implant Program, Department of Otolaryngology-Head and Neck Surgery, University of Michigan, Ann Arbor, Michigan
Retrolabyrinthine and Retrosigmoid Vestibular Neurectomy
Fred F. Telischi, MEE, MD, FACS Professor, Neurological Surgery, Biomedical Engineering, and Otolaryngology, University of Miami Miller School of Medicine; ������������������������������������������������ Vice Chairman and Director, University of Miami Ear Institute, Miami, Florida
Intraoperative Neurophysiologic Monitoring
Karen B. Teufert, MD House Ear Institute, Los Angeles, California
Congenital Malformation of the External Auditory Canal and Middle Ear; Transcochlear Approach to Cerebellopontine Angle Lesions
Spereotactic Radiosurgery of Skull Base Tumors
P. Daniel ������������������� Ward, MD, MS Resident, Department of Otolaryngology-Head and Neck Surgery, University of Michigan, Ann Arbor, Michigan
Retrolabyrinthine and Retrosigmoid V estibular Neurectomy
Frank M. Warren, MD Assistant Professor, Otolaryngology-Head and Neck Surgery, University of Utah, Salt Lake City, Utah
Facial Nerve Tumors
D. Bradley ������������������������������ Welling, MD, PhD, FACS Professor and Chair, Otolaryngology-Head and Neck Surgery, The Ohio State University, Columbus, Ohio
Surgery of the Endolymphatic Sac
Richard J. Wiet, MD Professor of Clinical Otolaryngology and Neurosurgery, Northwestern University, and Ear Institute of Chicago, Chicago, Illinois
Complications of Surgery for Chronic Otitis Media
Anders M.R. Tjellström, MD, PhD
Eric P. Wilkinson, MD
Department of Otolaryngology, Sahlgrenska University Hospital, Gőteborg, Sweden
Clinical Assistant Professor of Otolaryngology, University of Southern California Keck School of Medicine; Associate, House Clinic, Los Angeles, California
The Bone-Anchored Cochlea Stimulator (Baha)
Debara L. Tucci, MD Associate Professor, Division of Otolaryngology-Head and Neck Surgery, Department of Surgery, Duke University Medical Center, Durham, North Carolina
Extreme Lateral Infrajugular Transcondylar Approach for Resection of Skull Base Tumors
Canal W all Reconstruction Tympanomastoidectomy; Drainage Procedures for Petrous Apex Lesions; Management of Postoperative Cerebrospinal Fluid Leaks; Hypoglossal Facial Anastomosis
Acknowledgements Publication of a book of this scope requires a tremendous effort on the part of many, all of whom I wish to sincerely thank. First, my thanks to my lovely wife, Charlotte, who supports all my efforts and forgave my absence for the time devoted to this project. No less supportive are our four sons, David, Douglas, Mark, and Steven, who provide diversion and pleasure by taking me hunting and fishing. Since the publication of the second edition, four additional grandchildren have blessed Charlotte and me. Lauren, Nicholas, Sammy, Kaylee Daniel, and Megan provide immeasurable pleasure to us. What we have been told about the joy of grandparenting was underemphasized. My associate editors, Dr. Shelton and Dr. Arriaga, worked tirelessly to bring this volume to fruition. Anthony Pazos deserves special recognition. He spent countless hours in the temporal bone laboratory learning at first hand the various operations that he then illustrated. All the authors have appreciated his attention to detail and willingness to work with them until everything was “just right.” The publishers have been extremely supportive throughout the development of this book. From the beginning, they made a major commitment to ensure that this volume would be of the highest quality. I wish to thank particularly Rachel Yard and Scott Scheidt. Finally, I wish to thank our office staff, who work tirelessly on behalf of our patients and us. I would particularly like to thank Rita Koechowski, my surgery counselor, who not only schedules my surgery but also offers tremendous encouragement and support to all my patients. Her assistant, Patricia McGrath, has been a tremendous help to both of us in supporting our patients. Finally, I wish to thank my surgical assistants, Matthew Layner and Nancy Aguilar, and my executive assistant, Cathy Weise. She maintains my focus and orientation on a daily basis, and her efforts are greatly appreciated. Derald E. Brackmann, MD My family’s patience and support in this book and all my academic projects continues to motivate and encourage me. My wife Rosemary and children (Becca, Moi, and Toby) participated in this and my other professional activities through tolerating my time away, enduring family moves, and now even offering occasional editorial suggestions on substance and form. Moisés Agusto and Leticia are a continued inspiration. I particularly want to thank Derald Brackmann for his mentorship and stellar example of skill in otologic surgery and graceful balance of complex competing demands. The physicians and alumni of the House Ear Clinic provide an active network of otologic innovation whose concepts serve as a thread of continuity in this book. My Louisiana State University residents and partners have provided insightful questions and suggestions that have prompted some of the changes in this edition. Additionally, they have shown that even in the terrific upheaval of post-Katrina Louisiana, calm focus on basic principles ensures quality education and superb patient care as long as we remain flexible to use all resources available. Moisés A. Arriaga, MD, MBA, FACS
xvii
1
Otologic Instrumentation Jose N. Fayad, Barbara Stahl Hoskins, and James E. Benecke, Jr.
Sophisticated micro-otosurgical techniques mandate that the otologic surgeon and surgical team have an indepth understanding of the operating room (OR) layout and surgical instrumentation. This chapter describes in detail different surgical procedures. The OR setup and instruments necessary for the various types of otologic procedures are described. Appendix 1 provides a comprehensive list of instruments and equipment.
OPERATING ROOM The OR for otologic surgery requires features that differ from ORs used for nonotologic surgery. The following sections elaborate on the general environment of the OR designed for ear surgery. A word about the sterile field is in order. Respecting the sterile field is vital during routine otologic surgery, and takes on special significance during neurotologic procedures. Maintaining the proper environment means limiting traffic through the OR, and keeping the number of visitors to a minimum. It is preferable for observers to be in a remote room watching the procedures on video. Individuals allowed in the OR should be experienced in sterile technique and should wear jackets over scrubs so that all skin surfaces are covered (Fig. 1-1). Before entering the OR, the patient identifies the operative site. The correct ear is marked using a marking pen. The psychological environment of the OR must be respected because many otologic procedures are performed on awake patients under local anesthesia. Members of the surgical team and visitors must use discretion when making comments during surgery. The first piece of OR equipment to be discussed is the operating table. The surgeon must be comfortable while performing microsurgery. Adequate leg room under the table can be achieved with older OR tables by placing the patient 180 degrees opposite the usual position; in other words, the patient’s head is where the feet would normally be (Fig. 1-2). Newer electric tables easily accommodate the patient and surgeon. Because most otologists spin the OR table 180 degrees after the
induction of anesthesia, the new tables allow for spinning the table without unlocking it. Nonetheless, after the patient is properly positioned, the table must be firmly locked in place. All ORs are equipped with wall suction. Standard suction devices are acceptable for otologic surgery. It is preferable, however, to use a multiple-canister suction setup, minimizing the number of times the bottles must be emptied (Fig. 1-3). Suction systems have several locations where the amount of suction can be varied, but the surgeon should also employ a control clamp on the suction tubing on the sterile field (Fig. 1-4). The tubing that is attached to the suction tips and suction irrigators should be highly flexible. The readily available disposable tubing is not flexible enough for microsurgery, and places awkward torque on the surgeon’s hands. Suction setup problems are common in every OR. The prudent team troubleshoots the system in advance and has access to backup equipment. Electrocautery equipment should be in a ready-touse state on all procedures except perhaps stapes surgery. The patient must be properly grounded. It is advantageous to have unipolar and bipolar cautery on the field for all chronic ear and neurotologic procedures. Polytef (Teflon) tips are available for most cautery devices and are desirable. Surgeons have at their disposal a wide array of safe cautery devices, but they must be thoroughly familiar with these electric instruments before use. The surgical drill is another essential piece of equipment for otologic surgery. The vast array of available drills precludes an in-depth discussion of each system. Generally, otologic drills fall into two categories: air driven and electric. There are advantages and disadvantages to each type, and most surgeons have a distinct preference based on training and experience. For surgeons using air-driven drills, it is preferable to use a central source of nitrogen to power the drill, instead of using room tanks of the gas. Using a central source eliminates the need for changing tanks during long cases. High-speed drills capable of doing most of the bone work in the temporal bone include the Fisch, Midas Rex, and Anspach drill systems. These drills generally are
OTOLOGIC SURGERY
FIGURE 1-3. Multiple-canister suction setup.
FIGURE 1-1. Observer in jacket. FIGURE 1-4. Suction tubing with control clamp.
FIGURE 1-2. Operating table with patient’s head at foot of bed. FIGURE 1-5. Synergy microdrill for footplate work.
unsuitable for work in the middle ear, especially around the stapes footplate. For the latter purposes, a microdrill, such as the Skeeter drill or Synergy, is suitable (Fig. 1-5). Whatever drill is used in the middle ear, it must have a variable speed control and a wide array of drill bits. Most larger otologic drills are equipped with straight and angled handpieces. Most surgeons prefer straight handpieces for early gross removal of the mastoid cortex, switching to angled handpieces for working deeper in the temporal bone. The Anspach drill system has a handpiece that can be converted from straight to angled simply by rotating the connection. A full complement
of cutting and diamond burrs is mandatory. Figure 1-6 shows the Anspach drill system. Most drill systems have attachments that vary in shape, diameter, and length. It is the surgeon’s responsibility to be intimately familiar with the drill system and to have all of the attachments and burrs that might be needed. The otologic drill should be held in the hand like a pencil, with the hand resting comfortably on the sterile field. The side of the burr should be used to provide maximum contact between the bone and the flutes of the burr, affording safer and more efficient drilling (Fig. 1-7). The newer drills are remarkably reliable, but, similar
Chapter 1 • Otologic Instrumentation
FIGURE 1-6. Anspach drill system.
FIGURE 1-7. Proper holding of the drill.
to other tools, may malfunction. Drill systems require proper care and inspection before use. A backup system should be readily available. The introduction of the operating microscope revolutionized otologic surgery. Most otolaryngologists are familiar with the use of the microscope. Several brands of optically superior instruments are available; most are sufficiently similar to share the same general principles. The otologic surgeon must be familiar with the adjustments on the microscope, and must be prepared to troubleshoot the problems that may arise with the scope. The focal length of the objective lens is a matter of personal preference. Most otologists use a 200 mm or 250 mm objective. If a laser is attached to the microscope, one might consider a 300 mm objective. The objective lens should be selected, confirmed, and properly mounted before draping the microscope. Other adjustments, such as the most comfortable interpupillary distance, also should be done before the scope is draped. Par focal vision should be established so that the surgeon can change magnification without having to change focus. This is accomplished by first setting the diopter setting of both eyepieces to zero. The 40× magnification (or highest available setting) is selected. The locked microscope
is focused on a towel using the focus knob only. Without disturbing any of the settings, the magnification is now set at 6× (or the lowest available setting). The eyepieces are individually adjusted to obtain the sharpest possible image. The diopter readings are recorded for future use. The surgeon should have par focal vision when these appropriate eyepieces are used. The microscope should move easily. All connections should be adjusted so that the microscope does not wander by itself, yet permits movement to any position with minimal effort. Wrestling with the microscope during microsurgery is an extreme distraction. Proper posture at the operating table is crucial. To perform microsurgical procedures, rule number one is that the surgeon must be comfortable. The surgeon should be seated comfortably in a proper chair with the back support at the correct height. Both feet should be resting comfortably on the floor. Fatigue is avoided by assuming a restful position in the chair, rather than a rigid upright posture (Fig. 1-8). The overall OR setup for routine otologic surgery is shown in Figure 1-9. For neurotologic surgery, more space must be available for additional equipment. Middle cranial fossa procedures require some modifications to the OR setup (Fig. 1-10). Basically, the surgeon and the microscope trade places such that the surgeon is seated at the head of the table. Cooperation and careful orchestration between the surgeon, nursing personnel, and anesthesiologist are required for otologic surgery. The needs of the otologist are best served by having the anesthesiologist at the foot of the bed and the scrub nurse opposite the surgeon. Space for the facial nerve monitoring equipment and personnel is reserved.
STAPES SURGERY The following description of the instrumentation and operative setup for stapes surgery also provides information useful for other middle ear procedures. Under most circumstances, it is preferable to perform stapes surgery under local anesthesia, and surgeons who do so usually employ some type of preoperative sedation. Numerous regimens are available, and their description is beyond the scope of this text. If sedation is administered by the surgeon or nursing personnel, without the assistance of an anesthetist or anesthesiologist, the agents used should be short-acting and reversible. It is far safer for the patient to be psychologically prepared for the procedure than to be oversedated. At the House Clinic, the associates prefer to perform all local anesthesia cases (including stapes surgery) under monitored anesthesia care. This approach requires the presence of anesthesia personnel in the OR to sedate the patient, as is required for the operation, and to monitor vital functions. The surgeon is relieved from this duty, allowing total concentration on the microsurgery.
OTOLOGIC SURGERY
A
B
FIGURE 1-8. A, Proper posture for the surgeon. B, Wrong posture for the surgeon.
FIGURE 1-9. Usual otologic/neurotologic operating room setup.
About 30 minutes before the operation, the patient is brought to the preoperative holding area. If the surgeon routinely harvests a postauricular graft, this area is now shaved. A plastic aperture drape is applied to the operative site and trimmed so as not to cover the patient’s face (Fig. 1-11). An intravenous line is started, and the patient is now ready to go to the OR. When the patient is on the OR table, the monitors are placed on the patient by the nursing or anesthesia staff. Minimal monitoring includes pulse oximetry, automatic blood pressure cuff, and electrocardiogram electrodes. The ear and plastic drape are scrubbed
with an iodine-containing solution, unless the patient is allergic to iodine. A head drape is applied, and the ear is draped with sterile towels so as not to cover the patient’s face; this can be facilitated by supporting the drapes with a metal bar attached to the OR table, or by fixing the drapes to the scrub nurse’s Mayo stand (Fig. 1-12). The patient’s head is now gently rotated as far away from the ipsilateral shoulder as possible, and the table is placed in slight Trendelenburg position. These maneuvers increase the surgeon’s working room and help to straighten the external auditory canal (EAC). The EAC
Chapter 1 • Otologic Instrumentation
FIGURE 1-10. Operating room layout for middle fossa surgery.
FIGURE 1-11. Plastic drape applied for stapes surgery.
is gently irrigated with saline heated to body temperature. Vigorous cleaning of the canal is avoided until the ear is anesthetized. The local anesthesia is administered with a plastic Luer-Lok syringe that has finger and thumb control holes. A 11⁄2 inch, 27 gauge needle is firmly attached to the syringe. If the ear is injected slowly and strategically, excellent anesthesia and hemostasis can be achieved with a solution of 1% lidocaine with 1:100,000 epinephrine (e.g., 1:40,000). When using stronger concentrations of epinephrine, the patient’s blood pressure and cardiac status must be considered, in addition to the possibility of mixing errors. The canal is injected slowly in four quadrants starting lateral to the bony-cartilaginous junction. The final injection is in the vascular strip. If one routinely harvests fascia or tragal perichondrium, these areas are now injected. Before describing stapes surgical instruments, a few general comments are in order. All microsurgical
FIGURE 1-12. Patient draped in the operating room for stapes surgery.
instruments should be periodically inspected to ensure sharp points and cutting surfaces. The instruments for delicate work should have malleable shanks, enabling the surgeon to bend the instruments to meet the demands of the situation. If the surgeon prefers a total stapedectomy over the small fenestra technique, an oval window seal must be selected. If fascia is used, the tissue is harvested before exposing the middle ear. The tissue is placed on a Teflon block or fascia press to dry. If perichondrium is preferred, this may be harvested immediately before footplate removal. For the small fenestra technique, a small sample of venous blood is obtained when the intravenous line is started. This blood sample is passed to the scrub nurse and placed in a vial on the sterile field. Various ear specula should be available in oval and round configurations. Sizes typically range from 4.5 to 6.5 mm (Fig. 1-13). It is desirable always to work
OTOLOGIC SURGERY
through the largest speculum that the meatus permits, without lacerating canal skin. Some surgeons prefer to use a speculum holder for stapes and other middle ear procedures. The tympanomeatal flap is started with incisions made at the 6 and 12 o’clock positions with the No. 1, or sickle, knife. These incisions are united with the No. 2, or lancet, knife. This instrument actually undermines the vascular strip instead of cutting it. The strip is cut with the Bellucci scissors. The defined flap is elevated to the tympanic annulus with the large round knife, known as the large “weapon.” When properly identified, the annulus is elevated superiorly with the Rosen needle, and inferiorly with the annulus elevator,
FIGURE 1-13. Speculum array.
FIGURE 1-14. Stapes instruments
or gimmick. Figure 1-14 shows a typical set of stapes instruments, including suction tips. Adequate exposure usually requires removal of the bony ledge in the posterosuperior quadrant. This can be initiated with the Skeeter microdrill and completed with a stapes curette (Fig. 1-15). From this point on, the steps differ depending on the technique preferred by the surgeon. The diagnosis of otosclerosis should be confirmed on entering the middle ear, and a measurement should be taken from the long process of the incus to the stapes footplate with a measuring stick. The next step is to make a control hole in the footplate with a sharp pick-needle (Barbara needle) or the laser. The incudostapedial joint is separated with the joint knife, the tendon is cut with scissors or laser, and the superstructure is fractured inferiorly and extracted. For work on the footplate, the surgeon must have a variety of suitable instruments available. A stapedotomy can be created with a microdrill, laser, or needles and hooks. The 0.3 mm obtuse hook is useful for enlarging the fenestra. For total footplate extraction, a right angle hook or excavator (Hough hoe) is used. The harvested graft is guided into place with a footplate chisel. The prosthesis is grasped with a smooth alligator or strut forceps and placed on the incus. It is positioned on the graft, or into the fenestra, with a strut guide. The wire is secured onto the incus with a crimper, or wire-closing forceps. The McGee crimper is useful, especially if followed by a fine
Chapter 1 • Otologic Instrumentation
FIGURE 1-15. Stapes curette.
FIGURE 1-17. Rosen suction tubes with House adapter; Baron tubes.
TYMPANOPLASTY AND TYMPANOPLASTY WITH MASTOIDECTOMY
FIGURE 1-16. Crimpers and footplate hook.
alligator forceps for the last gentle squeeze. A small right angle hook may be necessary to fine-tune the position of the prosthesis (Fig. 1-16). Suction tubes for stapes surgery include Nos. 3 to 7 Fr Baron suctions plus Rosen needle suction tips (18 to 24 gauge) with the House adapter (Fig. 1-17). The Rosen tips are useful when working near the oval window, with the surgeon’s thumb off the thumb port. Ear packing after stapes surgery is accomplished with an antibiotic ointment to hold the flap in place. A piece of cotton suffices as a dressing, unless a postauricular incision has been made, in which case a mastoid dressing is applied. For all middle ear procedures, the surgeon should hold the instruments properly. The instrument should rest, like a pencil, between the index finger and thumb, allowing easy rotation around the shank. The hands should always be resting on the patient and the OR table. The middle and ring fingers should rest on the speculum so that the hand moves as a unit with the patient. Proper hand position and holding of instruments should afford the surgeon an unimpeded view (Fig. 1-18).
The preparation and draping for tympanoplasty with or without mastoidectomy are much the same as for stapes surgery. The major difference is the amount of hair shaved before draping. Usually, enough hair is shaved to expose about 3 to 4 cm of skin behind the postauricular sulcus. The plastic drape is applied to cover the remaining hair (Fig. 1-19). The patient is positioned on the OR table as described earlier. Whether the procedure is performed under local or general anesthesia depends on the extent of the surgery, the surgeon’s preference, and the desire of the patient. After appropriate sedation or induction of anesthesia, the ear and plastic drape are scrubbed with the proper soap or solution. Some surgeons place a cotton ball in the meatus if a perforation exists, preferring not to allow the preparation solution to enter the middle ear. The field is draped as described earlier, the head is rotated toward the contralateral shoulder, and the table is placed in slight Trendelenburg position (Fig. 1-20). The postauricular area, canal, and tragus (if necessary) are injected with 1% lidocaine with 1:100,000 epinephrine for local and general anesthesia cases. Most chronic ear procedures begin in a similar fashion. Through an ear speculum, vascular strip incisions are made with the sickle or Robinson knife and united along the annulus with the lancet knife. The vascular strip incisions are completed with a No. 64 or 67 Beaver blade. This same blade can be used to transect the anterior canal skin just medial to the bony-cartilaginous junction. The postauricular incision is made with a No. 15 Bard-Parker blade behind the sulcus. The level of the temporalis fascia is identified, and a small selfretaining (Weitlaner) retractor is inserted. The fascia is cleared of areolar tissue and incised. A generous area of fascia is undermined and removed with Metzenbaum scissors. The scrub nurse can assist by using a Senn retractor to elevate skin and soft tissues away from the
OTOLOGIC SURGERY
FIGURE 1-18. Proper holding of instruments as shown by Dr. William House.
FIGURE 1-19. Drape (3M 1020) applied for chronic ear surgery. FIGURE 1-20. Chronic ear surgery draping for local anesthesia.
fascia. The fascia is thinned on the Teflon block and dehydrated by placing it under an incandescent bulb, carefully monitoring its progress. The fascia may also be dehydrated by placing it on a large piece of Gelfoam and compressing this complex in a fascia press. Figure 1-21 shows the instruments used in the initial stages of chronic ear surgery. Continued postauricular exposure is obtained by incising along the linea temporalis with a knife or with the electrocautery. A perpendicular incision is made down to the mastoid tip. Soft tissues and periosteum are elevated with a Lempert elevator (Fig. 1-22), the vascular strip is identified, and a large self-retaining retractor is inserted. A very large retractor, such as an Adson cerebellar retractor with sharp prongs, is preferred. Next, under the microscope, the anterior canal skin is removed down to the level of the annulus with the large weapon. The plane between the fibrous layer of the drum remnant and the epithelium is developed with a sickle knife, and the skin is pulled free with a cup forceps. The
anterior canal skin is placed in saline for later use as a free graft. The ear canal is enlarged with the drill and suctionirrigators. An angled handpiece and medium to small cutting burr are used. Irrigation through the suctionirrigators is done with a physiologic solution such as TisU-Wol, lactated Ringer, or saline. Two large (3000 mL) bags of irrigant are hung and connected by way of a threeway stopcock to the delivery system (Fig. 1-23). For mastoidectomy surgery, the surgeon must have a full array of cutting and diamond burrs, and a complete set of suction-irrigators. It is advisable to have bone wax and absorbable knitted fabric (Surgicel) readily available. Cholesteatoma removal can be accomplished with middle ear instruments such as the gimmick, weapon, and fine scissors. Although the setup for closing and packing after chronic ear surgery varies with the specifics of the situation, a few generalities should cover most situations encountered by the otologist. To maintain the middle ear space, silicone elastomer (Silastic) sheeting works
Chapter 1 • Otologic Instrumentation
A
B FIGURE 1-21. A, Instruments for making a canal incision. B, Instruments for handling fascia.
FIGURE 1-22. Periosteal elevators.
well and is still readily available. This sheeting comes in various thicknesses, with and without reinforcement. For middle ear packing, absorbable gelatin sponge (Surgifoam) is the usual choice, soaked in saline or an antibiotic otic preparation. Surgifoam is also used to pack the EAC, although some surgeons prefer an antibiotic ointment, as described in the section on stapes surgery. For meatoplasty packing, 1-inch nonadhesive Curity packing
strip or nasal packing gauze is saturated with an antibiotic ointment and rolled around the tip of a bayonet forceps; this creates a plug that conforms to the new meatus and is easily removed (Fig. 1-24). Wound closure is accomplished in two layers with absorbable sutures. The skin is closed with a running intradermal suture of 1-0 polyglactin 910 (Vicryl) or polyglycolic acid (Dexon) on a cutting needle. Steri-Strips are
10
OTOLOGIC SURGERY
applied, and the wound is covered with a standard mastoid dressing. Some additional instruments that prove to be handy in many chronic ear procedures include an ossicles holder, Crabtree dissectors, Zini mirror, right angle hooks, and the House-Dieter malleus nipper (Fig. 1-25). It is impossible to describe instruments for every conceivable situation, but the foregoing should cover most of the needs of the otologist.
ENDOLYMPHATIC SAC SURGERY There are many well-described procedures on the endolymphatic sac. The purpose of this chapter is not to outline the surgical options, but rather to discuss the
FIGURE 1-23. Suction irrigation setup.
A FIGURE 1-24. A, Surgifoam packing. B, Meatoplasty packing.
methodology for performing sac surgery. The preparation and draping of the patient for endolymphatic sac surgery are essentially the same as for tympanoplasty with mastoidectomy surgery. In the preoperative holding area, the postauricular area is shaved, exposing at least 4 cm of skin behind the sulcus. Plastic adhesive drapes are applied, and the patient is transported to OR. Endolymphatic sac surgery is performed under general anesthesia. The field is scrubbed in the usual manner, and the patient is positioned as described for chronic ear surgery. This is a good time to mention briefly the use of intraoperative facial nerve monitoring and other forms of physiologic monitoring, including eighth cranial nerve and cochlear potentials. Many surgeons use facial nerve monitoring whenever the facial nerve might be in jeopardy. Electrodes for facial nerve monitoring or other forms of monitoring should be positioned before the preparation. After the preparation for endolymphatic sac surgery, the planned incision is injected with 1% lidocaine with 1:100,000 epinephrine. The incision is made 2 to 3 cm behind the sulcus. Periosteal incisions are made sharply or with the electrocautery. A Lempert elevator elevates soft tissues and periosteum up to the level of the spine of Henle. A House narrow (canal) elevator is used to delineate the EAC, and a large self-retaining retractor is inserted. With drill and suction-irrigator, a complete mastoidectomy is performed. The antrum is not widely opened, but is instead blocked with a large piece of absorbable gelatin sponge (Gelfoam) to prevent bone debris from entering the middle ear. Bone over the sigmoid sinus and posterior fossa dura is thinned with diamond burrs. The retrofacial air tract is opened widely to locate the endolymphatic sac. The sac is decompressed with a diamond burr. A stapes curette can be used to remove bone over the proximal sac. The occasional bleeding that occurs over the surface of the sac or surrounding dura is best controlled with bipolar cautery. Alternatively, unipolar cautery at a very low setting can be used. The cautery tip is touched to an insulated Rosen or gimmick that is in contact with the offending vessel (Fig. 1-26). Another
B
Chapter 1 • Otologic Instrumentation
FIGURE 1-25. Additional instruments used in chronic ear procedures
11
FIGURE 1-27. Endolymphatic sac instruments and materials.
(see text).
FIGURE 1-28. Drapes (3M 1000) applied for neurotologic surgery.
NEUROTOLOGIC PROCEDURES FIGURE 1-26. Insulated gimmick (top) and Rosen (bottom).
method used to control small areas of bleeding in endolymphatic sac and chronic ear surgery is to cover the area with pledgets of Gelfoam that have been soaked in topical thrombin. Before opening the sac, the wound is copiously irrigated with saline or bacitracin solution. Fresh towels are placed around the field. The sac is opened with a disposable Beaver ophthalmic blade (No. 59S, 5910, or 5920). The lumen is probed with a blunt hook or gimmick. The shunt tube preferred by the surgeon is now inserted. Thin Silastic sheeting (0.005 inch) can be used to fashion a shunt. Figure 1-27 shows the materials for the latter steps of endolymphatic sac surgery. As with chronic ear procedures, the wound is closed in layers, usually beginning with 2-0 chromic and finishing with 4-0 Vicryl or Dexon. A standard mastoid dressing is applied. This dressing either is prepared in the OR or is obtained as a prepackaged dressing (e.g., Glasscock dressing).
This section describes the OR layout for neurotologic procedures, the only exception being middle fossa surgery, which is discussed separately. For procedures involving intracranial structures, extraordinarily meticulous attention to detail is mandatory. The preparation for neurotologic surgery may begin the evening before surgery by having the patient wash his or her hair and scalp with an antiseptic shampoo. The day of surgery, the patient is seen by the surgeon in the holding area so that the ear to be operated on is positively identified. The surgical site is shaved so that at least 6 cm of postauricular scalp is exposed. The area is sprayed with an adhesive, and the plastic drapes are applied (Fig. 1-28). At the same time, the abdomen is shaved from below the umbilicus to the inguinal ligaments, in preparation for harvesting a fat graft. The fat donor site is surrounded by plastic drapes (Fig. 1-29). After anesthetic induction, a catheter is inserted, and arterial and central venous lines are placed when indicated. Electrodes for monitoring CN VII and VIII
12
OTOLOGIC SURGERY
FIGURE 1-29. Abdominal area prepared.
A
B
C FIGURE 1-30. A-C, Draping sequence for neurotologic surgery.
FIGURE 1-31. SK-100 Surgi-kit for holding instruments.
(and possibly other nerves) are positioned. The patient’s head is supported on towels or a “donut” as needed, and rotated toward the contralateral shoulder. The surgical sites are scrubbed, then blotted dry with a sterile towel. The areas are draped off with towels and then covered with plastic adhesive drapes (e.g., Steri-Drape, Ioban, Cranial-Incise). Some surgeons prefer to include another layer of towels around the cranial site, followed by either sheets or a disposable split sheet. It is important to have several layers of draping to prevent saturation of the drapes with fluids down to the level of the patient (Fig. 1-30). Because the scrub nurse must handle numerous items attached to tubes and cords, it is helpful to have fastened to the field a plastic pouch into which the drill, suction, and cautery tips can be placed (Fig. 1-31). Two Mayo stands are kept near the field: one for the neurotologic instruments and the other for the fat-harvesting tools (Fig. 1-32). The postauricular area is injected with the usual local anesthetic, and the plastic drape is cut away with scissors to expose the mastoid and lateral subocciput. As with other procedures, a skin incision is made, hemostasis is obtained, soft tissues and periosteum are elevated, and a large self-retaining retractor is inserted. Bone removal is accomplished using a drill and suction-irrigation. For neurotologic cases, bone removal is more extensive, exposing the sigmoid sinus and a considerable amount of posterior fossa dura behind the sigmoid. It is imperative that the surgeon has immediate access to bone wax and Surgicel. Many surgeons also insist on having immediate access to hemoclips and thrombin-soaked Gelfoam. The extent of bone removal varies depending on the surgeon’s preference and the nature of the procedure. Some surgeons decompress the sigmoid completely, whereas others leave a thin shell of bone over the sinus (Bill’s island). After appropriate bone removal, the retractor is removed, and the field is vigorously irrigated with bacitracin solution. Bacitracin solution can be prepared
Chapter 1 • Otologic Instrumentation
A
B FIGURE 1-32. A, Mayo stand setup for tumor. B, Mayo stand setup for fat graft.
FIGURE 1-33. Brackmann neurotologic instruments.
by dissolving 50,000 U of bacitracin in 1 L of normal saline. After wound irrigation, fresh towels are placed around the field. With a wound free of bone dust and debris, the dura can now be opened with a No. 11 Bard-Parker scalpel blade or with the tips of Jacobson scissors. The dura can be pulled away from underlying structures by using a corkscrew-like instrument included in some neurotologic instrument sets (Fig. 1-33). The subdural space is entered, taking care not to violate the arachnoid; this helps to avoid injury to vessels before adequate exposure. The dural flap is carefully developed with Jacobson scissors. Hemostasis is controlled with bipolar cautery. The arachnoid is carefully opened with a sharp hook or the
13
FIGURE 1-34. Brackmann fenestrated suction-irrigators.
tips of the scissors, allowing the egress of cerebrospinal fluid. Figure 1-33 shows the instruments for dural and arachnoid opening. After opening the arachnoid, one should switch to fenestrated (Brackmann) suction tips (Fig. 1-34). The cerebellum and other intracranial structures should be protected with moist neurosurgical cottonoids. A variety of cottonoids should always be on the stand. For vestibular neurectomy procedures, the plane between the cochlear and vestibular nerves can be developed with a blunt hook, or the gimmick. The nerve section itself can be completed with a sharp hook or microscissors (Fig. 1-35). The same instruments can be used to define the plane between an acoustic neuroma and the facial nerve. A sharp right angle hook palpates Bill’s bar and sections the superior vestibular nerve fibers along with the vestibulofacial fibers. After establishing the proper plane between the tumor and facial nerve, a blunt hook is used to continue the dissection, avoiding stretching of the facial nerve. Facial nerve monitoring has greatly assisted this part of the dissection. For small tumors, the previously mentioned technique might suffice for total tumor removal. Larger tumors are removed by gutting the tumor extensively, mobilizing the capsule, and removing the capsule in a piecemeal fashion; this is accomplished by morcellizing the tumor with a large crushing forceps, such as the Decker. The Urban rotary suction-dissector is used to extract the pieces (Fig. 1-36). Bayonet forceps direct the tumor into the suction port of the Urban suction-dissector. As the tumor is gutted, the capsule collapses and can be dissected from the brainstem. The Selector ultrasonic aspirator (Fig. 1-37) is another instrument that some surgeons prefer for gutting the tumor. Whatever tool is used, proper use of these sophisticated, potentially dangerous instruments must be learned from user manuals and appropriate training and courses. Hemostasis is vital during neurotologic surgery, and the surgeon must have immediate access to all possible items necessary to control bleeding from whatever the source. In addition to unipolar and bipolar cautery, bone wax and precut pieces of Surgicel should be on the Mayo stand. Microfibrillar collagen (Avitene) is another preferred hemostatic agent to have available. Pledgets
14
OTOLOGIC SURGERY
FIGURE 1-35. Hooks for neurectomy and tumor dissection.
FIGURE 1-36. Urban dissector.
FIGURE 1-37. Selector aspirator.
of Gelfoam soaked in topical thrombin are quite useful. Vascular clips and a reliable clip applicator are useful for controlling bleeding from the petrosal vein and its tributaries (Fig. 1-38). Infratemporal fossa and other approaches to the skull base are set up in much the same manner as has already been discussed. Incisions are generally long and may extend into the upper cervical region to access major
FIGURE 1-38. Clips and clip applicators.
neurovascular structures. Silastic vessel loops should be placed around these structures for control and easy identification. Ligatures of 0 silk and transfixion sutures of 2-0 silk need to be available for jugular vein ligation. Cardiovascular sutures (e.g., 5-0 and 6-0 polypropylene [Prolene]) should also be close by. The self-retaining retractors described earlier are usually insufficient for skull base surgery. The Fisch infratemporal retractor or pediatric rib retractor are better suited to these tasks, which often include anterior displacement of the mandible. If mandibulotomy is indicated, the appropriate oscillating saw needs to be available. Some instruments facilitate work on or near the facial nerve. For rerouting the facial nerve, bone is removed with a drill until an eggshell thickness remains. The remaining bone is gently removed with a stapes curette. The nerve can be mobilized with a dental excavator or microraspatory. If a segment of the nerve is to be excised, as in a facial neuroma, this should be done sharply with a fresh knife blade. Likewise, before any neurorrhaphy, the ends of the nerve and graft should be freshened. A 9-0 monofilament suture is used for nerve anastomosis.
Chapter 1 • Otologic Instrumentation
15
FIGURE 1-39. Nerve anastomosis equipment.
Also under the rubric of neurotologic surgery is cochlear implant surgery. Each presently available cochlear implant device has its own unique set of requirements and, possibly, instruments. The surgeon must have proper training and experience to perform cochlear implant procedures. He or she must have all of the necessary special equipment for electrode placement and internal receiver fixation (Fig. 1-40).
MIDDLE CRANIAL FOSSA SURGERY
FIGURE 1-40. Cochlear implant tools.
Appropriate needle holders and forceps must be available (Fig. 1-39). An alternative or adjunct to suturing is to use NeuraGen nerve guides. Before closing neurotologic and skull base wounds, abdominal fat is removed from the left lower quadrant, most of the dissection being done with electrocautery. The abdominal wound is closed (over a drain if necessary) in layers, with the skin being approximated with a running intradermal 4-0 Vicryl or Dexon suture. The fat is cut into strips and insinuated into the dural defect. Continuous lumbar drainage is rarely necessary to prevent cerebrospinal fluid leakage except in extensive intracranialextracranial resections. If the neck is opened, a suction drain is inserted into the depths of the wound before closure. Wounds are closed as in other otologic procedures and dressed with a standard mastoid dressing.
Middle fossa procedures are discussed separately from other neurotologic procedures because they involve a different OR setup and some different instruments. The most obvious deviation from other procedures is the position from which the surgeon operates. The surgeon and the microscope trade locations, so that the surgeon operates from the head of the bed facing caudally (see Fig. 1-10). As with other neurotologic procedures, middle fossa surgery is performed under general anesthesia. In the preoperative holding area, the ipsilateral scalp is shaved to a distance of 6 cm postauricularly and nearly to the midline of the head above the ear in the temporal fossa. Plastic adhesive drapes are applied, and the patient is taken to the OR. After anesthesia, the surgical site and plastic drapes are scrubbed and blotted dry. The area is covered with another plastic adhesive drape. Towels are positioned to block off the entire temporoparietal scalp, including the auricle and zygomatic arch. Sterile sheets complete the draping (Fig. 1-41). The abdomen is usually prepared as in other neurotologic surgeries. The incision is planned so that it begins in the preauricular incisura below the root of the zygoma. It extends cephalad to the area just above the superficial temporal
16
OTOLOGIC SURGERY
FIGURE 1-41. Patient draped for middle fossa surgery.
FIGURE 1-42. Adson periosteal elevator (“joker”).
line. A gentle curve facilitates exposure. Before the incision, as in other cases, the area is infiltrated with local anesthesia. The plastic drape is cut away to expose the skin. After the skin incision is made, the superficial temporal vessels are identified and ligated. After the temporalis fascia is identified, it is recommended that an inferiorly based temporalis muscle flap be created, instead of splitting the muscle. This flap is centered over the zygoma, is elevated from the calvaria, and is reflected caudally by suturing the end of the flap to the drapes. Preserving the muscle with its neurovascular bundle does not limit the surgeon’s exposure, and allows the use of this muscle if facial reanimation surgery should ever be necessary. The remaining temporalis muscle is reflected laterally, and a self-retaining retractor is inserted. A craniotomy is performed. The size of the bone flap removed is dictated by the amount of exposure necessary. For tumor removal, it is wise to err on the large side. The bone flap is carefully removed from the dura with an Adson periosteal elevator, or “joker” (Fig. 1-42). The bone flap is placed in bacitracin solution. The craniotomy
FIGURE 1-43. House-Urban middle fossa retractor.
edges are smoothed with a rongeur, and bleeding is controlled with bone wax. The joker is used to dissect the dura from the floor of the middle fossa. The surgeon is now ready to insert the House-Urban middle fossa retractor. The surgeon must be familiar with the mechanical workings of this device (Fig. 1-43). The retractor is locked under the bony edges of the craniotomy. The blade housing is positioned so that it allows good visualization of the field without placing excessive traction on the temporal lobe. This usually requires repositioning the retractor several times during the early stages of the dissection. Next, the retractor blade is inserted, and the extradural dissection proceeds. The blade can be tilted with the hand and advanced with the thumb, leaving the other hand free for suctioning. Bleeding can be troublesome from the floor of the middle fossa, especially near the middle meningeal artery. Bipolar cautery, bone wax, Surgicel, and other hemostatic agents should be readily available. The surgeon elevates the dura and temporal lobe until the arcuate eminence, superior petrosal sinus, and greater superficial petrosal nerve are visible. Bone over the internal auditory canal (IAC) and geniculate ganglion is removed with a large diamond burr. When the dura over the IAC has been completely skeletonized as far medially as the porus, the wound is irrigated with bacitracin solution, and fresh towels are placed around the field. The dura over the IAC is opened posteriorly (away from the facial nerve) with a sharp hook. For vestibular neurectomy, Bill’s bar is palpated with the same sharp hook that then transects the superior vestibular nerve. Fine microscissors (e.g., Malis, Jacobson) are used to remove a segment of the nerve in continuity with Scarpa’s ganglion. In a likewise fashion, the inferior vestibular and singular nerves are sectioned. For acoustic tumor removal, significantly more bone removal is required. Having established adequate exposure, the plane between the facial nerve and tumor is developed as in the translabyrinthine approach.
Chapter 1 • Otologic Instrumentation
17
and the middle fossa retractor is removed, allowing the brain to re-expand. The wound is irrigated again with bacitracin. Microplates are used to secure the bone flap in place (Fig. 1-44), and the wound is closed in layers, suturing the temporalis flap back to normal anatomic position. Some surgeons close the skin over a Penrose drain, which is removed the day after surgery. A mastoid dressing completes the closure.
CONCLUSION
FIGURE 1-44.
At the conclusion of the procedure, the defect over the IAC can be reconstructed by filling it with small pieces of muscle or abdominal fat and covering it with a small piece of the bone flap that has been cut and trimmed to an appropriate size. The field is inspected for hemostasis,
This chapter has provided a detailed description of the OR environment and instrumentation for most procedures that the otologist is likely to encounter. Although these descriptions do not exhaust all possibilities, they have proved to be satisfactory for many otologists. Appendix 1 lists instruments and equipment that have been presented in the text.
18
OTOLOGIC SURGERY
APPENDIX 1 INSTRUMENTS AND EQUIPMENT FOR OTOLOGIC SURGERY General Operating Room Equipment
1. 3M 1000 plastic aperture drapes 2. 3M 1020 aperture drapes 3. 3M Steri-Drape, Ioban drape, or Cranial-Incise drape 4. Pharmaseal preoperative skin preparation tray, No. 4480 5. Suction irrigation setup 6. Suction canisters 7. Electrocautery unit 8. Skytron operating table
Stapes Surgery 1. Assorted Farrior specula 2. Finger-control Luer-Lok syringe 3. 11⁄2 inch, 25 or 27 gauge needle 4. Small Weitlaner retractor 5. Sheehy fascia press 6. House cutting block 7. Scalpel, No. 15 Bard-Parker blade 8. Adson tissue forceps 9. Iris scissors 10. House-Baron suction tubes, No. 3 to 7 Fr 11. House suction tube adapter 12. Rosen suction tubes, 18 to 24 gauge 13. Sickle knife (No. 1 knife) 14. Lancet knife (No. 2 knife) 15. Robinson knife 16. Sheehy-House weapon (large and small) 17. Rosen needle 18. House elevator 19. Gimmick annulus elevator 20. House stapes curette 21. Incudostapedial joint knife 22. Bellucci scissors 23. Straight Barbara pick 24. Measuring struts, 4.0 to 5.0 mm 25. Measuring disk, 0.6 mm 26. Hough hoe 27. Obtuse, 30 degree, 0.25 mm hook 28. Pick, 0.3 mm, 90 degree 29. Strut guide 30. Footplate chisel 31. Skeeter drill; 1.0, 0.7, and 0.6 mm burrs 32. House strut forceps (nonserrated) 33. McGee wire closing forceps (crimper) 34. Antibiotic ointment 35. Cotton balls, Band-Aids, mastoid dressing 36. Speculum holder
Chronic Ear Surgery
1. Assorted Farrior specula 2. Finger-control syringe 3. 11⁄2 inch, 25 or 27 gauge needle 4. Small Weitlaner retractor 5. Large self-retaining retractor (Weitlaner, Adson cerebellar) 6. Scalpel, No. 15 Bard-Parker blade 7. No. 64 or 67 Beaver blade 8. House cutting block 9. Sheehy fascia press 10. House-Baron suction tubes, No. 3 to 7 Fr 11. Adson forceps 12. Iris scissors 13. Small Metzenbaum scissors 14. Sickle knife 15. Lancet knife 16. Robinson knife 17. Sheehy-House weapon (large and small) 18. Rosen needle 19. Gimmick 20. Crabtree dissector (large and small) 21. Lempert elevator 22. House narrow elevator 23. Pick, right angle, 0.6 mm 24. Pick, right angle, 1.5 mm 25. Pick, right angle, 3 mm 26. Bellucci scissors 27. Hartmann forceps 28. House alligator forceps 29. House cup forceps 30. House-Dieter malleus nipper 31. Zini mirrors 32. Sheehy ossicles holder 33. Speculum, endaural (or nasal) 34. Drill with cutting and diamond burrs 35. House suction-irrigators, No. 2.5 × 4 Fr through No. 8 × 12 Fr 36. Needle holder, Webster 37. Suture scissors 38. Suture, 2-0 chromic and 4-0 Vicryl (or Dexon) 39. Surgifoam (saline-soaked and antibiotic-soaked) 40. Curity Packing strip gauze 41. Silastic sheeting, 0.005 42. Gelfilm 43. Steri-Strips 44. Mastoid dressing 45. Bone wax 46. Surgicel 47. Sheehy bone dust collector
Chapter 1 • Otologic Instrumentation
Endolymphatic Sac Surgery
1. Finger-control syringe 2. 11⁄2 inch, 25 or 27 gauge needle 3. Scalpel, No. 15 Bard-Parker blade 4. Large self-retaining retractor 5. Lempert elevator 6. House narrow elevator 7. Drill and burrs 8. House suction-irrigators (assortment) 9. Brackmann suction-irrigators, No. 4 × 5 Fr, No. 5 × 7 Fr 10. Stapes curette 11. Gimmick 12. Insulated gimmick 13. Bone wax 14. Surgicel 15. Bipolar cautery 16. Bacitracin irrigation solution 17. Beaver ophthalmic blade (No. 59S, 5910, 5920) 18. Pick, right angle, 1.5 mm 19. Hook, right angle, blunt 20. Rosen needle 21. House alligator forceps 22. Silastic shunt material, 0.005 23. Suture, 2-0 chromic and 4-0 Vicryl (or Dexon) 24. Steri-Strips 25. Mastoid dressing 26. Cranial nerve monitoring equipment
Neurotologic Surgery 1. Finger-control syringe 2. 11⁄2 inch, 25 or 27 gauge needle 3. Scalpel, No. 15 Bard-Parker blade 4. Large self-retaining retractor 5. Lempert elevator 6. House narrow elevator 7. Drill and burrs 8. Assorted House suction-irrigators 9. Assorted Brackmann suction-irrigators 10. Stapes curette 11. Gimmick 12. Insulated gimmick 13. Bone wax 14. Surgicel 15. Bipolar cautery 16. Bacitracin irrigation 17. SK-100 Surgi-Kit (Ethox Corp.) 18. Suture scissors 19. House-Urban dissector 20. Pick, right angle, 1 mm 21. Pick, right angle, 1.5 mm 22. Hook, right angle, blunt, 1.5 mm 23. Bellucci scissors
19
24. House cup forceps 25. Blakesley nasal forceps (No. 1) 26. House alligator forceps 27. Myringoplasty knife 28. Jacobson scissors 29. Malis scissors 30. Allis forceps 31. Bayonet forceps 32. Adson tissue forceps 33. Microclip applicator 34. Assorted hemostats 35. Metzenbaum scissors 36. Senn retractor 37. U.S. Army retractor 38. House-Urban rotary dissector or Selector 39. Fisch infratemporal fossa retractor 40. Woodson elevator 41. Fisch microraspatory 42. Sagittal saw 43. Needle holder, Castroviejo 44. Needle holder, Crile-Wood 45. Needle holder, Webster 46. Fisch microscissors 47. Titanium needle holders, smooth (2) 48. Janetta forceps 49. Gerald forceps, with and without teeth 50. Avitene 51. Drains, Penrose and Jackson-Pratt 52. Vessel loops 53. Suture, 5-0 and 6-0 vascular Prolene 54. Suture, 0 and 2-0 chromic 55. Suture, 0 and 2-0 silk 56. Suture, 9-0 nylon or Prolene 57. Suture, 4-0 Dexon or Vicryl 58. Neurosurgical cottonoids 59. NeuraGen nerve guide 60. Steri-Strips 61. Mastoid dressing 62. Topical thrombin 63. Surgifoam 64. Special neurotologic instrument sets (e.g., Kartush, Benecke) 65. Cranial nerve monitoring equipment
Middle Cranial Fossa Surgery 1. Finger-control syringe 2. 11⁄2 inch, 25 or 27 gauge needle 3. Scalpel, No. 15 Bard-Parker blade 4. Large self-retaining retractor 5. Lempert elevator 6. House narrow elevator 7. Drill and burrs 8. Assorted House suction-irrigators 9. Brackmann suction-irrigators 10. Stapes curette
20
OTOLOGIC SURGERY
11. Gimmick 12. Insulated gimmick 13. Bone wax 14. Surgicel 15. Bipolar cautery 16. Bacitracin irrigation 17. SK-100 Surgi-Kit 18. Pick, right angle, 1 mm 19. Pick, right angle, 1.5 mm 20. Hook, right angle, blunt, 1.5 mm 21. Bellucci scissors 22. Fisch microscissors 23. House cup forceps 24. Metzenbaum scissors
25. House-Urban middle fossa retractor 26. Rongeur, Leksell 27. Adson tissue forceps 28. Microclip applicator 29. Assorted hemostats 30. Avitene 31. Cottonoids 32. Gelfoam 33. Suture, 0 and 2-0 chromic 34. Suture, 4-0 Vicryl (or Dexon) 35. Topical thrombin 36. Mastoid dressing 37. Special neurotologic instrument sets 38. Cranial nerve monitoring equipment
2
Canalplasty for Exostoses of the External Auditory Canal and Miscellaneous Auditory Canal Problems Joseph B. Roberson, Jr. and Rodney Perkins Videos corresponding to this chapter are available online at www.expertconsult.com.
Although clinical disease caused by exostoses of the external auditory canal (EAC) is infrequent, it occurs often enough that a method of surgical management should be in the armamentarium of the otologic surgeon. Because it is not a high-incidence problem or one that is life-threatening, many otolaryngologists use various independent approaches, which frequently result in elimination of or damage to the canal skin. These procedures frequently produce suboptimal results. A well-conceived approach addresses the problem of removal of exostoses, while maintaining the valuable residual skin of the EAC. This chapter begins with clinical observations regarding this condition and then describes an operative procedure that has been very successful in its management. The etiology of these benign growths of the tympanic bone is strongly associated with the frequency and severity of exposure to cold water.1 Frequently, these lesions are found in surfers, swimmers, or other individuals with frequent cold water exposure over several years. A widely held belief based on clinical information is that exostoses occur primarily during the years of growth, with their proliferation being enhanced or perhaps even caused by exposure to cold water during this period. This belief tends to be supported by historical information from patients with exostoses, who almost always indicate that they swam in cold water during their youth.2-4 This historical information is strongly corroborated by the high incidence of exostoses in avid surfers who spend hours in the water almost daily. In our clinical experience, this problem occurs almost exclusively in men, who are more likely than women of the same age to have had frequent cold water exposure during their youth. Most exostoses do not develop to a degree sufficient to cause clinical symptoms. Patients are frequently referred to otologists because the growths are observed, and not understood, by primary care physicians. This is particularly true with exostoses that have a more pedunculated form than the more subtle sessile configuration. When exostoses become more marked, however, they obstruct the natural elimination of desquamated epithelium from the ear canal, and patients usually present with recurrent episodes of external otitis. In their most prolific expression, exostoses can lead to hearing impairment by causing the collection
of epithelial debris that tamponades tympanic membrane movement, by impinging on and limiting the mobility of the malleus, or by markedly narrowing the aperture of the canal. These conditions may manifest as a conductive hearing impairment on audiometric examination. The EAC is part of the hearing pathway. Essentially, the EAC is a tube with resonant characteristics that amplify the incoming sound. The degree of amplification and the frequency at which it occurs are a function of the diameter and the length of the canal. When the diameter becomes small, it can interfere with the passage of sound and cause a hearing impairment. This effect does not become significant, however, until the aperture becomes very small. With apertures less than 3 mm, high-frequency sounds begin to diminish, and further compromise of the channel diameter results in increased impairment and lower frequency loss.
EXOSTOSES OF THE EXTERNAL AUDITORY CANAL Surgical Indications Surgery is indicated when chronic or recurrent external otitis exists, or a conductive hearing impairment develops. The presence of chronic and recurrent infection over an extended period seems to debilitate the canal skin, and can compromise the skin’s ability to re-epithelialize in a robust and healthy manner in the postoperative period. For this reason, surgical therapy should be considered when a pattern of recurrent external otitis has been established in these patients. Patients who have significant external canal exostoses without recurrent infection or hearing impairment should be observed periodically, and surgery should be avoided until these symptoms occur.
Preoperative Preparation Patient Preparation There are two components of patient preparation for otologic surgery performed under local anesthesia: psychological and pharmacologic. 21
22
OTOLOGIC SURGERY
Psychological Preparation To reduce anxiety and create rapport, the surgeon should provide the patient with a full explanation of the procedure and its objectives, benefits, and risks. In addition, a surgical nurse or medical assistant should explain what will happen to the patient in the operating room, and describe such things as the operating room environment, use of an intravenous line for medication delivery, placement of monitor electrodes, and draping. By informing the patient of these things and making him or her part of the process, the clinician reduces the patient’s anxiety, encourages cooperation, and may reduce bleeding. Beyond the technical advantages achieved by such preparation, there is an ethical responsibility to inform the patient. In addition, the likelihood of the patient’s becoming litigious because of a poor result is markedly reduced if he or she has been informed about the procedure and its risks and benefits, and has had an opportunity to discuss the risks and benefits with the surgeon before the surgery. Pharmacologic Preparation The pharmacologic preparation of the patient can be achieved in many ways. In the average adult who selects intravenous sedation with local anesthesia, we give fentanyl, 50 to 100 μg, and midazolam (Versed), 5 to 10 mg intramuscularly 1 hour before the surgical incision. An intravenous catheter is started in the arm opposite the ear to be operated on before the patient arrives in the operating room, and 5% dextrose in Ringer solution is started with a volutrol. Unless the patient appears very sedated, an additional 50 to 100 μg dose of fentanyl is placed in the volutrol and infused slowly over 30 to 45 minutes. As the surgery proceeds, alternating supplements of intravenous midazolam and fentanyl are infused as needed to maintain sedation. In most cases, general anesthesia is selected by either the surgeon or the patient. Patients with a history of claustrophobia, patients with poor language skills in the surgeon’s native tongue, and patients with difficult neck mobility are best approached under general endotracheal anesthesia. One advantage of use of general anesthesia is the ability to use facial nerve monitoring during the procedure. Patients receive cefazolin, 1 g intravenously, or another appropriate antibiotic 1 hour before incision.
Site Preparation The hair is shaved behind the ear to a distance of approximately 1.5 inches posterior to the postauricular fold. The auricle and the periauricular and postauricular areas are scrubbed with povidone-iodine (Betadine) solution or chlorhexidine gluconate (Hibiclens) for iodineallergic patients. A plastic drape is placed over the area with the auricle and the postauricular area exteriorized through the opening in the drape. This drape is placed
over an L-shaped bar that is fixed in the rail attachment of the operating table (Fig. 2-1). For patients under local anesthesia with sedation, a small, low-volume office fan is attached to the bar to provide a gentle cooling breeze to the patient’s face during the procedure. The plastic drape forms a canopy, allowing the patient to see from under the drape and reducing the feeling of claustrophobia. In addition, a foam earpiece from an insert speaker is put into the opposite ear. The earpiece is connected to a compact disk player and input microphone that allows the patient to listen to relaxing music and provides a pathway to converse with the patient, if desired.
Analgesia It is important not only to achieve analgesia, but also to maximize canal hemostasis with injections into the external auditory meatus. Using 2% lidocaine (Xylocaine) with 1:20,000 epinephrine solution in a ringed syringe with a 27 gauge needle, a classic quadratic injection is made such that each injection falls within the wheal of the previous injection. Another useful injection is an anterior canal injection, which is made with the bevel of the needle parallel to the bony wall of the external meatus (Fig. 2-2). In a patient with extensive exostoses, this injection is usually made into the lateral base of a large anterior sessile osteoma. After insinuation of the needle, it is advanced a few millimeters, and a few drops are injected extremely slowly. The solution infiltrates medially along the anterior canal wall and provides some analgesia to the auriculotemporal branch of CN V, which is usually unaffected by the quadratic injection and adds to the hemostasis anteriorly. The postauricular area is infiltrated with 2% lidocaine with 1:100,000 epinephrine solution mixed with equal parts of 0.5% bupivacaine.
Surgical Technique Most surgical approaches for removal of EAC exostoses are through the transmeatal route.5-7 This approach has two disadvantages. It usually results in significant loss of the remaining canal wall skin through damage by the drill, and it does not allow adequate visibility or instrument and drill access to remove the medial portion of the exostotic mass near the tympanic membrane safely. A large sessile anterior exostosis is almost uniformly present in these patients (Fig. 2-3). The approach described here is primarily postauricular and one that maximizes conservation of the canal wall skin and facilitates careful removal of the anterior exostosis, which is usually extremely close to the tympanic membrane. A curvilinear postauricular incision is made approximately 1 cm behind the postauricular fold (Fig. 2-4). The skin and subcutaneous tissues are elevated anteriorly to the area of the spine of Henle and the bony posterior canal, and a toothed, self-retaining retractor is placed (Fig. 2-5).
Chapter 2 • Canalplasty for Exostoses of the External Auditory Canal
Locating this area is facilitated by finding the plane of the lateral surface of the inferior border of the temporalis muscle and dissecting in this plane anteriorly to reach the meatus. When this area is reached, the skin overlying the lateral slope of the posterior exostosis is elevated from its
23
surface, and a Perkins bladed tympanoplasty retractor is inserted to hold elevated skin off the surface of the lateral portion of the bony mass (Fig. 2-6). Although there may be more than one posterior and anterior exostosis, predominant anterior and posterior
24
OTOLOGIC SURGERY
Chapter 2 • Canalplasty for Exostoses of the External Auditory Canal
exostoses are usually present along with others of lesser mass. These secondary masses may be handled similarly to the primary exostoses, or may be removed directly. To simplify the description here, this operation is divided into two major segments: removal of the posterior exostosis, and removal of the anterior exostosis.
Removal of Posterior Exostosis By use of a medium-sized cutting burr and an appropriately scaled suction-irrigator, the posterior exostosis is entered along its lateral sloping edge, and the bony removal is progressed medially, keeping a shell of bone over the area being burred anteriorly (Fig. 2-7). The remaining skin over the exostosis medial to the skin elevated earlier is protected from the burr. As this shell becomes thinner, it is advisable to switch to a diamond burr to prevent a sudden breakthrough to the skin, which might occur if one continues with the cutting burr on the excessively thinned bone. The bone removal is continued medially and posteriorly until the estimated normal posterior canal contour and dimension is achieved. As one approaches a medial depth consistent with the posterior annulus of the tympanic membrane (which usually cannot be seen directly at this point), care must be taken to avoid damage to the chorda tympani nerve and the posterior aspect of the tympanic membrane. The surgeon should also keep in mind that some patients’ facial nerve exists lateral to the tympanic annulus at its posteroinferior border. Facial nerve monitoring reduces the possibility of injury to the nerve in a patient unable to tolerate local anesthesia. The thinned bony shell is collapsed, and a small elevator reveals the inside surface of the posterior canal skin that was over the exostosis (Fig. 2-8). An incision is made midway along the posterior canal skin perpendicular to the long axis of the EAC (Fig. 2-9). The posterior canal skin medial to this incision is positioned onto the new contour of the posterior canal wall (Fig. 2-10). The transmeatal approach is then taken, and incisions are made with a sickle knife superiorly and inferiorly in the canal, extending from the ends of this previous incision laterally to the meatus, and creating a laterally based posterior canal skin flap. This flap is involuted back into the meatal portion of the canal and held there with the Perkins retractor (Fig. 2-11). Attention is turned to the anterior exostosis, which has now been revealed.
Removal of Anterior Exostosis By use of a round knife, an incision is made in the skin overlying the anterior exostosis from superior to inferior over the dome of the exostosis and as far medially as can be seen. This incision is connected to the incisions previously made superiorly and inferiorly in the canal that defined the posterior canal skin flap, and this anterior canal flap is elevated laterally (Fig. 2-12). Frequently,
25
the skin of the vascular strip can be left intact if the exostoses do not involve this portion of the canal. By use of a back-angled Perkins tympanoplasty elevator, this laterally based anterior canal skin flap is elevated further to the cartilaginous portion of the anterior canal and is smoothed so as to lie laterally near the posterior canal flap under the retractor (Fig. 2-13). With a cutting burr and small suction-irrigator, the anterior exostosis is removed in a manner similar to that of the posterior one, and a thin shell of bone that protects the canal skin is left over the anteromedial portion of the exostosis from the burr (Fig. 2-14). This bone removal is continued to the area of the anterior annulus of the tympanic membrane. The bony shell is collapsed and removed, leaving the intact anterior canal skin (Fig. 2-15). Usually, it is necessary to finish up and smooth an edge of bone that remains at the anterior extent of this dissection to have a smooth contour near the annulus area. To protect the elevated anterior sulcus skin from the burr, a small tympanic membrane–sized piece of silicone elastomer (Silastic) or suture packet foil is placed on the inside surface of the anterior canal skin to hold it against the tympanic membrane during drilling. This prevents the skin flap from getting involved with the burr, and prevents damage to the tympanic membrane that might occur with the burr being used in such close proximity to the membrane. Subsequently, the Silastic is removed, the medial anterior canal skin is placed on the bone, and all skin flaps are folded back into position on the new contours of the bony canal (Fig. 2-16). The medial flaps are packed into place with chloramphenicol (Chloromycetin)-soaked absorbable gelatin sponge (Gelfoam) pledgets, and the postauricular incision is closed with interrupted subcuticular 4-0 polyglactin 910 (Vicryl) suture. Through the transmeatal route, the laterally based canal skin flaps are packed into place with Gelfoam pledgets. A cotton ball is placed in the meatus, and a mastoid dressing is applied. The patient is returned to the outpatient recovery area and discharged after appropriate recovery.
Postoperative Care The patient is instructed to remove the mastoid dressing the next morning. The Gelfoam packing is removed using the stereomicroscope on the first office visit 1 week later. Antibiotic-steroid ear drops are prescribed for use twice daily for 1 week and once every 3 days for another 2 to 3 weeks. The second postoperative visit is at 1 month. If there is no evidence of infection, no additional ear drops are recommended. Because most of the patients in whom this procedure is done have had recurrent external otitis, and because time is needed for epithelialization of uncovered bone, the ear canal may remain moist for a longer time than in a typical tympanoplasty. Until the ear canal is completely dry and healed, the patient should be
26
OTOLOGIC SURGERY
Chapter 2 • Canalplasty for Exostoses of the External Auditory Canal
27
28
OTOLOGIC SURGERY
seen every few weeks to inspect and clean debris from the canal as needed. The canal skin has usually been exposed to numerous infections and has been stretched over the exostoses; it may not be as resilient as normal canal skin. Return to water exposure should be avoided until 1 to 2 months after complete healing has occurred. Frequently, avid surfers return to the water much sooner than instructed, however. Antibiotic drops given after water exposure reduce the risk of early postoperative infection. If the patient is still in the growth years, further repeated exposure of the ear to cold water should be moderated. The bone may reproliferate under these conditions, and further surgery may become necessary. In patients who want to return to frequent surfing or similar water exposure, earplugs should be worn to prevent water entrance. This problem lessens in older surfers because they may be beyond their rapid growth phase, and the economic exigencies of life tend to decrease their frequency of exposure. It is advisable to see the patient annually for 2 years to assess the tendency for the problem to recur, although recurrence is infrequent.
Problems and Complications Although this procedure is not fraught with serious complications, complications can occur during several aspects of the operation. As the medial extent of the canal is approached in the removal of the posterior exostosis, the course of the chorda tympani nerve must be kept in mind. This portion of the bone removal is done largely without definite landmarks: the surgeon must rely on mental estimation of the distances in arriving at the posterior annulus. The chorda tympani nerve is beneath the bone near this field of dissection and could sustain damage. Also, it is important to remember the course of the facial nerve, which passes posterior and inferior to the canal, although this area is farther from the immediate area of dissection than the chorda tympani nerve. When a burr is used very near the tympanic membrane and the malleus, a diamond burr should be used because it is less likely to run erratically than the cutting burr.
Summary Exostoses of the EAC usually manifest without attendant compromise in function or clinical disease. When recurrent external otitis or hearing impairment results, however, surgical removal is indicated. Canalplasty has significant advantages over commonly employed transmeatal approaches by maximizing conservation of canal skin and providing surgical access to the anterior medial zone of the canal. Complications are infrequent, but attention to the anatomy of the chorda tympani and facial nerve pathways and careful drill technique in the area of the tympanic membrane are important. Although surgical techniques involving the EAC have had little attention compared with other reconstructive
procedures, they should be in the armamentarium of all otologic surgeons. This technique has proved to be effective for the management of exostoses of the EAC.
MISCELLANEOUS EXTERNAL AUDITORY CANAL CONDITIONS Medial Third Stenosis For unknown reasons, some patients develop weeping epitheliitis over the medial third of the EAC. Treatment consists of antibiotic-steroid ear drops that supply broadspectrum bacterial coverage. Intense treatment, including débridement and the use of topical agents, is usually necessary to bring the process under control. Despite attempts at treatment, progression of the condition may follow a relentless course, resulting in dense fibrosis of the medial segment of the EAC with conductive hearing loss. The mesotympanum and ossicular chain are characteristically spared. Surgical repair may be necessary when conductive hearing loss produces a functionally significant deficit for the patient. Successful repair is frequently possible, although restenosis may occur, and this possibility should be included in the informed consent. Technically, a postauricular approach is used to allow complete resection of the fibrotic segment medial to noninvolved EAC skin where an incision has been previously created working through a transcanal route (Fig. 2-17). Removal of most of the fibrous layer of the tympanic membrane seems to reduce the chance of postoperative restenosis. Tympanoplasty is performed with a lateral graft or fasciaform technique. Coverage of the resultant exposed bone is mandatory and is provided with a free split-thickness skin graft. The posterior surface of the pinna provides skin of appropriate character within the operative field and can be taken with a No. 10 blade. Skin grafts should overlap the fascia used for tympanic membrane replacement, but should not extend to cover the lateral surface of the reconstructed drum. Antibiotic-containing absorbable packing is removed 7 to 14 days later and antibiotic-steroid ear drops are continued for 2 weeks beyond healing to be tapered over time. Close observation postoperatively is necessary to intervene with any signs of restenosis. Recurrent epitheliitis may occur months or years after successful repair.
Collapsing Canal Stenosis of the cartilaginous portion of the lateral EAC may produce symptoms for some patients. In severe cases, conductive hearing loss may result when closure to less than 2 mm occurs. More commonly, accumulation of debris and a warm, moist environment lead to recurrent external otitis. Although this condition occurs naturally, an iatrogenic component is frequently present. After a postauricular incision, the natural tension of the
Chapter 2 • Canalplasty for Exostoses of the External Auditory Canal
Canal incision
29
Tympanic membrane and Fibrosis
Noninvolved canal skin
M I S
FIGURE 2-17.
cartilaginous canal may be unopposed by inadequately reapproximated deep layers such as the mastoid periosteum. Gradual stenosis may occur until symptoms become evident many years after the surgical procedure. Operative repair includes removal of cartilage from the anterior concha and posterior cartilaginous canal from the postauricular area with imbrication of the deep tissue layers overlying the mastoid cortex, similar to imbrication of the subcutaneous musculoaponeurotic system in a facelift. The skin of the ear canal need not be violated in such a procedure. Postoperative stenting for 2 weeks also is helpful in restoring a normal contour to the canal. In some patients with only lateral soft tissue and cartilage involvement, the stenosis may be addressed via a transcanal route, avoiding a postauricular incision.
Keratosis Obturans Exuberant accumulation of desquamated skin may produce bony erosion and gradual expansion of the bony EAC.8 The process may progress to the point of erosion into structures adjacent to the canal, such as the temporomandibular joint or mastoid. Erosion lateral to the eardrum may cause loss of support of the fibrous annulus of the tympanic membrane and a characteristic “jump rope sign” inferiorly (which can also be seen after curetting for a stapes procedure more superiorly). Poor epithelial migration has been proposed as the cause of the disorder. Frequent cleaning may retard the process. Cleaning may be much easier if the typically inspissated and adherent material is softened with mineral oil for several days before the clinical appointment. Surgical intervention is rarely indicated, unless severe erosion exposes vital structures.
Osteonecrosis and Osteoradionecrosis of the Tympanic Bone Radiation and occasionally chronic vasculitis devascularize a portion of the tympanic bone, producing skin loss and bone exposure. The low-grade osteomyelitis can be managed conservatively with topical antimicrobials and mild débridement. Addition of oral antibiotics may improve the chance of healing lesions in the early phases. Frequently, bone involvement progresses, however, and can lead to further skin loss. A culture and sensitivity test is indicated before institution of topicals and later with deterioration of healing to look for resistant organisms. One must always consider malignancy in such a clinical situation, and biopsy is prudent in many cases. Operative repair is indicated for progression of bone exposure or associated cellulitis or both. Removal of all devitalized bone with the postauricular approach is necessary. The margins of the canal skin are freshened similar to what is performed in a tympanoplasty. Autogenous fascia is placed directly on the freshly drilled bone, and the skin is returned to anatomic position overlying it. The external canal is packed with antibiotic-containing absorbable sponge, which is removed in 7 to 10 days when antibiotic drops are initiated, which are continued until complete healing occurs.
Scutum Defects Cholesteatoma of the pars flaccida produces bone erosion in many patients. Repair of the EAC is necessary to prevent re-retraction and cholesteatoma formation through the canal defect. Small defects (<2 mm) may be repaired with double-layered fascia. Large defects must
30
OTOLOGIC SURGERY
Elevated skin and tympanic membrane
Cartilage and attached perichondrium
Cartilage and attached perichondrium
Lateral
M TM M S
I
Canal skin
Medial
FIGURE 2-18.
U-shaped fascial sleeve
Posterior portion Anterior portion
Autogenous bone graft Mastoidectomy, extension into epitympanum
U-shaped fascial sleeve Autogenous bone graft Bone removal of epitympanum M S I
Medial
FIGURE 2-19.
Chapter 2 • Canalplasty for Exostoses of the External Auditory Canal
be addressed for best long-term patient outcome. Repair may be accomplished with a composite cartilage graft or a fascial sleeve and bone pâté. Cartilaginous repair is possible with cartilage harvested from the base of the tragus. The perichondrium is left attached to one side of the cartilage, which is carved to match the bony defect similar to the piece of a puzzle. The composite graft is inserted in the defect after elevation of the canal skin and eardrum from the lateral surface of the handle of the malleus. The graft is positioned such that the perichondrium faces the elevated skin and eardrum, and the cartilage extends into the defect (Fig. 2-18). Alternatively, bone formation may be stimulated with placement of autogenous bone chips (bone pâté harvested with a microdrill and mixed with antibiotic) within a U-shaped fascial sleeve.9 This technique requires removal of bone lateral to the heads of the ossicles in the epitympanum and sculpting of the posterior surface of the bony external canal to allow placement of the fascia (Fig. 2-19).
Post-Traumatic Suture Dehiscence Temporal bone trauma may produce partial dislocation of either the temporosquamous or the temporomastoid suture lines. When visible on examination, these signs indicate a temporal bone fracture. Rarely, surgical intervention is necessary for entrapped epithelium.
31
REFERENCES 1. Kroon D F, Lawson M L , Derkay C S, et al: Surfer’s ear: External auditory exostoses are more prevalent in cold water surfers. Otolaryngol Head Neck Surg 126:499-504, 2002. 2. Adams W: The aetiology of swimmer’s exostoses of the external auditory canals and of associated changes in hearing: I. J Laryngol 65:133-153, 1951. 3. Harrison D: Exostosis of the external auditory meatus. J Laryngol 65:704-714, 1951. 4. Fowler E P Jr, Osmun PM : New bone growth due to cold water in the ears. Arch Otolaryngol Head Neck Surg 36:455-466, 1942. 5. Rauch S D: Management of soft tissue and osseous stenosis of the ear canal and canalplasty. In Nadol J B Jr, Schuknecht H F (eds): Surgery of the Ear and Temporal Bone. New York, Raven Press, 1993, pp 117-125. 6. Shambaugh G E Jr, Glasscock M E III: Operations on the auricle, external meatus, and tympanic membrane. In Shambaugh G E Jr, Glasscock M E III (eds): Surgery of the Ear. Philadelphia, Saunders, 1980, pp 194-215. 7. Dibartolomeo J R : Exostoses of the external auditory canal. Ann Otol Rhinol Laryngol 88(Suppl 61):2-20, 1979. 8. Piepergerdes JC, Kramer B M, Behnke E E : Keratosis obturans and external auditory canal cholesteatoma. L aryngoscope 90:383-390, 1980. 9. Althaus S R : Tympanomastoid surgery: A technique for repairing posterior osseous canal wall defects with autologous temporalis fascia and bone pâté. Otolaryngol Head Neck Surg 93:529-535, 1985.
3
Malignancies of the Temporal Bone—Limited Temporal Bone Resection Moisés A. Arriaga and John P. Leonetti Videos corresponding to this chapter are available online at www.expertconsult.com.
Although carcinoma of the temporal bone is uncommon and aggressive, the best outcome depends on careful evaluation and planning that result in a complete resection with pathologically clear surgical margins. The most common lesion is primary squamous cell carcinoma of the external auditory canal (EAC); however, direct extension by pinna and salivary gland lesions, metastatic lesions, and adenocarcinomas of the glandular adnexa of the ear canal and pinna are potential lesions. Tumors involve the temporal bone through primary growth, direct extension, and metastatic spread. The consequences of these lesions include morbidity owing to anatomic remodeling, intracranial extension, perineural spread, vascular encasement and dural invasion, and ultimately death. Historically, carcinoma of the temporal bone was an ominous diagnosis. Advancements in diagnostic and surgical technique have led to greatly improved outcomes. The successful resection of T1 disease has shown survival outcomes of 95%, and the treatment of T2 and T3 disease has shown survivals of 85% when a clear surgical margin is combined with postoperative radiation therapy.1 A multidisciplinary approach to the diagnosis and treatment of these lesions offers the best possible outcome. Surgery and postoperative radiation represent the principal treatment arms; however, interest in a possible combined role for chemotherapy is increasing.2 The type of resection needed is determined by the extent of disease. A sound oncologic resection can be achieved through a lateral temporal bone resection, subtotal temporal bone resection, or total temporal bone resection with such additional simultaneous procedures as parotidectomy, facial nerve resection, mandibulectomy, and cervical lymphadenectomy as indicated. When the surgery has been planned, reconstructive options should be explored. Great strides have been made over the past century in the surgical and nonsurgical management of temporal bone neoplasms, but they remain a significant treatment challenge.
TUMOR CONSIDERATIONS The incidence of tumors of the temporal bone is 200 new cases per year with a frequency of 6 cases per 1 million. Squamous cell carcinoma accounts for 86% of these tumors. Possible etiologic factors include industrial exposure to petroleum-based products, topical disinfectants, and chronic infection. Basal cell carcinoma, adenoid cystic carcinoma, adenocarcinoma, and ceruminous carcinoma occur less frequently.3-5 Tumors of mesenchymal origin are as rare, with rhabdomyosarcoma occurring most frequently.6 Salivary gland tumors can originate from ectopic rests of salivary tissue within the middle ear (pleomorphic adenoma has been described in that location), or from minor salivary glands within the EAC, but both are rare.7 Salivary gland tumors are more likely to involve the temporal bone through direct extension from the parotid gland. The anatomic relationship between the temporal bone and parotid gland is responsible for this tendency. The parotid gland is located in close proximity to the mastoid and tympanic portions of the temporal bone. It communicates directly with the cartilaginous EAC through the fissures of Santorini and foramen of Huschke.8 The stylomastoid foramen, carotid canal, jugular foramen, petrotympanic fissure, and eustachian tube provide avenues for intratemporal extension. In primary temporal bone carcinoma, these anatomic pathways are particularly significant as routes for extension beyond the temporal bone to the adjacent parotid tissue and soft tissue at the base of the skull base. In rare cases of temporal bone involvement by benign tumors of the parotid gland, symptoms of a facial mass, trismus, or compression of the parapharyngeal space by tumor usually precede temporal bone involvement.9 Additionally, benign masses have a tendency to compress or remodel adjacent tissue, rather than invade it. This tendency allows for extirpation of a benign parotid neoplasm, in some cases, without requiring a formal temporal bone resection. These lesions can be removed through 33
34
OTOLOGIC SURGERY
traditional parotidectomy techniques. Disarticulation of the mandibular condyle, resection of the EAC, or mastoidectomy may be necessary to assist with resection. Pleomorphic adenoma of the tail of the parotid gland can manifest as a subcutaneous mass of the floor of the lateral portion of the EAC. This situation is best managed with combined superficial parotidectomy with mastoidectomy and postauricular canalplasty. Temporal bone involvement by malignancies of salivary gland origin is aggressive and involves adjacent tissue more readily. Tumors can develop insidiously and relatively asymptomatically, with 75% manifesting with a painless mass and only 6% to 13% manifesting with facial nerve palsy. Symptoms may not be present until after temporal bone invasion has occurred. Pain, dysphagia, and dysphonia can occur after direct invasion of the skull or involvement of the lower cranial nerves at the jugular foramen. Direct extension into bone, fissures, and foramina are potential routes of spread. Mass lesions of the poststyloid parapharyngeal space can traverse the carotid canal and jugular foramen, and neurotrophic tumors, such as adenoid cystic carcinoma, follow the facial nerve as an avenue toward the stylomastoid foramen. Carcinoma of the auricle, anterior scalp, or face that has spread to the intraparotid lymphatics can involve the temporal bone in a similar fashion. In a review of 27 cases of advanced and recurrent parotid neoplasms requiring temporal bone resection, Leonetti and colleagues9 found that adenocarcinoma occurred most commonly followed by adenoid cystic carcinoma and mucoepidermoid carcinoma. Malignancies involving the temporal bone present a formidable problem. Without treatment, these lesions result in a high incidence of morbidity and almost certain death. Extension into the otic capsule and petrous bone can result in hearing loss, vestibulopathy, cranial neuropathies, and hemorrhage. Malignant spread into the middle and posterior fossa and extension into the petroclival region or cavernous sinus portend a grim prognosis even with aggressive surgical efforts. In cases of distant metastatic disease, a temporal bone resection may still be indicated for palliation.
HISTORY AND PHYSICAL EXAMINATION A high index of suspicion by the examiner is necessary for prompt diagnosis and treatment in temporal bone carcinoma. Symptoms of fullness, pain, or trismus without clear explanation or rapid resolution are suspicious for carcinoma. Refractory pain is a hallmark of temporal bone carcinoma. An unexplained mass of the pinna, ear canal, or middle ear should prompt closer evaluation. An insidious symptom of carcinoma that frequently causes a delayed diagnosis is persistent ear drainage. Before the advent of computed tomography (CT), most temporal bone carcinomas were diagnosed after mastoidectomy
for presumed chronic otitis media. Such nononcologic intervention frequently resulted in spread of the lesion to adjacent soft tissue structures.
BIOPSY The only way to diagnose temporal bone carcinoma definitively is by biopsy. Although it is possible to obtain a biopsy specimen of an exuberant lesion of the EAC in the office, we recommend imaging before performing any deep tissue biopsies or removing any soft tissue lesions of the middle ear to prevent inadvertent damage to the carotid, jugular bulb, or facial nerve. Falsenegative biopsy specimens are an important consideration in temporal bone carcinoma. These lesions are often secondarily infected, and superficial biopsy specimens may reveal only chronic inflammatory changes. If initial biopsy results are negative, we recommend performing deeper tissue biopsies in an operating room to ensure that an adequate sample has been obtained.
DIAGNOSTIC TESTS Imaging studies and audiometric testing are crucial. Magnetic resonance imaging (MRI) and CT provide accurate information helpful in the staging of disease and determining the extent of resection needed. A preoperative hearing assessment establishes a functional baseline, the need for postoperative middle ear reconstruction when appropriate, and the potential for postoperative deficits. Angiography should be performed in all cases in which the carotid artery is at risk. When carotid resection is anticipated, cerebral blood flow analysis should be used to determine resectability and the need for revascularization. Consideration should also be given to embolization when intraoperative hemorrhage is a concern. When the work-up is complete, proper TNM staging can occur, which provides a basis for discussing treatment options and prognosis with the patient. The close proximity of vital structures within and adjacent to the temporal bone requires an accurate assessment of the involved anatomy. CT and MRI are indispensable in this regard. The ability of CT to detail bony structures and the superb soft tissue contrast offered by MRI play complementary roles during the work-up. Together, CT and MRI are helpful in establishing tumor extent, the involvement of critical structures, and the best surgical plan.10 The use of high-resolution CT with fine cuts is recommended. Images with a thickness of 0.625 to 1.25 mm produce the best assessment of the skull base. Direct axial and coronal scans should be obtained. When advanced scanning technologies are available, reconstructed images provide resolution that is equivalent to the original data set. CT is essential to preoperative staging. Arriaga and
Chapter 3 • Malignancies of the Temporal Bone—Limited Temporal Bone Resection
colleagues11 showed that CT can “achieve 98% accuracy in predicting pathologic involvement in temporal bone resection specimens.” CT scans have limitations, however. Distinguishing mucosal inflammation from tumor and the extension of tumor without bony erosion remains difficult. In previously operated areas, positron emission tomography (PET) combined with CT has proven useful for distinguishing scar from neoplasm in identifying tumor recurrence and guiding treatment.12 MRI should be used to examine tumor relative to dura, brain, cerebrospinal fluid, and skeletal muscle. MRI also offers the advantage of imaging in multiple planes. Multiplanar imaging is useful when evaluating lesions that traverse the skull base through direct extent or perineural spread. T1 and T2 fat-saturated sequences should be obtained. T1 images help to determine spatial relationships and bone marrow involvement, whereas T2 images with fat saturation help to delineate tumors that enhance brightly. Postgadolinium T1-weighted images with fat saturation should also be obtained to determine the presence and extent of perineural involvement. This involvement can manifest as foraminal widening or enhancement, replacement of fat density, or increased signal intensity. Audiologic testing shows the functional status of the middle and inner ear. This information is useful during surgical planning and preoperative patient counseling. Conductive losses may be attributable to the presence of a malignancy in the external or middle ear. Anacusis, tinnitus, and vertigo suggest inner ear involvement. If a reduction or elimination of hearing is anticipated, the patient should be informed in advance. Ossicular chain reconstruction or a bone-anchored hearing aid may be indicated when conductive hearing is sacrificed, but sensorineural hearing is spared. A four-vessel angiogram with venous runoff should be used in cases of carotid encasement, or when disease mandates dissection of the petrous carotid. Arterial stenosis and contour irregularity at the site of the lesion suggest malignant involvement. Close inspection of the carotid canal on CT and CT angiography is useful for carotid assessment. Temporary balloon occlusion coupled with a cerebral blood flow quantification study helps to determine when carotid sacrifice would be tolerated. Quantification studies include PET, functional MRI, and xenon-CT. Xenon-CT is the best-studied modality.13 An occlusional cerebral blood flow less than 30 mL/100 g/min requires carotid bypass with prophylactic or intraoperative saphenous grafting or consideration of preoperative carotid stenting.14 Cerebral blood flow less than 30 mL/100 g/min or the development of neurologic symptoms during the test indicates a high risk of perioperative stroke. These findings either preclude the surgical option or require that prophylactic revascularization be done if surgery is performed. The venous side of imaging must not be ignored. If the torcular Herophili is not patent, permitting venous drainage from the ipsilateral
35
sigmoid-transverse system to the opposite sigmoidtransverse-jugular system, venous infarction may occur. Specific attention should be focused on the adequacy of contralateral venous drainage.
STAGING The American Joint Committee on Cancer states that a staging system should provide a sound basis for therapeutic planning for cancer patients by describing the survival and resultant treatment of different groups in comparable form. A lack of uniformity regarding preoperative staging and treatment plans made the study of temporal bone cancer difficult. This situation has changed with the demonstration that CT accurately predicts the degree of neoplastic involvement. The supplementation of these observations with clinical findings enabled the formation of a reliable system for staging carcinoma of the EAC.11 Attempts to modify this system have been made. Moody and coworkers15 proposed upstaging patients with preoperative facial nerve paralysis or paresis to T4 status when the primary tumor originates in the EAC. They proposed that the extension of tumor through tissue separating the facial nerve and EAC and tumor involving the horizontal segment of the nerve are ominous signs.15 Although this and other modifications to the original system by Arriaga and colleagues11 have been suggested, the original system has provided a useful framework for staging these lesions (Table 3-1), and validation of the various systems is still pending.
TABLE 3-1 Arriaga—University of Pittsburgh Tumor
Lymph Node Metastasis Staging System Proposed for Squamous Cell Carcinoma of the External Auditory Canal
T Status T1—Tumor limited to external auditory canal without bony erosion or evidence of soft tissue extension T2—Tumor with limited external auditory canal bony erosion (not full-thickness) or radiographic finding consistent with limited (<0.5 cm) soft tissue involvement T3—Tumor eroding osseous external auditory canal (full-thickness) with limited (<0.5 cm) soft tissue involvement, or tumor involving middle ear or mastoid, or patients presenting with facial paralysis T4—Tumor eroding cochlea, petrous apex, medial wall of middle ear, carotid canal, jugular foramen, or dura, or with extensive (>0.5 cm) soft tissue involvement N STATUS Involvement of lymph nodes is a poor prognostic finding and automatically places patient in advanced stage (i.e., stage III [T1, N1] or stage IV [T2, T3, and T4, N1] disease) M STATUS Distant metastasis indicates poor prognosis and immediately places patient in stage IV
36
OTOLOGIC SURGERY
This system can be used to plan the surgical management of the temporal bone when involved secondarily by a cancer of the parotid gland. Although the data are not as clear for salivary gland cancers or pinna lesions extending to the temporal bone as for primary temporal bone lesions, the same staging criteria should be applied to decisions regarding the extent of the anatomic resection and the need for radiation therapy. Patients can be counseled about prognosis and survival based on results with primary cancer arising in the temporal bone, while carefully evaluating the status of the primary source of origin and possible metastases to cervical lymphatics.
SURGICAL TREATMENT Surgery with postoperative radiation is the principal treatment for temporal bone carcinoma. The surgical procedure depends on the extent of disease because the medial extension of tumor dictates the aggressiveness: partial temporal bone resection, subtotal temporal bone resection, or total temporal bone resection. When tumors involve the temporal bone secondarily through direct extension, temporal bone resection accompanies the surgical management of the primary tumor. Similarly, if parotid gland involvement is likely from a primary temporal bone lesion, parotidectomy is also indicated. Partial temporal bone resection includes removal of the entire external auditory meatus. It is indicated for cancer of the temporal bone that is limited to the external canal. The facial nerve, stapes, and promontory provide the medial limit of resection. The incisions and soft tissue management for all temporal bone resections depend on the planned management of the pinna. If the pinna is extensively involved and is to be sacrificed, a postauricular incision combined with a preauricular incision permits excision of the pinna—usually the pinna is resected in continuity with the temporal bone resection and parotidectomy specimen. If the pinna can be preserved, a circumferential incision around the soft tissue of the ear canal is performed, and the opening is sutured closed to prevent contamination by tumor spillage during manipulation of the specimen (Fig. 3-1). The actual temporal bone resection begins by performing a complete mastoidectomy with accurate identification of the tegmen mastoideum and sigmoid sinus. The course of the facial nerve is dissected by exposing it with careful drilling of the fallopian canal from the lateral semicircular canal to the stylomastoid foramen. An extended facial recess approach provides wide access to the middle ear space by following the fibrous annulus inferiorly as the anterior limit of the dissection. Inferiorly and superiorly, the objective is to bring the dissection into the soft tissues of the glenoid fossa. As the inferior dissection proceeds anteriorly, the surgeon must consider the position of the jugular bulb and carotid artery. Usually the level of the fibrous tympanic annulus is safely
lateral to the important vascular structures; however, preoperative CT imaging and intraoperative vigilance are necessary to avoid injury to those vessels. Superiorly, the zygomatic air cells are also dissected, and drilling proceeds superiorly from the antrum toward the zygomatic root, and through the epitympanum to the soft tissue of the temporomandibular joint. Superior to the EAC, the dissection proceeds lateral to the incus and malleus, and the surgeon takes care to avoid middle fossa dura by remaining close to the superior aspect of the EAC. When the superior and inferior approaches to the glenoid fossa have been completed, the incudostapedial joint is separated; the tensor tympani is cut, and the ligamentous attachments of the ossicles are divided. At this point, the anterior portion of the external canal is its only remaining attachment (Figs. 3-2 and 3-3). This bone is fractured free of the carotid with gentle pressure or tapping with an osteotome. If a chisel is used, it is important to angle the instrument lateral to the carotid so that inadvertent injury of the vessel can be avoided. When the specimen has been removed, and the middle ear has been fully exposed, the eustachian tube is obliterated by filling it with muscle or fascia. During the procedure, it is important to remain cognizant of the tumor, taking care to avoid it during the dissection. While drilling the epitympanum, it is important to remain lateral to the geniculate ganglion and medial to the annulus. An intact tympanic membrane should be preserved with the specimen for tumors limited to the EAC; this reduces the likelihood of tumor spillage. When involvement of the facial nerve is present, specimens of the proximal and distal margins should be examined histologically with sufficient tissue resection until the margins are free of tumor. If the nerve is sacrificed, an interposition graft connects the proximal and distal tumor-free portions of the nerve. The sural nerve and the greater auricular nerve serve as excellent donor sites. The greater auricular nerve is used more often because of its proximity to the surgical field, but if there is malignant lymphadenopathy in the neck, the sural nerve should be used instead. Care is necessary during graft orientation to position the graft so that regenerating fibers are not lost through side branching, orienting the graft with the distal end of the graft in apposition with the proximal segment. If the parotid gland is involved secondarily from extension of a temporal bone carcinoma or as the primary lesion that has extended to the temporal bone, the involved parotid tissues should be removed in continuity. In contrast to most decision making in surgical otology, the primary objective is an oncologically complete resection with clear margins; functional considerations such as hearing function and facial nerve function are secondary in priority. Tumor involvement of the facial nerve, deep lobe of parotid, or glenoid fossa necessitates a more aggressive resection. Gross tumor involvement of the glenoid is best managed by simultaneous resection of
Chapter 3 • Malignancies of the Temporal Bone—Limited Temporal Bone Resection
37
38
OTOLOGIC SURGERY
Chapter 3 • Malignancies of the Temporal Bone—Limited Temporal Bone Resection
the mandibular condyle and zygomatic root, which facilitates dissection of the intratemporal and parapharyngeal spaces, and prevents the violation of soft tissue attachments between the parotid, temporomandibular joint, and EAC. Consideration should also be given to management of the cervical lymphatics. Generally, primary temporal bone cancers rarely metastasize to the cervical lymphatics. Secondary involvement of the parotid by a primary temporal bone carcinoma and primary cancers of salivary gland origin have inherently different lymphatic drainage than carcinoma limited to the temporal bone, however, making dissection of the cervical lymph nodes prudent. Subtotal temporal bone resection is necessary if the middle ear or facial nerves are involved. Involvement of the middle ear space requires a subtotal temporal bone resection. The dissection extends medially into the otic capsule and petrous portions of the temporal bone to obtain negative margins. The medial extent of dissection is defined by the internal auditory canal. At this extent of tumor involvement, adherence to the ideal of an en bloc resection is affected by the three-dimensional anatomy of the temporal bone. The technical steps of a subtotal temporal bone resection depend on the exact location of tumor involvement, but may result in piecemeal removal of the most medial extent of tumor with frozen section control. It is most practical to begin with an en bloc partial temporal bone resection, and then adjust the resection medially. If resection of the jugular bulb or extensive carotid dissection is planned, control of the internal carotid artery and jugular vein is obtained inferiorly. Anteriorly, the vertical portion of the petrous internal carotid artery is identified by drilling away the cochlea and hypotympanum. When resection of the sigmoid sinus and jugular bulb is necessary, careful planning of handling these vascular structures limits unnecessary blood loss. The sigmoid must be occluded proximally, and the jugular vein must be ligated distally. If dural entry is planned, the sigmoid can be ligated with sutures by exposing the dura anterior and posterior to the sigmoid sinus. Alternatively, bone can be preserved from the midportion of the sigmoid and proximally toward the sigmoid-transverse junction. Absorbable knitted fabric (Surgicel) can be packed extraluminally to occlude the sinus. Brisk bleeding from the inferior petrosal sinus and condylar vein occurs with opening of the jugular bulb, and the surgeon should be prepared with additional Surgicel packing. Although the initial packing involves significant material, the surgeon can gradually remove most of the packing leaving only the small portions occluding the openings of the inferior petrosal sinus and condylar vein in the bulb, and limit the risks to the lower cranial nerves in the pars nervosa of the jugular bulb from excessive packing and pressure. An additional strategy to prevent bleeding from the jugular bulb is preoperative coil embolization of the openings of
39
the inferior petrosal sinus and condylar vein.16 The final resection margins proceed along the floor of the middle fossa connecting the glenoid fossa, internal auditory canal, and posterior-superior mastoid. When the cancer has progressed beyond the limits of a subtotal temporal bone resection, a total temporal bone resection may be indicated. This topic is addressed in more detail in Chapter 4. Total temporal bone resection is rarely performed because of the high level of morbidity and a lack of well-documented survival benefit except in specific circumstances, such as verrucous squamous carcinoma, in which the aggressiveness is entirely local and indolent. The surgery involves freeing the temporal bone at the petrous apex. Before extirpation, the intratemporal portion of the internal carotid artery must be dissected to the foramen lacerum, and the internal jugular vein must be ligated. Sacrifice or reconstruction of the internal carotid artery may be performed based on the results of preoperative testing. A subtemporal craniotomy is performed exposing the transverse sinus. The dura covering the cerebellum is incised, and the transverse sinus is ligated and divided near the superior petrosal sinus, being particularly careful to avoid injury to the vein of Labbé. The tentorium is divided sharply along the petrous bone in a plane superior and parallel to the superior petrosal sinus. The seventh and eighth cranial nerves are divided at the internal auditory canal. With the cerebellum retracted medially, an intradural incision is used to divide CN IX, X, and XI, and free the specimen at its posterior dural attachment. When the surrounding structures have been freed, a chisel is placed in the foramen ovale and directed posteriorly in a trajectory lateral to foramen lacerum; this frees the temporal bone at its apex.
RECONSTRUCTION Reconstructive options should be explored preoperatively, but as in all oncologic surgeries, the reconstructive plan must never compromise the completeness of oncologic resection. Consideration should be given to the probable need for postoperative radiation, which should begin within 6 weeks of the resection. Additional concerns include cosmesis and the durability of the repair. Intraoperative findings might dictate a more complex reconstruction than anticipated. Careful planning improves the preparation for various potential defects. We prefer to involve a head and neck reconstructive surgeon for this portion of the procedure to avoid reconstructive consideration influencing the adequacy of the oncologic resection. Partial temporal bone resections may be reconstructed with split-thickness skin grafts lining the mastoid and middle ear and sutured to the remaining soft tissue of the meatus. The graft should be placed to line the ear canal and mastoid bowl. Its medial extent can be
40
OTOLOGIC SURGERY
carefully draped over the oval window or stapes remnant. Reconstruction of the ossicular chain can be done later. This reconstruction may require frequent postoperative care, however, and delayed healing can affect the timing of postoperative radiation. Vascularized temporal-parietal fascia or postauricular soft tissue (Palva flap) can be used to provide a vascularized bed for skin graft healing. Alternatively, the cavity can be obliterated with temporalis muscle or another rotational flap, such as a pectoralis major or sternocleidomastoid flap; however, this most likely would preclude reconstruction of the conductive hearing mechanism (Figs. 3-4, 3-5, and 3-6). Subtotal temporal bone resection and total temporal bone resection require added soft tissue to obliterate dead space, prevent the leakage of cerebrospinal fluid, and provide protection against complications related to radiation therapy, the most serious of which is osteo radionecrosis. Smaller wounds involving resection of the pinna and external canal may be closed by a posterior scalp or cervicofacial advancement flap. If additional bulk is needed, a muscle or myocutaneous flap can be rotated on a vascularized pedicle, or inset via microvascular anastomosis. Smaller defects may be amenable to closure using a temporalis muscle rotational flap. A trapezius, latissimus, or pectoralis major muscle rotational flap may become necessary as the size of the defect increases. Extensive defects may require a rectus abdominis muscle or perforator-based adipose free flap for closure. The arterial and venous anastomosis should be sutured to the external carotid artery and jugular vein. In wounds requiring additional skin, a split-thickness skin graft can be placed over the muscular portion of the reconstructive flap. If dura is resected, the dural defect and eustachian tube must be obliterated. This obliteration is accomplished with abdominal fat or free tissue transfer, and serves to shield the central nervous system from environmental insult.
COMPLICATIONS Hemorrhage is a common risk in temporal bone resection. The otologic surgeon should be very familiar with the different techniques to control unplanned venous bleeding from the sigmoid sinus, superior petrosal sinus, and jugular bulb. With the sigmoid sinus, gentle pressure with absorbable gelatin sponge (Gelfoam) or Cottonoid is usually adequate. Great care must be exercised to not allow packing to dislodge within the lumen of the sigmoid sinus, unless the jugular vein has already been ligated in the neck. The jugular bulb does not respond to the same hemostatic strategies as the sigmoid sinus because the vessel wall is much thinner. Venous bleeding can be responsible for massive blood loss, and losses of 2500 mL can be encountered. As mentioned previously, excessive bleeding most commonly occurs at the inferior petrosal sinus, but this bleeding can be controlled with
proper technique. The excessive use of packing at the jugular foramen can result in injury to CN IX, X, and XI, so this should be done with caution.17 Injury to the carotid artery should be managed with direct pressure and placement of temporary clips proximal and distal to the insult. Systemic and intra-arterial heparin should be given. The injury should be repaired with 8-0 monofilament suture. If minor leaks persist after closure, topical hemostatic agents can be applied. If successful repair of the vessel cannot be achieved, the preoperative balloon occlusion test should be used to guide management. Vascular surgery and neurosurgical consultation are imperative. If there is significant carotid involvement, preoperative stent placement has been suggested to avoid vascular complications.14 Facial nerve sacrifice or excessive manipulation of the nerve can result in facial asymmetry and inadequate eye closure. During the immediate postoperative period, the eye should be protected with lubrication, use of a moisture chamber, or mechanical lid closure. If the nerve is sacrificed, an interposition graft using the ipsilateral greater auricular nerve or sural nerve should be performed. If the greater auricular nerve is unavailable, the sural nerve can be used instead. Successful nerve interposition can restore function to a House-Brackmann scale grade III, but may take 12 to 18 months to occur. Placement of a gold weight implant or spring, lateral tarsorrhaphy, or tensing of the lower eyelid may be necessary during the reinnervation period. Facial-hypoglossal anastomoses can also help restore facial function. When lower cranial nerve deficits exist, this should be done with caution. A loss in the function of CN XII can have devastating consequences on an already impaired swallowing apparatus. Tumor involving the jugular foramen may result in preoperative or postoperative paresis or palsy of CN IX and X. Such lesions can manifest in the airway and cause nutritional difficulties. Temporary or permanent tracheotomy and the placement of a gastrostomy tube may be necessary to compensate for the associated deficits. The combination of an insensate supraglottis, vocal fold motion impairment, and inadequate swallow reflex can result in aspiration with fatal consequences. Vocal fold medialization may restore safe deglutition. In more severe cases of swallowing or laryngeal dysfunction, a laryngotracheal separation may be necessary. If dura has been excised or violated, the resulting defect must be repaired immediately. Primary closure should be instituted whenever possible. If necessary, the dural defect should be obliterated with fat, fascia, or free tissue transfer. Postoperative leakage of cerebrospinal fluid can be managed conservatively for 7 to 10 days with appropriate measures. Bed rest, elevation of the head of the bed, stool softeners, and placement of a lumbar drain can be useful in reducing intracranial pressure.18 If the leak persists beyond this period, the wound should be explored surgically, and closure of the dural defect should be reattempted.
Chapter 3 • Malignancies of the Temporal Bone—Limited Temporal Bone Resection
RADIATION Plans for the use of postoperative radiation should be made during the preoperative work-up. The use of adjuvant radiation should be anticipated whenever the middle ear is involved with disease. If intraoperative findings confirm the need for radiation, it should be administered in a timely fashion. Radiation therapy should not be delayed, or withheld in lieu of a recurrence, because this can result in decreased therapeutic efficacy. Closure of the surgical defect with viable, durable tissue is necessary to prevent development of osteoradionecrosis, which can be a devastating and sometimes fatal complication. In addition, careful postoperative hygiene and maintenance (cleaning and débridement with otomicroscopy) of the reconstructed area is essential to reduce likelihood of chronic infection that can predispose to osteoradionecrosis. Dosages typically range from 5000 to 6000 rad, and fields should be designed to include the area of resection and involved nodal groups. The rarity of temporal bone malignancy makes the study of treatment protocols difficult, but encouraging results have been reported. T1 lesions successfully treated with surgery alone have a 95% 5-year survival, with no benefit from the addition of radiation. Some T2 and T3 lesions treated with radiation after complete surgical removal have shown an 85% 5-year survival. The 5-year survival of patients with more advanced lesions decreases to less than 50%. Cancer requiring a total temporal bone resection carries a dismal prognosis, with a 50% 1-year survival and reports of 0% 2-year survival.5,6,19
CHEMOTHERAPY The role of chemotherapy in the treatment of temporal bone malignancy has not been determined, but interest in its use is growing. Because of the low incidence and the histologic diversity of temporal bone malignancies, few meaningful studies have been conducted. Temporal bone malignancies of salivary gland origin, similar to salivary malignancies in other sites, do not usually respond to currently available chemotherapeutic agents. Squamous cell carcinoma of the temporal bone may respond to platinum-based agents, but typically the results are not durable, and other modalities must be used. For these reasons, most patients with temporal bone cancer are treated with surgery, radiation, or a combination of the two, and chemotherapy is usually reserved for palliation in patients with distant metastases or recurrences. Nakagawa and coworkers2 published data that suggest a possible role for preoperative chemoradiation therapy in the treatment of T3 and T4 squamous cell carcinoma when staged according to the Arriaga system. The radiation-enhancing effect of chemotherapy may be useful in reducing the tumor burden and improving
41
control of the surgical margin. Nakagawa and coworkers2 showed improved survival rates for T4 lesions not involving dura, pyramidal apex, carotid canal, or lymph nodes that are as good as survival rates in patients with T3 lesions. The poor outcomes associated with the treatment of advanced temporal bone malignancy mandate further investigation into the possible role of chemotherapy as a treatment modality.
REHABILITATION OF LOWER CRANIAL NERVE DEFICITS The details of surgical rehabilitation of lower cranial nerve deficits are discussed elsewhere. Generally, in limited temporal bone resection, the function of CN IX, X, XI, and XII can be preserved. Their sacrifice has significant quality-of-life and function implications for speech, voice, and swallowing. In addition to involvement of a head and neck surgeon with expertise in laryngeal restoration, thyroplasty and palatal rehabilitation, team members should include speech-language pathologists with experience in diagnosing and treating lower cranial nerve dysfunction. Strategies such as modified barium swallow and flexible endoscopic evaluation of swallowing are important components of a comprehensive assessment. The impact of radiation sequelae such as xerostomia on swallowing should not be underestimated.20 Nonetheless, postoperative radiation is a crucial part of the comprehensive management of a patient with temporal bone carcinoma.
CONCLUSION Successful management of temporal bone carcinoma depends on adequate pretreatment evaluation and staging. Most patients require limited temporal bone resection followed by postoperative radiation. In contrast to other otologic procedures, the management priorities in malignancy require that the otologic surgeon should ensure that the resection is oncologically sound, and consider hearing, vestibular, and facial function as secondary priorities.
REFERENCES 1. Kinney S E: Malignancies of the temporal bone: Limited temporal bone resection. In Brackmann D E, Shelton C, Arriaga M A, et al (eds): Otologic Surgery. Philadelphia, Saunders, 1994. 2. Nakagawa T, Kumamoto Y, Natori Y, et al: Squamous cell carcinoma of the external auditory canal and middle ear: An operation combined with preoperative chemoradiotherapy and a free surgical margin. Otol Neurotol 27:242-249, 2006.
42
OTOLOGIC SURGERY
3. Moffat D, De la Cruz A, Evans PHR: Management of tumours of the temporal bone. In Evans PH R (ed): Principles and Practice of Head and Neck Oncology. London���, Martin Dunitz, 2003. 4. Brackmann D E, Arriaga M A : Posterior fossa skull base neoplasms. In Cummings C, Fredrickson J, Krause C, Schuller D (eds): Cummings Otolaryngology–Head and Neck Surgery, 4th ed. St. Louis, Mosby, 2005. 5. Prasad S, Janecka I P: Malignancies of the temporal bone: Radical temporal bone resection. In Brackmann D E, Shelton C, Arriaga M A, et al (eds): Otologic Surgery. Philadelphia, Saunders, 1994. 6. Marsh M M, Jenkins H A : Temporal bone neoplasms and lateral cranial base surgery. In Cummings C, Fredrickson J, Krause C, Schuller D (eds): Cummings Otolaryngology– Head and Neck Surgery, 4th ed. St. Louis, Mosby, 2005. 7. Peters B R , Maddox H E, Batsakia JG: Pleomorphic adenoma of the middle ear and mastoid with posterior fossa extension. Arch Otolaryngol Head Neck Surg 114:676-678, 1988. 8. Prasad M, Kraus D: Acinic cell carcinoma of the parotid gland presenting as an external auditory canal mass. Head Neck 26:85-88, 2004. 9. Leonetti J P, Smith PG, Anand V, et al: Subtotal petrosectomy in the management of advanced parotid neoplasms. Otolaryngol Head Neck Surg 108:270-276, 1993. 10. Glenn L : Innovations in neuroimaging of skull base pathology. Otolaryngol Clin North Am 38:613-629, 2005. 11. Arriaga M , Curtin H, Takahashi H, et al: Staging proposal for external auditory meatus carcinoma based on preoperative clinical examination and computed tomography findings. Ann Otol Rhinol Laryngol 99:714-721, 1990.
12. Shintani S A, Foote R L , Lowe V J, et al: Utilizing of PET/ CT imaging performed early after surgical resection in the adjunct treatment planning from head and neck cancer. Int J Radiat Oncol Biol Phys 70(2):322-329, 2008. 13. Janecka I P, Sekhar L N, Horton J A : General blood flow evaluation. In Cummings CW, Fredrickson J, Krause C, Schuller D (eds): Otolaryngology–Head and Neck Surgery Update II. St. Louis, Mosby–Year Book, 1990, pp 54-63. 14. Sanna M, Khrais T, Menozi R , et al: Surgical removal of jugular paraganglioma after stenting of the intratemporal internal carotid artery: A preliminary report. Laryngoscope 116:742-746, 2006. 15. Moody S A, Hirsch B E, Myers E N: Squamous cell carcinoma of the external auditory canal: An evaluation of a staging system. Am J Otol 21:582-588, 2000. 16. Carrier D, Arriaga M A, Dahlen R : Preoperative embolization of anastomosis of the jugular bulb: A new adjuvant in jugular foramen surgery. AJNR Am J Neuroradiol 18:1252-1256, 1997. 17. Brackmann D E, Arriaga M A: Surgery for glomus and jugular foramen tumors. In Brackmann D E, Shelton C, Arriaga M A, et al (eds): Otologic Surgery. Philadelphia, Saunders, 1994. 18. Nuss DW, Constantino PD: Cerebrospinal fluid leaks: Comprehensive diagnosis and management. Otolaryngol Head Neck Surg 108:539, 1993. 19. Pensak M, Willging JP: Tumors of the temporal bone. In Jackler R (ed): Neurotology. St. Louis, Mosby, 1994. 20. Koukourakis M I, Danieldis V: Preventing radiation induced xerostomia. Cancer Treat Rev 31:546-554, 2005.
4
Malignancies of the Temporal Bone—Radical Temporal Bone Resection Sanjay Prasad and Ivo P. Janecka
Primary malignancies of the temporal bone were first recognized in the late 18th century and histologically first confirmed in the 1850s. These lesions are uncommon, with only 250 cases having been reported in the English literature by 1974.1 The overall prevalence in the general population is 6 cases per 1 million.2 Because of their infrequent occurrence, these tumors are often misdiagnosed and treated as chronic external otitis or chronic mastoiditis. They are usually discovered at a later stage, when more radical treatment is required. The infrequent occurrence of the disease poses a challenging obstacle to any attempt at a clinical study regarding treatment. Secondary malignancies of the temporal bone from regional spread of parotid cancers occur far more commonly. The fissures of Santorini provide a conduit for regional spread through the anterior cartilaginous ear canal. Basal cell carcinoma and squamous cell carcinoma are the more common malignancies to affect the temporal bone. Basal cell carcinoma is thought to occur secondary to actinic exposure, and commonly involves the external ear or ear canal or both. Squamous cell carcinoma can arise primarily from the external canal or middle ear, or both, or spread into the temporal bone from a primary lesion in the parotid gland. In contrast to squamous cell carcinoma of the upper aerodigestive tract, squamous cell carcinoma of the temporal bone is not related to tobacco or alcohol use. Predisposing factors to these lesions are few. Chronic infection within the temporal bone is the most commonly cited factor. Adenoid cystic carcinoma can arise either from the ear canal and middle ear or from the parotid gland and spread secondarily into the temporal bone. These lesions have a tendency toward perineural spread. Ceruminous adenoma, adenocarcinoma, and mucoepidermoid carcinoma are other lesions that can affect the temporal bone. Regional spread to cervical nodes and distant metastases are uncommon. Depending on the extent of the disease, radical resection coupled with radiation treatment offers the best treatment. The efficacy of chemotherapy has not been clearly established, and it may play a role only in recalcitrant disease.
Radical temporal bone resection refers to one of three operations that can be offered to patients with this disease. A lateral temporal bone resection (LTBR) refers to the removal of the external auditory canal (EAC), tympanic membrane, malleus, and incus. Subtotal temporal bone resection (STBR) refers to the additional removal of the otic capsule, and total temporal bone resection (TTBR) refers to the additional removal of the petrous apex with or without the carotid artery. LTBR is well accepted for lesions that involve the EAC or tympanic membrane or both. Controversy arises in defining the optimal management of neoplasms that invade the mesotympanum. Some authors advocate LTBR with gross removal of middle ear disease followed by radiation therapy, whereas others prefer more radical surgery (STBR or TTBR), followed by radiation therapy. When the tumor has invaded the petrous apex, involvement of dura mater, brain parenchyma, or internal carotid artery (ICA) is usually present. This chapter focuses on preoperative diagnostic evaluation, the surgical techniques of STBR and TTBR, postoperative management, and potential complications. Rehabilitation and adjuvant treatment for recalcitrant disease are briefly discussed. Finally, we present a literature review of squamous cell carcinoma of the temporal bone in an effort to define the optimal management for middle ear disease and discuss the prognostic significance of dural, brain, and ICA involvement.
DIAGNOSTIC EVALUATION The diagnostic evaluation begins with a thorough history and physical examination, with special emphasis on the chronology of developing cranial neuropathies. The pathway of tumor spread occasionally can be deduced from a careful history. Facial nerve function, hearing, and balance function should be carefully documented. Examination includes palpation of the parotid gland and cervical lymph glands for the presence of local spread and regional metastases. Patients should be questioned and tested for temporal lobe signs (e.g., memory loss, 43
44
OTOLOGIC SURGERY
ysphasia, left-sided neglect, hemiparesis, and olfactory d hallucinations) and cerebellar signs (e.g., ocular dysmetria, truncal ataxia, nystagmus, and dysdiadochokinesia). Imaging allows determination of the extent of tumor involvement. High-resolution axial and coronal computed tomography (CT) imaging at 1.5 mm thickness can identify areas of bony involvement. Enhanced and unenhanced magnetic resonance imaging (MRI) can determine intracranial involvement. Histologic confirmation of the lesion is essential in further treatment planning. Biopsy specimens of lesions involving the EAC or periauricular skin can be easily obtained. Needle aspiration biopsy of parotid lesions can also be performed. Angiography is used when involvement of the major vessels is suspected on preoperative imaging, or when surgical exposure of the petrous carotid artery is anticipated. The venous phase of the study can provide important information regarding blood flow through the dural venous sinuses. Embolization of feeding vessels is rarely required because most lesions are relatively avascular. Cerebral blood flow evaluation is indicated when involvement of the ICA is present. Patency of the anterior and posterior communicating arteries on angiography is an inadequate evaluation for collateral flow. Our current method of preoperative carotid artery testing is described.3 A 30-minute temporary balloon occlusion of the ICA allows identification of patients who would most likely tolerate carotid artery sacrifice. Transfemoral introduction of a nondetachable intravascular balloon, inflated in the ICA, is performed in the patient while sensory, motor, and higher cortical functions are assessed. Patients who develop a neurologic deficit during temporary occlusion are at high risk for stroke after carotid sacrifice. Preoperative or intraoperative extracranialto-intracranial arterial bypass should be considered. Repeating the temporary balloon occlusion before surgical extirpation should also be considered. Conservative treatment options should also be discussed with these patients. Patients who tolerate a 30-minute balloon occlusion of the ICA are at low risk for development of a stroke, provided that a long “distal” stump is avoided. Permanent ICA occlusion can be performed angiographically. Hypotension and hypovolemia in the perioperative period should be avoided if permanent ICA occlusion is performed.
PREOPERATIVE PREPARATION Preoperative preparation sets the stage for the operative and postoperative course. The evening before surgery, the operative site is shampooed and scrubbed with hexachlorophene. Intravenous phenytoin or phenobarbital and cefuroxime are used for prophylactic anticonvulsant and antibiotic coverage. To facilitate intraoperative cranial
nerve monitoring, short-acting neuromuscular blocking agents are used only for the induction of anesthesia, and not during the operation. On the morning of the operation, sequential compression stockings are placed on both legs to help decrease the incidence of thromboembolic disease. After insertion of a central venous catheter and an arterial line, the patient is intubated, and the endotracheal tube is secured. The operating table is turned 90 degrees from the anesthesiologist, giving him or her access to the contralateral arm. The head is positioned on a horseshoe (Mayfield) head holder to allow repositioning during the course of the operation. Temporary bilateral tarsorrhaphies are placed to prevent corneal abrasions. The operative site, which includes the temporal fossa, lateral half of the face, postauricular area, neck, and ipsilateral thigh and lower leg (for the potential use of tensor fascia lata and sural nerve), is shaved and scrubbed with an iodine-based solution. Bipolar facial electromyographic electrodes are placed in areas where facial function exists.
SURGICAL PROCEDURE Incisions vary according to the extent of the tumor (Fig. 4-1). For lesions contained within the temporal bone, a C-shaped incision extending from the temporal fossa postauricularly into the neck is used. A blind-sac closure of the EAC helps contain the specimen. When tumor invasion of the conchal cartilage or periauricular skin is suspected, an appropriate skin island is incorporated into the overall design. The EAC skin is sutured shut to avoid tumor spillage. The outline of the incisions should preserve the blood supply to the remaining auricle. The anterior and posterior skin flaps are elevated (Fig. 4-2A). The superficial temporal fat pad is elevated with the anterior skin flap in a subperiosteal plane over the zygomatic arch. The superficial temporal and middle temporal arteries are ligated. The facial nerve can be handled differently, depending on tumor invasion of the parotid gland. When the gland is involved, peripheral branches of the facial nerve are identified with the help of the facial nerve monitor and then divided. The stumps of the anterior segments are secured to the anterior skin flap. The entire parotid gland is dissected off the masseteric fascia, provided that the latter is free of disease, and mobilized posteriorly, while the attachment to the EAC is maintained. When the parotid gland is suspected to be free of tumor, the facial nerve trunk is located in the usual manner at the tympanomastoid suture and divided. The parotid gland, along with the distal stump of the facial nerve, is dissected free of the EAC and mobilized anteriorly off the masseteric fascia. The jugulodigastric region is explored, and cervical lymph nodes are sent for frozen section pathologic analysis. Regional metastases determine the need for a formal
Chapter 4 • Malignancies of the Temporal Bone—Radical Temporal Bone Resection
45
FIGURE 4-1. Incisions vary according to whether the tumor is contained in the temporal bone. FIGURE 4-2. A, The facial nerve can be divided peripherally at the distal branches or centrally at the facial nerve trunk, depending on involvement of the parotid gland. B, After osteotomies and removal of the zygomatic arch and mandibular segments, dissection in the infratemporal fossa continues. EAC, external auditory canal.
46
OTOLOGIC SURGERY
cervical lymphadenectomy. The ninth cranial nerve, the greater auricular nerve, or cervical cutaneous nerves can be used as cable grafts for facial nerve reconstruction. CN IX, X, XI, and XII; the internal jugular vein; and the external carotid artery and ICA are dissected in an inferosuperior direction toward the temporal bone. The sternocleidomastoid and digastric muscles are detached from their attachment to the mastoid. The masseter is detached from the zygomatic arch, allowing exposure of the zygoma and mandible. Zygomatic and mandibular osteotomies (see Fig. 4-2A) can then be performed. The meniscus of the temporomandibular joint is separated from the glenoid fossa, and the chorda tympani nerve emerging from the petrotympanic fissure is divided. The stylomandibular and sphenomandibular ligaments are divided and allow removal of the mandibular segment (Fig. 4-2B). The temporalis muscle is elevated in a subperiosteal fashion and reflected inferiorly. The temporalis muscle must be separated from the lateral pterygoid muscle, and care should be taken not to injure the deep temporal arteries supplying blood to the temporalis muscle. The lateral and medial pterygoid muscles are resected either en bloc with the specimen or separately, depending on tumor invasion. In a subperiosteal manner, the contents of the infratemporal fossa are elevated off the floor of the middle fossa to expose the middle meningeal artery and vein in the foramen spinosum and the mandibular division of the trigeminal nerve in the foramen ovale. The contents of the foramen spinosum are bipolarly coagulated and divided. Frequently, the venous plexus of the foramen ovale requires bipolar coagulation and packing with oxidized cellulose. The lesser petrosal nerve can be seen emerging from the innominate canal, on its way to the otic ganglion. The stylohyoid, stylopharyngeus, and styloglossus muscles (Riolan’s bouquet) are detached from the styloid process, which is then rongeured away. The branches of the external carotid artery are dissected in the infratemporal fossa. The anterior tympanic and deep auricular branches of the internal maxillary artery are often divided before identification, and may require bipolar coagulation. The internal maxillary artery is preserved up to the branches of the deep temporal artery. When the internal maxillary artery must be sacrificed, brisk backflow from the anterior stump indicates that the temporalis muscle may derive its blood supply from reversed flow via the pterygoid system. If brisk backflow is not observed, the temporalis muscle cannot be relied on to reconstruct the surgical defect, and microvascular free flap options must be considered. The cartilaginous eustachian tube is divided, and the anterior end is sutured closed to prevent postoperative cerebrospinal fluid rhinorrhea (Fig. 4-3A). The ICA is dissected toward the carotid canal, and care is taken not to injure CN IX, which crosses its anterior surface. Kerrison rongeurs are used to uncover the vertical and horizontal petrous segments of the carotid
artery (Fig. 4-3B). Occasionally, bleeding from the pericarotid venous plexus requires bipolar coagulation. The caroticotympanic artery is also divided when the petrous carotid artery is separated from the specimen. The extent of petrous carotid mobilization depends on whether STBR or TTBR is performed. When STBR is performed, the vertical petrous carotid artery is mobilized from the carotid foramen and canal. When TTBR is performed, the vertical and horizontal petrous carotid artery is mobilized out of the carotid canal to the foramen ovale. A temporal craniectomy is performed, and the intracranial portion of the middle meningeal vessels is coagulated (see Fig. 4-3B). The patient is hyperventilated to keep the Pco2 at 25 mm Hg for adequate brain relaxation. Mannitol and furosemide can improve brain relaxation. Subtemporal dural elevation proceeds in a posteroanterior direction. The greater superficial petrosal nerve and accompanying petrosal artery are coagulated and divided to lessen traction on the geniculate ganglion. The lesser petrosal nerve and superior tympanic artery are similarly divided. Subtemporal dural elevation proceeds as far medially as possible to expose the superior petrosal sinus. When carcinomatous involvement of the middle fossa dura is suspected, an intradural approach keeps the involved dura attached to the specimen. The extent of the mastoidectomy depends on whether tumor is present in the mastoid air cells. Care is taken to avoid exposure of tumor in the mastoid. In this case, the confluence of the transverse, sigmoid, and superior petrosal sinuses is decorticated to expose posterior fossa dura on either side of the sigmoid sinus. When posterior fossa dural involvement is suspected, a presigmoid intradural approach with gentle retraction of the cerebellum (Fig. 4-4A) allows exposure of the vessels and nerves in the cerebellopontine angle that are keeping the involved dura attached to the specimen. When the mastoid air cells are thought to be free of tumor, a translabyrinthine approach to the IAC is used (Fig. 4-4B). The anterior inferior cerebellar artery is retracted after the labyrinthine artery is bipolarly coagulated and divided. The superior and inferior vestibular nerves, cochlear nerve, facial nerve, and nervus intermedius (nerve of Wrisberg) are divided. A segment of the proximal facial nerve stump can be sent for frozen section pathologic examination if it is suspicious for carcinoma. The dome of the jugular bulb must be separated from the specimen when the dural venous sinuses are spared. The surgical technique from here varies according to whether the dural venous sinuses or the ICA is preserved. When both are preserved, and STBR is being performed, the bone between the bony canal of the carotid artery at the junction of the vertical and horizontal segments (Fig. 4-5A) and the fundus of the IAC is removed with a highpowered drill. Separation of the petro-occipital synchondrosis sometimes requires insertion of an osteotome just above the jugular bulb. This step releases the specimen en bloc from attachment to the clivus. Packing oxidized
Chapter 4 • Malignancies of the Temporal Bone—Radical Temporal Bone Resection
47
FIGURE 4-3. A, Further dissection in the infratemporal fossa allows division and ligation of the eustachian tube and exposure of the petrous carotid artery. B, The petrous carotid artery is dissected further according to whether a subtotal or total temporal bone resection is performed. ICA, internal carotid artery; STBR, subtotal temporal bone resection.
48
OTOLOGIC SURGERY
FIGURE 4-4. A ( left and right), When tumor is thought to be present in the mastoid, a presigmoid intradural approach (enlarged view on the right) to the porus acusticus is used. B, When the mastoid is thought to be free of tumor, a translabyrinthine approach allows exposure of the internal auditory canal. AICA, anterior inferior cerebellar artery; ICA, internal carotid artery.
cellulose intraluminally toward the cavernous sinus controls bleeding from the inferior petrosal sinus. When both vascular structures are preserved, and TTBR is being performed, the bone between the posterior edge of the foramen ovale and spheno-occipital synchondrosis is removed (Fig. 4-5B). The petro-occipital
synchondrosis may require an additional osteotomy before the specimen is delivered. When the dural venous sinuses are sacrificed, the internal jugular vein is double ligated in the upper cervical area and mobilized toward the jugular bulb (Fig. 4-6). The superior petrosal sinus and sigmoid sinus are divided
Chapter 4 • Malignancies of the Temporal Bone—Radical Temporal Bone Resection
49
Drill aimed at fundus of IAC
Drill inserted at bony canal of internal carotid artery
Internal carotid artery mobilized
Drill inserted at posterior edge of foramen ovale
A FIGURE 4-5. Sigmoid sinus
B
FIGURE 4-6. FIGURE 4-5. A, A drill, inserted at the junction of the vertical and horizontal petrous carotid canal, is aimed toward the fundus of the internal auditory canal (IAC) to allow removal of the subtotal temporal bone specimen. B, A drill, inserted at the horizontal petrous carotid canal just posterior to the foramen ovale, is directed slightly posteriorly to avoid entry into the foramen lacerum. This releases the total temporal bone specimen.
FIGURE 4-6. When the dural venous sinus is sacrificed, the internal jugular vein is mobilized toward the cranial base as the sigmoid sinus is intraluminally packed.
50
OTOLOGIC SURGERY
and intraluminally packed with oxidized cellulose. Care is taken not to overpack the sigmoid sinus proximally to avoid obstruction of the vein of Labbé; this can cause hemorrhagic necrosis of the temporal lobe. Similarly, excessive packing of the inferior petrosal sinus can lead to cavernous sinus thrombosis. The sigmoid sinus is opened toward the jugular bulb and resected. Intraluminal rather than extraluminal packing of the inferior petrosal sinus is preferred to avoid injury to cranial nerves traversing the jugular foramen. The results of the temporary balloon occlusion determine whether the ICA can be safely resected. When the patient is at high risk for stroke after sacrifice of the ICA, the patient requires revascularization to ensure ade quate cerebral blood flow. An extracranial-to-intracranial arterial bypass, such as a superficial temporal artery– to–middle cerebral artery bypass, must be considered. The middle and posterior fossa dura is inspected for carcinomatous involvement. Any involved areas are resected, and margins are sent for frozen section pathologic examination. The temporal lobe and cerebellum are also examined, and limited involvement of the inferior temporal gyrus and cerebellum can be resected. When intradural hemostasis is achieved, the dura is closed in a watertight manner using pericranium, fascia lata, or cadaveric dura. Reconstruction includes restoration of facial nerve continuity and closing the dead space after tumor resection. The greater auricular nerve, cervical cutaneous nerve, or CN XI can be used as a cable graft from the facial nerve in the IAC to the peripheral branches. When greater length is required, the sural nerve can be harvested from the lower leg. Use of vascularized tissue, rather than autogenous fat, is preferred to fill in the surgical defect. The temporalis muscle can be rotated into the defect. Split-thickness skin grafts can be used for cutaneous defects. A small bolus dressing is placed to ensure adequate adhesion. When the temporalis is unavailable, plastic surgery teams can harvest microvascular free flaps. Jackson-Pratt drains are installed under the neck and scalp skin flaps to ensure tissue coaptation.
POSTOPERATIVE CARE After extubation in the operating room, the patient is examined for neurologic deficits. Unanticipated neurologic deficits are studied further by noncontrast CT imaging to evaluate the presence of intracranial edema or bleeding. Antibiotics are continued postoperatively until the drains are removed, or if an infectious process dictates otherwise. Dexamethasone is continued for 48 hours and then tapered over 5 days. When the ICA has been sacrificed, great care is taken to avoid even minor hypotensive and hypovolemic periods. These events can reduce the cerebral blood flow
through the remaining contralateral carotid artery and lead to cerebral ischemia. Intermittent spinal drainage, through a lumbar drain installed at the end of the procedure, reduces the pressure in the subarachnoid space and accelerates healing of dural defects. Continuous spinal drainage can cause overdrainage and pneumocephalus and is not used. Depending on patient tolerance, 35 to 50 mL of spinal fluid is drained every 8 hours for 48 hours. The drain is then clamped for 24 hours and removed.
POTENTIAL COMPLICATIONS Inadvertent carotid artery injury can be managed initially by local pressure to the site of entry and placement of temporary clips on either side. After systemic heparinization, the tear is examined, irrigated with heparinized saline, and repaired using 8-0 Novofil suture. Before placement of the last suture, the lumen is irrigated with heparinized saline to clear any clots. The hemoclips are released, and any minor leaks are managed with oxidized cellulose. If the tear is beyond repair, the results of the preoperative temporary balloon occlusion dictate further management.
ADJUVANT TREATMENT Treatment of temporal bone malignancies extending into the middle ear should consist of surgical resection of all visible disease followed by external-beam radiotherapy. Occasionally, large, aggressive tumors invading brain parenchyma, the dominant carotid artery, or dural venous sinus are encountered, and total tumor removal is impossible without serious neurologic deficits. In these instances, brachytherapy catheters can be inserted and used.
REHABILITATION Facial palsy after temporal bone resection requires careful assessment and treatment. A temporary tarsorrhaphy at the conclusion of the operation affords early corneal protection. When the periorbital edema subsides, a gold weight implant in the upper eyelid can offer long-term corneal protection. Elderly patients also may require lower eyelid tightening procedures. Facial nerve recovery can be expected at 12 to 18 months. The best facial function a patient can expect with a cable graft, in our hands, is House-Brackmann grade III. Excessive packing of the inferior petrosal sinus can lead to lower cranial nerve dysfunction (CN IX, X, and XI) that can be debilitating. Some patients require temporary tracheostomy and gastrostomy, and others can be treated with polytetrafluoroethylene (Teflon) injection of the vocal fold or an Isshiki thyroplasty.4
Chapter 4 • Malignancies of the Temporal Bone—Radical Temporal Bone Resection
51
FOLLOW-UP Patients are seen every month for the first year, and repeat imaging (CT and MRI) is obtained at 6 months and yearly thereafter. Enhancing tumors are best visualized by contrast-enhanced fat-suppression MRI, which helps differentiate transposed flaps from tumor recurrence. Any suspicious areas require biopsies either directly or by needle aspiration techniques with or without CT guidance.
RESULTS Temporal bone neoplasms occur infrequently. The limited experience makes analysis of treatment difficult. Because of this difficulty, we reviewed the literature for all cases of squamous cell carcinoma of the temporal bone,5 analyzed the extent of disease present, studied the different treatment strategies employed, and reported on outcome. Our findings are presented. All publications in the English language from 1915 to 1992 dealing with the treatment of squamous cell carcinoma of the temporal bone were reviewed; 96 publi cations1,6-100 were encountered, of which 26 articles75-100 contained enough information on 144 patients. Various parameters were then analyzed. The major reason for exclusion of a study was the lack of a descriptive table in which the extent of disease, type of treatment, and follow-up for each patient were carefully documented. Several conclusions about overall survival could be made. When disease was confined to the EAC, no statistically significant difference in 5-year survival was found between patients treated with LTBR (48.6%) and patients treated with STBR (50%). When disease extended into the middle ear, patients who had STBR had better 5-year survival (41.7%) than patients who had LTBR (28.6%) (Fig. 4-7). The experience with carcinoma that invaded the petrous apex was limited. Four patients treated with TTBR had a 50% 1-year survival and 0% 2-year survival. One patient treated with STBR was dead of disease at 1 year. The value of preoperative or postoperative radiation therapy was also analyzed. When disease was confined to the EAC, the addition of either preoperative or postoperative radiation therapy to LTBR did not significantly improve 5-year survival (48% with radiation therapy, 44.4% without radiation therapy) (Fig. 4-8). This was the only group in which a conclusion regarding radiation treatment could be made. The prognostic value of dural involvement was also studied. Resection of involved dura mater did not improve overall 5-year survival (11.1% with or without resection). Margin status was not always reported. Four patients had extension of disease to involve the ICA. Of the two patients treated with TTBR and ICA sacrifice, one died of postoperative cerebral ischemia, and the other died of disease at 14 months of regional and
FIGURE 4-7. Treatment-specific survival for patients with carcinoma extending to the middle ear. TBR, temporal bone resection. (From Prasad S, Janecka IP: Efficacy of surgical treatments for squamous cell carcinoma of the temporal bone. Otol Head Neck Surg 110:270-280, 1994.)
FIGURE 4-8. Survival of patients with carcinoma confined to the external auditory canal treated with lateral temporal bone resection (TBR) with or without preoperative or postoperative radiation therapy (RT). (From Prasad S, Janecka IP: Efficacy of surgical treatments for squamous cell carcinoma of the temporal bone. Otol Head Neck Surg 110:270-280, 1994.)
distant failure. The other two patients who were treated in a method that spared the ICA died of disease shortly after resection. Two patients with carcinomatous invasion of the temporal lobe treated with limited resection also died of disease shortly after resection. No patient was encountered who had resection of involved cerebral or cerebellar tissue. Site of failure was also studied. Of 54 patients who died of disease, 45 had local failure, 5 had locoregional failure, 3 had regional failure alone, and 1 had regional distant failure. Several other aspects of this disease could not be studied because of the lack of information provided by the authors. First, the histologic differentiation of the tumor and its relationship to overall survival could not be analyzed. Second, the method of temporal bone removal,
52
OTOLOGIC SURGERY
whether by en bloc resection, piecemeal resection, or a drillout, and its relationship to survival could not be ascertained. The status of the margins of resection and relationship to overall survival remains to be studied.
SUMMARY Radical resection of the temporal bone requires thorough knowledge of the intricate anatomy of the temporal bone and surrounding structures. There is no substitute for laboratory dissection before embarking on such an operation. Thorough preoperative imaging, delineation of the extent of tumor involvement, and preoperative carotid artery testing are imperative. The indications for the operation are slowly evolving as we gain experience and collate data for analysis. From our literature review regarding squamous cell carcinoma,5 it seems that cancerous involvement of the middle ear is best treated by STBR rather than LTBR and gross removal of middle ear disease. When the tumor involves the petrous apex, we believe that TTBR can allow total tumor extirpation and possibly prolonged survival, although the latter remains to be proved. Prospective, randomized studies are needed to define the value of margin-free dural, carotid artery, and brain parenchymal resection. The value of adjunctive radiation therapy for extensive lesions also requires further study.
REFERENCES 1. Johns M E, Headington JT: Squamous cell carcinoma of the external auditory canal: A clinicopathologic study of 20 cases. Arch Otolaryngol Head Neck Surg 100:45-49, 1974. 2. Kinney S E : Tumors of the external auditory canal, middle ear, mastoid, and temporal bone. In Thawley S E, Panje WR , Batsakis JG, Lindberg R D, et al (eds): Comprehensive Management of Head and Neck Tumors. Philadelphia, Saunders, 1987, p 182. 3. Janecka I P, Sekhar L N, Horton J A, Yonas H, et al: Cerebral blood flow evaluation. In Cummings CW, Frederickson J M, Harlor L A, et al (eds): Otolaryngology Head and Neck Surgery: Update II. St Louis, Mosby– Year Book, 1990, pp 54-63. 4. Sasaki CT, Leder S B, Petcu L , Freidman C D: Longitudinal voice quality changes following Isshiki thyroplasty type I: The Yale experience. Laryngoscope 100:849852, 1990. 5. Prasad S, Janecka I P: Efficacy of surgical treatments for squamous cell carcinoma of the temporal bone. Otol Head Neck Surg 110:270-280, 1994. 6. Campbell E, Volk B M, Burkland CW: Total resection of the temporal bone for malignancy of the middle ear. Ann Surg 134:397-404, 1951. 7. Coleman CC, Khuri A : A rational treatment for advanced cancer of the external ear and temporal bone. VA Med Monthly 86:21-24, 1959.
8. Hutcheon J R : Experiences with aural carcinoma over the past six years. Med J Aust 2:406-407, 1966. 9. Figi FA, Weisman PA : Cancer and chemodectoma in the middle ear and mastoid. JAMA 156:1157-1162, 1954. 10. Wang CC : Radiation therapy in the management of carcinoma of the external auditory canal, middle ear, or mastoid. Ther Radiat 116:713-715, 1975. 11. Sinha PP, Aziz H I : Treatment of carcinoma of the middle ear. Ther Radiat 126:485-487, 1978. 12. Boland J: The management of carcinomas of the middle ear. Radiology 80:285, 1963. 13. Holmes K S : The treatment of carcinoma of the middle ear by the 4-MV linear accelerator. Proc R Soc Med 53:242-244, 1960. 14. Yamada S, Schuh FD, Harvin J S, Perot PL : En bloc subtotal temporal bone resection for cancer of the external ear. J Neurosurg 39:370-379, 1973. 15. Hahn S S, Kim J A, Goodchild N, Constable WC : Carcinoma of the middle ear and external auditory canal. Int J Radiat Oncol Biol Phys 9:1003-1007, 1983. 16. Sorenson H : Cancer of the middle ear and mastoid. Acta Radiol 54:460-468, 1960. 17. Frazer JS: Malignant disease of the external acoustic meatus and middle ear. Proc R Soc Med 23:1235-1244, 1930. 18. Barnes E B : Carcinoma of the ear. Proc R Soc Med 23:1231-1234, 1930. 19. Conley J S : Cancer of the middle ear and temporal bone. N Y State J Med 74:1575-1579, 1974. 20. Kinney S E, Wood BG: Malignancies of the external ear canal and temporal bone: Surgical techniques and results. Laryngoscope 97:158-164, 1987. 21. Kinney S E : Squamous cell carcinoma of the external auditory canal. Am J Otol 10:111-116, 1989. 22. Graham M D, Sataloff RT, Kemink J, McGillicuddy J F: En bloc resection of the temporal bone and carotid artery for malignant tumors of the ear and temporal bone. Laryngoscope 94:528-533, 1984. 23. Clark L J, Narola A A, Morgan D A L , Bradley PJ: Squamous carcinoma of the temporal bone: A revised staging. J Laryngol Otol 105:346-348, 1991. 24. Corey J P, Nelson E, Crawford M, et al: Metastatic vaginal carcinoma to the temporal bone. Am J Otol 12:128131, 1991. 25. Hiraide F, Inonye T, Ishii T: Primary squamous cell carcinoma of the middle ear invading the cochlea: A histopathologic case report. Ann Otol Rhinol Laryngol 92:290-294, 1983. 26. Schusterman M A, Kroll S S : Reconstruction strategy for temporal bone and lateral facial defects. Ann Plastic Surg 26:233-294, 1983. 27. Haughey BH, Gates GA, Skerhut HE, Brown WE: Cerebral shift after lateral craniofacial resection and flap reconstruction. Otolaryngol Head Neck Surg 101:79-86, 1989. 28. Bergetedt H F, Lind MG: Temporal bone scintigraphy. Acta Otolaryngol (Stockh) 89:465-473, 1980. 29. Jahn A F, Farkashidy J, Berman J M : Metastatic tumors in the temporal bone—a pathophysiologic study. J Otolaryngol 8:85-95, 1979. 30. Ruben R J, Thaler SU, Holzer N: Radiation-induced carcinoma of the temporal bone. Laryngoscope 87:16131621, 1977.
Chapter 4 • Malignancies of the Temporal Bone—Radical Temporal Bone Resection 31. Katsarkas A, Seemayer TA : Bilateral temporal bone metastases of a uterine cervix carcinoma. J Otolaryngol 5:315-318, 1976. 32. Ramsden RT, Bulman C H, Lorigan B P: Osteoradionecrosis of the temporal bone. J Laryngol Otol 89:941-955, 1975. 33. Vize G: Laryngeal metastasis to the temporal bone causing facial paralysis. J Laryngol Otol 88:175-177, 1974. 34. Schuknecht H F, Allam A F, Murakami Y: Pathology of secondary malignant tumors of the temporal bone. Ann Otol Rhinol Laryngol 77:5-22, 1968. 35. Lewis J S : Temporal bone resection: Review of 100 cases. Arch Otolaryngol Head Neck Surg 101:23-25, 1975. 36. Arriaga M, Curtin H, Takahashi H, et al: Staging proposal for external auditory meatus carcinoma based on preoperative clinical examination and computed tomography findings. Ann Otol Rhinol Laryngol 99:714-721, 1990. 37. Adams WS, Morrison R : On primary carcinoma of the middle ear and mastoid. J Laryngol Otol 69:115-131, 1955. 38. Gacek R R : Management of temporal bone carcinoma. Trans Penn Acad Otolaryngol Ophthalmol 32:67-71, 1978. 39. Conley J, Schuller D E : Malignancies of the ear. Laryngoscope 86:1147-1163, 1976. 40. Greer J A, Body DTR , Weiland L H : Neoplasms of the temporal bone. J Otolaryngol 5:391-398, 1978. 41. Goodman M L : Middle ear and mastoid neoplasms. Ann Otol Rhinol Laryngol 80:419-424, 1971. 42. Arena S, Keen M : Carcinoma of the middle ear and temporal bone. Am J Otol 9:351-356, 1988. 43. Hilding D A, Selker R : Total resection of the temporal bone for carcinoma. Arch Otolaryngol Head Neck Surg 89:98-107, 1969. 44. Cundy R L , Sando I, Hemenway WG: Middle ear extension of nasopharyngeal carcinoma via the eustachian tube. Arch Otolaryngol Head Neck Surg 98:131-133, 1973. 45. Lewis J S : Squamous carcinoma of the ear. Arch Otolaryngol Head Neck Surg 97:41-42, 1973. 46. Sekhar L N, Pomeranz S, Janecka I P, et al: Temporal bone neoplasms: A report on 20 surgically treated cases. J Neurosurg 76:578-587, 1992. 47. Lessor RW, Spector GJ, Divinens VR : Malignant tumors of the middle ear and external auditory canal: A 20-year review. Otolaryngol Head Neck Surg 96:4347, 1987. 48. Kenyon G S, Marks PV, Scholtz C L , Dhillon R : Squamous cell carcinoma of the middle ear: A 25-year retrospective study. Ann Otol Rhinol Laryngol 94:273-277, 1985. 49. Arthur K : Radiotherapy in carcinoma of the middle ear and auditory canal. J Laryngol Otol 90:753-762, 1976. 50. Tucker WN: Cancer of the middle ear. Cancer 16:642650, 1965. 51. Conley JJ, Novack A J: The surgical treatment of malignant tumors of the ear and temporal bone. Arch Otolaryngol Head Neck Surg 71:635-652, 1960. 52. Conley JJ, Novack A J: Surgical treatment of cancer of the ear and temporal bone. Trans Am Acad Ophthalmol Otolaryngol 64:83-92, 1960.
53
53. Wagenfield DJ H, Keane T, Norstrand AWP, Bryce D P: Primary carcinoma involving the temporal bone: Analysis of 25 cases. Laryngoscope 90:912-919, 1980. 54. Lewis J S, Parsons H : Surgery for advanced ear cancer. Ann Otol Rhinol Laryngol 67:364-399, 1958. 55. Lewis J S, Page R : Radical surgery for malignant tumors of the ear. Arch Otolaryngol Head Neck Surg 83:56-61, 1966. 56. Lewis J S : Surgical management of tumors of the middle ear and mastoid. J Laryngol Otol 97:299-311, 1983. 57. Lederman M : Malignant tumors of the ear. J Laryngol Otol 79:85-119, 1965. 58. Frew I, Finney R : Neoplasms of the middle ear. J Laryngol Otol 77:415-421, 1963. 59. Goodwin WJ, Jesse R H : Malignant neoplasms of the external auditory canal and temporal bone. Arch Otolaryngol Head Neck Surg 106:675-679, 1980. 60. Lindahl JWS : Carcinoma of the middle ear and meatus. J Laryngol Otol 69:457-467, 1955. 61. Bradley WH, Maxwell J H : Neoplasms of the middle ear and mastoid: Report of 54 cases. Laryngoscope 54:533556, 1954. 62. Miller D: Cancer of the external auditory meatus. L aryngoscope 65:448-461, 1955. 63. Colledge L : Two cases of malignant disease of the temporal bone. J Laryngol Otol 58:251-254, 1943. 64. Peele JC, Hauser G H : Primary carcinoma of the external auditory canal and middle ear. Arch Otolaryngol Head Neck Surg 34:254-266, 1941. 65. Garnett-Passe E R : Primary carcinoma of the eustachian tube. J Laryngol Otol 62:314-315, 1948. 66. Spencer FR : Malignant disease of the ear. Arch Otolaryngol Head Neck Surg 28:916-940, 1938. 67. Means RG, Gersten J: Primary carcinoma of the mastoid process. Ann Otol Rhinol Laryngol 62:93-100, 1953. 68. Spector JG: Management of temporal bone carcinomas: A therapeutic analysis of two groups of patients and long-term follow-up. Otolaryngol Head Neck Surg 104:58-66, 1991. 69. Ariyan S, Sasaki CT, Spencer D: Radical en bloc resection of the temporal bone. Am J Surg 142:443-447, 1981. 70. Arena S : Tumor surgery of the temporal bone. Laryngoscope 84:645-670, 1974. 71. Brooker G B : Bilateral middle ear carcinomas associated with Waldenström’s macroglobulinemia. Ann Otol R hinol Laryngol 91:299-303, 1982. 72. Towson C E, Shofstall WH : Carcinoma of the ear. Arch Otolaryngol Head Neck Surg 51:724-738, 1950. 73. Robinson G A : Malignant tumors of the ear. Laryngoscope 41:467-473, 1931. 74. Kinney S E, Wood BG: Malignancies of the external ear canal and temporal bone: Surgical techniques and results. Laryngoscope 97:158-164, 1987. 75. Buckmann LT, Barre W: Carcinoma of the middle ear and mastoid. Ann Otol Rhinol Laryngol 52:194-201, 1943. 76. Liebeskind M M : Primary carcinoma of the external auditory canal, middle ear, and mastoid. Laryngoscope 61:1173-1187, 1951. 77. Stokes H B : Primary malignant tumors of the temporal bone. Arch Otolaryngol Head Neck Surg 32:1023-1030, 1990.
54
OTOLOGIC SURGERY
78. Rosenwasser H : Neoplasms involving the middle ear. Arch Otolaryngol Head Neck Surg 32:38-53, 1940. 79. Grossman A A, Donnelly WA, Smithman M F: Carcinoma of the middle ear and mastoid process. Ann Otol Rhinol Laryngol 56:709-721, 1947. 80. Mattick WL , Mattick JW: Some experience in management of cancer of the middle ear and mastoid. Arch Otolaryngol Head Neck Surg 53:610-621, 1951. 81. Wahl JW, Gromet MT: Carcinoma of the middle ear and mastoid. Arch Otolaryngol Head Neck Surg 58:121-126, 1953. 82. Crabtree J A, Britton B H, Pierce M K : Carcinoma of the external auditory canal. Laryngoscope 86:405-415, 1976. 83. Hanna DC, Richardson G S, Gaisford JC : A suggested technique for resection of the temporal bone. Am J Surg 114:553-558, 1967. 84. Adams G L , Paparella M M, Fiky FM : Primary and metastatic tumors of the temporal bone. Laryngoscope 81:1273-1285, 1971. 85. Sataloff RT, Myers D L , Lowry L D, Spiegel J R : Total temporal bone resection for squamous cell carcinoma. Otolaryngol Head Neck Surg 96:4-14, 1987. 86. Nadol J B, Schoknecht H F: Obliteration of the mastoid in the treatment of tumors of the temporal bone. Ann Otol Rhinol Laryngol 93:6-12, 1984. 87. Arriaga M, Hirsch B E, Kamerer D B, Myers E N: Squamous cell carcinoma of the external auditory meatus (canal). Otolaryngol Head Neck Surg 101:330-337, 1989. 88. Michaels L , Wells M : Squamous cell carcinomas of the middle ear. Clin Otolaryngol 5:235-248, 1980.
89. McCrea R S : Radical surgery for carcinoma of the middle ear. Laryngoscope 82:1514-1523, 1972. 90. Gacek R R , Goodman M : Management of malignancy of the temporal bone. Laryngoscope 87:1622-1634, 1977. 91. Scholl L A : Neoplasms involving the middle ear. Arch Otolaryngol Head Neck Surg 22:548-553, 1935. 92. Clairmont A A, Conley JJ: Primary carcinoma of the mastoid bone. Ann Otol Rhinol Laryngol 86:306-309, 1977. 93. Beal D D, Lindsay J R , Ward PH : Radiation-induced carcinoma of the mastoid. Arch Otolaryngol Head Neck Surg 81:9-16, 1965. 94. Coleman CC : Removal of the temporal bone for cancer. Am J Surg 112:583-590, 1966. 95. Parsons H, Lewis J S : Subtotal resection of the temporal bone for cancer of the ear. Cancer 7:995-1001, 1954. 96. Miller D, Silverstein H, Gacek R R : Cryosurgical treatment of carcinoma of the ear. Trans Am Acad Ophthalmol Otolaryngol 76:1363-1367, 1972. 97. Tabb HG, Komet H, McLaurin JW: Cancer of the external auditory canal: Treatment with radical mastoidectomy and irradiation. Laryngoscope 74:634-643, 1964. 98. Lodge WO, Jones H M, Smith M E N: Malignant tumors of the temporal bone. Arch Otolaryngol Head Neck Surg 61:535-541, 1955. 99. Ward G E, Loch WE, Lawrence W: Radical operation for carcinoma of the external auditory canal and middle ear. Am J Surg 82:169-178, 1951. 100. Newhart H : Primary carcinoma of the middle ear: Report of a case. Laryngoscope 27:543-555, 1917.
5
Congenital Malformation of the External Auditory Canal and Middle Ear Antonio De la Cruz and Karen B. Teufert Videos corresponding to this chapter are available online at www.expertconsult.com.
Congenital aural atresia is characterized by aplasia or hypoplasia of the external auditory canal (EAC), often associated with absence or deformity of the auricle (microtia) and the middle ear, with occasional inner ear abnormalities. Aural atresia occurs in 1 in 10,000 to 20,000 live births1-5; unilateral atresia is three times more common than bilateral atresia. This disorder occurs more commonly in males and on the right side.1 EAC atresia is more often bony rather than membranous, and bony atresia is regularly accompanied by malformation of the middle ear cavity and structures of the middle ear.6-9 More severe forms of congenital microtia are usually associated with EAC atresia; rarely, canal atresia may be seen in patients with a normal pinna.10 Generally, a more severe external deformity implies a more severe middle ear abnormality.11,12 Kiesselbach, in 1883, is often credited with the first deep operation attempting to correct this malformation.8 Lascaratos and Assimakopoulos13 noted that the Byzantine physician Paul of Aegina performed a surgical treatment of congenital aural atresia.14 The procedure done by Kiesselbach resulted in facial paralysis. Because of the lack of middle ear microsurgery and the high complication rate, congenital aural atresia surgery was considered dangerous and to be avoided for the most part. In 1914, Page15 reported hearing improvement after surgery in five of eight patients. This report was followed in 1917 by Dean and Gittens,16 who reported an excellent hearing result in a patient and reviewed the various types of operations that had been tried by other surgeons. The prevailing attitude toward surgical correction in these cases remained generally pessimistic, however, despite these and other occasional reports of successful operations, until 1947. That year, Ombredanne17 in France and Pattee18 in the United States each reported a series of patients successfully operated on to improve hearing. Pattee’s technique included removal of the incus to “mobilize” the stapes18; Ombredanne17 added fenestration of the lateral semicircular canal. With the advent of tympanoplasty techniques in the 1950s, interest in atresiaplasty increased as the teachings of Wullstein and Zollner carried over into surgery of the congenital ear.8 Larger series with greater success rates
were reported as surgeons attempted to improve their results, using ossiculoplasty, mastoidectomy, differing degrees of bone removal, and different types and techniques of graft placement.2,4,17,19-26 Ombredanne27 went on to report on more than 600 aplasia cases by 1971 and 1600 cases with major and minor malformations by 1976.8 Gill’s report of 83 cases in 19692 and Jahrsdoerfer’s 1978 article8 are considered landmarks. Crabtree,28 Jahrsdoerfer,29,30 Marquet,31,32 and De la Cruz6,33 and their colleagues all reported on large surgical series, with modifications of classification and operative techniques. Although techniques of canalplasty, meatoplasty, tympanoplasty, and ossiculoplasty have improved considerably, surgical correction of congenital aural atresia remains one of the most challenging operations performed by otologists. This is a complex surgical problem, requiring application of all tympanoplasty techniques and a thorough knowledge of the surgical anatomy of the facial nerve, oval window, and inner ear, and their congenital variants.1,6,8,17,21,26,27,29-31,33-40 The temporomandibular joint is displaced posteriorly by the lack of development of the EAC, narrowing the distance between the glenoid fossa and the anterior wall of the mastoid tip.19,41 Fusion of the incus and malleus is common, but because of its dual origin, the stapes footplate is usually normal.4,42 The surgical repair of aural atresia is recommended at age 6 years.6 The timing of repair must take into account any planned auricular reconstruction procedures. Criteria for patient selection must be stringent when attempting to achieve closure of the air-bone gap to within 20 to 30 dB. Preoperative counseling and several postoperative visits are essential for optimal results. In this chapter, we discuss these issues and provide guidelines for patient evaluation and selection, surgical techniques, and postoperative management.
EMBRYOLOGY A review of the normal embryologic development of the ear aids in understanding the myriad of possible combinations of malformations encountered in congenital 55
56
OTOLOGIC SURGERY
TABLE 5-1 Development of the Auricle First Branchial Arch
Second Branchial Arch
First hillock—tragus
Fourth hillock—antihelix
Second hillock—helical crus
Fifth hillock—antitragus
Third hillock—helix
Sixth hillock—lobule and lower helix
TABLE 5-2 De la Cruz Classification of Congenital
Aural Atresia
Minor Malformations 1. Normal mastoid pneumatization 2. Normal oval window/footplate 3. Good facial nerve–footplate relationship 4. Normal inner ear
aural atresia. The inner ear, middle ear, and external ear develop independently and in such a way that deformity of one does not presuppose deformity of another.9,43 Most frequently, abnormalities of the outer and middle ear are encountered in combination with a normal inner ear.44,45 Microtia is a result of first and second branchial arch anomalies. Growth of mesenchymal tissue from the first and second branchial arches forms six hillocks around the primitive meatus that fuse to form the auricle (Table 5-1). By the end of the third month, the primitive auricle has been completed. The external auditory meatus develops from the first branchial groove. During the second month, a solid core of epithelium migrates inward from the rudimentary pinna toward the first branchial pouch. This core, the precursor of the EAC, starts to hollow out and take shape in the sixth month. It canalizes in the seventh month, causing the developing mastoid to become separated from the mandible. Its subsequent posterior and inferior development carries the middle ear and facial nerve to their normal positions.1,43,46-48 Some of the literature supports the notion that microtia grade can indicate the status of middle ear development in aural atresia. The better developed the external ear is, the better developed is the middle ear. The first branchial pouch grows outward to form the middle ear cleft. The plaque of tissue where this cleft meets the epithelium of the EAC forms the tympanic membrane. While the pouch is forming the eustachian tube, tympanic cavity, and mastoid air cells, Meckel’s cartilage (first branchial arch) is forming the neck and head of the malleus and incus body. Reichert’s cartilage (second branchial arch) forms the remainder of the first two ossicles’ long processes and the stapes superstructure. The footplate has a dual origin from the second arch and the otic capsule. The ossicles attain their final shape by the fourth month. By the end of the seventh to eighth month, the expanding middle ear cleft surrounds the ossicles and covers them with a mucous membrane.1,45,49 The facial nerve is the nerve of the second branchial arch. At 4.5 weeks, this developing nerve divides the blastema, which is the condensation of the second arch mesenchymal cells, into the stapes, the interhyale (stapedius muscle precursor), and the laterohyale (precursor of the posterior wall of the middle ear). The nerve’s intraosseous course is dependent on this bony expansion.45,47 The membranous portion of the inner ear develops during
Major Malformations 1. Poor pneumatization 2. Abnormal or absent oval window/footplate 3. Abnormal course of facial nerve 4. Abnormalities of inner ear From De la Cruz A, Linthicum FH Jr, Luxford WM: Congenital atresia of the external auditory canal. Laryngoscope 95:421-427, 1985.
the third to sixth week from an auditory placode on the lateral surface of the hindbrain. The surrounding mesenchyme transforms into the bony otic capsule.50 Congenital aural atresia can range in severity from a thin membranous canal atresia to complete lack of tympanic bone, depending on the time of arrest of intrauterine development.1,51,52 The common finding of a normal inner ear is explained because the inner ear is formed by the time of external/middle ear development arrest. Facial nerve course abnormalities are often seen.
CLASSIFICATION SYSTEMS Of historical significance is a classification in congenital aural atresia developed in 1955 by Altmann.51 In this system, atresias are categorized into three groups, as follows: Group 1 (mild): Some part of the EAC, although hypoplastic, is present. The tympanic bone is hypoplastic, and the ear drum is small. The tympanic cavity is either normal in size or hypoplastic. Group 2 (moderate): The EAC is completely absent, the tympanic cavity is small, and its content is deformed, and the “atresia plate” is partially or completely osseous. Group 3 (severe): The EAC is absent, and the tympanic cavity is markedly hypoplastic or missing. Altmann’s classification system is purely descriptive, and most surgical candidates fall into groups 2 and 3. The De la Cruz classification includes surgical feasibility guidelines using only high-resolution computed tomography (CT), taking into consideration mastoid pneumatization, inner ear normality, and facial nerve and footplate relationship.6 The malformations are divided into minor and major malformations (Table 5-2). The
Chapter 5 • Congenital Malformation of the External Auditory Canal and Middle Ear TABLE 5-3 Jahrsdoerfer’s Grading System of
Candidacy for Surgery of Congenital Aural Atresia
Parameter
Points
Stapes present
2
Oval window open
1
Middle ear space
1
Facial nerve normal
1
Malleus-incus complex present
1
Mastoid well pneumatized
1
Incus-stapes connection
1
Round window normal
1
Appearance of external ear
1
Total available points
10
Rating
Type of Candidate
10
Excellent
9
Very good
8
Good
7
Fair
6
Marginal
≤5
Poor
From Jahrsdoerfer RA, Yeakley JW, Aguilar EA, et al: Grading system for the selection of patients with congenital aural atresia. Am J Otol 13:6-12, 1992.
clinical importance of this classification is that surgery in cases of minor malformations has a good possibility of yielding serviceable hearing, whereas cases of major malformations are frequently inoperable, but treatable with the bone-anchored hearing aid (BAHA) system.3 Jahrsdoerfer and colleagues53 developed a widely used point grading system to guide surgeons in preoperative assessment of the best candidates for hearing improvement (Table 5-3). This system takes into account the parameters of mastoid pneumatization, presence of the oval and round windows, facial nerve course, status of the ossicles, and external appearance. Point allocation is based primarily on the findings on high-resolution CT. Jahrsdoerfer and colleagues53 proposed that when the preoperative evaluation of the patient is itemized into this grading system, the best results (>80% success) are achieved with a score of 8 or better. A score of 7 indicates the patient is a fair candidate, a score of 6 indicates the patient is a marginal candidate, and with a score less than 5 the patient becomes a poor candidate. Ishimoto and associates54 evaluated the relationship between hearing level and temporal bone abnormalities in patients with microtia, using Jahrsdoerfer’s CT scoring system and high-resolution CT scans of the temporal bone. They found that the hearing level in microtic ears correlated with the formation of oval/round windows
57
and ossicular development, but not with the degree of middle ear aeration, facial nerve aberration, or severity of microtia. Schuknecht’s55 system of classification of congenital aural atresia is based on a combination of clinical and surgical observations. Type A (meatal) atresia is limited to the fibrocartilaginous part of the EAC. Meatoplasty is the surgical procedure of choice, and when performed in a timely fashion prevents formation of canal cholesteatoma and conductive hearing loss. In type B (partial) atresia, there is narrowing of the fibrocartilaginous and bony EAC, but a patent dermal tract allows partial inspection of the tympanic membrane. The tympanic membrane is small and partly replaced by a bony septum. Minor ossicular malformations exist, and hearing loss may be mild to severe. Type C (total) atresia includes all cases with a totally atretic ear canal, but a well-pneumatized tympanic cavity. There is a partial or total bony atretic plate, the tympanic membrane is absent, the heads of the ossicles are fused, there may be no connection to a possibly malformed stapes, and the facial nerve is more likely to have an aberrant course over the oval window. Type D (hypopneumatic total) atresia is a total atresia with poor pneumatization, common in dysplasias such as Treacher Collins syndrome. There are abnormalities of the facial nerve canal and the bony labyrinth. These patients are poor candidates for hearing improvement surgery. Chiossone’s56 classification is based primarily on the location of the glenoid fossa. In type I, the fossa is in the normal position; in type II, it is moderately displaced; in type III, the fossa overlaps the middle ear; and in type IV, in addition to the fossa overlapping the middle ear, there is lack of mastoid pneumatization. Patients with types I and II are ideal surgical candidates. Type III cases have a tendency toward graft lateralization. Patients with type IV are not surgical candidates. In atresiaplasty surgery, a classification scheme is useful for surgical planning, patient counseling, and comparison of outcomes.
INITIAL EVALUATION AND PATIENT SELECTION When aural atresia is noted in a newborn, several issues must be addressed. Where one congenital abnormality is found, others must be sought. A high-risk registry for hearing loss is helpful in this regard.57 After the degree of aural deformity is assessed by physical examination, evaluation of auditory function in unilateral and bilateral atresia should be performed using auditory brainstem response audiometry within the first few days of life. An 11% to 47% incidence of inner ear abnormality is associated with congenital aural atresia.12 Occasionally, in unilateral cases, there is a total sensorineural hearing loss (SNHL) on the side of the normal-appearing ear, which might otherwise be missed.6,33
58
OTOLOGIC SURGERY
In bilateral cases, a bone-conduction hearing aid should be applied as soon as possible, ideally in the third or fourth week of life. In unilateral cases in which the opposite ear hears normally, a hearing aid is unnecessary. A child with aural atresia and associated cephalic abnormalities (e.g., hemifacial microsomia and Treacher Collins, Crouzon, or Pierre Robin syndrome) should be recognized.57-61 In this subset, surgical correction has poor results,55 and a long-term bone-conduction hearing aid or BAHA in these nonoperable bilateral congenital aural atresia situations is beneficial (see later).3 Prompt and careful counseling of the parents of a child with sporadic (nonsyndromal) congenital aural atresia is necessary to alleviate concerns regarding possible occurrence in subsequent children (no more than the general population), to answer questions regarding future auricular reconstruction, and, most important, to ensure that proper hearing amplification is instituted in a timely fashion. The child should be enrolled in special education at an early age to maximize speech and language acquisition, in preparation for “mainstreaming” at preschool age. Radiologic and surgical evaluations are deferred until the child is 6 years old (see later). In the initial evaluation of an older individual with congenital aural atresia, the most crucial elements remain the functional and anatomic integrity of the inner ear. Audiometry and high-resolution thin-section (≤1 mm) CT in coronal and axial views are necessary. Prognosis for hearing improvement depends on the presence and degree of malformations. Auricular reconstruction must be done before hearing reconstruction to avoid interfering with the blood supply to the surrounding soft tissue, which is indispensable for microtia repair. Reconstruction with alloplastic materials, such as porous highdensity polyethylene (Medpor), can be done before or after atresiaplasty because intact blood supply is not indispensable. A patient with congenital aural atresia may present with an infected or draining ear or acute facial palsy; 14% have congenital cholesteatoma.6 The priority in these cases is removal of the cholesteatoma and resolution of the infection. Preoperative audiometry and highresolution CT scanning may be necessary at an earlier age in children with repeated infections in the atretic ear.6,55,62 There are two requirements for planning surgery in congenital aural atresia: radiographic three-dimensional evaluation of the temporal bone and audiometric evidence of cochlear function.8,63 Other conditions mandating prompt surgical intervention are congenital cholesteatoma, a draining postoperative atretic ear, and acute facial palsy. The CT scan should always be reviewed for cholesteatoma, which necessitates surgery at any age.55,62,64 It is not included in any of the grading systems because these are used only for predicting hearing results in elective atresiaplasty surgery.
TIMING OF AURICULAR RECONSTRUCTION AND ATRESIAPLASTY In bilateral or unilateral atresia, auricular reconstruction and atresiaplasty are recommended at 6 years of age. Before this age, there may be a tendency to form exostosis-like bony growth that may occlude the EAC, and there is less patient cooperation. By age 6, the costal cartilage has developed sufficiently to allow for reconstruction of the auricle, and the mastoid has become as pneumatized as possible. The microtia repair should be done first because the complex flaps and use of autologous rib graft demand excellent blood supply.65,66 Many surgeons do not perform cartilage microtia reconstruction on an ear with previous reconstruction attempts.64 The hearing restoration surgery is performed 2 months after the last step of the microtia repair. Rehabilitation of auricular defects can be done using an osseointegrated percutaneous mastoid implant prosthesis, with and without bone-conduction aids, such as BAHA.3,48,67-69 Alloplastic materials have been used for microtia repair in the past, but there is a high risk of extrusion associated with their use. Newer materials, such as Medpor, seem to be very well tolerated, however.70-72 As noted earlier, Medpor reconstruction can be done before or after atresiaplasty because intact blood supply is not indispensable. In unilateral cases, atresiaplasty surgery may be indicated in a patient with “minor” unilateral atresia with normal middle ear, ossicles, and facial nerve, and excellent pneumatization. In such patients, atresiaplasty may be offered in childhood with the parents’ consent.33 We often see older adults with unilateral atresia who request surgery when their normal ear begins having highfrequency hearing loss (presbycusis).
Bone-Anchored Hearing Appliances Implantable bone hearing aids available for clinical use were introduced in 1977 in Sweden. BAHA, using Branemark System implants in combination with a hearing aid, has proved to be a favorable means of providing hearing rehabilitation for certain groups of patients, including patients with a congenital ear malformation. BAHAs are suitable for patients who are poor atresiaplasty candidates because of the severity of their malformations. BAHA is a better alternative than a conventional bone hearing aid. Conventional bone hearing aids have several drawbacks: discomfort because of constant pressure from the steel spring, worse sound quality because of higher frequency attenuation by the skin, and poor esthetics and insecure positioning of the device. BAHA works without pressure on the skin and provides direct bone transmission without air interface. Other implants are discussed in detail elsewhere in this textbook. Surgery for BAHA can be performed under local or general anesthesia and is a one-stage procedure. To implant BAHA, a small transcutaneous titanium
Chapter 5 • Congenital Malformation of the External Auditory Canal and Middle Ear
a butment is implanted behind the ear, where it osseointegrates to the mastoid bone. After a 3-month healing period the BAHA processor can then be connected. Audiologic indications for BAHA include a pure tone average bone-conduction threshold better than or equal to 45 dB hearing level (HL) (average 500 Hz, 1000 Hz, 2000 Hz, and 3000 Hz). A maximum speech discrimination score better than 60% when using PB (Phonetically Balanced) word lists is recommended. From an audiologic point of view, the side with the best cochlear function (the best bone-conduction threshold) should be used. Patients with a bone-conduction threshold of 25 to 45 dB HL would be expected to improve, but might not achieve levels in the normal range. Patients with a bone-conduction threshold of less than 30 dB HL (similar to most atresias) can be expected to experience hearing improvements that restore hearing levels to normal ranges. Patients must be able to maintain the abutmentskin interface of BAHA. Careful consideration must be given to the patient’s psychological, physical, emotional, and developmental capabilities to maintain hygiene. Titanium implants can be installed in most patients because allergy to titanium is extremely rare. Alternative treatments should be considered for patients with a disease that might jeopardize osseointegration. For patients with bilateral hearing loss, BICROS (Bilateral Contralateral Routing of Signal) hearing aid is an additional microphone that can be used as an accessory to an implanted BAHA fixture and sound processor to overcome the head shadow effect. The accessory microphone is placed in the contralateral ear to the BAHA fixture. The signal is routed from the accessory microphone to the BAHA sound processor via a wire worn behind the neck. It is intended to improve hearing by eliminating the head shadow effect, but it does so with little success and is not well accepted by patients.73 We do not generally recommend implantable hearing aids in children because the surgical scars may preclude any future microtia repair. However, the BAHA titanium implant is ideally placed 5 to 6 cm behind and 3 cm above the ear canal in a hair-bearing area. This placement seems to allow for the possibility of transplanting costal cartilage to an area with unscarred skin for future microtia repair. On the other hand the titanium implants for a boneanchored ear epithesis are ideally placed 18 to 20 mm behind the (future) ear canal; this interferes with the skin of a future auricle, if reconstruction is contemplated.73 Numerous studies have reviewed results with BAHA in atresia patients.3,74-77 In one series, 45 atresiaplasty surgeries were compared with 39 BAHA surgeries.3 Of 44 ears followed for more than 2 years after atresiaplasty, hearing gain was less than 10 dB in 55% and 10 to 30 dB in 43%. Surgical outcome correlated with Altmann classification: the worse the Altmann classification stage, the worse the surgical outcome. Within 5 years, 24 ears were reoperated on. Of the 39 patients supplied with BAHA, 16 had had prior atresiaplasty
59
work. All BAHA patients in the series considered the implant to be superior to conventional bone-conduction hearing aids, and superior to hearing improvement obtained surgically, for the difficult cases for whom atresiaplasty had been unsuccessful. Van der Pouw and colleagues74 described experience with bilateral BAHA in four patients with bilateral inoperable congenital aural atresia, three of whom had Treacher Collins syndrome. Bilateral BAHA application resulted in improved performance in all tested audiologic parameters, including sound localization, speech recognition in quiet, speech recognition in noise, and a cued listening task. Kunst and associates75 reported on the audiologic outcome of BAHA in 20 patients with congenital unilateral conductive hearing impairment, with a mean air-bone gap of 50 dB. Small, although not statistically significant, improvements in localization scores were observed in favor of BAHA. Some patients with congenital unilateral conductive hearing impairment had such good directional hearing and speech-in-noise scores in the unaided situation that no overall significant improvement occurred after BAHA fitting. Compliance with BAHA use was remarkably high, however, suggesting patient perceived benefit. House and Kutz76 reported the incidence of complications associated with implantation of BAHA in 149 patients and the management of these complications. There were no intraoperative or perioperative complications. Significant postoperative complications requiring intervention occurred in 12.8% of patients. Skin overgrowing the abutment occurred in 7.4%, and all but one of these patients required revision in the operating room. Skin overgrowth was a late complication, occurring an average of 12 months (range 3 months to 2 years) after the initial procedure. Implant extrusion from failure to osseointegrate occurred in 3.4%. Two patients had local wound infections requiring oral antibiotics. One study evaluated hearing results in pediatric patients managed with EAC reconstruction or BAHA.77 The investigators also evaluated the medical costeffectiveness of each procedure. They reviewed 36 ears that underwent surgical canal reconstruction and 6 patients who underwent BAHA placement. Most (93%) of the patients undergoing EAC reconstruction required some form of amplification postoperatively. The investigators concluded that BAHA can achieve acceptable hearing (<15 dB) in school-aged children with normal bone curves, and it can match the bone curves for children with SNHL. The two-staged BAHA system placement may be provided at almost one third the cost to the medical system of surgical EAC reconstruction, on a decibel-for-decibel basis, with single-stage placement yielding even greater cost savings. Implantable hearing appliances seem to offer a good alternative for patients with inoperable atresia and for patients in whom the operative prognosis is poor.
60
OTOLOGIC SURGERY
PREOPERATIVE EVALUATION AND PATIENT COUNSELING Early diagnosis with auditory brainstem response and auditory enhancement with bone-conduction hearing aids in bilateral cases should be done in the first weeks of life.78 Corrective surgery begins at 6 years of age. Imaging is deferred until age 6. Microtia repair is performed before atresiaplasty. The size of the mastoid on physical examination can be estimated by palpation of the mastoid tip, suprameatal spine of Henle (if present), condyle, and zygomatic arch. This is a useful measure because the new ear canal is constructed at the expense of the mastoid air-cell system. The preoperative imaging study is a high-resolution CT scan of the temporal bone in coronal and axial planes (Fig. 5-1).5,12,63,79 The CT scan must be examined carefully by the otologic surgeon together with the otoradiologist.41,80 The four important imaging elements that are most helpful to the surgeon planning reconstruction of a congenitally malformed ear are (1) the degree of pneumatization of the temporal bone; (2) the course of the facial nerve, including the relationship of the horizontal portion to the footplate, and the location of the vertical segment; (3) the presence of the oval window and stapes footplate; and (4) the status of the inner ear.5,6,63 CT also provides information on thickness and form of the atretic bone, size and status of the middle ear cavity, and soft tissue contribution to the atresia,6 but these are less critical for the repair. Proponents of three-dimensional reconstruction CT for preoperative evaluation believe that three-dimensional imaging clearly shows the relationships between the condyle, zygomatic arch, temporal bone, temporomandibular joint, and fallopian canal. We have not found it to be of practical use, however. Lack of pneumatization is the major cause of inoperability in congenital aural atresia.6,33 Normal
I G H T
Y LIN R RED OM: 2 D: 400 V: 500
FIGURE 5-1. Congenital aural atresia, right ear. Coronal high-resolution CT shows atretic external auditory canal and normally developed mastoid system, with normal inner ear.
neumatization is present in most cases. The facial nerve p over the oval window may prevent ossiculoplasty and hearing improvement. If the oval window or footplate are absent, surgery with fenestration of the lateral semicircular canal or placement of a hearing aid is indicated. There is potential for facial nerve injury in atresia surgery.81 The nerve may describe a more acute angle rather than its usual 90 to 120 degrees at the mastoid genu, and often lies more lateral than usual (Fig. 5-2).82 On high-resolution CT, it is important not to “identify” mistakenly the vertical lie of the facial nerve in the marrow bone leading to the styloid process and the hypoplastic mastoid process (Fig. 5-3).83 Even in atretic ears in which the facial nerve does not have an abnormal course, a significantly reduced distance is found between the facial canal and the temporomandibular joint,41 and the facial canal and the posterior wall of the cavum tympani.47 To be considered a surgical candidate, the patient must have sufficient cochlear function as determined by auditory brainstem response or routine audiometry; a normal-appearing inner ear on CT scan; and, preferably, a well-developed mastoid, and good oval window/footplate and facial nerve relationship. The patient and parents are counseled regarding the success of atresiaplasty repair and the chances of successful hearing improvement. Patients with a degree of malformation equivalent to an 8 or better on the Jahrsdoerfer grading scale are given a greater than 80% chance of hearing improvement. The risk to the facial nerve is small, made more so by the use of the facial nerve monitor intraoperatively.33,82 Patients are informed that a split-thickness skin graft (STSG) from the hypogastrium is used to line the new EAC. Initially, frequent postoperative visits are necessary. The possibilities of a facial nerve paralysis (<1%), profound SNHL (approximately 7%, no dead ears), postoperative canal stenosis (<4%), tympanic membrane graft lateralization (<4%), and ossicular chain refixation (<4%) are noted.84,85
SURGICAL APPROACH Early atresiaplasty operations failed because of poor tympanoplasty techniques, large mastoidectomies, and faulty skin grafting techniques.18 Advances in meatoplasty added to the success rate of atresiaplasty surgery, however.19,86 Although fenestration remains a rare option in bilateral congenital absence of the oval window, modern methods of ossiculoplasty are used preferentially and yield superior results.6,8,19,23,26,28,31,34-36,87-89 General endotracheal anesthesia is used; muscle relaxants are avoided. Facial nerve monitoring is used in all cases. The patient is in the otologic position, and the head is turned away. A large shave of the postauricular area is performed, and the auricle and postauricular area are prepared with povidone-iodine and draped. Additionally, the hypogastrium is shaved, aseptically prepared, and draped for the skin graft donor site.
Chapter 5 • Congenital Malformation of the External Auditory Canal and Middle Ear
Oval window
61
Fused malleus-incus
Normal 95° to 120° 60°
A B FIGURE 5-2. Facial nerve in congenital aural atresia. A, Normal intratemporal facial nerve anatomy. B, Intratemporal facial nerve anatomy in congenital aural atresia.
Care is taken to avoid injury to an anomalous facial nerve exiting the temporal bone in this area. Temporalis fascia is harvested, trimmed of excess soft tissue if needed, and placed aside to dry. The temporomandibular joint space is explored to verify that the facial nerve or tympanic bone is not lying within it. The literature has fostered the belief that there are separate and distinct approaches to the bony work necessary in atresiaplasty: the anterior approach, the transmastoid approach, and modification of the anterior approach.6,8,24, 33,35,57,64 We believe that strict distinction between these surgical approaches is unnecessary because each can be used alone or in combination to facilitate the atresiaplasty.
Surgical Technique FIGURE 5-3. Pitfalls in congenital aural atresia surgery: facial nerve on coronal high-resolution CT. Arrows point to a vertical segment of the facial nerve in the left ear.
A postauricular temporo-occipital incision is made. In cases in which microtia repair has been done, care is taken not to expose the costal cartilage graft or alloplastic material used in the auricular reconstruction. Subcutaneous tissue is elevated anteriorly to the temporomandibular joint. An incision is made in the periosteum, which is elevated forward exposing the mastoid cortex and, anteriorly, the temporomandibular joint space (Fig. 5-4).
If a remnant of tympanic bone is present, drilling for the new ear canal begins at the cribriform area. If no such remnant exists, drilling of a 12 mm cylindrical-shaped canal begins at the level of the linea temporalis, just posterior to the glenoid fossa. Dissection proceeds anteriorly and medially, keeping in mind the lack of landmarks in atretic bone. The middle fossa plate is identified and followed to the epitympanum, where the fused malleus head/incus body mass is identified (Fig. 5-5). Care is taken to avoid exposure of the temporomandibular joint space or to avoid opening an excessive number of mastoid air cells. A radical mastoidectomy–like approach should be avoided. When the ossicular mass is identified,
62
OTOLOGIC SURGERY Mastoid periosteum reflected TMJ Cribriform area Temporalis fascia
FIGURE 5-4. Incision, harvesting temporalis fascia, T-incision, and elevation of periosteum. TMJ, temporomandibular joint.
12 mm Ant.
12 mm
Sup.
outer diameter
inner diameter
Inf.
Removal of atretic plate
Post.
FIGURE 5-5. Removal of atretic plate and adhesions, exposing the ossicles, and drilling for the new ear canal.
the atretic bone is removed with diamond microdrills and curettes to expose the ossicles completely (Fig. 5-6A). It is important to minimize drilling on the ossicular mass because transmission of high-speed drill energy to the inner ear may result in high-tone SNHL. An argon laser is used to free the malleus-incus complex from its soft tissue attachments to reduce potential drill trauma to the inner ear. The ossicular mass in the epitympanum is meticulously dissected free of the atresia plate and left
intact. Although the temporomandibular joint often limits anterior dissection, particular effort is made to try to create a space of 2 mm or greater, anterior to the ossicular mass (Fig. 5-6B). The horizontal facial nerve always lies medial to the ossicular mass. While dissecting the inferior and posterior aspect of the canal, an aberrant facial nerve may be encountered as it passes laterally through the atretic bone. Drilling for the new ear canal continues until it measures about 12 mm. This circumference is
Chapter 5 • Congenital Malformation of the External Auditory Canal and Middle Ear
63
A
A Atretic plate
I
S
S
MF plate
Mastoid genu of facial n. (Fallopian canal) 2 mm space
A
P
B
P
FIGURE 5-6. A and B, The ossicular mass is dissected free of the atretic plate, and a 2 mm space is created around the ossicles. MF, middle fossa.
most difficult to achieve medially, where the facial nerve, temporomandibular joint capsule, and middle fossa plate limit larger exposure. Reconstruction with the patient’s own ossicular chain is performed in almost all cases, even if the ossicular chain is deformed. If the ossicular chain is intact, it is left in place and used for ossicular reconstruction. When this is impossible, a total or partial ossicular reconstruction prosthesis to either a mobile footplate or the stapes head is used for reconstruction. The prosthesis is covered with cartilage before grafting the new drum. Occasionally, the stapes footplate may not be seen well because of anomalous facial nerve anatomy, making placement of an ossicular prosthesis difficult or dangerous. In these instances, gentle transposition of the facial nerve has been described, but more commonly hearing restoration can be accomplished with lateral semicircular canal fenestration or the patient can be counseled to use a hearing aid only.4,8,17,26,31,36,42 Drilling continues to create an EAC. It must be one and one third times the size of a normal canal (12 mm versus 9 mm) to allow for contracture healing in the postoperative period. While creating the EAC, care is taken not to open mastoid cells or enter the temporomandibular joint unnecessarily. A dermatome is used to obtain a 0.008-inch thick, 6 × 6 cm STSG from the hypogastrium. Pressure with a gauze sponge soaked in 1% lidocaine with epinephrine 1:100,000 and thrombin solution aids hemostasis of the donor site. When dry, the donor site is dressed with a sterile Tegaderm dressing.90 This dressing reduces most
of the pain associated with other methods of donor site care. One edge of the skin graft is cut in a zigzag fashion to create four or five triangular points (Fig. 5-7). To assist inspection of the skin graft in the final stages of the procedure, the tips of each point and the two corners on the opposite edge are colored with a skin marker. The dried temporalis fascia is trimmed to size, ideally a 20 × 15 mm oval, and a small 4 mm “tab” is cut into the anterior-inferior aspect of the graft in an attempt to prevent lateralization.91 Nitrous oxide is discontinued 30 minutes before grafting begins. The fascia is placed over the ossicular chain medial to the malleus, if present, or over the cartilage that covers the prosthesis (Fig. 5-8). The tab of the graft is placed medially into the protympanum to help prevent lateralization of the new eardrum. Next, the new ear canal is circumferentially lined with the STSG so that all the bone is completely covered (Fig. 5-9). The triangular corners of the skin graft are placed medially and partially overlap the fascia. The colored points help ensure that no skin lies folded on itself, and that the entire width of the graft is used. A single layer of antibiotic-soaked absorbable gelatin sponge (Gelfoam) holds the fascia and skin graft in place. To reproduce the anterior tympanomeatal angle, a disk of absorbable gelatin film (Gelfilm) is placed over the fascia and skin graft. Lateralization is the most common delayed cause of a poor hearing outcome, usually occurring in the first 12 months after surgery. When the graft starts lateralizing, it pulls away from the ossicles, and the original good hearing results slowly deteriorate. After the fascia
64
OTOLOGIC SURGERY
and skin graft are in proper position, three 0.020 inch silicone elastomer (Silastic) strips are placed to line the skin (Fig. 5-10). The Silastic strips help prevent adhesions between the skin graft and the ear canal packing, and ensure good contact between the skin graft and ear canal bone. A 9 × 15 mm Merocel wick is placed in the bony canal. A 12 mm meatus is created, anticipating that a third of its diameter will reduce because of normal healing. Before starting the meatoplasty, the ear is turned back, and the periosteum is brought back in place with one absorbable suture at the superior level of the new ear canal; this stabilizes the pinna and helps ensure that the meatus is at the same level as the new ear canal. Skin, subcutaneous tissue, and cartilage are removed in a 12 mm diameter core over the new meatus. An attempt is made to avoid exposing
6 cm
Lateral end Blue lines
Shape of STSG when in EAC
6 cm
Split-thickness skin graft
Medial end
Blue dots
FIGURE 5-7. Split-thickness skin graft (STSG) used to line new external auditory canal (EAC).
the grafted cartilage or other materials used for auricular reconstruction. The lateral edge of the STSG is brought through the meatoplasty, and the lateral portion of the new ear canal and the meatus are packed with a large 15 × 25 mm Ambrus wick (Ambrus Merocel, Medtronic Xomed Surgical Products, Jackonsiville, Florida) this applies diffuse pressure over the entire lateral skin graft and widely packs the meatoplasty. Five tacking sutures of 5-0 braided polyester (Ticron) attach the lateral edge of the skin graft circumferentially to the meatal skin. Finally, 6-0 fast-absorbing plain gut is used in a running manner between each Ticron suture (Fig. 5-11). The periosteum is sutured back into position. The postauricular incision is closed using absorbable sutures (3-0 polyglactin 910 [Vicryl]). Steri-Strips cover the incision, a mastoid dressing is applied, anesthesia is reversed, monitoring equipment is removed, and the patient leaves the operating room. Silastic sheets and Merocel wicks have helped decrease the incidence of tympanic membrane lateralization by maintaining the tympanic membrane graft in position and the incidence of postoperative EAC stenosis by covering all exposed areas of bone that may cause granulation tissue formation.85 The Merocel wicks, which are now available in many sizes for canal stenting, help prevent postoperative infection and granulation tissue by allowing administration of topical antibiotics to the entire EAC.
12 mm
I (F) A
P
S Temporalis fascia (F)
FIGURE 5-8. Temporalis fascia graft with tab used over an ossicular mass or a partial ossicular replacement prosthesis and cartilage. F, temporalis fascia.
Chapter 5 • Congenital Malformation of the External Auditory Canal and Middle Ear
65
Running suture
STSG
(Enlargement)
Fascia (F) Gelfilm (G)
G F
B
A FIGURE 5-9. A and B, Split-thickness skin graft (STSG) in place over fascia, lining new external auditory canal, and disk of Gelfilm over fascia and skin graft.
Transmastoid Approach (Canal Wall Down Technique) The transmastoid approach is of historical interest only (Fig. 5-12). It has not been used since 1975. In this approach, the drilling starts as posterior as possible from the temporomandibular joint. It starts at the linea temporalis and follows the sinodural angle to the ossicular mass. Mastoid air cell exenteration and lowering of the facial ridge to the facial nerve allows the creation of an EAC with a canal wall down technique. Bone pâté and soft tissue obliteration of the large mastoid cells is performed before undertaking meatoplasty and canal skin grafting. The transmastoid approach resulted in a large mastoid cavity with frequent postoperative drainage, and patients were not allowed to swim.
POSTOPERATIVE CARE The mastoid dressing is removed on the first postoperative day. The Steri-Strips are left in place over the postauricular incision for 7 days. The patient is counseled to keep the operative site dry and to change the cotton ball over the canal/meatus packing four times daily. The Tegaderm over the skin graft donor site is left on for at
least 3 weeks and requires no special attention. Epithelialization occurs under the plastic, and the typical pain associated with older methods of donor site dressing is absent.90 The patient is given a 5-day course of oral antibiotics. The patient is seen 1 week after surgery, at which time the postauricular strips and the tacking Ticron sutures at the meatus are removed. Any dried blood or crusting on the lateral end of the meatus pack should be trimmed. The donor site is inspected. The postauricular site can now be washed, but water precautions continue to apply to the ear canal. We continue to see the patient on a weekly basis. At 2 weeks, the Merocel packs and Silastic are removed, and the meatus is repacked with antibiotic-soaked Gelfoam. At this point, the patient should apply antibiotic suspension to the packing in the ear canal twice a day for 8 to 12 weeks. Usually by 3 weeks, the abdominal donor site has completed the initial re-epithelialization, and the Tegaderm may be removed. By 6 to 8 weeks postoperatively, nearly all of the Gelfoam is usually gone, and the canal is healing well. The first postoperative audiogram is obtained at that time and repeated in 6 months, at 1 year, and yearly thereafter.
66
OTOLOGIC SURGERY
6. 1. 2. 3. 4. 5. 6.
5. 4.
Fascia STSG Gelfilm Silastic sheets ´ 3 1st wick 9 ´ 15 mm 2nd wick 15 ´ 25 mm
Enlargement
4.
5.
4.
2.
4.
2. 3.
2.
2.
1.
3. 1.
A Ant.
Sup.
B
2.
Inf. 1.
4. 4.
Silastic strips
4.
C
Post.
FIGURE 5-10. A-C, Split-thickness skin graft (STSG) and Silastic strips with ear wicks to maintain contact with bony external auditory canal and wide meatus.
PITFALLS Clear communication between the surgeon and the anesthesiologist ensures that the patient is not paralyzed during the procedure, so that facial nerve monitoring, which is essential in these cases, can be effective. When the monitor is turned off transiently owing to the use of electrocautery, we monitor the facial nerve manually with
a hand on the patient’s face. In a poorly pneumatized mastoid, the otic capsule may be difficult to distinguish from surrounding bone, so care is taken not to fenestrate the semicircular canals accidentally. Postoperative care is crucial to achieve proper healing. Patients should understand beforehand that they must follow the postoperative instructions strictly, and that they must keep all scheduled postoperative visits. We check the circumference of
Chapter 5 • Congenital Malformation of the External Auditory Canal and Middle Ear
the pack each week and allow air to enter the EAC. If the meatus appears to be narrowing, usually at the third month, it should be dilated every 2 weeks and restented with a Merocel wick for a period of several months to 2 years. This practice usually eliminates the need of reoperation. Early identification and treatment of infection is also necessary to prevent graft failure and canal stenosis.
STSG sutured to new EAC
Wick
FIGURE 5-11. Skin graft sutured into place at meatus. EAC, external auditory canal; STSG, split-thickness skin graft.
RESULTS Numerous publications have described atresiaplasty results in more than 500 patients from our institution over time.6,33,84,85,92 Our most recent reviews were published in 2003 and 2004 and included 116 atretic ears operated at the House Ear Clinic.84,85 Closure of the airbone gap to 30 dB at short-term (<6 months) followup occurred in 58.5% of primary surgeries and 56% of revisions.84 The long-term (≥6 months) postoperative air-bone gap was 30 dB or less in 50.8% and 39.1% respectively. There was no significant change in air-bone gap from short-term to long-term follow-up for either primary or revision cases. Soft tissue stenosis occurred in 8% of primary surgeries and 3.4% of revisions, and tympanic membrane lateralization occurred in 3.4% and 3.4% respectively, an improvement from the previous series reported in 1995. The lower complication rates resulted from anchoring the fascia graft medial to the malleus handle, placing the tabs created in the graft into the hypotympanum and protympanum, using a Gelfilm disk to form an anterior tympanomeatal angle and to keep the graft in position, and using thinner STSG and Merocel Ambrus ear packs. One of the most common causes of postoperative hearing loss after congenital aural atresia surgery is ossicular chain refixation; this occurred in 11.5%
Transmastoid approach
Tympanic bone remnant
Middle fossa plate
67
TMJ
Deformed malleus Normal incus Stapes Horizontal canal
Posterior canal wall created Sinodural angle
FIGURE 5-12. Transmastoid approach: of historical interest only. TMJ, temporomandibular joint.
68
OTOLOGIC SURGERY
of primary cases and 6.9% of revision cases. To try to avoid refixation, we vaporize fibrous ligaments and bony adhesions with argon laser in the final phases of ossicular dissection. By 1994, all the modifications used in current practice (use of argon laser, thinner STSG, Silastic sheets in the EAC, and Merocel wicks) were routinely being employed. On this basis, the patients were divided into two groups: old cases performed before the changes in surgical technique (n = 36), and new cases that included surgeries performed after these changes were introduced (n = 80).85 We compared complication rates and hearing results before and after the introduction of the new techniques. Overall, the new group had better hearing results and lower rates of complications. Closure of the air-bone gap to 30 dB or less at short-term follow-up occurred in 63.1% of surgeries performed after modifications in the surgical technique and 44.5% of surgeries performed before the modifications. Soft tissue stenosis and bony growth of the EAC were seen in 3.8% of surgeries performed after and 13.9% of surgeries performed before the surgical technique changes. Tympanic membrane lateralization occurred in none of the cases performed after and in 11.1% of the surgeries performed before the surgical technique changes. Ossicular chain refixation occurred in 3.8% of surgeries performed after and 25% of surgeries performed before the modifications. There were no dead ears and no facial palsies in either group. Shih and Crabtree93 reviewed long-term surgical results for 39 ears. Hearing averages of 25 dB, 40 dB, and 46 dB were achieved for mild, moderate, and severe atresia. Restenosis occurred in 33%, and 31% had recurrent cavity/canal skin infections. This rate was reduced with the use of STSG. Digoy and Cueva94 reviewed their short-term (<1 year) and long-term (>1 year) surgical and hearing results for congenital aural atresia in 45 patients (54 ears). Approximately 50% of their patients achieved a speech reception threshold of 30 dB or better in the short term and long term. The average improvement in air-bone gap was 22 dB. Short-term and long-term outcomes were not significantly different. Patients with an intact ossicular chain did not seem to have a significant advantage in hearing compared with patients with a reconstruction prosthesis. They reported meatal stenosis in 7% and tympanic membrane lateralization in 18% of the cases. Persistent or recurrent conductive hearing loss after an initial satisfactory improvement can occur after this procedure and is usually secondary to lateralization of the tympanic membrane or ossicular chain refixation. Modifications used in the surgical technique, such as Silastic sheets, Merocel wicks, Gelfilm disks and tabs in the fascia graft placed in the protympanum, have helped decrease the incidence of these two complications and achieve better hearing results. Stenosis of the ear canal
is one of the most common postoperative complications. The subsequent routine use of STSG, covering all exposed bony and soft tissue surfaces, allows for quicker epithelialization and healing with a reduction in postoperative infection and stenosis. The use of Merocel packs and Silastic sheets has also helped decrease soft tissue stenosis. The Merocel wick placed in the EAC exerts enough pressure to hold the skin graft against the bone, and the Silastic sheets used to line the ear canal make the skin look smooth, with no bubbles, after the packing is removed.
COMPLICATIONS AND THEIR MANAGEMENT Complications of atresiaplasty at our institution for cases operated with the most current technique include lateralization of the tympanic membrane in 3.4%, stenosis of the meatus in 3.8%, high-tone SNHL in 7.5%, and facial nerve palsy in less than 1%.84,85 Other authors have reported meatal stenosis in 7% and tympanic membrane lateralization in 18% of their cases.94 Care is taken at the time of surgery to help minimize the incidence of lateralization. Modern techniques of tympanoplasty are applied during the reconstruction. Nitrous oxide is discontinued 30 minutes before grafting. The graft is anchored medially to the malleus, and tabs are placed into the protympanum. An accurately sized Gelfilm disk is used in an attempt to recreate an anterior tympanomeatal angle, minimizing blunting and keeping the graft in position. Silastic sheets and Merocel wicks placed in the EAC also help minimize the incidence of lateralization. The patient must be followed carefully for at least 24 months because lateralization has been known to occur up to 12 months postoperatively. The incidence of stenosis has been greatly reduced with the use of one-piece STSG covering all exposed bone, and Silastic sheets and Merocel wicks in the EAC.85 Local care and prevention of infection avoid graft failure and early restenosis. Careful inspection of the meatus and early restenting with large Merocel wicks can obviate reoperation. Bony EAC stenosis with a “bumpy” exostosis-like appearance tends to occur more frequently in young patients84; this is probably due to active bone growth at this age, resulting in an exuberant bone regrowth in the newly created EAC. This problem seems to be greatly underreported in the literature. Although laser is used in an attempt to reduce noise trauma to the inner ear and manipulation of the ossicular chain, cases with SNHL and noise-induced hearing loss are still observed. To reduce the risk of inner ear trauma, it is crucial to minimize drilling on the ossicular chain when dissecting it away from the atretic bone, but noise transmission during ear canal creation cannot be completely avoided.
Chapter 5 • Congenital Malformation of the External Auditory Canal and Middle Ear
With high-resolution CT, the oval window can be identified, and problems of SNHL resulting from oval window drill-out can be anticipated. Patients with severe malformations, in whom surgery would be fraught with problems, can be counseled against surgical intervention and fitted with hearing aids or BAHA. With CT, the facial nerve position and trajectory inside the temporal bone can be identified, and, along with facial nerve monitoring, use of this information can reduce the incidence of facial nerve paralysis further.
UNILATERAL ATRETIC EAR Controversy remains over whether children with unilateral atresia should undergo surgery. Jahrsdoerfer8 and De la Cruz and coworkers6 have argued for atresia surgery in selected patients with unilateral atresia. Other authors support this position.95 In the past, Schuknecht,96 Crabtree,28 and Bellucci52 recommended against operating on children with unilateral atresia. These authors argued that the benefit to be gained is minimal in the presence of a contralateral normal-hearing ear. Hearing results at that time were unpredictable and often did not approach the 20 dB air-bone gap needed for useful hearing in the atretic ear. Risks of surgery, including facial nerve injury, also precluded operating on the unilateral atretic ear. In a review of more than 1000 operations for aural atresia with and without cholesteatoma, however, Jahrsdoerfer and Lambert97 showed that the risk was minimal (1%). The incidence of major complications (total SNHL and facial nerve injury) and of other complications (tympanic membrane graft lateralization and restenosis) has decreased over the years, but these are still potential complications. The decision to operate on the unilateral atretic ear must weigh these potential complications along with the possibility of a draining ear. Nevertheless, with excellent preoperative imaging, improved surgical techniques, and advances in technology, we believe the results of atresia surgery are now more predictable. Closure of the air-bone gap to within 30 dB in a properly selected patient can be consistently achieved, and hearing remains stable over the length of follow-up.84 A review examining long-term stability of hearing results in patients operated on for aural atresia does show some drop-off in hearing thresholds (speech reception threshold) over time, however.84,85,98 In the hands of an experienced otologic surgeon with an anatomically favorable patient who (with the parents) understands the risks of potential complications and the need for postoperative care, atresiaplasty in the patient with unilateral atresia is a rewarding operation for the surgeon and the patient. Surgical correction of unilateral atresia offers the benefits of a clean, dry ear with binaural hearing, including sound localization and improved hearing in noise.
69
SUMMARY The treatment of congenital aural atresia poses a challenge. Early identification, appropriate hearing amplification, and speech and language therapy are crucial in bilateral cases. Cooperation with the auricular reconstruction surgeon allows for the best esthetic and functional success. Strict radiologic and clinical criteria are necessary to select appropriate candidates for atresiaplasty. Classification of patients into categories of minor and major malformations provides a guide for prognosis for hearing improvement and potential risks of surgery. A thorough understanding of the embryologic development of the ear and adherence to the surgical principles of tympanoplasty, canalplasty, and facial nerve surgery enable optimal and safe hearing restoration. Maintenance of good initial surgical results and avoidance of late compli cations require diligent postoperative office care. BAHAs provide a good alternative for patients who are poor atresiaplasty candidates.
REFERENCES 1. Federspil P, Delb W: Treatment of congenital malformations of the external and middle ear. In Ars B (ed): Congenital External and Middle Ear Malformations: Management. Amsterdam, Kugler Publications, 1992, pp 47-70. 2. Gill NW: Congenital atresia of the ear: A review of the surgical findings in 83 cases. J Laryngol Otol 83:551-587, 1969. 3. Granstrom G, Bergstrom K, Tjellstrom A : The bonea nchored hearing aid and bone-anchored epithesis for congenital ear malformations. Otolaryngol Head Neck Surg 109:46-53, 1993. 4. House H P: Management of congenital ear canal atresia. Laryngoscope 63:916-946, 1953. 5. Mehra YN, Dubey S P, Mann S B S, Suri S: Correlation between high resolution computed tomography and surgical findings in congenital aural atresia. Arch Otolaryngol Head Neck Surg 114:137-141, 1988. 6. De la Cruz A, Linthicum FH Jr, Luxford WM: Congenital atresia of the external auditory canal. Laryngoscope 95: 421-427, 1985. 7. Hiraide F, Nomura Y, Nakamura K : Histopathology of atresia auris congenita. J Laryngol Otol 88:1249-1256, 1974. 8. Jahrsdoerfer R A : Congenital atresia of the ear. Laryngoscope 88(Suppl 13):1-48, 1978. 9. Kelemen G D: Aural participation in congenital malformations of the organism. Acta Otolaryngol Suppl 321: 1-35, 1974. 10. Grundfast K M, Camilon F: External auditory canal stenosis and partial atresia without associated anomalies. Ann Otol Rhinol Laryngol 95:505-509, 1986. 11. Harada O, Ishii H : The condition of the auditory ossicles in microtia: Findings in 57 middle ear operations. Plast Reconstr Surg 50:48-53, 1972.
70
OTOLOGIC SURGERY
12. Hasso A N, Broadwell R A : Congenital anomalies. In Som PM, Bergeron RT (eds): Head and Neck Imaging. St Louis, Mosby, 1991, pp 960-966. 13. Lascaratos J, Assimakopoulos D: From the roots of otology: Diseases of the ear and their treatment in Byzantine times (324-1453 ad). Am J Otol 20:397-402, 1999. 14. Briau R : Chirurgie de Paul d’Egine. Paris, Masson, 1855. 15. Page J R : Congenital bilateral microtia with total osseous atresia of the external auditory canals: Operation and report of cases. Trans Am Otol Soc 13:376-390, 1914. 16. Dean LW, Gittens TR : Report of a case of bilateral, congenital osseous atresia of the external auditory canal with an exceptionally good functional result following operation. Trans Am Laryngol Rhinol Otol Soc 23:296-309, 1917. 17. Ombredanne M : Chirurgie de la surdité: Fenestration dans les aplasies de l’oreille avec imperforation du conduit: Resultats. Otorhinolaryngol Int 31:229-236, 1947. 18. Pattee G L : An operation to improve hearing in cases of congenital atresia of the external auditory meatus. Arch Otolaryngol Head Neck Surg 45:568-580, 1947. 19. Bellucci R J: The problem of congenital auricular malformation, I: Construction of the external auditory canal. Trans Am Acad Ophthalmol Otolaryngol 64:840-852, 1960. 20. Meurman Y: Congenital microtia and meatal atresia. Arch Otolaryngol 66:443-463, 1957. 21. Nager GT: Aural atresia: Anatomy and surgery. Postgrad Med 29:529-541, 1961. 22. Ombredanne M : Malformations des osselets dans les embryopathies de l’oreille. Acta Otorhinolaryngol (Belg) 20:623-652, 1965. 23. Ruedi L: The surgical treatment of the atresia auris congenita: A clinical and histological report. Laryngoscope 64:666-684, 1954. 24. Scheer A A : Correction of congenital middle ear deformities. Arch Otolaryngol 85:269-277, 1967. 25. Shambaugh G E Jr: Developmental anomalies of the sound conducting apparatus and their surgical correction. Ann Otol 74:873-887, 1952. 26. Woodman DG: Congenital atresia of the auditory canal. Arch Otolaryngol 55:172-181, 1952. 27. Ombredanne M : Chirurgie des surdites congenitales par malformations ossiculaires. Acta Otorhinolaryngol Belg 25:837-869, 1971. 28. Crabtree J A : Congenital atresia: Case selection, complications, and prevention. Otolaryngol Clin North Am 15:755-762, 1982. 29. Jahrsdoerfer R A, Cole R R , Gray L E : Advances in congenital aural atresia. Adv Otolaryngol Head Neck Surg 5:1-15, 1991. 30. Jahrsdoerfer R A : Clinical aspects of temporal bone anomalies. AJNR Am J Neuroradiol 13:821-825, 1992. 31. Marquet J: Homogreffes tympano-ossiculaires dans le traitement chirurgical de l’agenesie de woreille: Rapport preliminaire. Acta Otorhinolaryngol Belg 25:885-897, 1971. 32. Marquet J E, Declau F, De Cock M, et al: Congenital middle ear malformations. Acta Otorhinolaryngol Belg 42:117-302, 1988.
33. Molony TB, De la Cruz A: Surgical approaches to congenital atresia of the external auditory canal. Otolaryngol Head Neck Surg 103:991-1001, 1990. 34. Crabtree J A : Tympanoplastic techniques in congenital atresia. Arch Otolaryngol 88:89-96, 1968. 35. Minatogawa T, Nishimura Y, Inamori T, Kumoi T: Results of tympanoplasty for congenital aural atresia and stenosis, with special reference to fascia and homograft as the graft material of the tympanic membrane. Laryngoscope 99:632-638, 1989. 36. Schuknecht H F: Reconstructive procedures for congenital aural atresia. Arch Otolaryngol 101:170-172, 1975. 37. Bellucci R J: Congenital auricular malformations: Indications, contraindications, and timing of middle ear surgery. Ann Otol Rhinol Laryngol 81:659-663, 1972. 38. Linthicum FH Jr: Surgery of congenital deafness. Otolaryngol Clin North Am 4:401-409, 1971. 39. Patterson M E, Linthicum FH Jr: Congenital hearing impairment. Otolaryngol Clin North Am 3:201-219, 1970. 40. Ruben R J: Management and therapy of congenital malformations of the external and middle ears. In Alberti PW, Ruben R J (eds): Otologic Medicine and Surgery. New York, Churchill Livingstone, 1988, pp 1135-1154. 41. Jahrsdoerfer R A, Garcia ET, Yeakley JW, Jacobson JT: Surface contour three-dimensional imaging in congenital aural atresia. Arch Otolaryngol Head Neck Surg 119: 95-99, 1993. 42. Ombredanne M : Absence congenitale de fenetre ronde dans certaines aplasies mineures. Ann Otolaryngol (Paris) 85:369-378, 1968. 43. Aase J M : Microtia: Clinical observations. Birth Defects 16:289-297, 1980. 44. Sando I, Shibahara Y, Takagi A, et al: Congenital middle and inner ear anomalies. Acta Otolaryngol (Stockh) Suppl 458:76-78, 1988. 45. Van de Water TR , Maderson PF, Jaskoll TF: The morphogenesis of the middle and external ear. Birth Defects 16:147-180, 1980. 46. Barrios-Montes J M : Malformaciones auriculares. Acta Otorhinolaryngol Esp 27:17-46, 1976. 47. Savic D, Jasovic A, Djeric D: The relations of the mastoid segment of the facial canal to surrounding structures in congenital middle ear malformations. Int J Pediatr Otorhinolaryngol 18:13-19, 1989. 48. Tjellstrom A, Bergstrom K : Bone-anchored hearing aids and prostheses. In Ars B (ed): Congenital External and Middle Ear Malformations: Management. Amsterdam, Kugler Publications, 1992, pp 1-9. 49. Gill NW: Congenital atresia of the ear. J Laryngol Otol 85:1251-1254, 1971. 50. Melnick M : The etiology of external ear malformations and its relation to abnormalities of the middle ear, inner ear, and other organ systems. Birth Defects 16:303-331, 1980. 51. Altmann F: Congenital atresia of the ear in man and animals. Ann Otol Rhinol Laryngol 64:824-858, 1955. 52. Bellucci R J: Congenital aural malformations: Diagnosis and treatment. Otolaryngol Clin North Am 14:95-124, 1981. 53. Jahrsdoerfer R A, Yeakley JW, Aguilar E A, et al: Grading system for the selection of patients with congenital aural atresia. Am J Otol 13:6-12, 1992.
Chapter 5 • Congenital Malformation of the External Auditory Canal and Middle Ear 54. Ishimoto S, Ito K, Karino S, et al: Hearing levels in patients with microtia: Correlation with temporal bone malformation. Laryngoscope 117:461-465, 2007. 55. Schuknecht H F: Congenital aural atresia and congenital middle ear cholesteatoma. In Nadol J B Jr, Schuknecht H F (eds): Surgery of the Ear and Temporal Bone. New York, Raven Press, 1993, pp 263-274. 56. Chiossone E : Surgical management of major congenital malformations of the ear. Am J Otol 6:237-242, 1985. 57. Sando I, Suehiro S, Wood RP II: Congenital anomalies of the external and middle ear. In Bluestone CD, Stool SE (eds): Pediatric Otology. Philadelphia, Saunders, 1983, pp 263-274. 58. Fernandez AO, Ronis M L : The Treacher Collins syndrome. Arch Otolaryngol 80:505-520, 1964. 59. Rapin I, Ruben R J: Patterns of anomalies in children with malformed ears. Laryngoscope 86:1469-1502, 1976. 60. Ruben R J, Toriyama M, Dische M R , et al: External and middle ear malformations associated with mandibulo facial dysostosis and renal abnormalities: A case report. Ann Otol Rhinol Laryngol 78:605-624, 1969. 61. Sanchez-Corona J, Garcia-Cruz D, Ruenes R, Cantu J M : A distinct dominant form of microtia and conductive hearing loss. Birth Defects 18:211-216, 1982. 62. Miyamoto RT, Fairchild TH, Daugherty H S : Primary cholesteatoma in the congenitally atretic ear. Am J Otol 5:283-285, 1984. 63. Jahrsdoerfer R A, Yeakley JW, Hall JW III, et al: Highresolution CT scanning and auditory brain stem response in congenital aural atresia: Patient selection and surgical correlation. Otolaryngol Head Neck Surg 93:292-298, 1985. 64. Jahrsdoerfer R A, Hall JW: Congenital malformations of the ear. Am J Otol 7:267-269, 1986. 65. Brent B : The correction of microtia with autogenous cartilage grafts, I: The classic deformity. Plast Reconstr Surg 66:1-12, 1980. 66. Brent B : Auricular repair with autogenous rib cartilage grafts: Two decades of experience with 600 cases. Plast Reconstr Surg 90:355-374, 1995. 67. Gates G A, Hough JV, Gatti WM, Bradley WH : The safety and effectiveness of an implanted electromagnetic hearing device. Arch Otolaryngol 115:924-930, 1989. 68. Hakansson B, Liden G, Tjellstrom A, et al: Ten years of experience with the Swedish bone-anchored hearing system. Ann Otol Rhinol Laryngol Suppl 151:1-16, 1990. 69. Niparko J K, Langman AW, Cutler DS, Carroll WR : Tissue-integrated prostheses in the rehabilitation of auricular defects: Results with percutaneous mastoid implants. Am J Otol 14:343-348, 1993. 70. Naumann A : Plastic surgery to correct deformities of the ear. MMW Fortschr Med 147:28-31, 2005. 71. Romo T 3rd, Presti PM, Yalamanchili H R : Medpor alternative for microtia repair. Facial Plast Surg Clin North Am 14:129-136, 2006. 72. Kelley PE, Scholes M A : Microtia and congenital aural atresia. Otolaryngol Clin North Am 40:61-80, 2007. 73. Tjellstrom A , Hakansson B : The bone-anchored hearing aid: Design, principles, indications, and long-term clinical results. Otolaryngol Clin North Am 28:53-72, 1995.
71
74. van der Pouw KTM, Snik A FM, Cremers CWR J: Audiometric results of bilateral bone-anchored hearing aid application in patients with bilateral congenital aural atresia. Laryngoscope 108:548-553, 1998. 75. Kunst SJ, Leijendeckers J M, Mylanus E A, et al: Boneanchored hearing aid system application for unilateral congenital conductive hearing impairment: audiometric results. Otol Neurotol 29:2-7, 2008. 76. House JW, Kutz JW Jr: Bone-anchored hearing aids: incidence and management of postoperative complications. Otol Neurotol 28:213-217, 2007. 77. Evans A K , Kazahaya K : Canal atresia: “Surgery or implantable hearing devices? The expert’s question is revisited.” Int J Pediatr Otorhinolaryngol 71:367-374, 2007. 78. Sortini A J: Hearing aids for children with bilateral congenital ear canal atresia. Hear Instrument 6:20-23, 1981. 79. Zalzal G H, Shott S R , Towbin R , Cotton RT: Value of CT scan in the diagnosis of temporal bone diseases in children. Laryngoscope 96:27-32, 1986. 80. Andrews JC, Anzai Y, Mankovich NJ, et al: Threedimensional CT scan reconstruction for the assessment of congenital aural atresia. Am J Otol 13:236-240, 1992. 81. Crabtree J A : The facial nerve in congenital ear surgery. Otolaryngol Clin North Am 7:505-510, 1974. 82. Linstrom C I, Meiteles L Z : Facial nerve monitoring in surgery for congenital auricular atresia. Laryngoscope 103:406-415, 1993. 83. Chandrasekhar S S, De La Cruz A, Lo WWM, Telischi FJ: Imaging of the facial nerve. In Jackler R A, Brackmann D E (eds): Neurotology. St. Louis, Mosby–Year Book, 1993, pp 341-359. 84. De la Cruz A, Teufert KB: Congenital aural atresia surgery: Long-term results. Otolaryngol Head Neck Surg 129:121-127, 2003. 85. Teufert K B, De la Cruz A: Advances in congenital aural atresia surgery: Effects on outcome. Otolaryngol Head Neck Surg 131:263-270, 2004. 86. Chole R A : Meatoplasty using inferiorly based island pedicle flap for congenital aural atresia. Laryngoscope 93:954-955, 1983. 87. Colman B H : Congenital malformations of the ear— aspects of management. J Otolaryngol Soc Austral 4: 197-200, 1978. 88. Wigand M E : Tympano-meatoplastie endurale pour les atresies congenitales severes de l’oreille. Rev Laryngol 99:14-28, 1978. 89. Ombredanne M : Transposition des osselets dans certaines “Aplasies Mineures.” Ann Otolaryngol (Paris) 83:273-280, 1966. 90. Weymuller E A Jr: Dressings for split-thickness skin graft donor sites. Laryngoscope 91:652-653, 1981. 91. Sheehy J L : Surgery of chronic otitis media. In English G E (ed): Otolaryngology, Vol 1. Hagerstown, MD, Harper & Row, 1977. 92. Chandrasekhar S S, De la Cruz A, Garrido E : Surgery of congenital aural atresia. Am J Otol 16:713-717, 1995. 93. Shih L , Crabtree J A : Long-term surgical results for congenital aural atresia. Laryngoscope 103:1097-1102, 1993. 94. Digoy G P, Cueva R A : Congenital aural atresia: Review of short- and long-term surgical results. Otol Neurotol 28:54-60, 2007.
72
OTOLOGIC SURGERY
95. Trigg DJ, Applebaum E L : Indications for the surgical repair of unilateral aural atresia in children. Am J Otol 19:679-684, 1998. 96. Schuknecht H F: Congenital aural atresia. Laryngoscope 99:908-917, 1989.
97. Jahrsdoerfer R A, Lambert PR : Facial nerve injury in congenital aural atresia surgery. Am J Otol 19:283-287, 1998. 98. Lambert PR : Congenital aural atresia: Stability of surgical results. Laryngoscope 108:1801-1805, 1998.
6
Surgery of Ventilation and Mucosal Disease Bradley W. Kesser, M. Jennifer Derebery, and Rick A. Friedman
Bilateral myringotomy with placement of ventilation tubes is the most common surgical procedure performed in the United States. An estimated 1.05 million tympanostomy tube procedures are performed annually in the United States.1 In addition, otitis media is the most common diagnosis of patients who make office visits to physicians in the United States—the diagnosis increased from about 10 million visits in 1975 to 25 million in 1990.2 The annual visit rate for children younger than 2 years statistically increased by 224% during one study period.3 Otitis media with effusion (OME) incurs approximately $5 billion annually in direct and indirect costs.1 Because of the magnitude of the disease and its impact on society, and conflicting reports over the most appropriate and cost-effective management of the problem,1,3-5 consensus on the treatment of OME has been difficult to achieve. Attempts have been made to devise an algorithm for the management of OME in young children,6 and these guidelines have been recently reviewed and updated (see later).7,8 This chapter reviews the terminology, epidemiology, pathophysiology, and medical and surgical treatment of OME.
TERMINOLOGY The term otitis media, in its broadest sense, refers to any inflammatory process in the middle ear. The etiology of the inflammation can be (and usually is) infectious in nature, but it can also involve rarer systemic inflammatory diseases (e.g., Wegener’s granulomatosis or type I Gel and Coombs hypersensitivity). The inflammation can be marked by the presence or absence of an effusion, or fluid in the middle ear space. The fluid can be serous (thin, watery, often golden), purulent (pus), or mucoid (thick, viscid “glue”).
Acute Otitis Media without Effusion Acute otitis media (AOM) without effusion is characterized by an inflamed middle ear mucosa and tympanic membrane in the absence of an effusion; this can be seen in the early stages of AOM or during its resolution. The tympanic
membrane appears dull, erythematous, hypervascular, and inflamed; normal landmarks are often lost. In infants and children, AOM without effusion is usually caused by the same organisms that are isolated from acute OME.9 Treatment principles are the same and are discussed later.
Acute Otitis Media with Effusion Acute OME occurs most frequently in infants. Redness with or without bulging of the tympanic membrane, fever, irritability, and pain are the hallmark signs and symptoms. An older child with acute OME has a red tympanic membrane and middle ear effusion, but may not have pain or fever. The middle ear effusion is generally purulent. Casselbrant and associates10 reported a cumulative incidence of acute OME of 43% in a study of 198 newborns followed monthly until the age of 2 years. In the Greater Boston Otitis Media Study Group, infants had an average of 1.2 and 1.1 episodes per year, with 46% of children having had 3 or more episodes by the age of 3 years.11 Recurrent AOM refers to frequent bouts of AOM. The child most likely has intercurrent, persistent (chronic) OME. The effusion becomes infected, and the child develops AOM. Recurrent AOM is an indication for surgical intervention (see later).
Otitis Media with Effusion Otitis media with effusion simply refers to fluid in the middle ear without signs or symptoms of ear infection. Asymptomatic OME can be classified as acute (<3 weeks), subacute (3 weeks to 3 months), or chronic (>3 months).12 Acute and chronic refer to the temporal course of the disease, not to severity. Synonyms of OME include secretory otitis media, nonsuppurative otitis media, or serous otitis media; the most commonly used term is OME.
Chronic Suppurative Otitis Media Chronic suppurative otitis media (CSOM) is a stage of ear disease in which there is chronic infection of the middle ear and mastoid, and in which a central perforation 73
74
OTOLOGIC SURGERY
of the tympanic membrane (or a patent tympanostomy tube) and discharge (otorrhea) are present.13 To meet the requirement for “chronic,” the otorrhea should be present for 6 weeks or longer. The infection involves the mastoid and the middle ear, and usually drains through a central perforation. Chronic otorrhea through a nonintact tympanic membrane (perforation or ventilation tube) may or may not be accompanied by cholesteatoma. Cholesteatoma may or may not result in CSOM. CSOM should not be confused with chronic OME; in the latter, no perforation is present, and the fluid is not purulent.
Idiopathic Hemotympanum The clinical hallmark of idiopathic hemotympanum is the dark blue–appearing tympanic membrane. There is usually no antecedent history of trauma, but trauma can induce this condition. Idiopathic hemotympanum represents a tissue response of the temporal bone to the presence of a foreign body—cholesterol crystals. Three etiologic factors are thought to be responsible: interference with drainage, hemorrhage, and obstruction of ventilation. Chronic OME is the principal precursor. Cholesterol granuloma is the histopathologic correlate. These cholesterol cysts can take an aggressive course with bone erosion and osteitis. Treatment is generally surgical drainage (see later).
EPIDEMIOLOGY Teele and colleagues11,14,15 found that 13% of children in their study groups had at least one episode of AOM by age 3 months; that percentage increased to 67% by 12 months. By age 3 years, 46% of children had three or more episodes of AOM. The highest incidence of AOM was found in children 6 to 11 months old. Most children with multiple recurrences of otitis media have their first episode before age 12 months. An episode of AOM is a significant risk factor for the development of OME. Many investigators have documented persistent middle ear effusion after a single episode of AOM.14-17 Middle ear effusion has been shown to persist after an episode of AOM for 1 month in 40% of children, 2 months in 20%, and 3 or more months in 10%.15
Risk Factors Risk factors for OME include male gender, recent upper respiratory infection, allergic rhinitis, first-degree relative with allergy, bottle feeding, cigarette smoke in the house, increased number of siblings in the house, and, probably the most important, daycare.18 Children in a public daycare facility have a fivefold increase in otitis media at age 2 years compared with children in home
care.19 Whites and Hispanics are more susceptible than African Americans; Native Americans and Inuit are at even greater risk. Skeletal and anatomic factors also predispose to OME. Cleft palate—either overt or submucous—is a significant risk factor. Other craniofacial anomalies, including Treacher Collins syndrome, Apert’s syndrome, Down syndrome and the mucopolysaccharidoses, put children at greater susceptibility to middle ear disease, presumably because of immaturity, dysfunction, and anatomic course of the Eustachian tube. Children with immunodeficiencies are also at greater risk. IgG subclass deficiencies, acquired immunodeficiency syndrome, complement deficiencies, and immunosuppression secondary to medication all predispose to otitis media. Ciliary dysfunction and cystic fibrosis are also known risk factors.
Microbiology Bluestone and coworkers20 obtained aspirates of middle ear effusions by tympanocentesis in infants and children with AOM or OME. Of aspirates from ears with AOM, 35% grew Streptococcus pneumoniae, 23% grew Haemophilus influenzae, and 14% grew Moraxella catarrhalis.20 S. pneumoniae remains the most common bacterium causing AOM.21-23 Introduction of the pneumococcal vaccine may significantly reduce the incidence of pneumococcal disease, including otitis media.24-26 Asymptomatic middle ear effusion (OME) had been previously thought to be sterile. Newer, more sensitive cultures and the introduction of polymerase chain reaction (PCR) testing have shown bacteria and bacterial DNA in asymptomatic middle ear effusions.27 These investigators found that 77% of middle ear effusions had evidence of the three major organisms by PCR (with or without being culture positive), whereas only 28% were culture positive. The most common bacteria were H. influenzae (54.5%), M. catarrhalis (46.4%), and S. pneumoniae (29.9%).27 By comparison, in an earlier study of ears with OME, 30% of aspirates did not grow bacteria, 45% grew “other” strains, 15% had H. influenzae, 10% had M. catarrhalis, and 7% grew S. pneumoniae.20 Other bacteria include Staphylococcus aureus and gram-negative enteric bacilli. In infants younger than 6 weeks, gram-negative bacilli cause about 20% of AOM episodes.21 The bacteriology of CSOM with or without cholesteatoma is different. Most frequently isolated bacteria include Pseudomonas aeruginosa (most common), S. aureus, Corynebacterium, Klebsiella pneumoniae, and anaerobes.23 With better culture techniques, anaerobes have been increasingly isolated from chronic suppurating ears; these organisms include Bacteroides spp., Peptococcus spp., Peptostreptococcus spp., and Propionibacterium acnes.23
Chapter 6 • Surgery of Ventilation and Mucosal Disease
PATHOPHYSIOLOGY Acute Otitis Media Retrograde reflux of material from the nasopharynx through the Eustachian tube is thought to account for the introduction of microorganisms into the middle ear. Bacteria colonize the nasopharynx, but infect the host as a result of a breakdown in barrier or protective factors in the nasopharynx, Eustachian tube, and middle ear. AOM is principally a sequela of a viral upper respiratory infection. The upper respiratory infection impairs ciliary motility and breaks down mucosal barriers that prevent bacterial adherence and growth. Poor clearance of secretions results in stasis and allows bacteria to infect the host. Pathogenic bacteria that appear in the nasopharynx after an upper respiratory infection are the same as the bacteria cultured from middle ear effusions (S. pneumoniae and H. influenzae).28 The adenoid seems to be the source of infecting bacteria in middle ear disease; Pillsbury and associates29 showed higher bacterial colony counts in the adenoids of children with recurrent otitis media than in children undergoing adenoidectomy for adenoid hypertrophy without otitis media. During an upper respiratory infection, sneezing, blowing the nose, and swallowing in the presence of nasal obstruction may create a pressure differential between the nasopharynx and middle ear, forcing bacteria through the Eustachian tube into the middle ear. Finally, Eustachian tube dysfunction is held accountable for OME. The Eustachian tube has three functions: (1) clearance of secretions from the middle ear into the nasopharynx, (2) protection of the middle ear from nasopharyngeal pathogens, and (3) equalization of pressure between the atmospheric pressure (in the nose) and middle ear pressure. The middle ear is an aerated “sinus.” It too must be ventilated and cleared of secretions—the Eustachian tube serves this capacity. In children, the Eustachian tube is short, horizontal, and relatively flaccid. As a result, the protective function of the tube is compromised, and retrograde reflux of secretions into the middle ear occurs. During an acute infection, ciliary function is also compromised, further leading to stasis of secretions and persistence of effusion (see next section).
Chronic Otitis Media with Effusion Two theories have been proposed to account for the persistence of middle ear effusion in the absence of acute infection. As shown by the Boston Collaborative Group, persistent effusion is a natural sequela of acute middle ear infection.15 Effusion persists for 1 month in 40% of children after an episode of AOM, 2 months in 20%, and 3 or more months in 10%.15 Because pathogenic bacteria and bacterial DNA have been recovered from “nonacutely infected” fluid in the middle ear,21,27,30-31 it seems
75
that Eustachian tube obstruction and retained secretions in these cases are the result of the acute infection, rather than the cause. Eustachian tube dysfunction may be a primary disorder that causes acute and chronic OME. Primary Eustachian tube dysfunction results in underaeration and poor ventilation of the middle ear space; this leads to negative pressure in the middle ear with resultant transudation of fluid. Negative middle ear pressure also causes hypoxia and hypercapnia of the middle ear mucosa, resulting in goblet cell hyperplasia and hypersecretion.32,33 The result is a sterile fluid that becomes secondarily infected. The fluid resolves only after adequate ventilation is restored, either by return of Eustachian tube function or by placement of alternative ventilation, such as a ventilation tube. According to Gates,34 the available evidence lends support to the theory that the secretory changes in the middle ear that exist in cases of chronic OME are the histologic sequelae of chronic infection, rather than a separate pathologic disorder. Most cases of chronic OME begin as acute infection of the middle ear; postinflammatory alterations in the mucosa of the middle ear and Eustachian tube lead to persistent effusion. Obstruction of the Eustachian tube is secondary to the infection and not the cause of it. Eustachian tube obstruction prevents clearance of secretions, impedes ventilation and drainage, and perpetuates the inflammatory process. Allergic rhinitis has been recognized as a risk factor in the development of chronic OME. The actual prevalence of “allergic” chronic OME has been reported in the literature with a very broad range of 10% to 90%. Although it is beyond the scope of this chapter to provide extensive details of the type I hypersensitivity response, we provide a short summary. Atopic disease is initially characterized by antigen exposure and specific sensitization. Subsequent re-exposure results in mast cell degranulation and the release of numerous inflammatory mediators, including histamine, cysteinyl leukotrienes, and cytokines, and the infiltration of eosinophils, mast cells, basophils, and other inflammatory cells.35 These cells further the release of histamine and cysteinyl leuko trienes, resulting in increased mucosal blood flow, vascular permeability, and mucus production, and continued recruitment of inflammatory cells. As in the nose, continued allergic inflammation in the middle ear is associated with a marked increase in the number of mucosal mast cells and cell bound IgE.36 In addition to degranulation, mast cells may also present antigen to B lymphocytes, dendritic cells, and monocytes, resulting in the further release of proinflammatory36 cytokines, including tumor necrosis factor, interleukin (IL)-1, and IL-6.36 IL-4 is released by activated mast cells and circulating basophils, and naïve T cells. IL-4 eventually mediates an isotypic switch from a predominantly TH1 lymphocytic profile of naïve T cells to the proinflammatory TH2 cells that are the hallmark of atopic
76
OTOLOGIC SURGERY
disease. The further release of IL-5, IL-9, and IL-13 by TH2 lymphocytes results in the recruitment and activation of more basophils, mast cells, and eosinophils, insuring the development of a chronic allergic milieu.33 The nasal mucosa, middle ear space, nasopharynx, and Eustachian tube, which together comprise the middle ear system, could each potentially serve as the target organ of an allergic reaction resulting in the production of chronic OME. As noted subsequently, not only have allergic effector cells been isolated in each area, but also the intimate anatomic relationships ensure that disruption of the normal functions in one area may have an undesirable effect in another. We now review the evidence of allergic reactivity in each area.
Nose The nose is the target organ most commonly involved in the allergic reaction, and numerous publications have supported the concept that allergic inflammation resulting in nasal congestion may result in significant Eustachian tube obstruction.37-40 The sensitized mucosa of the nose, adenoid bed, or nasopharynx exposed to allergens releases cytokines, resulting in increased mucus secretion, edema, and inflammation around the torus tubarius. The resulting Eustachian tube obstruction has a negative effect on middle ear ventilation, resulting in the production of chronic OME.
Middle Ear Space Although it is intuitively tempting to consider the middle ear the most likely target organ of an allergic reaction resulting in the production of chronic OME, there is likely little antigen exposure to the middle ear mucosa. The Eustachian tube is normally closed, unless one is swallowing or yawning, which would make it much less likely than the nose to allow physical contact with inhaled aeroallergens. Supporting this limited role, Bernstein39 suggested, based on a study determining the local production of IgE in middle ear mucosal biopsy specimens of atopic children undergoing pressure equalization tube insertion, that the middle ear mucosa likely serves as the target organ of an allergic reaction less than 10% of the time. Other authors disagree, however. Eustachian tube dysfunction has been reported from transtympanic challenge on antigen in sensitized animals.41,42 Ebert and colleagues43 showed a similar finding with intratympanic histamine mast cells found to be present in the middle ear effusions and biopsy specimens of children with chronic OME, suggesting local inflammation and reaction.44-47 The middle ear mucosa seems to be capable of being involved directly in an allergic reaction. Authors agree, however, that this direct involvement is likely uncommon. The role of possible local involvement triggered by a food antigen has yet to be defined.
Eustachian Tube Several mechanisms for ��������������������������������� Eustachian tube dysfunction ����� have been proposed. First, congestion of the nasal mucosa may produce a retrograde spread of edema and Eustachian tube dysfunction. Second, poor mucociliary function, either innate or resulting from allergic or other inflammatory etiologies, may lead to retention of secretions resulting in obstruction of the Eustachian tube.48,49 Third, inhalation of aeroallergens with subsequent direct allergic inflammation within the Eustachian tube may produce venous engorgement and hypersecretion of mucus, with the subsequent obstruction affecting gas exchange in the middle ear space.50 The resulting negative pressure allows a transudation of fluids into the middle ear space by the interruption of cellular tight junctions.39,50 Later, persistent obstruction of the Eustachian tube with mucus results in chronic middle ear inflammation, with resulting mucosal metaplasia and increased glandular activities of goblet cells.39,50 Increasingly, physicians who treat allergic patients embrace the concept of one airway, whereby the inflammatory response in one portion of the airway may also lead to inflammatory changes in other portions. The classic example is the recognition of the intimate relationship existing between the upper and lower airways. In a study of atopic subjects, inhaled allergen challenge isolated to the nose produced inflammatory changes in the upper and lower airways. These changes included increased adhesion molecules, increased bronchial hyperreactivity, and eosinophil infiltration.51 Although it is important that we continue research to define the immunology involved in the subset of patients with chronic OME who have allergy as an underlying etiology, we should keep in mind that allergy does affect the common airway, and that lessening significant allergic inflammation in the entire anatomic area would be beneficial.
DIAGNOSIS Idiopathic Hemotympanum Long-standing cases of chronic OME that develop granulomatous deposits in the middle ear and mastoid can lead to idiopathic hemotympanum. Symptoms of idiopathic hemotympanum are those of OME—hearing loss with a plugged or pressure sensation in the ear. Idiopathic hemotympanum is more common in adults. It is characterized by a dark blue–appearing tympanic membrane; fluid at myringotomy is dark brown and syrupy in consistency. Histologically, cholesterol crystals are seen—hence the pathologic term cholesterol granuloma. It is theorized that a small mucosal hemorrhage in the absence of adequate ventilation and drainage results in deposition of hemosiderin, iron, and blood breakdown products into the submucosa. The contents can be walled off, with resultant cyst development. The cyst slowly expands, causing bone thinning and erosion.
Chapter 6 • Surgery of Ventilation and Mucosal Disease
TREATMENT AND PATIENT SELECTION Acute Otitis Media and Recurrent Acute Otitis Media For a single episode of AOM, antimicrobial therapy targets the most common offending pathogens: S. pneumoniae, H. influenzae, and M. catarrhalis. We recommend a 10-day course of amoxicillin as first-line empirical therapy. Clinical practice guidelines from the Agency for Healthcare Research and Quality (AHRQ) argue for observation without antibiotic therapy in selected patients with AOM (Table 6-1).7 Studies have shown an increase in the β-lactamase-producing organisms, H. influenzae and M. catarrhalis.19,52 β-Lactamase renders the organism that produces it resistant to penicillin (and ampicillin). Persistent or recurrent AOM may be secondary to a β-lactamase-producing organism and requires a broader spectrum antibiotic53; good choices in this setting include cefuroxime, erythromycin-sulfisoxazole, trimethoprimsulfamethoxazole, amoxicillin-clavulanate, and cefaclor. Antipyretics (but not aspirin) are also indicated for children with AOM. A child (or adult) with an infectious complication of otitis media requires more aggressive therapy, including intravenous antibiotics and possible surgical intervention. This subject is beyond the scope of this chapter and is discussed in Chapter 19. Children with recurrent AOM may exhibit normal middle ear examinations between episodes, or may retain persistent effusions and fall into the category of chronic OME. The goal of any treatment of a patient with recurrent AOM is long-term prevention of further episodes of otitis media. Placement of tympanostomy tubes is effective treatment in the prevention of recurrent otitis media. Many authorities accept four episodes of AOM in 6 months as a criterion for tympanostomy tube placement. Gebhart54
77
was the first to show a reduction in the number of new episodes of AOM after insertion of tympanostomy tubes. The role of adenoidectomy in the treatment of recurrent AOM is controversial. Although Paradise and coworkers55 found a significant reduction (28% and 35%) in the incidence of AOM in the first and second years after adenoidectomy, a formal study examining the role of adenoidectomy in the treatment of recurrent AOM has not been done. Results of studies of chronic OME and adenoidectomy may or may not be applicable for patients with recurrent AOM. For patients with recurrent AOM and persistent effusion, adenoidectomy is an appropriate surgical treatment (see the following section on chronic OME).
Chronic Otitis Media with Effusion As mentioned, 10% of children with AOM have persistent middle ear effusion 3 or more months after resolution of the acute infection.15 Most children clear their effusion within 1 to 2 months; these patients need no further therapy. The few patients who retain fluid in the middle ear longer than 3 months are at risk for other sequelae, including hearing loss, language delay, vertigo or imbalance, tympanic membrane changes (including atelectasis or retraction pockets or both), further middle ear pathology (including ossicular problems and adhesive otitis), and discomfort with nighttime wakefulness and irritability. Numerous treatment strategies have been proposed for chronic OME: antimicrobial therapy, antihistamines/ decongestants, corticosteroids, tympanostomy tubes with or without adenoidectomy, and mastoidectomy. Updated clinical practice guidelines from the AHRQ do not recommend antibiotics, antihistamines, decongestants, or corticosteroids for the treatment of chronic OME.8 These modalities and allergic strategies are now reviewed.
Antimicrobial Therapy TABLE 6-1 Criteria for Initial Antibacterial Agent
Treatment or Observation in Children with Acute Otitis Media
Age
Certain Diagnosis
Uncertain Diagnosis
<6 mo 6 mo–2 yr
Antibacterial therapy Antibacterial therapy
≥2 yr
Antibacterial therapy if severe illness; observation option if nonsevere illness
Antibacterial therapy Antibacterial therapy if severe illness; observation option if nonsevere illness Observation option
From American Academy of Pediatrics Subcommittee on Management of Acute Otitis Media: Diagnosis and management of acute otitis media. Pediatrics 113:1451-1465, 2004.
More sensitive techniques (e.g., PCR) have shown bacterial DNA in middle ear effusions previously thought to be “sterile” or culture negative. Prolonged antibiotic therapy theoretically eradicates the organism and eliminates the chronic source of effusion. Some studies have shown the efficacy of antibiotics in OME.21,56 Despite these studies, theoretical and practical arguments can be made against their use in chronic OME. Clinical experience indicates that the utility of antibiotics is reduced as the number of treatment courses increases. Children receiving four or more courses of antibiotics over a 3 to 4 month period are most likely not going to resolve their effusion with medical management. Other adverse effects of prolonged antimicrobial therapy include development of anaphylaxis and allergic reactions; hematologic disorders; and the emergence of resistant organisms, a serious worldwide
78
OTOLOGIC SURGERY
problem best shown by the development of resistance to penicillin by S. pneumoniae. Finally, Rosenfeld and Post found through a large meta-analysis of existing studies that the benefit of antimicrobial therapy in chronic OME is slight.
Antihistamines and Decongestant Therapy Oral antihistamine and decongestant combinations and monotherapy have not been shown to be beneficial in the treatment of chronic OME.56 The AHRQ clinical practice guideline does not recommend these agents for chronic OME.8 A possible exception may be in an adult patient with allergen-induced Eustachian tube dysfunction. Stillwagon and colleagues57 investigated the effects of pharmacotherapy on allergen-induced Eustachian tube dysfunction. In this study, adults with a history of seasonal allergic rhinitis to ragweed pollen received either a combination of antihistamine and decongestants or placebo for 7 days followed by an intranasal challenge with ragweed pollen. Eustachian tube obstruction occurred in fewer patients receiving active treatment than in patients receiving placebo. They concluded that pre-exposure treatment with antihistamines in patients with allergic rhinitis may help decrease the risk of developing Eustachian tube dysfunction. There are no studies to date on a possible role of intranasal antihistamines on treatment of middle ear effusion.
Antileukotrienes Antileukotrienes have not been well studied for a possible role in the treatment of chronic OME. Combs58 found a significant decrease, however, in the duration of middle ear effusion in otitis-prone children treated with montelukast after AOM compared with a control group. The reader is advised to take into account the paucity of research and the expense of this relatively safe medication when debating its possible merit in treatment.
Corticosteroid Therapy Steroid therapy for chronic OME has been controversial. Lambert59 found no difference in outcomes between the steroid group and the control group with chronic OME. At this time, the AHRQ guideline does not recommend steroid therapy for chronic OME.8
Treatment for Food Allergy The younger the child with suspected allergies, the more likely the antigen will be ingested rather than inhaled. Juntti and associates60 found that 34% of children with allergic rhinitis or asthma who also had cow’s milk allergy had recurrent OME compared with 13% of children who had allergic rhinitis and no diagnosed milk allergy.
Food allergy is usually suspected in a young child with OME or recurrent AOM and a family history of allergy, and the frequent association of rhinitis with or without asthma or eczema or both. Objective confirmation may be obtained by in vitro or skin testing to a panel of food antigens, followed by an oral challenge/elimination diet as needed.
Immunotherapy Many studies have suggested that more than 70% of children with chronic OME are considered atopic, based on skin tests or in vitro testing.61-64 Immunotherapy for inhalant allergies has been found to be very efficacious in symptom reduction and in lessening the progression of allergic disease. There have been no studies to date of the same rigor on the possible role of immunotherapy for treatment of allergic middle ear disease. The studies that do report a very high (≥75%) rate of resolution of OME with immunotherapy or dietary elimination virtually never have a sufficient (or any) control group, and have varying definitions of allergy and mode of diagnosis, as opposed to studies that have been published on the role of immunotherapy in treatment for allergic rhinitis.37,61,62,65,66 Although it is challenging to desensitize young children with injectable immunotherapy, these studies need to be done to define what role this modality should play in evidence-based practices of the future. It is possible that sublingual immunotherapy, currently widely practiced in Europe, but not yet approved for use in the United States, may more easily allow the design of clinical trials involving young atopic children in the future.
Surgical Therapy Armstrong67 introduced ventilation tube placement in 1954 as a treatment for OME. The ventilation tube acts as an artificial Eustachian tube, aerating the middle ear and equilibrating middle ear pressure with atmospheric pressure. The pathophysiology of chronic OME involves Eustachian tube dysfunction and reflux of nasopharyngeal organisms. Ventilation tubes are aimed at correcting Eustachian tube dysfunction. Children with tubes in place can still get otitis media; the acute infection is not painful because the infected effusion is allowed to pass through the tube and out of the middle ear. The effusion also is not associated with hearing loss; correction of hearing loss is one of the most important goals of surgical therapy. Tube insertion with or without adenoidectomy has been shown to improve conductive hearing loss secondary to OME, and to decrease the amount of time spent with middle ear effusion.21 Placement of tubes is often a clinical judgment based on experience and is addressed on a case-by-case basis. Nevertheless, the AHRQ clinical practice guidelines do offer
Chapter 6 • Surgery of Ventilation and Mucosal Disease
evidence-based recommendations for tympanostomy tube placement (see later).8
Adenoidectomy Nasopharyngeal reflux of secretions and microorganisms into the middle ear plays a large role in the pathophysiology of chronic OME. Adenoidectomy is designed to remove the source of the infecting microorganisms. Three landmark studies have shown the efficacy and low morbidity associated with adenoidectomy.4,55,68 Adenoidectomy is effective treatment for chronic OME, and significantly reduces its morbidity. Its effect is independent of adenoid size. It is argued that the small, “smoldering” adenoid chronically harbors bacteria and is a major contributor to OME. The decision for adenoidectomy should be based on the severity and persistence of the middle ear disease, not the size of the adenoid. Nasal obstruction with adenoid hypertrophy stands alone as an indication. Given the increased cost and slightly increased risk to the patient, Paradise and Bluestone69 have argued for adenoidectomy only in recurrent cases. Much of the literature published on the role of adenoidectomy in OME has been in children 4 to 8 years old.8 Nevertheless, adenoidectomy has been shown to be safe in children older than 18 months,5 and may be effective in younger, high-risk children. In the San Antonio study, the effect of adenoidectomy was greater for younger children.4 We recommend adenoidectomy for recurrent cases—cases in which the child (>4 years old) needs a second set of tubes.
Mastoidectomy Rarely, mastoidectomy is required for chronic OME. The continuously draining ear with secretory tissue in the mastoid (serous mastoiditis) benefits most from opening the aditus ad antrum and facial recess to increase aeration of the middle ear/mastoid air cell system. Removal of secretory tissue or granulation tissue also improves symptoms. Because of the rare necessity for mastoidectomy in chronic OME, no systematic study has been undertaken to prove its efficacy. Decision to proceed with mastoid surgery is based on clinical experience and judgment. We have often left a small Penrose drain in the mastoid and carried it out through the postauricular incision. The drain is removed after the drainage has stopped. Mastoidectomy should be reserved for cases with abnormal mucosa or cholesterol granuloma in the mastoid; it is more commonly indicated for idiopathic hemotympanum.
PATIENT COUNSELING Benefits, limitations, and risks and complications all should be addressed preoperatively with the patient and the parents.
79
Benefits Hearing improvement and reduction in the number of subsequent episodes of ear infections are the chief benefits of tympanostomy tube placement. Hearing improvement speeds and sharpens language and developmental maturation. If the child develops otitis media, tympanostomy tubes also eliminate pain because the infected fluid is allowed to drain out of the middle ear space. Studies have also shown improvement in vestibular function after tympanostomy tube placement.79-81 Finally, reducing the number of secondary problems of recurrent ear infections means less time lost from work for the parents, fewer (if any) courses of antibiotics, and reduction in the cost to the parents of multiple courses of antibiotics. Surgery has been shown to be a cost-effective treatment for children in whom medical therapy fails.1 Benefits of adenoidectomy include improved nasal airway and breathing, and removal of the probable source of the offending pathogens causing middle ear infection. Adenoidectomy is associated with a reduction in the number of new episodes of otitis media. It can also improve sleep, especially in a child with obstructive sleep apnea. As discussed, size of the adenoid has no influence on the incidence of chronic OME; it is theorized that a smaller, chronically infected adenoid may lead to more middle ear problems.
Limitations Tympanostomy tube placement is only a temporizing measure—the tube ventilates the middle ear, but eventually extrudes. The goal is that the tube serves as an artificial Eustachian tube until the child’s own Eustachian tube matures and functions properly. Some children need second and third sets of tubes before their own Eustachian tube functions well enough to ventilate the middle ear. Some Eustachian tubes never work well, and the child may develop further ear disease. Nevertheless, parents should know that tubes are placed to buy time for their child’s own Eustachian tube to function normally. Tubes can also become obstructed. Adenoidectomy does not sterilize the nasopharynx, and it does not prevent otitis media. It can improve the nasal airway, but may not cure obstructive sleep apnea.
Risks and Complications All potential risks and complications, including the risk of anesthesia, should be explained to the parents. Risks of tympanostomy tube insertion include persistent otorrhea, tympanic membrane perforation, and hearing loss. We advise water precautions in children with tubes in place. Eardrum perforation is related to bore of the tube, length of time in place, number of intercurrent infections, and previous history of tubes. Incidence of perforation can be 1% to 15%.
80
OTOLOGIC SURGERY
Bleeding occurs in less than 1% of cases of adenoidectomy. It can require a trip back to the operating room. Temporary velopharyngeal insufficiency has been reported in less than 5%; permanent disability is rare in the absence of problems with palatal clefting (must check for submucous cleft palate by palpation and inspection before proceeding with adenoidectomy). Nasopharyngeal stenosis with subsequent nasal airway and speech problems is a rare complication, but should be mentioned preoperatively.
Idiopathic Hemotympanum We generally recommend tympanostomy tube placement as first-line treatment for idiopathic hemotympanum. This treatment is usually inadequate, however, and the patient requires mastoidectomy. Preoperatively, we order a high-resolution temporal bone computed tomography (CT) scan to examine the air cell pattern, evidence of erosion, and extension of the cholesterol granuloma. Intact canal wall mastoidectomy with removal of diseased mucosa aerates the mastoid and middle ear; aeration is the goal of surgery. The greatest limitation is the risk of recurrence, which is reported to be 50%. Risks of hearing loss, facial nerve injury, and dizziness are low, but are also discussed with the patient.
SURGICAL TECHNIQUE Preoperative Preparation The child is kept NPO after midnight the night before surgery. As a consequence, surgery on children should be the first cases when possible. NPO status 4 to 6 hours before the administration of anesthesia is generally adequate. We do not use perioperative antibiotics for tubes or mastoidectomy. We attempt to keep the child with the parent as long as possible, depending on the child’s age, the anesthesiologist, and hospital policy. Anesthesia is delivered via mask induction and maintenance. A general history and physical examination with screening for anesthetic risks is done for patients needing mastoidectomy. This is done in concert with the anesthesiologist and hospital policy. Mastoidectomy usually requires a general anesthetic.
Surgical Site Preparation and Draping Tympanostomy Tubes Sterilization of the external auditory canal is unnecessary; thorough cleaning of the canal is important, however, for visualization of the tympanic membrane and for postoperative care. The child lies in the supine position with the anesthesiologist delivering anesthetic by holding the mask over the face. The head can be turned to gain optimal visualization. A drape is placed over the child’s
body, but not over the head. The operating microscope is brought into position directly below (inferior to) the surgeon, next to the patient bed.
Mastoidectomy The patient is placed in the supine position with the head at the foot end of the table. The table is turned such that the anesthesiologist and equipment are located at the patient’s feet. Long tubing is used to span the distance. The head is turned away from the affected side; a shoulder roll is placed under the child to improve the surgeon’s angle of view. Straps or heavy tape across the chest and pelvis are used to avoid patient sliding as the table is rotated. The postauricular area is shaved, and clear plastic drapes with sticky edges are applied to the skin edges after the skin is dried and coated with tincture of benzoin. These drapes keep unwanted hair out of the wound, and blood and bone dust out of the hair. An antibacterial soap followed by povidone-iodine is used to scrub the ear, postauricular area, and sticky drapes. A second layer of sticky blue drapes or blue towels is placed around the postauricular area, followed by a standard ENT split sheet that is placed in such a way as to frame the operative site. A trough is created with the split sheet and clips to catch the irrigant and direct it away from the field and into a floor trash can.
Adenoidectomy For adenoidectomy, we turn the table 90 degrees from the anesthesiologist. The patient is placed in the Rose position with shoulder roll. The head is brought to the edge of the table. A towel wrap is placed around the head and secured with a towel clip. This wrap leaves the nose and mouth exposed. Care is taken to ensure that the eyes are closed and taped shut properly before the head wrap is placed. A body drape is placed at the level of the shoulders and unfolded down over the body. A single Mayo stand with all necessary instruments is used over the body.
Instruments Tympanostomy Tubes A set of metal specula, cerumen curettes, and several sizes of suction cannulas (Baron Nos. 3, 5, and 7) are necessary. The operating microscope is also essential. Sterile myringotomy knives come in various shapes and angles. It is useful to choose a knife with a blade width the size of the tube to aid in making the correct dimensions of the incision. Tubes are placed with a cup or alligator forceps and positioned with a Rosen needle. Placement of longstemmed T-tubes is facilitated by the use of an inserter in which the tube is positioned with the short arms of the tube folded forward (Fig. 6-1B). Care is taken to
Chapter 6 • Surgery of Ventilation and Mucosal Disease
minimize trauma to the external auditory canal, drum, or ossicles. The tremendous variety of tube shapes and sizes is testament to the success of the operation. Choice of tube is dictated by surgeon preference on a case-by-case basis. The following basic principles guide tube choice: 1. Short, wide-bore tubes offer little resistance to water entry into the middle ear compared with the longer shaft tubes. 2. Longer shaft tubes can be more easily removed in the office; removal of short tubes with rigid flanges may require an anesthetic. 3. For longer middle ear intubation, long-stemmed Ttubes are preferred. 4. The Richards T-tube with flanges that rest against the middle ear side of the tympanic membrane stay in longest and can be permanent. 5. Risk of perforation increases with increasing duration of intubation, increasing number of infections, and increasing size of the tube. For short-term intubation, short grommets are the best choice—they can last several months to 2 years. For long-term intubation, T-tubes are preferred.
Adenoidectomy Any of the available mouth gags/retractors are adequate with proper technique. Adenoid curettes are available in several sizes and configurations. Angled instruments are easier to use than straight ones. Red rubber catheters placed through the nares and brought out the mouth are used to retract the soft palate. Laryngeal mirrors are useful to inspect the adenoid pad, any residual adenoid tissue, bleeding sites, and final operative site. Tonsil packs are used to pack the nasopharynx for hemostasis. A malleable suction cautery can also be helpful for hemostasis; we urge caution using this device, however, because it can lead to nasopharyngeal stenosis. A bulb irrigator and suction are also used to flush the nose and nasopharynx.
Mastoidectomy The ideal operating room table has motorized controls for adjustment of height and side-to-side rotation. The surgeon’s stool has casters, easy height adjustment, and a flexible back support that permits the surgeon to lean backward as necessary while still receiving full back support. The operating microscope should be adjusted to the surgeon’s refraction and interpupillary distance. A standard set of ear instruments is necessary, including Rosen needle, annulus elevator, round knives, flat knives, and right angle dissectors. Bovie electrocautery can be used in the subcutaneous tissue of the postauricular incision, but it is not recommended around the middle or
81
inner ear. Drill systems (e.g., Ototome, Midas Rex, MedNext, Anspach) with various sizes of cutting and diamond burrs and suction irrigators are also necessary. Finally, facial nerve monitoring may be necessary. Several facial nerve monitors are available; the device should have the capacity for continuous monitoring and for stimulating the nerve for localization during the case.
Technical Details Tympanostomy Tube Insertion The ear canal is gently cleaned of all wax and debris. Contact with the anterior bony canal wall is avoided because of risk of bleeding. The tympanic membrane is inspected, and the short process of the malleus is identified. This is a constant landmark and may be the only one available in cases of acute infection. The tympanic membrane is incised anteroinferiorly by using an incision that parallels the fibrous annulus (see Fig. 6-1). Use of a radial incision is satisfactory, but may be limited by an overhanging anterior canal wall. Posterior incisions should be avoided because they place the ossicles at risk. The incision is gently spread open. Care is taken to avoid any major vessel in the tympanic membrane to prevent hemorrhage into the layers of the eardrum. This bleeding into the drum is thought to predispose to tympanosclerosis. The middle ear should be evacuated by using a smalldiameter (5 Fr) suction cannula. Occasionally, gluey material is too viscous to pass through the cannula. We do not recommend using anything larger than the 5 Fr cannula. In these cases, the middle ear and ear canal can be irrigated with warm sterile saline. This usually breaks the viscous material up enough to pass through the cannula. Not all of the effusion needs to be evacuated; as long as the middle ear has a near-normal airspace to place the tube, the remainder of the effusion is carried into the Eustachian tube or drains out the tube. Culture of the effusion rarely is done. It is important to position the tube such that the lumen is in line with the surgeon’s line of sight, facilitating postoperative examination of the middle ear mucosa in the office and cleaning of the tube should it become plugged later on. When using T-tubes, the surgeon should ensure that the short arms of the tube are completely unfolded. Ototopical drops are placed if there is an acute infection. A small cotton ball is placed in the meatus.
Laser Myringotomy The laser has become a useful, albeit expensive, tool in the management of chronic OME. Advantages of the laser include office-based application, ease of use, and the ability to place a controlled perforation in the tympanic membrane that stays open for a medium length of time (2 to 6 weeks). Using the CO2 laser at 12 W with a single
82
OTOLOGIC SURGERY
FIGURE 6-1. A, Incision in tympanic membrane. B, Placement of T-tube within the inserter. C, Proper position of the tube. FIGURE 6-2. A, Technique of adenoidectomy using the curette. It is important to keep the handle in the sagittal or parasagittal plane to prevent injury to the torus tubarius. B, Use of the suction cautery with mirror for hemostasis.
Chapter 6 • Surgery of Ventilation and Mucosal Disease
100 ms pulse through a 200 mm objective, Goode73 reliably placed 1.5 to 2 mm perforations in the tympanic membranes of 10 subjects. Ten of the 11 ears healed within 6 weeks. Tube placement was avoided. Marchant and Bisschop74 performed 20 consecutive CO2 laser myringotomies on ears with chronic OME. All myringotomies closed within 4 weeks, with an average closing time of 17 days; 60% of cases of chronic OME were cured after 3 months. CO2 laser myringotomy has application in clinical situations when middle ear ventilation is needed for a medium length of time (weeks) without having to place a ventilation tube. Disadvantages include cost, need for extra machinery that can be bulky, required maintenance, instruction on use and technique, and office space. Local anesthesia (iontophoresis or topical phenol application) is still required. Most otologists prefer simple cold-steel myringotomy with or without tube placement, but CO2 myringotomy is an alternative; only time and experience will tell whether this technique will have widespread application and use.
Adenoidectomy For an adenoidectomy, the patient is given a general anesthetic, and the airway is secured via endotracheal intubation. The patient is placed in the Rose position with the neck extended over a shoulder roll and draped, as described earlier. The mouth gag is inserted and suspended from the Mayo stand located over the body of the patient. The soft palate is retracted with red rubber catheters. The hard and soft palates are palpated for the presence of a submucous cleft palate. The adenoid pad is inspected with the curved laryngeal mirror. The adenoid is excised with curved curettes of various sizes (Fig. 6-2A). The curette is seated high in the nasopharynx, and the adenoid pad is resected with a down-sweeping motion on the curette. Care must be taken to avoid injury to the prevertebral fascia and muscles, which may cause excessive bleeding. The nasopharynx is palpated for residual adenoid tissue; a second or third pass may be necessary. Curved biting forceps are useful to remove tissue inaccessible by the curette. The mirror is again used to inspect the site. Curettage of the tissue in the fossa of Rosenmüller is not done because it may lead to scar tissue formation and contracture that might result in stenosis or Eustachian tube reflux or both. Direct injury to the Eustachian tube also is avoided. The goal of surgery is the complete removal of the midline adenoid pad to achieve smooth re-epithelialization of the nasopharynx. Bleeding usually stops quickly; tonsil sponges are used to pack the nasopharynx for hemostasis. The nasal cavities and nasopharynx are irrigated with warm saline. A malleable suction cautery can be used for precise coagulation, but its use is cautioned because of the risk of stenosis (see Fig. 6-2B).
83
Mastoidectomy The ear canal and postauricular areas are initially injected with 1% lidocaine with 1:100,000 concentration of epinephrine. Vascular strip incisions are started medially at the fibrous annulus and carried laterally along the tympanomastoid and tympanosquamous suture lines (approximately 12 and 8 o’clock positions for a right ear and 12 and 4 o’clock positions for a left ear). The incisions should come over the bony-cartilaginous junction laterally. The incisions are connected medially around the annulus with the round knife, and the vascular strip is elevated from medial to lateral. A cotton ball soaked in 1:100,000 epinephrine solution (with or without lidocaine) is placed in the canal, and attention is turned to the postauricular area. The postauricular incision is based about 1 fingerbreadth behind the postauricular crease, roughly paralleling the free margin of the helix (Fig. 6-3). The further posterior the incision, the greater the ease of inspecting the middle ear and Eustachian tube through the facial recess. The incision is carried slightly anterior in its inferior dimension to allow the ear to be retracted forward easily. In a child, care is taken not to extend the incision beyond the mastoid tip, which is more superior than in an adult. Carrying the incision more inferior or anterior puts the facial nerve at risk. Superiorly, the temporalis fascia is identified, and a piece of fascia is harvested if needed. The fascia identifies the plane of dissection. The ear is held forward with a self-retaining retractor. The linea temporalis is palpated, and an incision is made down to the bone along this line from anterior to posterior. A second incision is made perpendicular to the first in a curvilinear fashion down to the mastoid tip. The Lempert elevator is used to elevate the periosteum to identify the cribriform area and posterior canal wall. A small elevator is next used to elevate the vascular strip out of the canal. The vascular strip is held forward with the ear under the self-retaining retractor. The tympanic membrane is carefully elevated, and the middle ear is inspected. A large cutting burr and continuous suction-irrigation are used to remove the lateral mastoid cortex. Important landmarks to identify include the posterior bony canal wall anteriorly, the tegmen mastoideum superiorly, the sigmoid sinus posteriorly, and the digastric ridge inferiorly. Care is taken not to expose the dura of the middle cranial fossa or the sigmoid sinus. When these lateral landmarks have been identified, the dissection is continued medially under the microscope. Körner’s septum is opened medially, and the mastoid antrum is identified. This dissection is carried anteriorly to open the aditus ad antrum and attic. The fossa incudis and short process of incus are carefully uncovered. The short process of the incus should be seen refracted through water. The short process marks the level of the facial recess. All air cells of the mastoid cortex should be taken down to reduce
84
OTOLOGIC SURGERY
the surface area of the system. The bony plates over the posterior and middle fossa dura are skeletonized to form a smooth surface (Fig. 6-4). Care is taken to avoid exposing dura. The mucosa regenerates into a single large cavity. The descending segment of the facial nerve is identified (using a diamond burr and copious suction-irrigation) by gentle dissection from superior to inferior using the fossa incudis, lateral semicircular canal, and digastric ridge as essential landmarks. The nerve and blood vessels on the nerve can be seen through bone. Care is taken not to expose the nerve. The facial recess is opened into the middle ear by the use of progressively smaller diamond burrs. The plane of the short process of the incus leads to the facial recess (Fig. 6-5). When the fallopian canal and chorda tympani nerve are identified, the dissection is carried between them medially to open into the middle ear. Coupled with the tympanotomy, all parts of the mesotympanum can be inspected. The ossicular chain is palpated, and any hyperplastic mucosa, granulation tissue, or secretory tissue is removed. If bone of the middle ear/promontory is exposed, a piece of absorbable gelatin film (Gelfilm) is placed through the facial recess and across the promontory toward the Eustachian tube at the end of the procedure to prevent adhesions and to keep the recess open. When all hyperplastic mucosa has been removed, the tympanic membrane is folded down back over the bony annulus and grafted if necessary. Cortisporin-soaked absorbable gelatin sponge (Gelfoam) packing is placed in the medial canal. The vascular strip is returned to the posterior canal wall, and the periosteal flap is resutured to the native periosteum with 2-0 chromic suture. The vascular strip is inspected transcanal and carefully placed back so that no edges are rolled under. The remainder of the canal is packed with Cortisporin-soaked Gelfoam. The postauricular incision is closed meticulously with 3-0 undyed (Vicryl) in the subcutaneous layer, avoiding the need for skin sutures. A small Penrose drain can be placed in the mastoid and carried out through the inferior aspect of the incision if the mastoid is very weepy. The drain is removed when the mastoid no longer drains. A cotton ball is placed in the meatus, Steri-Strips are applied to the postauricular incision, and a sterile dressing consisting of Telfa, gauze, fluffs, and a mastoid (or cup) wrap is placed.
POSTOPERATIVE CARE AND FOLLOW-UP Tympanostomy Tubes A cotton ball is inserted into the ear canal to absorb any drainage. Parents are asked to change the cotton ball once or twice daily for a couple of days, until the drainage stops. If the ear was acutely infected at the time of surgery, a topical antibiotic/steroid suspension is used twice a day for 5 days. The first follow-up visit is at 10 to
14 days to ensure tube placement and resolution of infection. Children are seen every 6 months for tube check thereafter. Water precautions are instituted. Parents are instructed to treat any new otorrhea with the antibiotic/ steroid suspension. If the drops do not clear the infection in 2 to 3 days, an oral antibiotic is prescribed. Should this fail, the child is seen in the office for aural hygiene and cleaning, and possibly intravenous antibiotics if the infection fails to resolve. Occasionally, the tube elicits an allergic reaction with granulation tissue; removal of the tube often resolves the reaction. Tubes generally extrude within 1 to 2 years.
Adenoidectomy Otalgia is common after adenoidectomy, and the parents should be counseled as such. Acetaminophen is usually adequate. The child should be started on a liquid diet initially, and if tolerated, advanced. Transient nasal speech may occur in a small percentage of cases, but regurgitation of liquids through the nose is rare. Palatal and pharyngeal wall compensation occurs quickly. If the child is old enough, chewing gum may speed the process and strengthen the pharyngeal and palatal musculature. Children return to the office 10 to 14 days after surgery for a checkup.
Mastoidectomy Mastoidectomy is generally well tolerated. Patients receive pain medication; antibiotics are given in cases of acute infection. If a drain was placed, it is removed when the drainage has ceased—usually on the first or second postoperative day. Most patients do not require drains and are discharged the evening after surgery. Some require overnight observation for prolonged recovery from general anesthesia. We advise patients not to lift anything heavier than 5 to 10 lbs and not to blow their nose. The postauricular area should be kept dry. Patients return to the office 7 to 10 days after surgery for wound and canalpacking inspection. Drops are started 1–2 weeks after surgery to dissolve the packing; patients return again 8 weeks after surgery to check and clean the canal and for a postoperative audiogram. Drops can be started earlier if drainage or infection develops.
SURGICAL PITFALLS Tympanostomy Tubes Great care should be taken to avoid trauma to the external auditory canal during tube insertion. The resultant bleeding, although minor in amount, leads to a clot that is difficult to remove and obscures visibility in the office. Irrigation with saline and application of a topical vasoconstrictor such as phenylephrine or oxymetazoline usually bring this bleeding under control.
Chapter 6 • Surgery of Ventilation and Mucosal Disease
85
FIGURE 6-3. Skin incision and the second incision for the Palva flap. FIGURE 6-4. Mastoidectomy in progress. The sigmoid plate is skeletonized, and the bulge of the lateral semicircular canal in the antrum is visible. Note the positioning of the large retractor and dural hooks to keep the Palva flap rotated forward.
FIGURE 6-5. Completed mastoidectomy in the right ear with the facial recess opened. The dimensions of the facial recess are exaggerated by the artist to display the middle ear structures that may be seen by the surgeon after multiple repositionings of the viewing angle of the microscope.
86
OTOLOGIC SURGERY
Bleeding within the tympanic membrane is commonly seen when a vessel is included in the incision. Such bleeding dissects between the layers of the drum and may result in the formation of a tympanosclerotic plaque. Careful placement of the incision avoids this potential problem. Dislodgment of the tube into the middle ear may be a problem with a large myringotomy incision and small tube, or too small a myringotomy where the tube is pushed through. Tubes usually fall toward the Eustachian tube orifice; widening a small myringotomy allows the surgeon to retrieve the tube. Early extrusion of the tube usually occurs because the incision is too large or the tube is only partially inserted. Placement of a tube in an atelectatic ear can be difficult. The best area to place the tube is around the Eustachian tube orifice (anterior) because this area usually contains the most aeration and allows tube placement. Ossicular injury is rare if the incision is kept anterior, and tube manipulation is gentle.
Adenoidectomy The chief surgical pitfalls to avoid during adenoidectomy are trauma to the torus tubarius, which protects the opening of the Eustachian tube, and deep removal of the posterior wall of the nasopharynx, which leads to excessive bleeding. Palpation and careful inspection of the nasopharynx with a mirror avoid inadequate removal of adenoid tissue. Bleeding usually can be controlled with packing of the nasopharynx and saline irrigation. Bleeding is usually from residual adenoid tissue; mirror examination can be helpful to identify the source and apply precise suction cautery or to remove adenoid remnants. Before placing a patient with Down syndrome in the Rose position with the neck extended, the cervical spine should be cleared.
Mastoidectomy The chief pitfall of mastoidectomy is injury to the facial nerve when opening the facial recess. This trauma is rare in experienced hands, especially with continuous monitoring of the nerve during the procedure. Dural exposure in the tegmen mastoideum should be avoided because of the risk of later encephalocele. Care also should be taken around the sigmoid sinus; opening of the sinus can usually be controlled with precise packing with absorbable cellulose packing (Surgicel). Contact of the drill with the incus may result in mild to moderate high-frequency sensorineural hearing loss owing to vibratory energy transmitted to the cochlea. Drill contact should also be minimized on labyrinthine bone. Blue lining or opening a semicircular canal should be addressed immediately by gently closing the opening with bone wax. A blue-lined canal should be recognized and avoided. Wound infection is uncommon; copious
irrigation during and after the procedure aids in removal of unwanted debris or bone dust that might serve as a nidus for infection.
RESULTS Goals of surgery include hearing improvement, reducing the time spent with middle ear effusion, and reducing the number of recurrences of middle ear effusion. Tube insertion with and without adenoidectomy has been shown to improve conductive hearing loss secondary to chronic OME and to decrease the amount of time spent with effusion.21 Gates and colleagues4 assigned 491 children, 4 to 9 years old with OME persisting 60 days or more after repeated medical therapy, to various combinations of myringotomy, tympanostomy tube insertion, and adenoidectomy for chronic OME. They found that time with recurrent middle ear effusion was decreased by 29% in the tympanostomy tubes–only group, by 38% in the adenoidectomy-plus-myringotomy group, and by 47% in the adenoidectomy-plus-tympanostomy tubes group. Hearing was equivalent in all groups except the myringotomy-alone group; hearing in this group was significantly worse. Surgical retreatments were necessary more often in children initially treated with myringotomy alone (36%) or tympanostomy tubes alone (20%) than in children treated by adenoidectomy and myringotomy (10%) or adenoidectomy plus tympanostomy tubes (10%). The number of repeat operations in the two adenoidectomy groups was significantly less than in the two groups without adenoidectomy (P < .001).4 Paradise and associates55 randomly assigned patients who had previously undergone tympanostomy tube placement and had recurrent middle ear disease into either an adenoidectomy or control group. During the first and second years of follow-up, the adenoidectomy group had 47% and 37% less time with otitis media than the control patients. Although most cases of idiopathic hemotympanum resolve satisfactorily, surgical management (tube placement or mastoidectomy or both) is generally successful. Often it takes months for the ear to aerate and for satisfactory hearing to return. Nonetheless, persistence and patience are often rewarded.
COMPLICATIONS AND MANAGEMENT Tympanostomy Tubes Otorrhea The most common sequela of tympanostomy tubes is purulent otorrhea. In the Gates study,4 otorrhea occurred one or more times in 22% of the myringotomy-alone group, 29% of the tympanostomy tubes group, 11% of the adenoidectomy-myringotomy group,
Chapter 6 • Surgery of Ventilation and Mucosal Disease
and 24% of the adenoidectomy-tympanostomy tubes group. Some cases of otorrhea are due to water contamination; others are the result of AOM. Some cases also involve an inflammatory reaction to the tube itself. Treatment is the same: a topical polymicrobial-steroid suspension with or without an oral antibiotic, along with aural hygiene in the office. Most cases clear quickly. In recalcitrant cases, the tube is removed, and cultures are done. Failure to clear persistent otorrhea after maximal medical therapy is an indication for tympanoplasty with mastoidectomy.
Persistent Perforation Tympanostomy tubes extrude within 1 to 2 years of insertion. Depending on the tube, 1% to 15% of cases result in a persistent perforation. Older children may be able to tolerate attempts to close the perforation in the office with freshening of the edges and placement of a paper patch. Other children are good candidates for a fat plug myringoplasty75 or more formal myringoplasty/tympanoplasty under anesthesia.
Adenoidectomy Bleeding The most common complication of adenoidectomy is postoperative bleeding. The incidence is low, however: of 250 cases done by 13 surgeons, only 1 child required operative treatment for bleeding, and none needed transfusion.4 Helmus and colleagues85 noted that only 4 patients in 1000 (0.4%) bled after outpatient adenoidectomy; all instances occurred in the first 6 postoperative hours and were managed without transfusion. Return trip to the operating room for bleeding involves the same positioning as for routine adenoidectomy. Irrigation with suction cautery generally controls bleeding.
Velopharyngeal Insufficiency Other, less common complications include nasopharyngeal stenosis and velopharyngeal insufficiency. Stenosis results from excessive tissue destruction, including excessive use of cautery, excessive curettage of the fossa of Rosenmüller, and removal of the lateral pharyngeal bands. Stenosis can require reoperation for scar tissue removal or pharyngeal flap reconstruction. The best treatment is prevention. Transient velopharyngeal insufficiency may occur after removal of a large adenoid, but resolves quickly in most cases. If the child is old enough, gum chewing strengthens and reconditions the injured pharyngeal musculature. Persistent velopharyngeal insufficiency is the most feared complication because it requires either a prosthesis or a secondary procedure (pharyngeal flap) for reconstruction. Most cases are due to an undetected
87
s ubmucous cleft palate. Preoperative palpation and inspection (note bifid uvula) identify patients at risk.
Mastoidectomy Facial Paralysis Complications after mastoidectomy are rare. Facial paresis occurs rarely in experienced hands. Intraoperative facial nerve monitoring has decreased this complication further. Heat injury should not occur if continuous irrigation is used. Intimate knowledge of the anatomy and anatomic relationships is the best prevention of facial nerve injury. Injury, if it involves more than 25% of the nerve, has historically been repaired with direct anastomosis after mild decompression and freshening of the edges. Review of our results reveals that recovery to a maximum House-Brackmann grade III was similar for primary anastomosis and cable graft.86
Hearing Loss High-frequency sensorineural hearing loss may be caused by drill trauma around the ossicles, especially the incus. Careful inspection of the aditus through water identifies the incus and minimizes drill trauma. Drilling on labyrinthine bone should be kept to a minimum. Inadvertent opening of a semicircular canal should be sealed with bone wax with no suctioning around the fistula.
Recurrent Hemotympanum Recurrent idiopathic hemotympanum is common. Treatment should be with a large-bore tympanostomy tube. Occasionally, a second-look mastoid operation is indicated. Hearing amplification helps with hearing loss.
ALTERNATIVE TECHNIQUES Children with persistent effusion for whom medical and surgical treatment have failed should be evaluated for auditory trainer, hearing aid use, or other form of hearing rehabilitation, especially when the child is in school. Preferential seating in class is strongly encouraged.
ALLERGY TREATMENT Children with symptomatic food or inhalant allergy deserve therapy whether they have OME or not. Because most patients with OME have had prior nasal infection, and children with nasal allergy have a higher prevalence of infection, allergy evaluation is appropriate for children with OME who also have nasal symptoms. Gates and colleagues4 found a lower incidence of allergy in their subjects with OME, however, than in the general population. Although a cause-and-effect relationship between
88
OTOLOGIC SURGERY
nasal allergy and OME has not been shown, the surgeon should inquire about allergic symptoms to provide proper therapy to patients with dual problems.
BIOFILMS It has been known for more than a decade that culturenegative OME is not sterile, or bacteria negative.78-80 Sophisticated techniques including in situ hybridization, PCR, and blotting techniques have shown bacterial DNA, RNA, and proteins in “culture-negative” middle ear fluid.81 A logical explanation for this phenomenon is bacterial biofilm formation. Ninety-nine percent of bacteria in nature live in complex aggregates or communities called biofilms. Biofilm formation begins with the adhesion of bacteria in an aqueous environment to any organic or inorganic surface. Irreversible adsorption of the bacteria to the surface is followed by bacterial colony formation. The colony forms a complex polysaccharide coating, an extracellular polysaccharide matrix that protects the colony from outside stresses. The colony enlarges and forms water channels through which the bacteria gain nutrients and excrete waste. The community is extremely complex, where some bacteria are responsible for waste excretion, other bacteria are responsible for detecting changes in the outside environment, and other bacteria are responsible for producing the extracellular matrix. The bacteria within this complex community communicate, a process called quorum sensing, through various mechanisms, including differential gene expression and release of local cytokines. The biofilm has a characteristic appearance on scanning electron microscopy. The final step in the biofilm life cycle is the breaking off of pieces of the community to travel “downstream” to set up a new biofilm community. Biofilms are implicated in chronic infections of indwelling catheters, including line sepsis. The treatment is removal of the foreign body. In otolaryngology, biofilms have been proposed to cause chronic adenoiditis/ tonsillitis, chronic rhinosinusitis, and chronic OME.82 Tympanostomy tube otorrhea has been directly linked to biofilm formation on the tube.83 After the demonstration that biofilms form in the middle ear in an animal model of OME,84 biofilm formation has more recently been shown in the middle ear mucosa of children with chronic OME.85 Resistant to antibiotics, biofilms are a proposed mechanism for chronic OME. Biofilms are very difficult to eradicate because the extracellular polysaccharide matrix makes the biofilm impervious to white blood cells and antibodies. Not only is the biofilm resistant to cellular and humoral immunity, but it also blocks antibiotic penetration. Strategies to eradicate biofilm formation include tympanostomy tubes coated with antibiotic to prevent adhesion of bacteria, detergents to dissolve the polysaccharide matrix,
robiotics (replacing bad with good bacteria), and disp ruption of quorum sensing. Macrolide and quinolone antibiotics seem to be the most effective agents against biofilms. Further research into biofilms may uncover the pathophysiology of chronic OME and elucidate new strategies to prevent biofilm formation and to eradicate biofilms to reduce the impact and sequelae of OME.
GUIDELINES FOR THE MANAGEMENT OF ACUTE OTITIS MEDIA AND OTITIS MEDIA WITH EFFUSION In 2004, AHRQ and the American Academies of Pediatrics, Family Medicine, and Otolaryngology–Head and Neck Surgery published clinical practice guidelines for the management of AOM and chronic OME. The guidelines for the management of AOM recommend judicious use of antibiotics and promote observation as a viable management strategy (see Table 6-1).7 The AHRQ and Academies of Pediatrics, Family Medicine, and Otolaryngology–Head and Neck Surgery also published clinical practice guidelines for the management of children 2 months through 12 years with OME.8 These guidelines stress the importance of distinguishing OME from AOM; the role of pneumatic otoscopy in the diagnosis of OME; and the importance of documenting laterality, duration, and presence and severity of associated symptoms at each assessment of a child with OME. The guidelines recommend that clinicians should distinguish the child with OME who is at risk for speech, language, or learning problems from other children and recognize their need for more prompt intervention. Children not at risk can be managed with watchful waiting for 3 months from the onset of effusion; antihistamines, decongestants, antimicrobials, and steroids do not have long-term efficacy, and are not recommended for the routine treatment of OME. Hearing and language testing are recommended for children with OME that persists longer than 3 months, or at anytime a significant hearing loss or language delay is suspected.8 Children with persistent OME should be observed at 3 to 6 month intervals until the effusion clears, significant hearing loss (>30 dB hearing level in the better hearing ear) is identified, or structural abnormalities of the eardrum or middle ear are suspected. When a child becomes a surgical candidate, tympanostomy tube insertion is the preferred initial procedure. Repeat surgery should include adenoidectomy with myringotomy, with or without tube insertion. No recommendations were made with regard to alternative medicine or allergy management.8 These evidence-based evaluations and recommendations are expected to sharpen the focus on managing AOM and OME, and are likely to result in more careful use of antimicrobial therapy with less bacterial resistance and better patient outcomes.
Chapter 6 • Surgery of Ventilation and Mucosal Disease
Acknowledgments We would like to thank George A. Gates, M.D., author of this chapter in the previous edition. We would also like to thank Liz Gnerre for her assistance on this chapter.
REFERENCES 1. Gates G A : Cost-effectiveness considerations in otitis media treatment. Otolaryngol Head Neck Surg 114: 525-530, 1996. 2. Schappert S M : Office visits for otitis media: United States, 1975-90. Adv Data 214:1-19, 1992. 3. Kleinman LC, Kosecoff J, Dubois RW, Brook R H : The medical appropriateness of tympanostomy tubes proposed for children younger than 16 years in the United States. JAMA 271:1250-1255, 1994. 4. Gates G A, Avery C A, Prihoda TJ, Cooper JC : Effectiveness of adenoidectomy and tympanostomy tubes in the treatment of otitis media with effusion. N Engl J Med 317:1444-1451, 1987. 5. Gates G A, Muntz H R , Gaylis B : Adenoidectomy and otitis media. Ann Otol Rhinol Laryngol 101:24-32, 1992. 6. Stool S E, Berg AO, Berman S, et al: Managing Otitis Media with Effusion in Young Children: Quick Reference Guide for Clinicians. AHCPR Publication No. 940623. Rockville, MD, Agency for Health Care Policy and Research, Public Health Service, U.S. Department of Health and Human Services, July 1994. 7. American Academy of Pediatrics Subcommittee on Management of Acute Otitis Media: Diagnosis and management of acute otitis media. Pediatrics 113:1451-1465, 2004. 8. American Academy of Family Physicians: American Academy of Otolaryngology-Head and Neck Surgery, American Academy of Pediatrics Subcommittee on Otitis Media with Effusion: Otitis media with effusion. Pediatrics 113:1412-1429, 2004. 9. Bluestone C D: Otitis media: A spectrum of diseases. In Lalwani A K, Grundfast K M (eds): Pediatric Otology and Neurotology. Philadelphia, Lippincott-Raven, 1998, pp 233-240. 10. Casselbrant M L , Mandel E M, Kurs-Lasky M, et al: Otitis media in a population of black American and white American infants, 0-2 years of age. Int J Pediatr Otorhinolaryngol 33:1-16, 1995. 11. Teele DW, Klein JO, Rosner B A : Greater Boston Otitis Media Study Group: Epidemiology of otitis media during the first seven years of life in children in Greater Boston: A prospective cohort study. J Infect Dis 160:83-94, 1989. 12. Senturia B H, Bluestone C D, Klein JO, et al: Report of the Ad Hoc Committee on Definition and Classification of Otitis Media and Otitis Media with Effusion. Ann Otol Rhinol Laryngol 89(Suppl 68):3-4, 1980. 13. Kenna M A : Otitis media with effusion. In Bailey B J (ed): Head and Neck Surgery-Otolaryngology. Philadelphia, Lippincott, 1993, pp 1592-1606. 14. Teele DW, Klein JO, Rosner B : Greater Boston Otitis Media Study Group: Middle ear disease in the practice of pediatrics: Burden during the first five years of life. JAMA 249:1026-1029, 1983.
89
15. Teele DW, Klein JO, Rosner B : Epidemiology of otitis media in children. Ann Otol Rhinol Laryngol 89(Suppl 68):5-6, 1980. 16. Schwartz R H, Rodriguez WJ, Grundfast K M : Duration of middle ear effusion after acute otitis media. Pediatr Infect Dis J 3:204-207, 1984. 17. Shurin PA, Pelton S I, Turczyk VA, et al: Persistence of middle ear effusion after acute otitis media. N Engl J Med 300:1121-1123, 1979. 18. Fireman P: Otitis media and Eustachian tube dysfunction: Connection to allergic rhinitis. J Allergy Clin Immunol 99:787-797, 1997. 19. Henderson FW, Giebink G S : Otitis media among children in day care: Epidemiology and pathogenesis. Rev Infect Dis 8:533-538, 1986. 20. Bluestone C D, Stephenson J S, Martin L M : Ten-year review of otitis media pathogens. Pediatr Infect Dis J 11(8 Suppl):S7-S11, 1992. 21. Bluestone C D, Klein JO: Otitis media, atelectasis, and Eustachian tube dysfunction. In Bluestone C D, Stool S E (eds): Pediatric Otolaryngology. 3rd ed. Philadelphia, Saunders, 1996, pp 388-582. 22. Kenna M A, Bluestone C D: Microbiology of chronic suppurative otitis media. Pediatr Infect Dis J 5:223-225, 1986. 23. Papastavros T, Giamarellou H, Varlejides S: Role of aerobic and anaerobic microorganisms in chronic suppurative otitis media. Laryngoscope 96:438-442, 1986. 24. Grijalva CG, Griffin M R : Population-based impact of routine infant immunization with pneumococcal conjugate vaccine in the USA. Expert Review of Vaccines 7(1):83-95, 2008. 25. Greenberg D, Hoffman S, Leibovitz E, Dagan R : Acute otitis media in children: association with day care centers--antibacterial resistance, treatment, and prevention. Paediatr Drugs 10(2):75-83, 2008. 26. Pichichero M E, Casey J R : Evolving microbiology and molecular epidemiology of acute otitis media in the pneumococcal conjugate vaccine era. Pediatr Infect Dis J 26(10 Suppl):S12-6, 2007. 27 Post JC, Preston R A, Aul JJ, et al: Molecular analysis of bacterial pathogens in otitis media with effusion. JAMA 273:1598-1604, 1995. 28. Howie VM, Ploussard J H : Simultaneous nasopharyngeal and middle ear exudate cultures in otitis media. Pediatr Digest 13:31-35, 1971. 29. Pillsbury HC III, Kveton J F, Sasaki CT, Frazier W: Quantitative bacteriology in adenoid tissue. Otolaryngol Head Neck Surg 89:355-363, 1981. 30. Bluestone C D, Paradise J L, Beery QC: Physiology of the Eustachian tube in the pathogenesis and management of middle ear effusions. Laryngoscope 82:1654-1670, 1972. 31. Liu YS, Lang RW, Lim DJ: Microorganisms in chronic otitis media with effusion. Ann Otol Rhinol Laryngol 85:245-249, 1976. 32. Sade J: The natural history of the secretory otitis media syndrome. In Sade J (ed): Secretory Otitis Media and Its Sequelae. New York, Churchill Livingstone, 1979, pp 89-101. 33. Tos M : Production of mucus in the middle ear and Eustachian tube: Embryology, anatomy, and pathology of the mucous glands and goblet cells in the Eustachian tube and middle ear. Ann Otol Rhinol Laryngol 83(Suppl 11): 44-58, 1974.
90
OTOLOGIC SURGERY
34. Gates G A : Surgery of ventilation and mucosal disease. In Brackmann D E, Shelton C, Arriaga M A (eds): Otologic Surgery. Philadelphia, WB Saunders, 1994, p 86. 35. Borish L : Allergic rhinitis: Systemic inflammation and implications for management. J Allergy Clin Immunol 112:1021-1031, 2003. 36. Viegas M, Gomez E, Brooks J, Davies R J, et al: Changes in nasal mast cell numbers in and out of the pollen season. Int Arch All Appl Immunol 82:275-276, 1987. 37. Tomonaga K, Kurono Y, Mogi G: The role of nasal allergy in otitis media with effusion: A clinical study. Acta Otolaryngol Suppl (Stockh) 458:41-47, 1988. 38. Baroody FM : Allergic rhinitis: Broader disease effects and implications for management. Otolaryngol Head Neck Surg 128:616-631, 2003. 39. Bernstein J M : Allergic disease and the middle ear. In Krouse J H, Chadwick SJ, Gordon B R , Derebery M J, et al (ed): Allergy and Immunology. An Otolaryngologic Approach. Philadelphia, Lippincott Williams & Wilkins, 2002, pp 192-200. 40. Doyle WJ, Friedman R , Fireman P, Bluestone C D: Eustachian tube obstruction after provocative nasal antigen challenge. Arch Otolaryngol 110:508-511, 1984. 41. Hardy S M, Heavner S B, White D R , et al: Late-phase allergy and Eustachian tube dysfunction. Otolaryngol Head Neck Surg 125:339-345, 2001. 42. Pollock HW, Ebert C S, Dubin MG, et al: The role of soluble interleukin-4 receptor and interleukin-5 antibody in preventing late-phase allergy-induced Eustachian tube dysfunction. Otolaryngol Head Neck Surg 127:169-176, 2002. 43. Ebert C S Jr, Pollock HW, Dubin MG, et al: Effect of intranasal histamine challenge on Eustachian tube function. Int J Pediatr Otorhinolaryngol 63:189-198, 2002. 44. Hurst DS, Venge P: Levels of eosinophil cationic protein and myeloperoxidase from chronic middle ear effusion in patients with allergy and/or acute infection. Otolaryngol Head Neck Surg 114:531-544, 1996. 45. Hurst DS, Venge P: Evidence of eosinophil, neutrophil, and mast-cell mediators in the effusion of OME patients with and without atopy. Allergy 55:435-441, 2000. 46. Sobol S E, Taha R , Schloss M D, et al: T(H)2 cytokine expression in atopic children with otitis media with effusion. J Allergy Clin Immunol 110:125-130, 2002. 47. Wright E D, Hurst D, Miotto D, et al: Increased expression of major basic protein (MBP) and interleukin-5 (IL5) in middle ear biopsy specimens from atopic patients with persistent otitis media with effusion. Otolaryngol Head Neck Surg 123:533-538, 2000. 48. Bernstein J M : Role of allergy in Eustachian tube blockage and otitis media with effusion: A review. Otolaryngol Head Neck Surg 114:562-568, 1996. 49. Lazo-Saenz LG, Galvan-Aguilera A A, Martinez-Ordaz V.A., et al: Eustachian tube dysfunction in allergic rhinitis. Otolaryngol Head Neck Surg 132:626-631, 2005. 50. Bernstein J M, Doyle WJ: Role of IgE-mediated hypersensitivity in otitis media with effusion: Pathophysiologic considerations. Ann Otol Laryngol Suppl 16:15-19, 1994. 51. Braunstahl GJ, Overbeek S E, Kleinjan A, et al: Nasal allergen provocation induces adhesion molecule expression and tissue eosinophilia in the upper and lower airways. J Allergy Clin Immunol 107:469-476, 2001.
52. Kovatch A L , Wald E R , Michaels R H : β-Lactamaseproducing Branhamella catarrhalis causing otitis media in children. J Pediatr 102:261-264, 1983. 53. McCracken G H : Management of acute otitis media with effusion. Pediatr Infect Dis J 7:442-445, 1988. 54. Gebhart D E : Tympanostomy tubes in the otitis mediaprone child. Laryngoscope 91:849-866, 1981. 55. Paradise J L , Bluestone C D, Rogers K D, et al: Efficacy of adenoidectomy for recurrent otitis media in children previously treated with tympanostomy tube placement. JAMA 263:2066-2073, 1990. 56. Mandel E M, Rockette H E, Bluestone C D, et al: Efficacy of amoxicillin with and without decongestantantihistamine for otitis media with effusion in children. N Engl J Med 316:432-437, 1987. 57. Stillwagon PK, Doyle WJ, Fireman P: Effect of an antihistamine/decongestant on nasal and Eustachian tube function following intranasal pollen challenge. Ann Allergy 58:442-446, 1987. 58. Combs JT: The effect of montelukast sodium on the duration of effusion of otitis media. Clin Pediatr 43:529-533, 2004. 59. Lambert PR : Oral steroid therapy for chronic middle ear effusion: A double-blind crossover study. Otolaryngol Head Neck Surg 95:193-199, 1986. 60. Juntti H, Tikkanen S, Kokkonen J, et al: Cow’s milk allergy is associated with recurrent otitis media during childhood. Acta Otolaryngol 119:867-873, 1999. 61. Hall L J, Lukat R M : Results of allergy treatment on the Eustachian tube in chronic serous otitis media. Am J Otol 3:116-121, 1981. 62. Draper WL : Secretory otitis media in children: A study of 540 children. Laryngoscope 77:636-653, 1967. 63. Nsouli TM, Nsouli SM, Linde RE, et al: Role of food allergy in serous otitis media. Ann Allergy 73:215-219, 1994. 64. Barenkamp SJ, Kurono Y, Ogra PL , et al: Recent advances in otitis media, 5: Microbiology and immunology. Ann Otol Rhinol Laryngol Suppl 194:60-85, 2005. 65. Fernandez A A, McGovern J P: Secretory otitis media in allergic infants and children. South Med J 58:581-586, 1965. 66. Psifidis A, Hatzistilianou M, Samaras K, et al: Atopy and otitis media in children. In Ruben R J, Karma PH (eds): Proceedings of the Seventh International Congress of Pediatric Otorhinolaryngology. Amsterdam, Elsevier Science, 1998, pp. 205. 67. Armstrong BW: A new treatment for chronic secretory otitis media. Arch Otolaryngol 69:653-654, 1954. 68. Maw A R : Chronic otitis media with effusion (glue ear) and adenotonsillectomy: A prospective randomized controlled study. BMJ 287:1586-1588, 1983. 69. Paradise J L , Bluestone C D: Adenoidectomy and chronic otitis media [Letter]. N Engl J Med 318:1470-1471, 1988. 70. Koyuncu M, Saka M M, Tanyeri Y, et al: Effects of otitis media with effusion on the vestibular system in children. Otolaryngol Head Neck Surg 120:117-121, 1999. 71. Golz A, Westerman T, Gilbert L M, et al: Effect of middle ear effusion on the vestibular labyrinth. J Laryngol Otol 105:987-989, 1991. 72. Jones NS, Radomsky P, Princhard ANJ, Snashell SE: Imbalance and chronic secretory otitis media in children: Effect of myringotomy and insertion of ventilation tubes on body sway. Ann Otol Rhinol Laryngol 99:477-481, 1990.
Chapter 6 • Surgery of Ventilation and Mucosal Disease 73. Goode R L : CO2 laser myringotomy. Laryngoscope 92:420-423, 1982. 74. Marchant H, Bisschop P: Value of laser CO2 myringotomy in the treatment of seromucous otitis. Ann OtoL aryngol Chir Cervico-Fac 115:347-351, 1998. 75. Gross CG, Bessila M, Lazar R H, et al: Adipose plug myringoplasty: An alternative to formal myringoplasty techniques in children. Otolaryngol Head Neck Surg 101:617-620, 1989. 76. Helmus C, Grin M, Westfall R : Same-day-stay adenotonsillectomy. Laryngoscope 100:593-596, 1990. 77. Green J D Jr, Shelton C, Brackmann D E : Surgical management of iatrogenic facial nerve injuries. Otolaryngol Head Neck Surg 111:606-610, 1994. 78. Aul JJ, Anderson KW, Wadowsky R M, et al: Comparative evaluation of culture and PCR for the detection and determination of persistence of bacterial strains and DNAs in the Chinchilla laniger model of otitis media. Ann Otol Rhinol Laryngol 107:508-513, 1998. 79. Liederman E M, Post JC, Aul JJ, et al: Analysis of adult otitis media: Polymerase chain reaction versus culture for bacteria and viruses. Ann Otol Rhinol Laryngol 107: 10-16, 1998.
91
80. Post JC, Aul JJ, White GJ, et al: PCR-based detection of bacterial DNA after antimicrobial treatment is indicative of persistent, viable bacteria in the chinchilla model of otitis media. Am J Otolaryngol 17:106-111, 1996. 81. Kenna MA: Bacteriology of otorrhea: Polymerase chain reaction versus cultures. In Proceedings of the 6th International Symposium on Recent Advances in Otitis Media. Toronto, BC Decker, 1996, pp 428-430. 82. Post JC, Stoodley P, Hall-Stoodley L , Ehrlich G D: The role of biofilms in otolaryngologic infections. Curr Opin Otolaryngol Head Neck Surg 12:185-190, 2004. 83. Jang C H, Cho YB, Choi C H : Structural features of tympanostomy tube biofilm formation in ciprofloxacin-resistant Pseudomonas otorrhea. Int J Pediatr Otorhinolaryngol 71:591-595, 2007. 84. Ehrlich G D, Veeh R , Wang X, et al: Mucosal biofilm formation on middle-ear mucosa in the chinchilla model of otitis media. JAMA 287:1710-1715, 2002. 85. Hall-Stoodley L, Hu FZ, Gieseke A, et al: Direct detection of bacterial biofilms on the middle-ear mucosa of children with chronic otitis media. JAMA 296:202-211, 2006.
7
Diagnosis and Management of the Patulous Eustachian Tube Dennis S. Poe and Ophir Handzel Videos corresponding to this chapter are available online at www.expertconsult.com.
A patient with a patulous or abnormally patent eustachian tube (ET) can experience symptoms that lead him or her to seek medical advice and possibly undergo surgery. This condition can be overlooked or mistakenly thought of as not having sufficient bearing on a patient’s quality of life to warrant efforts for its identification and correction. The ET has a pivotal role in the maintenance of proper volume and pressure homeostasis of the middle ear cleft, in expelling middle ear secretions, and in protecting the middle ear from reflux of sound and material from the pharynx.1 Dysfunction of the ET may be divided into two groups: dilatory dysfunction with failure to open the valve of the ET adequately, and patulous dysfunction with failure to close the valve adequately. This chapter addresses the diagnosis and treatment of an excessively open ET.
ANATOMY AND PATHOPHYSIOLOGY OF THE PATULOUS EUSTACHIAN TUBE The ET is an organ composed of a skeleton made of cartilage and bone and associated muscles, fat, connective and lymphoid tissues, nerves, and blood supply.2 The normal ET is maintained at a closed resting position, and opens by voluntary and nonvoluntary maneuvers, such as swallowing; yawning; and deliberate manipulations of the palate, pharynx, muscles of mastication, and mandible. The tube may also open passively by changes in ambient air pressure or by forcing passage of air, such as with a Politzer maneuver. As mentioned, tubal opening can be a by-product of movements initiated for other purposes (e.g., swallowing), but it can also be triggered by direct reflex stimuli. Gas is exchanged between the middle ear space, surrounding mucosa, and blood vessels. There is a net absorption of gas into the circulation; when the ET is closed, this causes an increasingly negative pressure beyond that of the normal resting pressure of the middle ear. The absorption rates of the constituents of air differ with nitrogen being the predominant gas, but having relatively slow absorption compared with oxygen and carbon dioxide. Therefore, the gas composition varies over time
after each dilation of the ET. It is thought that deviation from the set-points of pressure and gas composition may be detected by a combination of baroreceptors and chemoreceptors that initiate the opening of the ET to convey air into the middle ear and re-establish homeostasis.3 Active opening of the ET usually lasts about 400 ms. It is the result of a coordinated effort of four muscles, the most important of which is the tensor veli palatini (TVP), laterally dilating the anterolateral membranous wall.1,4,5 Although the opening mechanism of the ET has been studied extensively, closing of the ET has received less attention. The part of the ET that is normally closed at rest and open on dilation is referred to as the valve. The valve constitutes the approximately 5 mm long segment of apposing mucosal surfaces within the middle of the cartilaginous portion of the ET.6 Closure of the ET is thought to be the result of the following factors: recoiling memory properties of the cartilaginous ET, relaxing bulk of the TVP muscle, and pressure of neighboring paraluminal tissue. The closed lumen is likely maintained by all of these factors, and aided by the surface tension of the apposed wet mucosal surfaces. Transnasal endoscopy and surgery of patulous ETs has revealed that the underlying pathology is a longitudinal concave defect in the anterolateral wall of the valve (Fig. 7-1). This wall normally has a convex bulge into the lumen of the ET in the relaxed position. The defect represents a lack of tissue volume that can be a deficiency of the lateral cartilaginous lamina, Ostmann’s fat, TVP muscle bulk, submucosa, or mucosa. Examples of these deficiencies, either isolated or in combination, are seen in clinical cases. Temporary or permanent obliteration of the defect, even with some mucus, immediately relieves patients’ symptoms. The cartilaginous canal is anchored to the basisphenoid bone in its lateral side. The medial cartilaginous lamina is mobile, principally by action of the levator veli palatini muscle. Superiorly, the medial and much smaller lateral cartilaginous laminae come together to form a junction rich in elastin fibers that contributes to the springlike quality of the cartilage. The recoiling qualities of the cartilaginous canal play an important role in ET closure. 93
94
OTOLOGIC SURGERY
REF LEX DECAY ml Y 226H z ON: 10.0 35 .00 HL
.16 ml
I
1000 Hz OFF: 3.5
TEST 1 L d a Pa : – 1 7 0
.08
.00
.16 15 sec DECAY TIME (S E C ) :
8
FIGURE 7-1. Tympanometry tracing shows superimposition of a sawtooth pattern of tympanic membrane excursions resulting from breathing while the eustachian tube (ET) is patulous. Tympanometry shows an abnormality only while the patient is actively symptomatic.
Because the cartilaginous backbone of the ET is open in its medial-anterior aspect, the lumen patency also depends on the pressure of the abutting soft tissues. The relaxed paratubal muscles (mainly the TVP) and Ostmann’s fat pad contribute to lumen closure. Glandular tissue and the size of Ostmann’s fat pad were found to be smaller in patients with a patulous ET compared with normal controls as measured by computed tomography (CT) scans.7 The mucosal and submucosal tissue layers increase in thickness as they are relieved of tension with muscle relaxation, and are in themselves a factor in closure of the valve. A report of fluctuating patulous symptoms in hemodialysis patients describing relief of symptoms during fluid retention and exacerbation after excretion showed the close connection between periluminal mass and symptoms of patulous ET.8
ETIOLOGY AND CLINICAL PRESENTATION An overly patent ET allows for the free passage of air and the sound it carries from the nasopharynx to the middle ear, creating a pathologic acoustic and pressure environment. The most prominent symptoms of a patulous ET are aural fullness and autophony of a patient’s own breathing noises and voice. Symptoms are mostly related to free passage of air and sound, but not reflux of material. Symptoms compatible with patulous ET have been described since the second half of the 19th century. The constellation of symptoms compatible with those seen in patients with patulous ET was reported by Jago in 1858,9 who later described his personal experience with the condition. Schwartze recognized the connection between the patulous ET and excursions of the tympanic membrane with respiration. Zollner and Shaumbaugh emphasized the important symptom of autophony in a series of patients with patulous tubes (as reported by Bluestone and Magit10).
Autophony is described by patients as hearing their own voice and breathing noises. A patient’s own voice sounds to them as if they are talking very loudly into a barrel, typically prominent on pronunciation of “M” and “N.” Clinicians can simulate the symptoms by talking and breathing into the diaphragm of their stethoscopes. Because the auditory feedback can be disturbingly loud, there is a wide range of cognitive and emotional responses ranging from asymptomatic to suicidal ideation. Autophony can be disturbing to the extent of leading to depression and the need for psychiatric medication or intervention. The perception of pressure changes in the middle ear with breathing can be loud windlike noise, constant ear blockage or pressure, and the sense of the medial and lateral excursions of the tympanic membrane. When patients forcibly blow their nose, there can be pain from the increased middle ear pressure on the tympanic membrane. Aural fullness and the sensation of ear blockage can be misdiagnosed as dilatory dysfunction of the ET, especially because the patient usually describes the ear as “blocked.” The differential diagnosis for aural fullness also includes temporomandibular joint or related musculoskeletal dysfunction, Minor’s syndrome of semicircular canal dehiscence, and endolymphatic hydrops (see Fig. 7-3). Because of the typical complaint of “ear blockage,” many patients are initially treated with medications directed against dilatory dysfunction, such as nasal decongestants, steroid topical sprays, and topical or systemic antihistamines, all of which fail to help or even aggravate the symptoms. Symptoms are mitigated by maneuvers and conditions that support tubal closure by increasing venous congestion in tubal tissues, causing inflammation of the tissues, or restoring some hydration or mucus onto the luminal surfaces. Placing the head in a dependent position (supine or between the legs), pressure on the ipsilateral internal jugular vein, upper respiratory tract infections and allergies, and sniffing inward to “lock” the ET closed by creating negative middle ear pressure all are capable of generating temporary symptom relief in most patients. Symptoms are typically relieved overnight and may start sometime after arising in the morning. They are often initiated by exercise, possibly because of dehydration or epinephrine-like hormones causing nasal decongestion, or by prolonged speaking and singing, possibly from desiccation of the mucosa from air movement. The etiology for the loss of tissue creating a patulous ET is uncertain. Possibly associated factors such as weight loss and tissue atrophy from chronic disease account for only two thirds of the patients.11 Loss of tissue volume from within the tubal lumen is cited as the most common pathogenesis and most commonly reported with weight loss.12 Shea (personal communication, 2007) noted that one third of patients have a significant weight loss, and one third have some sort of rheumatologic condition. The authors’ experience has been similar.
Chapter 7 • Diagnosis and Management of the Patulous Eustachian Tube
95
Aural Fullness– Differential Diagnosis
Retracted tympanic membrane • Negative pressure or effusion • Type B or C tympanogram • Relief of symptoms with myringotomy +/– tube
Normal tympanic membrane
Patulous excursion of TM while symptomatic
Eustachian tube dilatory dysfunction
No TM excursion while symptomatic
Patulous ET
Minor’s Syndrome
TMJ Dysfunction
Inner Ear Hydrops
• Semicircular canal dehiscence syndrome • Conductive hearing loss • Supranormal bone conduction • +/– Vertigo • +/– Tullio’s phenomenon • CT confirmation • VEMP-abnormal low thresholds
• Tenderness in or around joint capsule with mouth wide open • Intraoral examination of lateral pterygoids for spasms or tenderness • Chronic or fluctuating pain/pressure, normal TM • Clenching or bruxism • Malocclusion
• Fluctuating low-frequency sensorineural hearing loss • +/– Episodic vertigo
FIGURE 7-2. Diagnostic algorithm for the evaluation of aural fullness.
Simonton13 grouped etiologies according to positive and negative contributing factors. Positive factors were defined as factors actively reducing tissue volume, such as with scarring from previous procedures, inflammation, and radiation.14,15 Negative factors were due to a passive loss of tissue around the pharyngeal orifice, loss of tonic action of the TVP muscle, and conceivably reduced coiling properties of the ET. Hormonal factors include pregnancy,16 high-dose oral contraceptives, and estrogen treatment for prostate cancer. Estrogen can reduce the viscosity of tubal secretions, reduce the elastic properties of the tubal cartilage, and elevate the level of surfactant via a change of prostaglandin levels. All these factors would support opening of the ET. Reflux of gastric contents and allergy were reported to be the cause in 3 of 11 patients in one series.17 Other contributors reported include abuse of nasal decongestants and cocaine, poliomyositis, multiple sclerosis and other neuromuscular disorders, cerebrovascular accidents, craniofacial abnormalities, temporomandibular joint malfunction, and malocclusion. Occasional association between patulous ET and palatal myoclonus has been reported.18 The habit of sniffing or Valsalva’s maneuver can be acquired in patients attempting to aerate their middle ears at times of otitis media, but may ultimately result in a patulous tube.
DIAGNOSIS The diagnosis is certain when medial and lateral excursion of the tympanic membrane with breathing is observed. Otoscopy should be done with an otoscope or microscope and the patient seated. If the patient does not have autophony at the time of the examination, it may be induced by physical activity. Excursions of the tympanic membrane can be enhanced by closure of the contralateral nostril during nasal breathing. The improved inspection and manipulation of the nasopharynx brought about with the introduction of endoscopic sinus surgery tools greatly improved the ability to diagnose and correct patulous ET. High-quality endoscopy allows for direct and detailed examination of the ET valve. Done with either a rigid or a flexible endoscope, it may show a concave defect in the anterolateral part of the valve, instead of the normal convexity.19 The defect prevents normal closure of the orifice in the resting position, leaving a slitlike opening in the lumen of the orifice. Ancillary tests may be of some assistance in the diagnosis of patulous ET. Impedance tympanometry while the patient experiences autophony may reveal fluctuation in tracing in line with tympanic membrane excursion (Fig. 7-2). Tympanometry can also record middle ear negative
96
OTOLOGIC SURGERY
pressure or hypermobility of the tympanic membrane. Sonotubometry can directly measure ET patency.20,21 During the examination, a sound is emitted in the nasal cavity and is recorded by a microphone located in the external auditory canal of the examined ear. As the ET opens, the sound recorded in the external auditory canal intensifies, and an increase of 5 dB or more is considered to reflect opening of the ET reliably.22 Severity of subjective autophony was found to be correlated with objective measurements of ET function by sonotubometry.23 This test is not available routinely in the clinical setting. It has been suggested more recently to make use of this principle in performing audiometry masked by a nasally presented noise. In subjects with patulous tubes, thresholds of low tones were significantly elevated more than in normal controls because sound presented in the ear was masked by the noise escaping the nasal cavity to the middle ear through the patent ET. The thresholds normalized in most patients after corrective measures were taken.24 Imaging is not routinely used for the diagnosis of patulous ET. Until more recently, CT scanners required a recumbent examinee, a position that commonly results in closure of the patulous tube. The availability of scanners allowing examinations in the seated position has allowed researchers to record patency of the ET in resting and under Valsalva maneuver.25 Imaging-based measurements of a group of patients with patulous ETs were statistically significant compared with normal controls. To date, the diagnosis of patulous ET is mainly a clinical one and is based on history and physical examination. Patulous excursions of the tympanic membrane during nasal breathing are pathognomonic for the condition, but this is generally seen only while symptoms are active. Patients may be asked to run up and down stairs for several minutes in an attempt to generate the symptoms when not initially present. Patulous symptoms and signs (e.g., tympanic membrane excursions) are present only while the ET valve is stuck in the open position. If patients complain that they actively have their fullness and autophony symptoms, but there are no patulous tympanic membrane excursions and no visible ET valve defect at that moment, another cause for the symptoms must be sought. Ancillary tests may be important to distinguish this pathology from others sharing similar symptoms. The other causes for aural fullness or ear blockage should be systematically considered (Fig. 7-3). ET dilatory dysfunction causes negative pressure in the middle ear or effusion that should be visible on otologic examination and tympanometry. Endoscopic examination reveals evidence of an obstructive problem with mucosal swelling or inflammation, or a dynamic disorder with failure of muscular dilation.19 The diagnosis of semicircular canal dehiscence syndrome, or Minor’s syndrome, must be considered before making a final diagnosis of patulous ET (of which most are dehiscences of the tegmen into the superior
semicircular canal).26 These two very different pathologies can manifest with similar symptoms of autophony of the voice, but autophony of breathing sounds is much more pronounced in patulous ET.17 Differential diagnosis is made more difficult by the fact that superior semicircular dehiscence (SSCD) may manifest without vestibular symptoms.27 Older series of patients assumed to have a patulous ET most likely included patients who had Minor’s syndrome. Almost 20 years before the first description of SSCD syndrome by Minor and associates,28 O’Connor and Shea14 described some patients thought to have patulous ET who experienced vertigo accompanied by nystagmus induced by a sharp increase in nasopharyngeal and intratympanic pressure. They attributed symptoms and signs to stapedial or round window membrane movement, although in retrospect, it is highly likely that the patients had Minor’s syndrome. Of the senior author’s series of patients with SSCD, 94% complained of autophony of their voice, and half of them found relief by placing their head in a dependent position.17 Relief of autophony while lying supine has long been thought to be pathognomonic for patulous ET, but changes with intracranial or intralabyrinthine pressure with the head dependent probably also influence the impedances of the dehiscences that cause Minor’s syndrome. Patients with Minor’s syndrome tend to have unremitting autophony when it becomes present, as opposed to symptoms of patulous ET, which are generally intermittent. Vertigo and nystagmus induced by sound or pressure, supranormal bone conduction, and conductive hearing loss with intact stapedial reflexes all should raise the suspicion of SSCD. The diagnosis is confirmed by demonstration of the bone dehiscence with high-resolution temporal bone scan reconstructed in the Poschel and Stenvers planes, and by lower than normal thresholds in Vestibular Evoked Myogenic Potential. Patients with aural fullness complaints but normal tympanic membrane findings should be asked about and evaluated for temporomandibular joint and musculoskeletal disorders. Aural fullness with otalgia is especially suggestive of these problems. The joint capsule is examined with the patient opening the mouth as widely as comfortable; the space between the condyle and the glenoid fossa is palpated deeply looking for any tenderness. The mouth is moved side to side during a portion of the examination, again palpating for areas of tenderness, especially superiorly and posteriorly to the condyle. Deep palpation is made in the temporalis and masseter muscles surrounding the temporomandibular joint, examining for spasms and tenderness. Bimanual examination of the lateral pterygoid muscle is done to look for spasms and tenderness. If a patient does not have active symptoms during the office visit, it is likely that no findings will be elicited, so the patient is instructed in how to do a self-examination during episodes of future symptoms, especially if there is aural pain. Inner ear conditions such
Chapter 7 • Diagnosis and Management of the Patulous Eustachian Tube
as Meniere’s disease can cause aural fullness. Patients would characteristically have a fluctuating sensorineural hearing loss and vertigo. Appropriate studies would be indicated if Meniere’s disease is suspected.
TREATMENT Treatment should be designed in a stepwise fashion. In some patients, the specific etiology can be determined, and treatment can be initiated. Often a correctable etiology cannot be found. For most patients, reassurance suffices. Others find relief by swallowing, yawning, or positioning their heads in a dependent position. Patients are advised to discontinue decongestants and nasal steroids, observe good hydration, and use nasal saline drops, which are more effective than sprays. For the subset of patients experiencing unwarranted weight loss, appropriate evaluation is recommended. When indicated, treatment is directed toward augmenting the periluminal tissue, specifically filling the defect in the valve. This can be done by adding bulk to the native tissue by inducing congestion or by introducing mass from other sources (autografts, allografts, xenografts, or synthetic implants). Numerous agents affecting the mucosal lining of the valve can be used with good, albeit temporary, results. In an off-label use, a conjugated estrogen preparation (Premarin) can be administered as a nasal solution resulting in mucosal edema.29 A dose of 25 mg of intravenous Premarin is diluted in 30 mL of isotonic sodium chloride to be taken as nasal drops three times a day. Side effects include epistaxis and nasal irritation. Saturated solution of potassium iodide (SSKI), an expectorant designed to enhance viscosity of the mucus, is taken three times a day as 8 to 10 drops diluted in water or juice. Powder of boric and salicylic acid (4:1 ratio) insufflated to the nasopharynx or instilled with a catheter causes local irritation and edema offering temporary relief.12 Insufflation of mucosal irritants can also be used as a diagnostic test or temporary treatment for patulous ET. Some local irritants reported have included the application of silver nitrate, nitrate acid, and phenol.11,29 These irritants can also scar the mucosa and restrict the opening of the ET. Although many of these treatments brought about improvements in patients’ symptoms, the improvements were transient. Surgical intervention is reserved for the few patients with symptoms significantly affecting their quality of life, when less intrusive methods of treatment fail. Myringotomy with the placement of ventilation tubes is probably the most common surgical intervention, reported to aid approximately 50% of patients.30 Ventilation tubes alleviate mostly symptoms related to middle ear pressure changes and tympanic membrane movements, but less so reflux of sound—autophony.17 Various materials have been injected in an effort to fill the orifice defect responsible for creation of patulous
97
ET. In 1937, Zollner31 proposed infiltrating paraffin anterior to the ET orifice, in an attempt to recreate the missing mass abutting the orifice. Pulec12 reported only 1 patient out of 26 not helped by injecting polytef (Teflon) to the anterior part of the orifice; 19 patients had complete relief from symptoms after one injection of 0.75 to 1.5 cm3 of Teflon paste. None of the patients had otitis media after the procedure. The same material was used by O’Connor and Shea,14 who injected Teflon anterior to the orifice. Use of Teflon for this indication was discontinued because of complications, including cerebral thrombosis and death, possibly as a result of injection of Teflon into the immediately adjacent internal carotid artery. Ogawa and colleagues32 reported their successful, albeit short-lived, experience with injection of absorbable gelatin sponge (Gelfoam) mixed with glycerin to the lumen of the ET. Brookner and Pulec33 studied a dog model and pointed to the anterior wall of the pharyngeal orifice of the ET as the correct location for augmentation. Injection of Teflon paste into this area resulted in an increased resistance in airflow through the ET without initiating otitis media. Injecting the posteromedial wall of the posterior cushion (torus tubarius) resulted in elevated resistance of the tube as well, but three of four injected dogs developed serous otitis media. Impediment of lymphatic flow was thought to be the likely cause of this side effect. Complete blockage of the ET cures patients from symptoms of patulous ET, but usually results in the need for long-term or permanent ventilation tubes. Simonton13 reported partial success in three cases using conization of the ET cartilage and suturing the mucosa closed. Doherty and Slattery11 used circumferential electric coagulation and application of a fat graft to seal the tube successfully with symptomatic improvement in two patients. A different approach to obstruct the patent ET is by placing an intraluminal catheter. Plugging the ET with a catheter sealed with bone wax that was introduced through a tympanotomy or myringotomy approach resulted in long-term resolution of symptoms.34 An anteriorly based tympanomeatal flap was raised, and if required the bony annulus would be drilled to facilitate exposure of and approach to the ET orifice. An intravenous catheter filled with bone wax was introduced into the lumen of the ET through the middle ear. Nine patients, who failed previous medical and surgical interventions, were treated by insertion of an intraluminal catheter through the tympanotomy approach and placement of tympanostomy tubes, and were followed for 2 to 15 years. Six patients had marked improvement in symptoms. Two patients extruded their catheter and tympanostomy tubes, but still obtained relief of symptoms. Two patients extruded the tympanostomy tubes, and although the catheters were in place did not experience middle ear effusion, probably because of sufficient leak of air around the catheter.10 Intraoperative manometric measurement was employed by Bluestone1 with a tympanometer noting
98
OTOLOGIC SURGERY
that opening pressure of the ET had been increased to 500 to 700 mm (Figs. 7-4 through 7-7). Shea Jr. and associates at the Shea clinic offer a similar transtympanic procedure: introducting of one or more Teflon catheters of various appropriate sizes. According to information available from the Shea clinic’s website www.sheaclinic. com, of 40 patients operated on, 90% were improved, and 5% were made worse; in 5%, the symptoms were unchanged. Release of the TVP muscle by transposing it medially to the hamulus relieved symptoms for at least 6 months in 9 of 10 patients undergoing this procedure.35 The incidence of postoperative otitis media was not reported. Virtanen and Palva15 published their experience with pterygoid hamulotomy. Release of the tendon of the TVP was supplemented at times with cutting its tendon if deemed necessary. Clinical relief was achieved in 11 of 16 of the operated patients, although sonotubometry recording normalized in only 9 patients.
TECHNIQUE OF TRANSNASAL AND TRANSORAL OCCLUSION OF THE PATULOUS EUSTACHIAN TUBE DEFECT Repair of the patulous defect can be done less invasively through the nasopharyngeal orifice without tympanotomy (Figs. 7-8 through 7-14). An intravenous catheter can be inserted into the tubal lumen wedged into the isthmus, the narrowest segment of the ET for long-lasting relief of patulous symptoms while generally maintaining the patency of the tube when it dilates normally to the open position. The physiologic function of the ET can be preserved in most cases, while repairing the patulous defect in the valve. The ipsilateral nasal cavity is sprayed with pseudoephedrine-containing nasal drops. The mandible is retracted with a tonsillectomy mouth gag. A soft red rubber catheter is passed through the contralateral nasal passage and used to retract the soft palate. The nasopharynx is seen with a 45 degree rigid telescope introduced through the ipsilateral naris. An intravenous catheter, 14 or 16 gauge (2.1 or 1.7 mm diameter), is prepared, cutting it to 35 to 38 mm length, preserving the tapered tip; the lumen is obliterated full length with bone wax by pouring it melted into the lumen before cutting the catheter to size. The length is determined by the size of a patient’s head and by CT scan. The CT scan is inspected for any dehiscence of the internal carotid artery into the ET. The catheter is loaded into a custom-made piston within a tube introducer (see Fig. 7-13).17 The catheter is held in place by a small piece of bone wax having been initially placed into the introducer and against the mobile piston. The introducer with the catheter is passed through the oral cavity and oropharynx, and is brought to the proximity to the nasopharyngeal orifice of the ET under view with the endoscope. The catheter is pushed
into the lumen of the ET with the piston-like movement of the introducer; engagement of the catheter into the isthmus is felt as heightened resistance. Care is taken to ensure smooth intraluminal placement without any mucosal injury. Creation of a false passage into the soft tissues could risk inadvertent penetration of the internal carotid artery. Otoscopy at the end of the procedure with a microscope or a 0 degree endoscope is done to ascertain that the catheter is not visible within the middle ear. If it is seen within the middle ear, it is withdrawn further into the nasopharynx, and the ear is reinspected to ensure it is no longer visible. The catheter should protrude slightly from the nasopharyngeal orifice, but not so long as to protrude into the palate when it elevates with speaking and swallowing. The catheter should not protrude into the nasopharynx beyond the inferior edge of the anterior cushion, the free end of the anterolateral wall. Alternatively, the defect in the ET valve can be augmented. Knowledge of the exact area of the defect is the key to good results. Correction of this defect can restore the competence of the tubal valve, avoiding overcorrection and otitis media with effusion. The patulous defect has been identified as a loss of bulk in the valve within the anterolateral wall, superiorly. The senior author has injected various materials (e.g., fat, collagen, hydroxyapatite) into this location in patients. Introduction of 0.3 cm3 resulted in immediate alleviation of symptoms in all patients, but most results tend to be temporary with return of symptoms 2 weeks to several months later in most patients.17 Because hydroxyapatite is radiopaque, CT scans of patients injected with this material have illuminated the fate of injectable materials. When injected into the anterolateral defect of the valve, hydroxyapatite gained access to the musculofascial plane superficial to the TVP, and spread superiorly and inferiorly up to the basisphenoid and down to the parapharyngeal space. This finding offered explanations for the usual transient relief experienced by injected patients. Injecting the posteromedial part of the orifice (posterior cushion), the hydroxyapatite remained in place. Correcting the defect of the orifice with this type of injection is difficult, however, requiring a larger volume of material than the submucosal tissue is likely to hold. Additionally, inadvertent passage of the needle through the medial cartilaginous lamina within the posterior cushion brings the injected material to the internal carotid artery. In search of a material that can be safely used to fill the valve’s defect with stable long-term results, the senior author used acellular dermal matrix (AlloDerm) and cartilage grafts to augment the defect in a patulous ET reconstruction (PETR) procedure.17 AlloDerm is typically resorbed by about 65% and is no longer used because of a higher long-term failure rate than cartilage grafts. The graft is placed into a luminal, submucoperichondrial pocket made in the superior half of the ET orifice.
Chapter 7 • Diagnosis and Management of the Patulous Eustachian Tube
99
100
OTOLOGIC SURGERY
Skin incision
Cochal cartilage graff 1 mm
3–4 mm
FIGURE 7-8. Patulous eustachian tube reconstruction with cartilage grafts. Thick cartilage is harvested from the nasal septum or inferomedial aspect of the conchal cartilage. It is cut into wedges 1 mm wide at the apex, 3 mm wide at the base, and 3 to 4 mm long.
3 mm
Superior
Lateral cartilaginous lamina
Adenoid
3:00 Medial
FIGURE 7-9. Superior semicircular incision is made on the leading edge of the eustachian tube nasopharyngeal orifice from 9 to 3 o’clock.
Lateral
9:00
Posterior cushion (torus tubaris)
The procedure is done using transoral and endoscopic transnasal approaches, with the patient supine and under general anesthesia. The nose is decongested by topical application of phenylephrine solution, and a mixture of lidocaine 1% and epinephrine 1:100,000 is injected by an
Medial cartilaginous lamina
angled needle into the nasopharyngeal orifice of the ET. The cartilage graft can be taken from the tragus (with the posterior perichondrium attached), concha, or nasal septum (without perichondrium). The mouth is opened with a mouth gag, and the soft palate retracted with a soft
Chapter 7 • Diagnosis and Management of the Patulous Eustachian Tube
rubber catheter, introduced through the contralateral nose. The nasopharynx is viewed with a 45 degree rigid endoscope with a camera. The endoscope is attached to
FIGURE 7-10. Sharp and blunt dissection is done to create an intraluminal submucoperichondrial pocket. Great care is taken to avoid tears in the flap. A suction dissector is particularly helpful in developing this flap well into the eustachian tube valve area inferior/distal to the bonycartilaginous isthmus.
101
a gag-mounted endoscope holder. All instruments other than the endoscope are passed through the oral cavity. The incision in the mucosa is made with an otologic diode pumped KTP laser (Iridex Corporation, Mountain View, CA) set at 2500 mW/1000 mS. Standard otologic fiberoptic laser probes (200 μ diameter) are used with the probe gently curved to reach the ET orifice through the mouth. The incision is made into the mucosa of the anterior cushion extending superiorly from 9 to 3 o’clock. The incision is carried down until the superior cartilage is encountered. A submucoperichondrial plane is developed superiorly and carried deeply into the valve, avoiding tearing the flap. The plane is followed superiorly and distally into the level of the tubal valve. The flap may include a segment of the ET cartilage, which for this purpose is separated from the basisphenoid bone and from the lateral and medial cartilaginous laminae. If the cartilage is excessively thick, it is thinned with a sharp knife or laser dissection. The cartilage grafts are cut into wedges approximately 1 mm wide at the apex and 3 to 4 mm wide at the base and 5 to 8 mm long. The grafts are firmly packed deeply into the pocket, usually using two to four grafts. There should be sufficient mucosal coverage of the grafts to allow the edges to be sutured together.
3 2
1
Grafts
Cartilaginous ET skeleton
FIGURE 7-11. Cartilage grafts are wedged, apex first, into the pocket, creating a convexity to the anterolateral wall and depressing the superior wall inferiorly. Wedges are inserted and rotated to make them stable in position. Delayed extrusion may occur if the grafts are held in place under tension against the suture closure. ET, eustachian tube.
102
OTOLOGIC SURGERY
A
Olive tipped curved antral suction utilized as a knot pusher.
B
C FIGURE 7-12. A, Technique of suture closure using needle driver to engage the medial mucosal edge. B, Suture is grasped just behind the needle’s swedge to allow the needle to be toggled into position for insertion into the lateral mucosal edge. C, Olive-tipped maxillary antral suction is used as a knot pusher.
The incision is closed with two or three 4-0 polyglactin 910 (Vicryl) sutures on a TF needle shaped into a J toward the tip to allow for easy passage through the lateral mucosal edge after first traversing the medial edge. At least one suture is passed through one of the cartilage
grafts. The suture is tied through the oral cavity, passing one limb of the suture into a curved olive-tipped maxillary antral suction while the vacuum is engaged. The suction tubing is removed, and the threaded suture limb is maintained on tension as the suction catheter is
Chapter 7 • Diagnosis and Management of the Patulous Eustachian Tube
103
A
B FIGURE 7-13. Nasopharyngeal eustachian tube catheter insertion. A, Eustachian tube insertion tool is loaded with a precut angiocatheter prefilled with bone wax that is held within the insertion tool by a small plug of bone wax against the internal piston. Catheter is smoothly inserted, with some resistance noted as it engages the isthmus. B, Catheter is inserted within the orifice just past the posterior cushion to avoid contact with the cushion or palate during swallowing and speaking. It is pushed into place with a curved empty laser catheter guide or a frontal sinus ball-tipped probe.
advanced to serve as a knot pusher. If the patient is to fly within 6 weeks of the procedure, a ventilation tube may be placed in the drum. Initially taking about 3.5 hours, operative time was later reduced to 1.5 to 2 hours (see Figs. 7-8 through 7-14). The results of PETR performed in 14 ETs of 11 patients have been published.17 All patients had longstanding (1 to 38 years) patulous ET that failed previous therapies. Eight patients had a trial of myringotomy and placement of tubes, and two had undergone previous ET procedures at other centers. The most common and most disturbing symptom was autophony. All patients experienced immediate and complete relief after the operation. Follow-up duration ranged from 3 to 30 months (average 15.8 months). Symptoms were completely relieved in one patient, significantly improved to
satisfaction in five patients, and significantly improved but with patient’s dissatisfaction in seven patients; in only one patient were symptoms unchanged. No complications were experienced. One patient required a supplementary hydroxyapatite injection, following which complete relief of autophony persisted for another 22 months. Presently, we offer the minimally invasive catheter insertion procedure as the primary procedure (see Fig. 7-14). In the event of failure, we may offer insertion of a larger or a second catheter or a cartilage PETR. The catheter approach can also be used after failure of the PETR. Ultimately, obliteration of the ET and long-term tympanostomy tube can be done for repeated failures; we prefer an endoscopic method for occlusion similar to that described by Doherty and Slattery,11 but we also suture
104
OTOLOGIC SURGERY
but patients may need permanent middle ear ventilation via tubes. PETR with a subocclusive catheter relieves most patients of their symptoms with very mild and temporary side effects. If the catheters fail, PETR with cartilage graft can be placed under a submucoperichondrial flap with generally excellent long-term benefits. Some patients require additional adjunctive procedures over time, suggesting loss of graft volume or ongoing loss of ET valve tissue.
REFERENCES A
B FIGURE 7-14. A, Preoperative patulous eustachian tube. B, Postoperative eustachian tube after patulous eustachian tube reconstruction.
the lumen closed over the fat graft. This technique may prove to be effective for control of cerebrospinal fluid leakage after skull base operations.
CONCLUSION Patulous ET can cause autophony of one’s own voice and breathing sounds and aural fullness. When a cause can be identified, correction of patulous ET can relieve the symptoms. A concave defect in the anterior wall of the nasopharyngeal orifice of the ET can be seen in these patients. The lack of tissue in this area prevents normal closure of the ET valve in its resting position. Most patients do not need further treatment other than reassurance. A small subset of patients find their symptoms debilitating to such an extent that they seek medical and surgical attention. Before issuing such interventions, SSCD, inner ear hydrops, or temporomandibular joint dysfunction must be ruled out. Various materials have been injected in the defect to allow for tubal closure, but success is not uniform and is temporary. Complete closure of the ET lumen solves the problem of patulous ET,
1. Bluestone C D: eustachian Tube Structure, Function, and Role in Otitis Media. Hamilton, Ontario, Decker, 2005, pp 25-62. 2. Sade J: Introduction to eustachian tube function. In The eustachian Tube, Kugler, Amsterdam, 1985, pp 3-28. 3. Rockley TJ, Hawke WM : The middle ear as a baroreceptor. Acta Otolaryngol 112(Stock):816-823, 1992. 4. Proctor B : Embryology and anatomy of the eustachian tube. Arch Otolaryngol 86:51-60, 1967. 5. Misurya VK : Functional anatomy of tensor palatini and levator palatini muscles. Arch Otolaryngol 102:265-270, 1976. 6. Poe D: Opening and closure of the fibrocartilaginous eustachian tube. In Ars B (ed): Fibrocartilaginous eustachian Tube—Middle Ear Cleft, Kugler, The Hague, 2003, pp 49-56. 7. Yoshida H, Kobayashi T, Morikawa M, et al: CT imaging of the patulous eustachian tube—comparison between sitting and recumbent positions. Auris Nasus Larynx 30:135-140, 2003. 8. Kawase T, Hori Y, Kikuchi T, et al: Patulous eustachian tube associated with hemodialysis. Eur Arch Otorhinolaryngol 264:601-605, 2007. 9. Jago J: On the function of the tympanum. Proc R Soc Lond B Biol Sci 9:134-140, 1858. 10. Bluestone C D, Magit A E : The abnormally patulous eustachian tube. In Brackmann D E, Shelton C, Arriaga M A (eds): Brackmann’s Otologic Surgery. 2nd ed. Philadelphia, Saunders, 2001, pp 82-87. 11. Doherty J K, Slattery WH : Autologous fat grafting for the refractory patulous eustachian tube. Otolaryngol Head Neck Surg 128:88-91, 2003. 12. Pulec J L : Abnormally patent eustachian tubes: Treatment with injection of poly-tetrafluoroethylene (Teflon) paste. Laryngoscope 77:1543-1554, 1967. 13. Simonton K M : Abnormal patency of the eustachian tube—surgical treatment. Laryngoscope 67:342-359, 1957. 14. O’Connor A F, Shea JJ: Autophony and the patulous eustachian tube. Laryngoscope 91:1427-1435, 1981. 15. Virtanen H, Palva T: Surgical treatment of patulous eustachian tube. Arch Otolaryngol 108:735-739, 1982. 16. Suehs O: The abnormally open eustachian tube. Laryngoscope 70:1418-1426, 1960. 17. Poe S D: Diagnosis and management of the patulous eustachian tube. Otol Neurotol 28:668-677, 2007.
Chapter 7 • Diagnosis and Management of the Patulous Eustachian Tube 18. Hazell JW, Tinnitus II: Surgical management of conditions associated with tinnitus and somatosounds. J Otolaryngol 105:832-835, 1990. 19. Poe S D, Pyykko I, Valtonen H, Silvola J: Analysis of eustachian tube function by video endoscopy. Am J Otol 21:602-607, 2000. 20. van der Avoort S JC, van Heerbeek N, Zielhuis G A, Cremers CWR J: Sonotubometry: eustachian tube ventilatory function test: State-of-the-art review. Otol Neurotol 26:538-543, 2005. 21. van der Avoort SJC, van Heerbeek N, Zielhuis G A, Cremers CWR J: Validation of sonotubometry in healthy adults. J Laryngol Otol 120:853-856, 2006. 22. Di Martino E FN, Thaden R , Antweiler C, et al: Evaluation of eustachian tube function by sonotubometry: Results and reliability of 8 kHz signals in normal subject. Eur Arch Otorhinolaryngol 264:231-236, 2007. 23. Hori Y, Kawase T, Oshima T, et al: Objective measurements of autophony in patients with patulous eustachian tube. Eur Arch Otorhinolaryngol 264:1387-1391, 2007. 24. Hori Y, Kawase T, Hasegawa J, et al: Audiometry with nasally presented masking noise: Novel diagnostic methods of patulous eustachian tube. Otol Neurotol 26:596599, 2006. 25. Kikuci T, Oshima T, Ogura M, et al: Three-dimensional computer tomography imaging of the sitting position for the diagnosis of patulous eustachian tube. Otol Neurotol 28:199-203, 2007. 26. Zhou G, Gopen Q, Poe S D: Clinical and diagnostic characterization of canal dehiscence syndrome: A great otologic mimicker. Otol Neurotol 28:920-926, 2007.
105
27. Mikulec A A, McKenna M J, Ramsey M J, et al: Superior semicircular canal dehiscence presenting as conductive hearing loss without vertigo. Otol Neurotol 25:121-129, 2004. 28. Minor L B, Solomon D, Zeinreich J S, Zee DS : Soundand/or pressure-induced vertigo due to bone dehiscence of the superior semicircular canal. Arch Otolaryngol Head Neck Surg 24:249-258, 1998. 29. Dyer R K, McElveen JT: The patulous eustachian tube: Management options. Otolaryngol Head Neck Surg 105:832-835, 1991. 30. Luxford WM, Sheehy J L : Myringotomy and ventilation tubes: A report of 1,568 ears. Laryngoscope 92:12931297, 1982. 31. Zollner F: Die Klaffende ohrtompete, storungen dadurch und vorschlage zu ihrer behebung. Z Hals Nasen Ohr 42:287-298, 1937. 32. Ogawa S, Satoh I, Tanaka H : Patulous eustachian tube. Arch Otolaryngol 102:276-280, 1976. 33. Brookner K H, Pulec J L : Auditory tube patency after injection of Teflon paste. Arch Otolaryngol 90:60-64, 1969. 34. Bluestone C D, Cantekin E I : Management of the patulous eustachian tube. Laryngoscope 91:149-152, 1981. 35. Stroud M H, Spector GJ, Maisel R H : The patulous eustachian tube syndrome. Arch Otolaryngol 99:419421, 1974.
8
Office Management of Tympanic Membrane Perforation and the Draining Ear Peter S. Roland, Brandon B. Isaacson, and Joe W. Kutz
Otorrhea can arise from many causes: acute and chronic external otitis, acute and chronic otitis media, chronic myringitis, and secondary to a cerebrospinal fluid (CSF) fistula.
OTORRHEA FROM CEREBROSPINAL FLUID FISTULA CSF is a rare cause of persistent or intermittent otorrhea. CSF otorrhea may be classified as spontaneous or acquired. Temporal bone trauma is the most common cause of acquired CSF otorrhea, but it may also arise secondary to neoplasms, infections, and iatrogenic causes. Brodie and Thompson1 observed that CSF otorrhea resulting from temporal bone fractures resolved within 1 week of onset in 95 of 122 subjects (77.8%). They also showed that leaks persisting beyond 7 days had a much higher incidence of meningitis (23% versus 3%). Iatrogenic CSF otorrhea may occur immediately after surgery or may be delayed for days or years. Spontaneous CSF otorrhea may result from congenital temporal bone abnormalities, or can arise in the setting of normal temporal bone morphology as a consequence of arachnoid granulations, or increased intracranial pressure.2,3 More recent studies have shown that subjects with spontaneous CSF otorrhea are often morbidly obese and often have empty or partially empty sella on radiographic examination.3,4 Patients who are eventually diagnosed with CSF otorrhea are often initially identified during an evaluation of persistent unilateral middle ear effusion. The diagnosis is often confirmed when persistent watery drainage follows myringotomy. A careful history may elicit symptoms of intermittent positional nasal drainage. The diagnosis is usually established by history and physical examination, but may be confirmed with β2-transferrin if enough fluid can be collected. High-resolution temporal bone computed tomography (CT) and magnetic resonance imaging (MRI) with cisternogram, FIESTA, or CISS images are complementary studies often obtained in evaluating patients with CSF otorrhea.
Traumatic CSF otorrhea is often managed conservatively with stool softeners, head of bed elevation, serial lumbar punctures, or lumbar drainage. Occasionally, patients require operative management if conservative measures fail. The surgical approach is dictated by the fracture location, and hearing status in patients with temporal bone trauma. Patients with little or no residual hearing can be managed with a transmastoid or translabyrinthine approach, whereas patients with residual hearing can be managed with a middle fossa approach. Spontaneous CSF otorrhea requires surgical repair of cases caused by fistula between the subarachnoid space and temporal bone air cell system. The middle fossa approach is often used because these patients commonly have multiple tegmen defects. A transmastoid approach may be used if the defect is solitary and is sufficiently lateral in the temporal bone, or if the patient has significant medical comorbidities. CSF otorrhea resulting from congenital inner ear anomalies may be managed via a transcanal or transmastoid approach. Autologous (e.g., bone, cartilage, fascia) and nonautologous (e.g., acellular dermal matrix [AlloDerm], bone cement) materials have been used to repair these defects. Bone defects with communication into the posterior fossa are usually repaired with a transmastoid approach.
CHRONIC SUPPURATIVE OTITIS MEDIA Definition There is no widely agreed on definition of chronic suppurative otitis media (CSOM). Nonetheless, most practicing otologists would agree that the following elements should be present: 1. The drainage must be purulent, mucoid, or mucopurulent. 2. The otorrhea must arise from the middle ear space through a tympanic membrane perforation or tympanostomy tube. 3. Drainage must persist beyond 3 weeks despite appropriate medical management.5,6 107
108
OTOLOGIC SURGERY
Although some otologists refer to an infected cholesteatoma as a form of CSOM, others prefer to limit the term only to ears without cholesteatoma. Although many of the management principles discussed in this chapter may be applicable to individuals with an infected cholesteatoma, surgery, and not office management, is the mainstay of treatment when cholesteatoma is present. Consequently, ears with cholesteatomas are not a focus in this discussion.
Etiology Many, perhaps most, cases of CSOM arise from acute infections that have not resolved. Whether or not all cases of CSOM have an infectious etiology is unclear. Antonelli and colleagues5,6 have suggested that some cases may be due simply to “bacterial colonization and overgrowth in an ear that remains moist because of tubal pathology.” Many authors believe that allergy alone can result in chronic drainage through tympanic membrane perforation, at least occasionally. Although tympanic membrane perforations are invariably present (by definition), the etiologic role of membrane perforations in the development of CSOM is also unclear. Sometimes the tympanic membrane perforation is simply a result of the initial middle ear infection that never resolves. Alternatively, there are some cases in which a long-standing, dry perforation becomes infected, and the infection persists either because of lack of treatment or inadequate or inappropriate treatment. Most tympanic membrane perforations do not result in infected middle ears, however, and, when they do, most such infections, during the acute phase, respond briskly to appropriate treatment. As noted by Bluestone and others,7,8 a nonintact tympanic membrane (perforation or tympanostomy tube or both) eliminates the “middle ear cushion” and facilitates eustachian tube reflux. By eliminating the middle ear cushion, a tympanostomy tube or tympanic membrane perforation allows air to escape from the middle ear space, which eases the retrograde reflux of nasopharyngeal secretions into the middle ear. Eustachian tube reflux may be an especially important etiologic factor in populations who are otherwise prone to it (e.g., aboriginal North Americans). Many cases of CSOM arise from inadequately or incompletely treated cases of acute otitis media (AOM). Gibney and coworkers9 showed that aggressive treatment of AOM in aboriginal children in Australia reduces the incidence of CSOM. It is usually unclear why an acute infection fails to resolve and becomes chronic. We do know that AOM causes mucosal sloughing, causes impairment of ciliary clearance, and exposes microbial binding sites.5.6 Even so, only a few episodes of AOM evolve into CSOM. The onset of CSOM is characterized initially by increased vascularity of the mucosa and submucosa.
As CSOM persists, the proportion of chronic inflammatory cells increases. This increase in cells leads to osteitis and mucosal edema with ulceration, and two important pathophysiologic events follow: (1) capillary proliferation, which results in formation of granulation tissue and polyps, and (2) a rarifying osteitis, which ultimately produces new bone formation and fibrosis. Osteitis is present in virtually 100% of CSOM patients, a finding that distinguishes CSOM from more transient pathologic alterations in the middle ear cleft.10
Bacteriology CSOM is predominantly a gram-negative infection, although staphylococcal species occur with sufficient frequency that they must be taken into consideration when any microbial therapy is considered. Pseudomonas aeruginosa is generally the most common gram-negative organism, but other gram-negative organisms are commonly encountered, especially species of Enterobacteriaceae.11 The extent to which anaerobic organisms or fungi or both are pathogenically involved is controversial. Anaerobic organisms are often present, but whether or not they are cultured depends on how rigorously they are sought.12,13 The contributions of anaerobes to pathophysiology remain unclear. Fungi are commonly recovered, but the extent to which they are pathogens as opposed to saprophytes is unresolved, and probably variable.
Pathophysiology Granulation tissue is almost an invariant accompaniment of CSOM. It can develop quickly in a draining ear, and is already present in many infections of less than 6 weeks’ duration. The presence of granulation tissue, especially when it is abundant, may be a factor that contributes to treatment failure of acute infections, and the evolution of AOM into CSOM. The formation of granulation tissue in the middle ear begins with a break in the basement membrane of the surface epithelial cells. Inflammatory cells in the underlying lamina propria traverse through the broken basement membrane and enter the lumen of the middle ear space. The rupture of the basement membrane and epithelial lining cell is caused by bacterial toxins, inflammatory mediators produced by ruptured liposomes, and the accumulation of subepithelial fluid and vacuoles, all of which exert pressure on the surface epithelium.14 The next step in the formulation of granulation tissue occurs when a small piece of the herniated lamina propria extrudes through the ruptured area between epithelial cells. Initially, the extruded lamina propria pushes into the middle ear without any epithelial covering. Angiogenic growth factors incite capillary budding, vascular hyperpermeability, and fibroblast recruitment—that is, granulation tissues form. If the
Chapter 8 • Office Management of Tympanic Membrane Perforation and the Draining Ear
growth of granulation tissue is vigorous and aggressive, polyps develop.15 Meyerhoff and colleagues10 evaluated temporal bone from subjects with CSOM and reported that granulation tissue develops in 90% of all CSOM, and in 100% of cases of CSOM that develop intracranial complications.
ACUTE OTITIS MEDIA WITH AN OPEN TYMPANIC MEMBRANE Inadequately treated AOM with a nonintact tympanic membrane is a frequently cited cause of CSOM.16 Acute infections in an ear with a nonintact tympanic membrane are frequently inadequately or inappropriately treated (especially in children) because it is not universally recognized that in many ways an acute infection occurring in an open middle ear cavity is significantly different from classic AOM. Most importantly, the microbiology is significantly different. Although some cases of acute infection in an open middle ear arise from the typical AOM pathogens—Streptococcus pneumoniae, Haemophilus influenzae, and Moraxella catarrhalis—most, especially in patients older than 2 years, are caused by species of Staphylococcus, Pseudomonas, or other gram-negative organisms.17 Yeasts (Candida spp.) are occasionally encountered when the eardrum is not intact, but rarely when the middle ear space is closed.17 The oral antibiotics usually prescribed for AOM are poorly active against Staphylococcus and completely inactive against gramnegative organisms. Individuals treated with these systemic antibiotics consequently are often treated inappropriately or inadequately. Aggressive, appropriate treatment of acute middle ear infections in individuals with a nonintact tympanic membrane is an important way of preventing the development of CSOM. Another meaningful difference between AOM with an intact tympanic membrane and AOM in an open middle ear is that the natural history of these two infections seem to be different. It is well established that 80% or more of untreated children with AOM and intact eardrums recover spontaneously, and that suppurative complications are unlikely. Ruohola and associates18,19 found that AOM in children with tympanostomy tubes is significantly different in this respect: at the end of their study, 60% of nontreated children still had infection and drainage.
MANAGEMENT OF OTORRHEA There are three principal components to effective management of the draining ear: 1. Effective aural toilet 2. Use of an appropriate antimicrobial agent 3. Management of granulation tissue
109
Aural Toilet Effective aural toilet mechanically decreases the bacterial burden, eliminates debris that nurtures bacterial growth, and removes obstacles to the effective delivery of antimicrobial agents. The most effective aural toilet is achieved using an operating microscope and microinstrumentation. When microscopic débridement cannot be performed, the use of irrigation solutions to remove debris and mucopurulent exudate is a reasonable alternative. Irrigation of the external auditory canal can be performed at home using a small bulb syringe. When used in combination with antimicrobial drops, irrigation should precede the instillation of drops by 20 to 30 minutes, allowing time for the instilled irrigation solution to drain out of the ear completely. Various solutions have been advocated for this purpose, including 3% hydrogen peroxide, vinegar, and isopropyl alcohol. Many otologists dilute these solutions 50:50 with sterile water. Water alone should not be used. The use of water alone increases the ambient humidity; may alkalize the ear canal or middle ear space or both; and, if not sterile, can introduce pathogenic organisms. Isopropyl alcohol is not recommended because it can be painful on instillation and is potentially ototoxic. “Dry mopping” refers to the use of twisted pieces of cotton or tissue pushed into the external auditory canal to “mop up” and remove any secretions within the canal. Its effectiveness is unknown, but it may be better than nothing, especially if the principal component to be removed is liquid and not desquamated epithelial debris or inspissated mucopus.
Antimicrobial Therapies Elimination of infection is crucial to the successful management of a draining ear. Consequently, appropriate antimicrobial treatment is pivotal. Infection that is limited to the external auditory canal or an open mastoid cavity can sometimes be managed with the use of antiseptics or acidifying agents, or both, alone. When effective, these agents have several advantages: they are inexpensive, are easy to procure, do not promote bacterial resistance, and are generally effective against bacteria and fungi. The relative merits of one of these preparations versus another is generally unknown, although Tom20 showed that Cresylate is more effective in vitro than other antiseptics against fungal organisms. The acidifying agents and antiseptics are unattractive agents when the middle ear is open. Almost all have been shown to be potentially ototoxic. Most of these preparations are acidic (ototoxic in itself) and may be quite painful when they come in contact with the middle ear mucosa, which is of neutral pH. When antibiotics are used, topical delivery in the form of otic drops is more effective and safer than the use of systemic antibiotics, either orally or parenterally.21-25,27
110
OTOLOGIC SURGERY
By using a topical agent, the physician can deliver a concentration of medication that is several orders of magnitude higher in affected tissues than the concentration that can be achieved by using a systemic agent. A 3 to 5 drop dose of a 0.3% solution of antibiotic contains only 90 to 150 mg of medication, but its concentration is 3000 μg/mL, which exceeds the minimum inhibitory concentration of any known relevant pathogen. In contrast, consider the typical drug levels in middle ear fluid that can be achieved by three systemic antibiotics: Cefuroxime 1 to 2 μg/mL Amoxicillin 8 to 10 μg/mL Ceftriaxone 25 to 30 μg/mL Because the concentration of medicine delivered to infected tissue is so much higher than can be achieved with systemic antibiotics, antibiotic sensitivities as determined by clinical laboratories are irrelevant. These sensitivity determinations made by clinical laboratories are based on what are believed to be realistically achievable tissue levels after systemic administration of a drug. An organism with a minimal inhibitory concentration (MIC) of 4 to ciprofloxacin would generally be considered to be a “resistant” organism because it is unlikely that a tissue level of 4 μg/mL of ciprofloxacin could be achieved by oral or parenteral administration of ciprofloxacin. If otic drops with a concentration of 3000 μg/mL (a concentration available in commercial preparations) is delivered, however, the organism would succumb. Consequently, sensitivity determinations made in clinical laboratories can be safely ignored when topical therapy is used. A second important consequence of such high concentrations is that the risk of the emergence of resistant organisms is minimized. Neither resistant clones already present nor newly emerging resistant mutants would survive in such very high concentrations of antibiotic. Despite the very high concentration of antibiotic in topical ear drops, the risk of systemic side effects or adverse reactions is very low because the total dose administered (approximately 1 mg per dose) is low, and systemic absorption is limited. Skepticism has been voiced that topical drops would not provide sustained levels of antibiotic inside the middle ear space, especially if the perforation is small or delivery is through a tympanostomy tube. Ohyama and associates26 documented that topical antibiotic drops can be, and often are, effectively delivered into the middle ear space and have long dwell times. These investigators measured the persistence of a single dose of topically applied 0.3% ofloxacin in otorrhea fluid, serum, and middle ear mucosa at various time intervals after administration. They found a very high level of antibiotic in the otorrhea fluid several hours later. Perhaps the most surprising finding was the high concentration of drug in biopsy specimens of middle ear mucosa approximately 1 hour after the administration of ear drops. As expected, drug concentrations in serum were very low or nondetectable.
The superior efficacy of topically applied antibiotics is well documented in the literature. Esposito and colleagues28 studied 60 subjects with CSOM who were randomly assigned to receive one of three treatment regimens: one group received oral ciprofloxacin twice a day, another received three drops of ciprofloxacin solution twice a day, and a third group received oral and topical ciprofloxacin doses twice a day. The topical group had a clinical response rate of 100% and a bacteriologic cure rate of 95%, and the topical/oral group had response rates of 95% and 85%. By contrast, the group that received oral antibiotics alone had a clinical response rate of 65% and a rate of bacterial eradication of only 40%. These differences were statistically significant.28 Two years later, Esposito and colleagues29 published the results of their comparison of topical ciprofloxacin versus intramuscular gentamicin in 60 adults. All patients had CSOM, and 40 of them had previously undergone systemic therapy. Half of the group received four drops of ciprofloxacin solutions twice a day, and the other half received twice-daily injections of 80 mg of gentamicin. Favorable clinical responses were seen in 87% of the topically treated patients and only 67% of patients who received systemic gentamicin. Only 43% of the patients receiving systemic gentamicin achieved microbiologic eradication compared with 83% of the group receiving topical drops. Two studies have compared the relative efficacy of topical drop regimens with the use of systemic antibiotics with amoxicillin/clavulanic acid.30,31 Both studies showed enhanced efficacy with the topical treatment regimen. These studies were performed in children with acute otorrhea after insertion of a tympanostomy tube. Because Pseudomonas and Staphylococcus aureus are more prominent pathogens in CSOM, the superiority of a topical regimen over amoxicillin/clavulanic acid should be even greater because S. aureus is a more frequent pathogen in CSOM than an acute otitis media with a tympanostomy tube (AMOT). In 1994, Yuen and coworkers32 published results of a prospective trial comparing 0.3% ofloxacin otic drops with oral amoxicillin/clavulanic acid in 56 patients. One week after completion of treatment, 76% of the ofloxacin group had dry ears compared with only 26% of the other group. de Miguel-Martinez and associates33 divided 125 patients into five treatment arms. The participants received one of the following treatments: 1. Oral ciprofloxacin, 500 mg every 12 hours 2. Topical ciprofloxacin/hydrocortisone, 3 drops every 8 hours 3. Topical 0.2% ciprofloxacin 3 drops every 8 hours 4. Combination of oral and topical ciprofloxacin 5. Topical neomycin/polymyxin B The best outcomes were achieved in the group that received 0.2% ciprofloxacin. In 2000, the Cochrane database published a review by Acuin and coworkers.24 The review concluded that
Chapter 8 • Office Management of Tympanic Membrane Perforation and the Draining Ear
111
topical antibiotics are superior to systemic antibiotics for the treatment of CSOM, and that a combination of systemic plus topical antibiotic offers no therapeutic advantage. Fluoroquinolone topical antibiotics were seen to be more effective than other types of topical antibiotic drops. The systemic review emphasized the superior microbiologic eradication rates achieved with topical therapy, and emphasized the importance of bacterial eradication. The addition of a steroid to a topical antibiotic was compared with the ototopical antibiotic alone in three studies by Roland and colleagues.15,34,35 Both studies showed that the steroids hastened the resolution of otorrhea, and in one study the final cure rate was higher in the group treated with a topical antibiotic that included a steroid. The efficacy of combination ototopical drops that contain an antibiotic and a steroid has not been specifically studied in the treatment of CSOM. The improved effectiveness of steroid-containing drops in the treatment of post-tympanostomy tube otorrhea and the management of granulation tissues has led many clinicians, however, to adopt them as the topical treatment of choice in individuals with CSOM. Combination powders are widely used by experienced clinicians in the treatment of CSOM and in the management of draining mastoid cavities. Powders dry moist surfaces and stick to moist tissue tenaciously. They consequently have long dwell times after a single application. Various preparations are available. Most contain a combination of an antibiotic, antifungal agent, and steroid. Although none of these preparations has been systematically evaluated in controlled clinical trials, and although none is approved by the U.S. Food and Drug Administration (FDA), many experienced clinicians rely on them and is convinced they are more efficacious than otic drops. Despite the superior efficacy of topically delivered medications, topical delivery does have some disadvantages, as follows:
and excipients mixed with antibiotics in topical antibiotic solutions (e.g., propylene glycol) can be irritating. It is believed that the steroid component present in some commercial preparations may minimize the inherent irritating properties of other ingredients. 3. The aminoglycoside antibiotics and polymyxin B are devastatingly ototoxic in animals. A single dose of neomycin/polymyxin B/hydrocortisone instilled into the external auditory canal of a chinchilla passes easily through tympanostomy tube and produces a significant loss of hair cells in the basal turn of the cochlea. Although the extent to which these animal data can be extrapolated to humans is arguable, researchers at the University of Toronto have shown that vestibular toxicity may be a significant risk when aminoglycoside drops are used. In contrast, animal data indicate that fluoroquinolones are not ototoxic. Although antibiotic drops other than quinolones are available, all are potentially ototoxic, and none is FDA approved for use in the middle ear space. A Consensus Panel of the American Academy of Otolaryngology–Head and Neck Surgery has advised against the use of potentially ototoxic drops when the middle ear is open on the grounds that because they convey no apparent benefit (other than low cost), the additional risk is unwarranted.36 4. Aminoglycosides, and especially neomycin, have a propensity to cause topical sensitization. Although severe hypersensitivity reactions are usually easy to recognize, hypersensitivity may take a more subtle form. Often, the only manifestations of hypersensitivity are persistent erythema, edema, and otorrhea. When sensitization assumes this more subtle form, it may be impossible to distinguish between hypersensitivity reactions and clinical treatment failures.37,38 5. The biggest liability associated with topical therapy is that drops must come into direct contact with infected tissues if they are to be effective. Failure of delivery may arise from various causes (see later).
1. Topical application can be uncomfortable. The mucosa of the middle ear space is much more sensitive than the skin of the external auditory canal. When very acidic drops (e.g., neomycin/polymyxin B/hydrocortisone preparations, whose pH may range between 3 and 3.5) are placed in the middle ear cavity, they can cause intense burning pain. Likewise, alcoholcontaining drops sting and burn intensely, and their use should be avoided if they would enter the middle ear space. Discomfort can be caused simply because the drops are cold, especially in children. Children may have difficulty distinguishing between pain and the unpleasant sensation of cold medication. When possible, medication should be delivered close to body temperature. It is recommended that otic drops be warmed before instillation. 2. Topical medications can have direct inflammatory effects. Some antibiotics have inflammatory properties,
If the infection fails to resolve after 1 week to 10 days of topical therapy, it is not unreasonable to consider a course of systemic therapy. In contrast to topical therapy, systemic therapy should be based on the sensitivity of the infecting organism to available antibiotics. Oral therapy is preferred, unless the appropriate antibiotic is available only in a parenteral form, or is significantly more effective when delivered parenterally. Oral quinolones achieve the same blood and tissue levels when administered orally as when administered parenterally. Theoretically, systemic antibiotics should be effective only in cases in which effective delivery of topical drops cannot be achieved.
Granulation Tissue Granulation tissue is an important component of the pathophysiology of CSOM. There are four components to the successful management of granulation tissue: aural
112
OTOLOGIC SURGERY
toilet, appropriate antibiotic therapy, use of steroids, and débridement. Aural toilet and antibiotic therapy have been discussed previously. Otolaryngologists have long held as an article of faith that steroids are important in suppressing, eliminating, and preventing granulation tissue. Several animal studies have found that steroids are effective in controlling keloids and hypertrophic scarring, and reducing angiogenesis with subsequent formation of granulation tissue.39-43 More recently, evidence has shown that steroids are effective in humans as well. A randomized, controlled, double-blind drug comparison study of 599 children with patent tympanostomy tubes and AOM with otorrhea of 3 weeks or less duration provides such evidence. Children with granulation tissue associated with tympanostomy tubes were treated with either a quinolone plus a steroid or a quinolone alone. Not only did the combination product prove to be significantly more effective in establishing clinical and microbiologic cure in all patients, but also it was significantly more effective in eliminating granulation tissue.15,44 A more potent steroid seems to be more effective than a less potent steroid. Office débridement of granulation tissue can be performed using the microscope and microinstrumentation, but it is most commonly performed using chemical cautery. Silver nitrate is the agent most commonly used, although trichloroacetic acid is used by some clinicians. Chemical cautery is, in effect, a chemical burn, and control over the depth and severity of that burn is limited.45 If care is not exercised, vital structures (i.e., facial nerve) can be permanently injured. Mechanical removal should be done sharply, with clear visualization of exactly what tissues are being incised. Avulsion of aural polyps is not usually recommended because when the polyp is removed, the otolaryngologist might find that the stapes is attached to it. Using sharp techniques under direct visualization avoids inadvertent injury to important structures. Tympanomastoid surgery for chronic otitis media is an aggressive form of débridement and is often very effective.
Preventing Bacterial Resistance The FDA and major infectious disease societies of the United States have urged clinicians to use treatment regimens that minimize the opportunity for the development of bacterial resistance. Acquired resistance refers to a change in the susceptibility of an organism to a particular antibiotic. Topical therapy is less likely to result in emergence of resistant strains for two reasons: 1. The very high concentration of antibiotic delivered to infected tissues when a topical therapy is used 2. The relatively shorter duration of therapy recommended when a topical treatment or regimens are used
Applying the term resistant to a bacterial species is based on two determinations: 1. The MIC for a given organism to a given antibiotic 2. The amount of antibiotic that can reasonably be expected to reach the site of infection after systemic administration (intramuscular, intravenous, or oral). The MIC is an inherent biologic characteristic of an individual bacterial strain. The organism is resistant to an antibiotic, however, if, and only if, the amount of antibiotic that can be delivered to the site of infection does not exceed the particular MIC of that organism to the antibiotic used. Consequently, although MIC is a biologic characteristic of the organism, the designation of either sensitive or resistant is relative. As a practical matter, increases in bacterial resistance to a specific antibiotic are manifest by increasing MICs for that particular antibiotic. An analysis of clinical failures from several large clinical trials verified that changes in MIC have not occurred as a result of treatment using topical antibiotic drops.35
Causes of Failed Therapy Although the principal virtue of a topical antibiotic therapy is the very high concentrations of drugs that can be delivered, the weakness of topical therapy is that efficacy depends on effective delivery. When one considers the MICs of infective organisms (virtually never >250 μg/mL) and the concentration of antibiotic in topical drops (3000 μg/mL), it becomes apparent that a cure would be achieved if the medication is effectively delivered to infected tissues. Virtually all failures of topical therapy are failures of delivery. There are many reasons why successful delivery may not be achieved, including the following: 1. Noncompliance 2. Poor delivery technique 3. Voluminous drainage 4. Mucosal edema 5. Granulation tissue Noncompliance can take several forms. The medicine may never be purchased, and the treatment regimen never begun. The potential efficacy and potency of topical antibiotic therapy is not widely recognized by the public. The medicine may never be purchased because it is too costly. Inquiries into whether or not the patient can afford medication are important, especially if more expensive brand-name products are chosen. More typically, noncompliance takes the form of inconsistent, erratic dosing or early termination of treatment or both. Proper technique requires that the drops are delivered into the internal auditory canal, which is often difficult with a squirming child. “Tragal” pumping (rhythmic
Chapter 8 • Office Management of Tympanic Membrane Perforation and the Draining Ear
pressure just anterior to the tragal cartilage) is important in ensuring that drops penetrate through a tympanostomy tube or tympanic membrane perforation into the middle ear space.46 Issues related to management of granulation tissue have been discussed previously. Occasionally, mucosal edema is sufficient to prevent adequate delivery of antibiotics into and through the middle ear space. When this is the case, patience is required. The use of a steroid-containing antibiotic drop generally results in progressively increasing middle ear mucosal edema over several days. Voluminous drainage is managed as mentioned earlier by effective aural toilet. Occasionally, therapy fails because there is a sequestered nidus of infection in a portion of the middle ear or mastoid, inaccessible to topical drops. Sometimes such a nidus is focused around a small area of devascularized or dead bone. When this is the case, topical therapy may never be effective. Such cases occasionally respond to systemic antibiotics, however. Consequently, a treatment regimen of systemic antibiotic therapy in individuals who failed topical therapy is logical and occasionally effective, even though, as noted earlier, topical treatment regimens are generally more effective than systemic treatment. Appropriate imaging studies occasionally can identify such areas. A fine cut CT scan of the temporal bone may provide useful information if systemic therapy is contemplated or necessary. An occult cholesteatoma is a common source of persistent drainage that does not respond to topical therapy. Cholesteatomas can be extremely difficult to visualize in an infected, purulent ear. Effective resolution of otorrhea can often be achieved with topical therapy even if a cholesteatoma is present. When topical therapy is effective, and when granulation tissue, edema, and purulence are eliminated, physical examination often allows easy identification of the cholesteatomatous process. It may be impossible to eliminate infection, however, when this source is an infected cholesteatoma. Imaging studies are often capable of identifying a cholesteatomatous process even in the face of significant infection. An experienced radiologist can often detect the pattern typical for cholesteatoma, and differentiate a cholesteatoma-driven from a purely infectious process. When drainage cannot be resolved, the patient should be evaluated for disease in remote locations. Most specifically, in children, the tonsils, adenoids, and sinuses need to be considered. Chronic, unremitting rhinosinusitis can produce constant reseeding of the middle ear space, especially in young children with eustachian tube reflux. Infected adenoids can be another ongoing source for persistent infection. Whether or not infected tonsils can repeatedly reseed the middle ear space has not been clearly shown or refuted. Occasional patients have persistent middle ear and mastoid infection because of depressed or diminished immune capacity. Topical antibiotic therapy is almost certain to be insufficient treatment for individuals who
113
have diseases at remote sites or compromised immune systems. Biofilms have been shown not only on tympanostomy tubes, but also in the middle ear and mastoid with suppurative otitis media. The biofilms seen in patients with CSOM resemble those found on the surface of the middle ear mucosa of chinchillas after inoculation with nontypable Haemophilus influenzae. Although the role of biofilms in the pathogenesis of culture-negative otitis media with effusion is outside the scope of this chapter, the presence of biofilms in individuals with persistent/ recurrent otologic infections does suggest a possible pathogenic role for biofilms.47,48 It is widely recognized that patients who fail combined topical/systemic therapy often resolve the infection after the removal of a tympanostomy tube. Effective treatment of biofilms depends on restoration of the normal physiology of the middle ear combined with supratherapeutic doses of antibiotics. Whether or not sufficiently high doses of antibiotic therapy can be achieved with topical regimens remains unclear. The draining mastoid cavity is a problem occasionally seen in patients who have undergone open cavity techniques, usually for the management of cholesteatoma. Drainage persists in such cavities for several reasons. The most common cause of unremitting drainage is either persistent infection in a sequestered air cell or a small area of osteitis. In such cases, the cavity as a whole heals up well except for a small area that remains covered with granulation tissue. Occasionally, aggressive, nonsurgical management of the small area of granulation tissue permits brief episodes of epithelial coverage, but the epithelium breaks down again as infection wells up beneath it from the area of osteitis. When this is the case, the only solution is the removal of the involved area. Sometimes this removal can be done with the patient under local anesthesia in the clinic using small curettes to scrape away the involved area. More frequently, it requires a brief return to the operating room, where surgical drills can be used to eliminate infected bone. If the area of the osteitis is large, and postoperative otorrhea has persisted for months or years, consideration should be given to skin grafting. Skin grafts are occasionally used in cavities that have developed mucosal as opposed to squamous epithelial linings on one or more occasions. Sometimes persistent or recurrent drainage is caused by residual cholesteatoma. The only viable solution for recurrent cholesteatoma is reoperation with removal of residual disease. Sometimes persistent or recurrent drainage is caused by persistent otitis media as a result of eustachian tube obstruction or reflux. Generally, infection is clearly limited to the middle ear, and the patient has a draining perforation. The mastoid cavity may seem healthy, or may be repeatedly contaminated by mucopurulent exudate flowing out of the middle ear, which can result in persistent mastoid infection or granulation tissue or both. Therapy
114
OTOLOGIC SURGERY
Chapter 8 • Office Management of Tympanic Membrane Perforation and the Draining Ear
should be directed toward resolving the persistent infection of otitis media. Revision surgery can eliminate foci of persistent infection in the middle ear space and successfully close recurrent tympanic membrane perforations leading to a clean, dry, healed ear. Pharmacologic management of the draining cavity follows the same principles as for other chronic middle ear and mastoid infections. Topical therapy is the treatment of choice—in the setting of a draining cavity, powders are particularly useful. If the middle ear space is closed (i.e., the tympanic membrane graft has healed), the antiseptics are more attractive because issues relating to ototoxicity are irrelevant.
OFFICE TECHNIQUES FOR CLOSURE OF TYMPANIC MEMBRANE PERFORATIONS Two techniques are used for closure of tympanic membrane perforations in the outpatient office setting: a fat graft tympanoplasty and paper patch tympanoplasty. Paper patching is used much more commonly in the outpatient setting than fat graft tympanoplasty.
115
Cauterization and Paper Patching The use of paper patches, placed in an outpatient setting (a general anesthetic may be necessary for children), has been widespread for many decades. Paper patches are designed to promote healing of perforated tympanic membranes, while avoiding a formal surgical procedure. The technique is based on the recognition that one cause of failure to heal is epithelialization of the margin of a perforation. Removing the healed epithelial edge transforms the perforation into an open wound, and allows the regenerative process to begin anew. Rose and Politzer were among the first to use silver nitrate cauterization of the margins of a perforation to promote healing.49 Dunlap and Schuknecht reported that closure generally requires repeated application of cautery with removal of dried exudates every 2 weeks; closure was complete, on average, only after 6 to 12 months.49 Jeurs and Wright reported success in closing perforations by marginal stimulation in 80% to 90% of cases; multiple treatment episodes were required.50 Derlacki51,52 was able to close 75% of 131 perforations he treated, with an average of 14 treatments required. Although Derlacki51,52
116
OTOLOGIC SURGERY
had some success in larger perforations, most successes are reported for small to medium-sized perforations. Repeated cauterizations are always necessary. Trichloroacetic acid has become the agent of choice for removal of the old epithelial margin. It is applied to the margins of the perforation using a very small cotton swab (Calgiswab or handmade). Effective chemical cauterization turns the edge of the perforation white about 1 or 2 mm from the rim. A small paper patch can be applied to act as a scaffolding for the regenerating epithelium. Paper patches are often made from either cigarette or filter paper. A commercially available “hole punch” creates a patch of ideal size. The patient needs to be seen at weekly intervals for recauterization and patch refreshing until the perforation is closed. There is a general consensus that cauterization and patching techniques are impractical when perforations exceed 25% to 50% of the tympanic membrane, and are contraindicated in the presence of frank or suspected cholesteatoma, chronic drainage, and ossicular disruption. Some otologists believe that traumatic tympanic membrane perforations must be managed acutely. They believe the best results are achieved when torn areas of drum are immediately reapproximated, often over absorbable gelatin sponge (Gelfoam) placed through the perforation into the middle ear to support the reapproximated portions of drum (Figs. 8-1 through 8-6). Special emphasis is often placed on elevating segments of drum that are folded back on themselves after an implosiontype injury. Accurate reapproximation of drum fragments requires microinstrumentation and a microscope. Some authors recommend combining fragment reapproximation with paper patching.
Fat Graft Tympanoplasty The technique of fat graft tympanoplasty was first described by Ringenberg in 1962,53 but differed substantially from the technique currently recommended in two respects: (1) Ringenberg turned a tympanomeatal flap and explored the middle ear as part of the procedure, and (2) the fat graft was placed lateral to (and not through) the perforation. Donor fat is generally harvested from the earlobe. Ringenberg53 evaluated fat from several anatomic sites, and determined that earlobe fat was more compact and contained more connective tissue than fat harvested from the abdomen or the buttocks. Other advantages include the ease with which lobule fat from the lobe can be obtained, and that enough donor fat can be taken from a single lobule to graft more than one tympanic membrane perforation.54 The technique as currently described is quite simple (Figs. 8-7 and 8-8). The epithelial margin of the perforation is removed so that the edges of the perforation are fresh and open. This is sometimes referred to as “rimming” the perforation. A globule of fat slightly larger than the perforation is pushed through the perforation so that
a portion of the inserted fat is medial to the tympanic membrane, and a portion is lateral. Histologically, the graft seems to epithelialize over 2 to 3 weeks. The graft atrophies in the first couple of weeks, but at 3 to 4 weeks a significant volume of fat can still be seen within the epithelialized medial and lateral surface. This fat atrophies almost completely in the following 3 to 6 weeks. Very little of the fat remains within the middle layer of the now healed tympanic membrane 6 to 8 weeks after grafting.53,55 It is widely agreed that the technique is most suitable for small perforations involving no more than 25% to 30% of the tympanic membrane.54,56-59 The technique should be limited to individuals whose hearing is normal or close to normal, and in whom there is little or no possibility of associated ossicular abnormality.54,56,57,60,61 Because this technique is most frequently used in pediatric patients for small perforations that persisted after extrusion of tympanostomy tubes, most cases have been done in a day surgery setting. The technique is suitable for office outpatient use, however, and can be performed with local anesthesia. Because the middle ear space is not entered, it is reasonable to perform the procedure bilaterally.54,57 Closure rates have been reported as 76% to 92% of cases.54,56-62 In a small series of 15 tympanostomy tubes removed for persistent unremitting otorrhea, Liew and associates63 reported a success rate of 100%.
REFERENCES 1. Brodie H A, Thompson TC : Management of complications from 820 temporal bone fractures. Am J Otol 18:188-197, 1997. 2. Gacek R R : Evaluation and management of temporal bone arachnoid granulations. Arch Otolaryngol Head Neck Surg 118:327-332, 1992. 3. Prichard CN, Isaacson B, Oghalai JS, et al: Adult spontaneous CSF otorrhea: Correlation with radiographic empty sella. Otolaryngol Head Neck Surg 134:767-771, 2006. 4. Scurry WC Jr, Ort S A, Peterson WM, et al: Idiopathic temporal bone encephaloceles in the obese patient. Otolaryngol Head Neck Surg 136:961-965, 2007. 5. Antonelli PJ, Juhn S K, Goycoolea MV, Giebink G S : Middle ear susceptibility to Pseudomonas infection during acute otitis media. Ann Otol Rhinol Laryngol 102:531536, 1993. 6. Antonelli PJ, Juhn S K, Le CT, Giebink G S : Acute otitis media increases middle ear susceptibility to nasal injection of Pseudomonas aeruginosa. Otolaryngol Head Neck Surg 110:115-121, 1994. 7. Bluestone C D: Pathogenesis of otitis media: Role of eustachian tube. Pediatr Infect Dis J 15:281-291, 1996. 8. Buchman CA, Doyle WJ, Swarts JD, Bluestone CD: Effects of nasal obstruction on Eustachian tube function and middle ear pressure. Acta Otolaryngol 119:351-355, 1999. 9. Gibney K B, Morris PS, Carapetis J R, et al: The clinical course of acute otitis media in high risk Australian aboriginal children: A longitudinal study. BMC Pediatr 5:16, 2005.
Chapter 8 • Office Management of Tympanic Membrane Perforation and the Draining Ear 10. Meyerhoff WL , Kim C S, Paparella M M : Pathology of chronic otitis media. Ann Otol Rhinol Laryngol 87(6 Pt 1):749-760, 1978. 11. Roland PS : Chronic suppurative otitis media: A clinical overview. Ear Nose Throat J 81(8 Suppl 1):8-10, 2002. 12. Brook I : Microbiology and management of chronic suppurative otitis media in children. J Trop Pediatr 49:196199, 2003. 13. Brook I : The role of anaerobic bacteria in acute and chronic mastoiditis. Anaerobe 11:252-257, 2005. 14. Hashimoto I, Nakanishi H, Shono Y, et al: Angiostatic effects of corticosteroid on wound healing of the rabbit ear. J Med Invest 49:61-66, 2002. 15. Roland PS : The formation and management of middle ear granulation tissue in chronic ear disease. Ear Nose Throat J 83(Suppl 1):5-8, 2004. 16. Bluestone C D: Epidemiology and pathogenesis of chronic suppurative otitis media: Implications for prevention and treatment. Int J Pediatr Otorhinolaryngol 42:207-223, 1998. 17. Roland PS, Parry D A, Stroman DW: Microbiology of acute otitis media with tympanostomy tubes. Otolaryngol Head Neck Surg 133:585-595, 2005. 18. Ruohola A, Heikkinen T, Meurman O, et al: Antibiotic treatment of acute otorrhea through tympanostomy tube: Randomized double-blind placebo-controlled study with daily follow-up. Pediatrics 111(5 Pt 1):1061-1067, 2003. 19. Ruohola A, Meurman O, Nikkari S, et al: Microbiology of acute otitis media in children with tympanostomy tubes: Prevalences of bacteria and viruses. Clin Infect Dis 43:1417-1422, 2006. 20. Tom LW: Ototoxicity of common topical antimycotic preparations. Laryngoscope 110:509-516, 2000. 21. Acuin J: Chronic suppurative otitis media. Clin Evid 15:772-787, 2006. 22. Roland PS : Ototopical agents are superior to systemic therapy for the treatment of acute and chronic otitis media. Ear Nose Throat J 83(9 Suppl 4):9-12, 2004. 23. Macfadyen C A, Acuin J M, Gamble C : Systemic antibiotics versus topical treatments for chronically discharging ears with underlying eardrum perforations. Cochrane Database Syst Rev (1):CD005608, 2006. 24. Acuin J, Smith A, Mackenzie I : Interventions for chronic suppurative otitis media. Cochrane Database Syst Rev (2):CD000473, 2000. 25. Macfadyen C A, Acuin J M, Gamble C : Topical antibiotics without steroids for chronically discharging ears with underlying eardrum perforations. Cochrane Database Syst Rev (4):CD004618, 2005. 26. Ohyama M, Furuta S, Ueno K, et al: Ofloxacin otic solution in patients with otitis media: An analysis of drug concentrations. Arch Otolaryngol Head Neck Surg 125:337-340, 1999. 27. Hannley MT, Denneny JC III, Holzer S S : Use of ototopical antibiotics in treating 3 common ear diseases. Otolaryngol Head Neck Surg 122:934-940, 2000. 28. Esposito S, D’Errico G, Montanaro C : Topical and oral treatment of chronic otitis media with ciprofloxacin: A preliminary study. Arch Otolaryngol Head Neck Surg 116:557-559, 1990. 29. Espisito S, Noviello S, D’Errico G, et al: Topical ciprofloxacin vs intramuscular gentamicin for chronic otitis media. Arch Otolaryngol Head Neck Surg 118:842-844, 1992.
117
30. Dohar J, Giles W, Roland P, et al: Topical ciprofloxacin/ dexamethasone superior to oral amoxicillin/clavulanic acid in acute otitis media with otorrhea through tympanostomy tubes. Pediatrics 118:e561-e569, 2006. 31. Goldblatt E L , Dohar J E, Noza R J, et al: Topical ofloxacin versus systemic amoxicillin/clavulanate in purulent otorrhea in children with tympanostomy tubes. Int J Pediatr Otorhinolaryngol 146:91-101, 1998. 32. Yuen ��� P W, Lou SK, Chau PY, et al: Ofloxacin eardrop treatment for active chronic suppurative otitis media: Prospective randomized study. Am J Otol 15:670-673, 1994. 33. de Miguel-Martinez I, Ramos-Macias A, Martin-Sanchez A M : Otitis media due to Corynebacterium jeikeium. Eur J Clin Microbiol Infect Dis 18:231-232, 1999. 34. Roland PS, Anon J B, Moe R D, et al: Topical ciprofloxacin/ dexamethasone is superior to ciprofloxacin alone in pediatric patients with acute otitis media and otorrhea through tympanostomy tubes. Laryngoscope 113:2116-2122, 2003. 35. Roland PS, Kreisler L S, Reese B, et al: Topical ciprofloxacin/dexamethasone otic suspension is superior to ofloxacin otic solution in the treatment of children with acute otitis media with otorrhea through tympanostomy tubes. Pediatrics 113:e40-e46, 2004. 36. Roland PS, Stewart MG, Hannley M, et al: Consensus panel on role of potentially ototoxic antibiotics for topical middle-ear use: Introduction, methodology, and recommendations. Otolaryngol Head Neck Surg 130(3 No. 3S):S51-S56, 2004. 37. Schapowal A : Contact dermatitis to antibiotic ear drops is due to neomycin but not to ciprofloxacin [Abstract]. Allergy 56(Suppl 68):148, 2001. 38. VanGinkel C J, Bruintjes TD, Huizing E H : Allergy due to topical medications in chronic otitis externa and chronic otitis media. Clin Otolaryngol 20:326-328, 1995. 39. Sobol S E, Keswani S, Parvadia J K, et al: Effect of corticosteroids-antibiotic agents on granulation tissue in a murine model. Arch Otolaryngol Head Neck Surg 131:330-335, 2005. 40. Park S N, Yeo SW: Effects of antibiotics and steroids on middle ear mucosa in rats with experimental acute otitis media. Acta Otolaryngol 121:808-812, 2001. 41. Cutler J L , Wall G M, Labadie R D: Effects of ototopic steroid and NSAIDs in clearing middle ear effusion in an animal model. Otolaryngol Head Neck Surg 135:585589, 2006. 42. Florea A, Zwart J E, Lee CW, et al: Effect of topical dexamethasone versus rimexolone on middle ear inflammation in experimental otitis media with effusion. Acta Otolaryngol 126:910-915, 2006. 43. Alper C M, Je Dohar, Gulhan M, et al: Treatment of chronic suppurative otitis media with topical tobramycin and dexamethasone. Arch Otolaryngol Head Neck Surg 126:165-173, 2000. 44. Bertone A L : Management of exuberant granulation tissue. Vet Clin North Am Equine Pract 5:551-562, 1989. 45. Wachter BG, Leonetti J P, Lee J M, et al: Silver nitrate injury in the rat sciatic nerve: A model of facial nerve injury. Otolaryngol Head Neck Surg 127:48-54, 2002. 46. Hebert R L 2nd, Vick M L , King G E, Bent J P 3rd: Tympanostomy tubes and otic suspensions: Do they reach the middle ear space? Otolaryngol Head Neck Surg 122:330333, 2000.
118
OTOLOGIC SURGERY
47. Post JC, Hiller N L , Nistico L , et al: The role of biofilms in otolaryngologic infections: Update 2007. Curr Opin Otolaryngol Head Neck Surg 15:347-351, 2007. 48. Dohar J E, Hebda PA, Veeh R , et al: Mucosal biofilm formation on middle-ear mucosa in a nonhuman primate model of chronic suppurative otitis media. Laryngoscope 115:1469-1472, 2005. 49. Stenfors LE, Carlsӧӧ B, Salen B, et al: Repair of experimental tympanic membrane perforations. Acta Otorhinolaryngol 90:332-341, 1980. 50. Glasscock M, Levine S, McKennan K : Materials used in tympanoplasty. In Paparella M, et al: Otolaryngology. Philadelphia, Saunders, 1991, pp 1441-1447. 51. Derlacki E L : Residual perforations after tympanoplasty: Office technique for closure. Otolaryngol Clin North Am 15:861-867, 1982. 52. Derlacki E L : Office closure of central tympanic membrane perforations: A quarter century of experience. Trans Am Acad Ophthalmol Otolaryngol 77:53-66, 1973. 53. Ringenberg JC : Fat graft tympanoplasty. Laryngoscope 72:188-192, 1962. 54. Mitchell R B, Pereira K D, Younis RT, et al: Bilateral fat graft myringoplasty in children. Ear Nose Throat J 75(652):655-656, 1996. 55. Imamoglu M, Isik AU, Acuner O, et al: Fat-plug and paper-patch myringoplasty in rats. J Otolaryngol 27:318321, 1998.
56. Mitchell R B, Pereira K D, Lazar R H : Fat graft myringoplasty in children—a safe and successful day-stay procedure. J Laryngol Otol 111:106-108, 1997. 57. Ozgursoy O B, Yorulmaz I : Fat graft myringoplasty: A cost-effective but underused procedure. J Laryngol Otol 119:277-279, 2005. 58. Terry R M, Bellini M J, Clayton M I, Gandhi AG: Fat graft myringoplasty—a prospective trial. Clin Otolaryngol Allied Sci 13:227-229, 1998. 59. Fiorino F, Barbieri F: Fat graft myringoplasty after u nsuccessful tympanic membrane repair. Eur Arch Otorhinolaryngol 264�������������������������������������������� (10):1125-1128, 2007. 60. Landsberg R , Rishman G, DeRowe A, et al: Fat graft myringoplasty: Results of a long-term follow-up. J Otolaryngol 35:44-47, 2006. 61. Hagemann M, Hausler R : Tympanoplasty with adipose tissue. Laryngorhinootologie 82:393-396, 2003. 62. Ayache S, Braccini F, Facon F, Thomassin J M : Adipose graft: An original option in myringoplasty. Otol Neurotol 26:554-555, 2005. 63. Liew L , Daudia A, Narula A A : Synchronous fat plug myringoplasty and tympanostomy tube removal in the management of refractory otorrhoea in younger patients. Int J Pediatr Otorhinolaryngol 66:291-296, 2002.
9
Tympanoplasty—Outer Surface Grafting Technique Jose N. Fayad and James L. Sheehy Videos corresponding to this chapter are available online at www.expertconsult.com.
The aims of tympanoplasty are elimination of disease and restoration of function. Restoration of function requires a tympanic membrane; an air-containing, mucosal-lined middle ear (so that the membrane vibrates); and a secure connection between the tympanic membrane and the inner ear fluids. This chapter presents one of the three major techniques of tympanic membrane grafting: the outer surface, or onlay, procedure. This technique is used with rare exceptions by physicians of the House Clinic. Before describing the surgical procedure, the evolution of tympanic membrane grafting techniques, patient selection, and evaluation and counseling before surgery are discussed.
HISTORICAL ASPECTS Systematic reconstruction of the tympanic membrane, the sine qua non of the modern era of reconstructive ear surgery, had its beginning with reports by Wullstein1 and Zollner.2 Split-thickness or full-thickness skin was placed over the de-epithelialized tympanic membrane remnant. The initial results were very encouraging, but graft eczema, inflammation, and perforation were common. As a result of these experiences, most surgeons had begun changing to undersurface (underlay) connective tissue grafts by the late 1950s (see Chapters 11 and 12). The House Clinic physicians continued using an onlay technique, but changed to “canal skin,”3,4 which actually was periosteum graft covered by canal skin. This change was made in 1958, and resulted in an immediate improvement in results. Draining ears and total perforations continued to have a failure rate reaching 40%. In 1961, Storrs5 published the results of a small series of cases in which temporalis fascia had been used as an outer surface graft. Changing to this technique resulted in a dramatic improvement in results over the next 3 years: greater than 90% graft take.6-8
PATIENT SELECTION AND EVALUATION A patient with chronic otitis media may consult a physician because of a hearing impairment or because of discharge from the ear. Occasionally, the patient may have symptoms of more advanced chronic ear pathology, such as pain, vertigo, or facial nerve paralysis. Careful evaluation of the symptoms and findings allows the otologist to determine the need for surgery, its urgency, and the anticipated result. Only by completing the evaluation can the patient be advised properly. A good surgical result should not be a disappointment to the patient if there is proper counseling. Let us assume, for purposes of this chapter, that the patient has a dry central perforation. The ear may drain briefly with upper respiratory infections or if water is allowed to get into the ear. This discharge responds promptly to local medication. The preoperative treatment of the draining ear is discussed in depth in Chapter 16. When one is dealing with a dry central perforation, or inactive disease, surgery is elective, and the patient (or family) should be so informed. Assuming that the problem is unilateral, with only a mild hearing impairment, the only indication for surgery is to avoid further episodes of otorrhea. In children, it is best (from a psychological standpoint) to avoid elective surgery of any type between the ages of 4 and 7 years. In ear surgery, it is wise to wait until after age 7 so as not to lay the groundwork for serous otitis media. If the problem is bilateral, there is a hearing problem, and the ears do not drain often, fitting with hearing aids in each ear may be preferable for children younger than 8 years; however, the parents have to make the decision. At age 8 and thereafter, the patient (and child) should be allowed to make the decision. When contemplating tympanoplasty, the House Clinic physicians do not usually test to determine the status of the eustachian tube.9 The philosophy has been that tubal malfunction per se is not a contraindication 119
120
OTOLOGIC SURGERY
to tympanoplasty, but that the operation would not be successful unless tubal function is re-established. Many patients showing no tubal function by various available tests used in the past have been operated on to eliminate a chronic drainage problem. When the ear heals, the drum is usually mobile. Re-exploration in some of these patients has shown normal mucosa in the tubotympanum, where before surgery the mucosa was of a very poor quality. It would seem that the surgery, in eliminating infection and sealing the ear, is in itself the best treatment for the obstructed tube.
PATIENT COUNSELING What is the outlook with surgery, and what are the risks and complications? A surgeon must relate his or her own experience. House Clinic physicians explain that the likelihood of obtaining a permanently healed, dry ear, which may be treated normally, is better than 90%. “The only complication that happens with any degree of regularity, and is serious, is a total loss of hearing in the operated ear. That likelihood is no more than 1%. All of the other things listed here are either very remote or are temporary.” (“Listed here” refers to the Risk and Complications section of a patient Discussion Booklet. The Risk and Complications Sheet is given to the patient at the time the surgery is scheduled, which allows the patient to review the sheet leisurely. Appendix 1 is the Risk and Complications Sheet.)
PREOPERATIVE PREPARATION If the patient is a child, the preoperative visit occurs the day before surgery. Surgery is done under general anesthesia the following morning. The child is released to the parents’ care in the afternoon. For adults, the preoperative visit is sometimes the morning of surgery. The patient goes to the hospital for afternoon surgery, which is done under local or general anesthesia. The patient may be kept in the hospital overnight, depending on many circumstances.
PREPARATION IN THE OPERATING ROOM The smoothness with which the operation proceeds depends not only on the ability of the surgeon, but also on the organization of the team (anesthesiologist and surgical nurse) and arrangements in the operating room. The patient’s hair is shaved 3 cm above and behind the ear. The skin is cleansed with an iodine-based soap and rinsed with water. A sterile plastic adhesive drape is placed after applying tincture of benzoin. The mattress of the operating table is taped securely to the table to prevent it from slipping when the table is tipped from
side to side or into the Trendelenburg position. The patient is placed on the table with his or her head at the foot of the table, which allows the circulating nurse or anesthesiologist free access to the table controls, which are at the feet of the patient. The patient’s head and shoulders should be as near to the surgeon’s side of the table as possible. A pillow is placed under the patient’s knees. The Bovie plate goes under the patient’s buttocks.
Anesthesia Ear surgery can be performed with the patient under local anesthesia in many cases. The decision depends on the age of the patient and other factors. Lidocaine with 1:100,000 epinephrine is used for postauricular and meatal incisions. Some of the injection material should find its way under the skin of the posterior superior wall of the canal (the vascular strip) and into the middle ear. General anesthesia is used in children, in most mastoid surgery, and in procedures that require more than 1.5 hours of operating time. The anesthesiologist should be at the feet of the patient (which is actually the head of the table) to allow the surgeon and scrub nurse complete freedom at the patient’s head. A few useful suggestions for the anesthesiologist are as follows: 1. Have extra-long tubing for the gas machine to allow seating at the foot of the patient. 2. Start the intravenous infusion in the forearm, and extend the tubing to the foot of the table. 3. Place the blood pressure cuff on the arm opposite the ear to be operated on. 4. Be certain that the patient is secured to the table with wide adhesive tape. The eyes should be taped shut.
Arrangement and Instrumentation Particular attention should be paid to operating room arrangement (Fig. 9-1). The anesthesiologist is at the patient’s feet, far removed from the operating field, and the scrub nurse is directly across from the surgeon, where the nurse may be of the most assistance. The same arrangement, minus the anesthesiologist, is used for procedures done with the patient under local anesthesia. It is important for the comfort of the surgeon that the patient is in a satisfactory position. The table is usually placed in a few degrees of Trendelenburg position and rolled slightly toward the surgeon. The patient’s head is adjusted as necessary, usually flexed slightly onto the opposite shoulder. The surgeon should be comfortably seated on a chair with a back support. The surgeon should use the back support and be in a comfortable position so that all back and arm muscles are relaxed.
Chapter 9 • Tympanoplasty—Outer Surface Grafting Technique
121
122
OTOLOGIC SURGERY
SURGICAL TECHNIQUE The lateral surface grafting technique involves eight steps: (1) transmeatal canal incisions, and elevations of the vascular strip; (2) postauricular exposure, and removal and dehydration of the temporalis fascia; (3) removal of canal skin; (4) enlargement of the ear canal by removal of the anterior (and inferior) canal bulge; (5) de-epithelialization of the tympanic membrane remnant; (6) placement of the rehydrated fascia on the outer surface of the remnant, but under the manubrium; (7) replacement of canal skin; and (8) closure of the postauricular incision and replacement of the vascular strip transmeatally.10
Transmeatal Incisions Incisions are made along the tympanomastoid and tympanosquamous suture lines, demarcating the vascular strip with a No. 1 (sickle) knife (Fig. 9-2). The vascular strip is the area of the canal skin that covers the superior and posterior portions of the ear canal between these two suture lines. It is easily demarcated from the skin of the remainder of the ear canal because of its thickness, and the fact that it balloons up when local anesthesia is injected into the area. The vascular strip is elevated from the bone, from within outward using a round knife (Fig. 9-3). A semilunar incision is made in the outer third of the ear canal, using a Beaver knife with a No. 64 blade, connecting the two incisions already made along the border of the vascular strip (Fig. 9-4). The knife blade is angled toward the bone to thin the 1 or 2 mm section of the membranous canal included.
Postauricular Exposure and Removal of Fascia The skin incision must provide adequate exposure for the operative field. It should extend far enough forward superiorly and inferiorly to allow adequate exposure of the bony meatus when the ear is retracted forward. Failure to do this may result in difficulty seeing structures in the posterior part of the middle ear. The superior portion of the incision begins at the most anterior extent of, and 1 cm above, the postauricular fold. It is continued into the postauricular fold at the level of the lower border of the muscle and extends inferiorly under the lobule of the ear. Exposure of the temporalis fascia is facilitated if ample local anesthetic has been injected to balloon the area. A retractor is inserted to retract the skin margins in this area and to obtain hemostasis. By lifting up on the retractor, one may pull the areolar tissue away from the fascia, facilitating the dissection and ensuring that all loose areolar tissue is lifted off the fascia before the fascia’s removal. Local anesthesia is injected under the fascia to elevate it slightly from the underlying muscle. A 2 × 2 cm piece
of fascia is removed. The fascia is spread on a polytef (Teflon) block, undersurface upward, and any adherent muscle is removed. The fascia is placed on a fascia press, absorbable gelatin sponge (Gelfoam) is placed on the fascia, and the press is closed. The press is opened after 5 minutes, and the Gelfoam is removed; the fascia, now smooth and partially dehydrated, is left attached to the press. The press, with the attached fascia, is placed under an electric lamp to complete the dehydration process. An incision is made through the soft tissue above the meatus, from the root of the zygoma, horizontally, along the linea temporalis. This horizontal incision is extended posteriorly to the level of the skin incision. The incision is extended inferiorly, below the linea temporalis, following the postauricular incision, incising the periosteum until the incision curves forward, down to the level of the floor of the ear canal. The periosteum is elevated superiorly (under the temporalis muscle), posteriorly and anteriorly, using a Lempert elevator, to obtain adequate exposure of the mastoid cortex. A self-retaining retractor is inserted to retract the auricle and vascular strip forward, exposing the ear canal.
Removal of the Canal Skin The periosteum and canal skin are elevated from the bone as far as the annular ligament (Fig. 9-5). Care should be taken not to elevate the ligament and the remnant of the middle fibrous layer. The dissection is superficial to the fibrous layer of the remnant in such a way that the remnant is de-epithelialized in continuity with the canal skin, if possible. It is often easier to begin the final removal and de-epithelialization by starting anterosuperiorly, using a cup forceps (Fig. 9-6). Removal of the canal skin and de-epithelialization are continued inferiorly and posteriorly. The periosteum and canal skin are removed from the ear and kept moist in Tis-U-Sol irrigating solution. In elevating the periosteum and the canal skin, one works perpendicular to the annular ligament and remnant, keeping the instrument on the bone at all times, until the dissection is completed to the level of the remnant. The dissection is continued parallel to the annular ligament to avoid elevating it and the remnant (Fig. 9-7).
Enlargement of the Ear Canal Through the use of a drill and continuous suctionirrigation, the ear canal is enlarged by removal of the anterior and inferior canal bulges (Fig. 9-8). The importance of this step in the lateral surface grafting technique cannot be overemphasized. Removal of this bone enlarges the field of surgery. The anterior and inferior sulci are thoroughly exposed to allow de-epithelialization and satisfactory graft placement. The acute angle that exists anteriorly is opened, helping to prevent postoperative
Chapter 9 • Tympanoplasty—Outer Surface Grafting Technique
123
124
OTOLOGIC SURGERY
blunting. There is no area hidden from postoperative observation. Enlarging the ear canal is routine in all lateral surface grafting procedures, and is the main reason for removal of canal skin.
De-epithelialization of the Tympanic Membrane Remnant The lateral surface grafting technique demands a complete de-epithelialization of the remnant. Although the graft may take without all the skin having been removed, postoperative epithelial cysts may develop. Attention is directed first to the ear canal bone immediately adjacent to the bony annulus. Particular attention is paid to the anteroinferior bone 1 mm lateral to the annulus, where a small vessel and nerve perforate the bone, and where there is a particularly tight attachment to the skin. Next, the tympanic membrane remnant is checked carefully for skin. If there is a question about whether de-epithelialization has been thorough, a portion of the remnant is removed to be certain. The size of the perforation is of no consequence in regard to graft take in the lateral surface technique.
Preparation of Packing The surgical nurse should have begun preparing Gelfoam packing before or shortly after the beginning of the operation. Uncompressed Gelfoam is cut into an ample number of various-sized pieces and soaked in antibioticcortisone solution. Pieces of Gelfoam are removed from the solution and compressed (on a tongue blade or a paper Gelfoam packet) until most of the solution has been removed. The Gelfoam is put aside and allowed to dry more until needed by the surgeon.
Placement of Fascia When the perforation is large, or the fascia is unusually thin, it is helpful to fill the middle ear with Gelfoam packing before placing the graft. The Gelfoam serves as an artificial remnant and facilitates graft placement. The fascia is placed under the malleus handle. When the manubrium is surrounded by remnant (small perforation), the remnant is separated from the malleus handle to allow proper placement of the fascia. The dehydrated fascia is trimmed to an oval shape measuring approximately 1.3 × 1.5 cm. A slit is cut in the fascia to allow placement under the manubrium (Fig. 9-9). The two cut ends are grasped with the forceps, and the fascia is immersed for a few seconds in Tis-U-Sol irrigating solution to hydrate it. The fascia is placed over the perforation and immediately slipped under the manubrium (Fig. 9-10). One ensures that the apex of the slit in the fascia comes into contact with the tensor tendon. The fascia is adjusted to the remnant anteriorly and inferiorly, with care
being taken that it does not extend onto the bony wall anteriorly, unless there is no remnant at all, then it should extend only for 1 mm at the most. The anterior flap is turned back over the exposed manubrium, resulting in a better appearance of the membrane when healed (Fig. 9-11). When the malleus is absent, and there is only a small remnant present, it is necessary to insert the fascia in a different way to result in the stabilized graft (Fig. 9-12). The fascia is cut twice, creating a flap that can be tucked under the lateral wall of the epitympanum. The anterosuperior edge of the fascia is swung posteriorly to overlap the upper edge of the graft and secure the seal of the middle ear (Fig. 9-13).
Replacement of Canal Skin The canal skin is replaced to cover the bone from which it was removed. It is positioned only slightly more medially, allowing it to overlap the fascia by 1 mm (Fig. 9-14), which helps promote rapid epithelialization. Epithelialization is particularly important in preventing blunting in the anterosuperior sulcus. There must be no edges of epithelium turned under, or small epithelial cysts may develop on the surface of the fascia during healing. The first piece of packing is a small, dry, rolled-up (cigar-shaped), tightly compressed piece of absorbable Gelfoam, which is placed in the sulcus anteriorly. The canal is packed tightly with pledgets of slightly moist Gelfoam, leaving room posterosuperiorly for the vascular strip.
Closure and Replacement of the Vascular Strip The retractors are released, and the vascular strip is pushed anteriorly to lie over the packing. One suture is placed subcutaneously postauricularly to stabilize the auricle. Transmeatally, the vascular strip is replaced in the ear canal in the exact position from which it came (Fig. 9-15). Gelfoam packing in the canal is completed, and a plug of cotton is placed in the outer meatus. The postauricular incision is closed with subcutaneous sutures, and a mastoid dressing is applied.
POSTOPERATIVE CARE The mastoid dressing is removed the day after surgery. The patient is given a postoperative instruction card (Appendix 2) just before entering the hospital. It is important to review some aspects of this information with the patient. The patient should be reminded not to blow the nose and not to get water in the ear. An antibiotic is prescribed and should be taken as directed on the prescription label. There will be discomfort for a few days, and the patient should take aspirin or acetaminophen
Chapter 9 • Tympanoplasty—Outer Surface Grafting Technique
four times a day regularly for the first few days to keep the pain under control. Nothing need be done with the cotton in the ear, but it may be changed if it becomes soiled. The patient is asked to touch the edge of the auricle; the physician points out that the ear is numb, and that it is going to take a few months for the numbness to fade. There is also tenderness on the incision behind the ear. This tenderness diminishes rapidly, but it may be 6 months before it is totally gone. Finally, the patient should be reminded of the first postoperative appointment, which
125
should be scheduled 7 to 10 days later in the physician’s office. At the first postoperative visit, the cotton plug in the ear canal is removed, and the ear is inspected. The Gelfoam should appear firm. A piece of Gelfoam should be removed to show to the patient so that the patient understands that it will eventually turn to a liquid and will run out of the ear. The patient is instructed to begin using ear drops (of one type or another) 3 weeks after the date of surgery, twice daily. The drops may be started sooner if the ear begins to drain, an indication of liquefaction of
126
OTOLOGIC SURGERY
the Gelfoam. The second postoperative visit is scheduled for 6 to 8 weeks after the date of surgery. At this point, 80% to 90% of the ear should be totally healed.
PROS AND CONS OF OUTER SURFACE TECHNIQUE One of the problems faced by a novice surgeon is that there are many techniques and prostheses recommended as “the best—always works well.” How well a technique or prosthesis works for the individual depends on the technical ability of that individual—his or her judgment and manual dexterity.11 The outer surface grafting technique has numerous advantages and disadvantages. The advantages are well known to all who have used the technique. First, the exposure is excellent—one can see everything necessary without moving the microscope. Second, one may remove as much remnant as necessary to eliminate the disease. There is no need to scrape in many different places. Third, the graft rate take is high. Fourth, it is one technique that can be used in all cases. There are disadvantages that may outweigh the advantages for some individuals. First, the technique requires very precise surgery to avoid problems. Second, the healing time is longer than with the undersurface technique. Third, if the operation is not done extremely skillfully, the patient may develop either blunting in the anterior sulcus or lateral healing, both of which may result in a healed ear with worse hearing.
Healing Problems One of the disadvantages of the outer surface grafting technique, as already noted, is that although there is a very high graft take rate regardless of how the operation is performed, there are healing problems. These healing problems may outweigh the advantages for some individuals.10 These healing problems became evident soon after the technique was started in the early 1960s: lateralization of the graft, blunting in the anterior sulcus, excessive membrane thickness, epithelial cysts between the remnant and the fascia, and epithelial pearls on the drum surface and ear canal. Most of these have ceased to be major problems, but they still occur in a small percentage of cases.
Lateralization of the Tympanic Membrane The very first problem that was noticed with this technique was lateralization of the membrane (Fig. 9-16). This lateralization usually did not become apparent until 6 to 12 months after surgery, and resulted from the fascia not being placed under the malleus handle when the technique was first introduced (see Fig. 9-10). When lateralization has occurred, the patient’s hearing is reduced, but often not as much as anticipated. The appearance is
of an eardrum smaller than normal-sized, mobile, and at a direct right angle to the line of vision. Treatment of this problem (if needed) requires reoperation and the placement of the new graft underneath the malleus handle.
Blunting in the Anterior Sulcus Blunting in the anterior sulcus, particularly anterosuperiorly, occurs to a minor degree in most lateral surface cases, but is of no consequence and results from the formation of excess fibrous tissue. Blunting is probably the most common healing problem encountered by the novice surgeon. It can interfere greatly with the hearing result if it is great enough to involve the malleus handle when the ossicular chain is intact (Fig. 9-17). To prevent blunting, one should remove the anterior canal bulge so that the anterior angle is open. One should not place the fascia onto the anterior canal bone unless there is no alternative. Finally, one should ensure that the replaced canal skin overlaps the graft slightly anterosuperiorly. Finally, placing the rolled-up piece of dried Gelfoam on the anterior sulcus as the first piece of packing also helps prevent blunting. When severe blunting occurs, the manubrium becomes indistinguishable, and the anterior half of the tympanic membrane is immobile and takes on a concave appearance with no clear-cut distinction between the edge of the membrane and the bony wall. The posterior half of the membrane may show fair mobility. If this appearance persists after 6 months, and there is a hearing problem, reoperation is required to correct it.
Other Problems Two varieties of epithelial cysts may be noted. One cyst is common and appears as a small pearl on the tympanic membrane or ear canal. It is the result of turning under the skin edges when replacing the canal skin. Spontaneous rupture and healing are common. The cyst might be marsupialized under the microscope in the office if desired. An epithelial cyst may occur between the remnant and the fascia, and enlarge slowly over 1 to 2 years (Fig. 9-18). This is an uncommon problem that results from inadequate de-epithelialization of bone and the remnant adjacent to the bone. The only place where this is likely to occur is anteroinferiorly where the small vessel and nerve enter the ear canal 1 mm lateral to the drum. As opposed to blunting, in which there is a concave appearance, the appearance here is convex, and it occurs anteroinferiorly. When it is recognized, it can be corrected by incising the cyst and evacuating it.
Acknowledgments Many of the illustrations are modified from Sheehy JL, Brackmann DE: Surgery of chronic otitis media. In English GM (ed): Otolaryngology. Philadelphia, Lippincott, 1994.
Chapter 9 • Tympanoplasty—Outer Surface Grafting Technique
REFERENCES 1. Wullstein H: Theory and practice of tympanoplasty. Laryngoscope 66:1076-1093, 1956. 2. Zollner F: Principles of plastic surgery of the soundconducting apparatus. J Laryngol Otol 69:637-652, 1955. 3. House WF, Sheehy J L : Myringoplasty: Use of ear canal skin compared with other techniques. Arch Otolaryngol Head Neck Surg 73:407-415, 1961. 4. Plester D: skin and mucous membrane grafts in middle ear surgery. Arch Otolaryngol Head Neck Surg 72:718721, 1960. 5. Storrs L A : Myringoplasty with use of fascia graft. Arch Otolaryngol Head Neck Surg 74:45-49, 1961. 6. Sheehy J L : Tympanic membrane grafting: Early and long-term results. Laryngoscope 74:985-988, 1964.
127
7. Sheehy J L , Glasscock M E : Tympanic membrane grafting with temporalis fascia: A report of four years’ experience. Arch Otolaryngol Head Neck Surg 86:391-402, 1967. 8. Sheehy J L , Anderson RG: Myringoplasty: A review of 472 cases. Ann Otol Rhinol Laryngol 89:331-334, 1980. 9. Sheehy J L : Testing Eustachian tube function. Ann Otol Rhinol Laryngol 90:562-564, 1981. 10. Sheehy J L , Brackmann D E : Surgery of chronic otitis media. In English G M (ed): Otolaryngology. Philadelphia, Lippincott, 1994. 11. Sheehy J L , Brackmann D E : Surgery of chronic ear disease: What we do and why we do it. In: Instructional Courses. Vol. 6, St. Louis, Mosby, 1993.
128
OTOLOGIC SURGERY
APPENDIX 1: RISKS AND COMPLICATIONS OF MYRINGOPLASTY, TYMPANOPLASTY, MASTOID SURGERY, AND OTHER OPERATIONS FOR CORRECTION OF CHRONIC EAR INFECTIONS (Operations to eliminate middle ear or mastoid infection, to repair the eardrum or the sound transmission mechanism)
Ear Infection Ear infection with drainage, swelling, and pain may persist after surgery or, rarely, may develop after surgery because of poor healing of the ear tissue. If this is the case, additional surgery may be necessary to control the infection.
Loss of Hearing Further permanent impairment of hearing develops in 3% of patients because of problems in the healing process. In 2%, this loss of hearing may be severe or total in the ear that was operated on. Nothing further can be done in these instances. When a two-stage operation is necessary, the hearing is usually worse after the first operation.
Tinnitus Should the hearing be worse after surgery, tinnitus (head noises) likewise may be more pronounced.
Dizziness Dizziness may occur immediately after surgery because of irritation of the inner ear structures. Some unsteadiness may persist for 1 week postoperatively. Prolonged dizziness is rare, unless there was dizziness before surgery.
Taste Disturbance and Mouth Dryness Taste disturbance and mouth dryness are common for a few weeks after surgery. In some patients, this disturbance is prolonged.
Facial Paralysis A rare postoperative complication of ear surgery is temporary paralysis of one side of the face. This may occur as a result of an abnormality or a swelling of the nerve, and usually subsides spontaneously. Very rarely, the nerve may be injured at the time of surgery, or it may be necessary to excise it to eradicate infection. When this happens, a skin sensation nerve is removed from the upper part of the neck to replace the facial nerve. Paralysis of the face under these circumstances lasts 6 months to 1 year, and there would be a permanent residual weakness. Eye complications requiring treatment by a specialist could develop.
Hematoma A hematoma (collection of blood) develops in a small percentage of cases, prolonging healing. Reoperation to remove the clot may be necessary if this complication occurs.
General Anesthesia Complications Anesthetic complications are very rare, but can be serious. You may discuss these with the anesthesiologist if you desire.
Complications Related to Mastoid Surgery A cerebrospinal fluid leak (leak of fluid surrounding the brain) is a very rare complication. Reoperation may be necessary to stop the leak. Intracranial (brain) complications, such as meningitis or brain abscess, or even paralysis, were common in cases of chronic otitis media before the antibiotic era. Now these are extremely rare complications.
Chapter 9 • Tympanoplasty—Outer Surface Grafting Technique
129
APPENDIX 2: POSTOPERATIVE INSTRUCTION FOLDER: MYRINGOPLASTY, TYMPANOPLASTY, AND MASTOIDECTOMY Precautions 1. Do not blow your nose until your physician has indicated that your ear is healed. Any accumulated secretions in the nose may be drawn back into the throat and expectorated if desired. This is particularly important if you develop a cold. 2. Do not “pop” your ears by holding your nose and blowing air through the eustachian tube into the ear. If it is necessary to sneeze, do so with your mouth open. 3. Do not allow water to enter the ear until advised by your physician that the ear is healed. Until such time, when showering or washing your hair, lamb’s wool or cotton may be placed in the outer ear opening and covered with Vaseline. If an incision was made in the skin behind your ear, water should be kept away from this area for 1 week. 4. Do not take an unnecessary chance of catching cold. Avoid undue exposure or fatigue. Should you catch a cold, treat it in your usual way, reporting to your physician if you develop ear symptoms. 5. You may anticipate a certain amount of pulsation, popping, clicking, and other sounds in the ear, and a feeling of fullness in the ear. Occasional sharp shooting pains are not unusual. Sometimes it may feel as if there is liquid in the ear. 6. Do not plan to drive a car home from the hospital. Air travel is permissible 2 days after surgery. When changing altitude, you should remain awake and chew gum to stimulate swallowing.
Dizziness Minor dizziness may be present on head motion and need not concern you unless it increases.
Hearing An improvement in hearing is rarely noted immediately after surgery. Hearing may even be worse temporarily because of swelling of the ear tissues and packing in the ear canal. An improvement may be noted 6 to 8 weeks after surgery. Maximum improvement may require 4 to 6 months.
Discharge A bloody or watery discharge may occur during the healing period. The outer ear cotton may be changed if necessary, but generally, the less done to the ear, the better. A yellow (infected) discharge at any time is an indication to call to make an appointment to see your physician. Discharge with foul odor should also be reported.
Pain Mild, intermittent ear pain is not unusual during the first 2 weeks. Pain above or in front of the ear is common when chewing. If you have persistent ear pain, not relieved by a few aspirins, call to make an appointment to see your physician.
Ear Drops If you were given a prescription for ear drops, begin using these 3 weeks after surgery. Place a few drops in the ear twice daily to loosen the packing, which will run out of the ear as a liquid. Tip the head to the side, place two drops in the ear, and allow them to remain for 5 minutes. Then tip the head in the opposite direction to allow the ear drops to run out. Continue doing this twice daily until you have finished the drops or until advised otherwise by your physician.
10
Cartilage Tympanoplasty John L. Dornhoffer Videos corresponding to this chapter are available online at www.expertconsult.com.
The use of cartilage in middle ear surgery is not a new concept; it has been recommended on a limited basis to manage retraction pockets for many years.1-5 More recently, the use of cartilage has been increasingly described for the reconstruction of large portions of the pars tensa of the tympanic membrane in cases of recurrent perforation, atelectasis, and cholesteatoma.6-8 Although one might anticipate a significant conductive hearing loss with cartilage because of its thickness and rigidity, several studies have reported results to the contrary, suggesting hearing results with cartilage to be no different than results with fascia.6-8 It has been shown in experimental and clinical studies that cartilage is well tolerated by the middle ear, and long-term survival is the norm.9-12 Cartilage grafts seem to be nourished largely by diffusion and become well incorporated in the tympanic membrane.3 Human and animal studies have shown that although some softening occurs with time, the matrix of the cartilage remains intact, but with development of empty lacunae, showing degeneration of the chondrocytes.13,14 The cartilage graft retains its rigid quality and resists resorption and retraction, even in the milieu of continuous eustachian tube dysfunction. Two distinct techniques are commonly employed for cartilage reconstruction of the tympanic membrane: the perichondrium/cartilage island flap, which uses tragal cartilage, and the palisade technique, which uses cartilage from the tragus or cymba. The choice of technique is typically dictated by the specific middle ear pathology or, in cases where the tympanic membrane reconstruction is in conjunction with ossiculoplasty, the status of the ossicular chain. The palisade technique is preferred in cases of cholesteatoma, and when ossicular reconstruction is needed in the situation with the malleus present. In this situation, an exact fit is necessary to prevent cholesteatoma recurrence. The perichondrium/cartilage island flap is preferred for management of atelectatic ears and high-risk perforations. This chapter describes the two techniques in detail, followed by descriptions of modifications in response to specific surgical indications.
PATIENT SELECTION Generally, cartilage is used as a graft material in any ear considered to be at high risk for failure with traditional techniques using temporalis fascia or perichondrium. Included in this group would be high-risk perforations, atelectatic ears, and cases of cholesteatoma. High-risk perforation comprises a revision surgery, a perforation anterior to the annulus, a perforation draining at the time of surgery, a perforation larger than 50%, or a bilateral perforation, all of which have been shown to be associated with increased failure rates using traditional techniques.15,16 The atelectatic ear is one of the most important indications for cartilage tympanoplasty, and numerous reports have established its efficacy over fascia in this situation.1-3 For similar reasons, the use of cartilage to reconstruct and reinforce the scutum and posterior half of the eardrum in cholesteatoma surgery has reduced the incidence of recurrent atrophy and retraction pockets in these difficult cases. Cartilage tympanoplasty has proven to be efficacious in pediatric and adult patients, with special precautions used in the former group. The general approach to pediatric patients is to avoid repairing the tympanic membrane during the otitis-prone years (<3 years old). If the contralateral ear is normal, routine tympanoplasty is performed at age 4.17 If the contralateral ear is abnormal at this time, adenoidectomy is considered, and tympanoplasty is generally deferred until age 7.18,19 If contralateral disease is still present at this time, cartilage tympanoplasty is performed on the worse ear because a perforation in the contralateral ear has been shown to be associated with a high risk for failure.20 As part of the preoperative preparation, all patients are encouraged to perform the Valsalva maneuver (or use the Otovent [Invotec International, Jacksonville, FL] in younger children). Patients unable to insufflate the ear are placed on nasal steroids 6 weeks before surgery, and these are continued in the postoperative period until an aerated middle ear cleft is documented. Although we have 131
132
OTOLOGIC SURGERY
found no difference in graft take between patients who can and cannot perform the Valsalva maneuver, we have found a slightly increased need (7%) for postoperative tube insertion in the Valsalva-negative group. Likewise, an attempt is made to optimize concomitant sinonasal disease (allergy, chronic sinusitis) before ear surgery, and smoking cessation is encouraged where applicable. The draining ear is treated with antibiotic/steroid-containing topical solutions and aural toilet for 6 to 8 weeks before surgery. Although every attempt is made to dry an ear before surgical intervention, it is not considered a prerequisite for tympanoplasty.
SURGICAL TECHNIQUE
FIGURE 10-1. Harvest of cartilage, leaving small rim of cartilage in
General Considerations
dome for cosmesis (right ear). (From Dornhoffer JL: Cartilage tympanoplasty: Indications, techniques, and outcomes in a 1000-patient series. Laryngoscope 113:1844-1856, 2003.)
The surgical approach for cartilage tympanoplasty does not differ from the approach for traditional otologic surgery and is dictated by the extent and location of the disease. The postauricular approach is used in most patients because these cases, by definition, tend to have more extensive middle ear pathology. A small, localized, posterior retraction or perforation can be performed through a transcanal or endaural incision, but great care must be taken to ensure that the depths of the retraction can be reached. Likewise, the placement of the cartilage graft uses the underlay technique, so no special tympanomeatal flaps or skin incisions are required. Some general observations should be made regarding the differences in cartilage that occur with aging. The cartilage thickness of the tragus and cymba are not appreciably different between children and adults, but the perichondrium seems to be more adherent in children. For this reason, when the perichondrium is removed from one side when fashioning the graft (described later), care must be taken to ensure the correct plane is dissected, especially in children. Likewise, cartilage is more pliable in children, making it slightly easier to work with than adult cartilage, which can become brittle in patients older than 65 years. For this reason, during the fashioning of the graft, it is generally a good practice to manipulate and hold the cartilage with fingers instead of forceps, and toothed forceps should never be used to grasp the cartilage to avoid fracture. The perichondrium generally is left attached to the side of the cartilage that faces the ear canal, regardless of which technique is used. It is believed that the perichondrium improves graft stability and facilitates ingrowth of fibrous tissue and epithelialization. With the palisade technique, however, perichondrium occasionally has been removed from both sides of the cartilage because of untoward curling of the graft, which typically curves toward the side with the perichondrium. No ill effect has been noted, other than increased fragility of the graft during formation and placement.
Perichondrium/Cartilage Island Flap The general technique of reconstruction using the perichondrium/cartilage island flap begins with harvest of the cartilage from the tragal area (see Video Clip 10-1).21 This cartilage is ideal because it is thin, flat, and in sufficient quantities to permit reconstruction of the entire tympanic membrane. The cartilage is used as a fullthickness graft and is typically slightly less than 1 mm thick in most cases. Although it has been suggested that a slight acoustic benefit could be obtained by thinning the cartilage to 0.5 mm,22 this advantage is offset by the unacceptable curling of the graft, which occurs when the cartilage is thinned, and perichondrium is left attached to one side. An initial cut through skin and cartilage is made on the medial side of the tragus, leaving a 2 mm strip of cartilage in the dome of the tragus for cosmesis (Fig. 10-1). The cartilage, with attached perichondrium, is dissected medially from the overlying skin and soft tissue by spreading a pair of sharp scissors in a plane that is easily developed superficial to the perichondrium on both sides. It is necessary to make an inferior cut as low as possible to maximize the length of harvested cartilage. The cartilage is grasped and retracted inferiorly, which delivers the superior portion from the incisura area. The superior portion is dissected out while retracting, which produces a piece of cartilage typically measuring 15 × 10 mm in children and larger in adults. The perichondrium from the side of the cartilage farthest from the ear canal is dissected off, leaving the thinner perichondrium on the reverse side. A perichondrium/cartilage island flap is constructed in the following manner. Using a round knife, cartilage is dissected from the graft to produce an eccentrically located disc, approximately 7 to 9 mm in diameter, which is used for total
133
Chapter 10 • Cartilage Tympanoplasty 1.
2.
2.
3.
3. 4.
FIGURE 10-2. Prepared perichondrium/cartilage island graft, showing strip of cartilage removed to facilitate malleus. (From Dornhoffer JL: Cartilage tympanoplasty: Indications, techniques, and outcomes in a 1000patient series. Laryngoscope 113:1844-1856, 2003.)
1. 2. 3. 4.
Canal skin (external auditory) Perichondrium Cartilage Malleus
FIGURE 10-3. Lateral line drawing showing proper placement of graft.
tympanic membrane reconstruction. A flap of perichondrium is produced posteriorly that eventually drapes over the posterior canal wall. A complete strip of cartilage 2 mm wide is removed vertically from the center of the graft to accommodate the entire malleus handle (Fig. 10-2). The creation of two cartilage islands in this manner is essential to enable the reconstructed tympanic membrane to bend and conform to the normal conical shape of the tympanic membrane. When the ossicular chain is intact, an additional triangular piece of cartilage is removed from the posterosuperior quadrant to accommodate the incus. This excision prevents the lateral displacement of the posterior portion of the cartilage graft that sometimes occurs because of insufficient space between the malleus and incus. The entire graft is placed in an underlay fashion, with the malleus fitting in the groove and pressing down into and conforming to the perichondrium (Fig. 10-3). The cartilage is placed toward the promontory, with the perichondrium immediately adjacent to the tympanic membrane remnant, both of which are medial to the malleus. Failure to remove enough cartilage from the center strip causes the graft to fold up at the center instead of lying flat in the desired position. Likewise, if the strip is insufficient, the cartilage may be displaced medially instead of assuming a more lateral position in the same plane as the malleus. Absorbable gelatin sponge (Gelfoam) is packed in the middle ear space underneath the anterior annulus to support the graft in this area, and the posterior flap of perichondrium is draped over the posterior canal wall. Middle ear packing is avoided on the promontory and in the vicinity of the ossicular chain. One piece of Gelfoam is placed lateral to the reconstructed tympanic membrane, and antibiotic ointment is placed in the ear canal (Fig. 10-4).
(From Dornhoffer JL: Cartilage tympanoplasty: Indications, techniques, and outcomes in a 1000-patient series. Laryngoscope 113:1844-1856, 2003.)
FIGURE 10-4. Postoperative ear with perichondrium/cartilage island graft (left ear). (From Dornhoffer JL: Cartilage tympanoplasty: Indications, techniques, and outcomes in a 1000-patient series. Laryngoscope 113:18441856, 2003.)
Palisade Technique When the palisade technique is used for reconstruction of the tympanic membrane, cartilage can be harvested from either the tragus or the cymba (see Video Clip 10-2). Cartilage from the cymba area of the conchal bowl is used if the surgical approach involves a postauricular incision. Tragal cartilage is used if the approach is transcanal or endaural. The cartilage of the cymba is similar
134
OTOLOGIC SURGERY Lateral attic defect (cholesteatoma)
TM defect
A Elevated TM flap
FIGURE 10-5. A and B, Schematic of palisade techCartilage palisade
B
to the cartilage of the tragus in that it has an acceptable thickness of about 1 mm, in contrast to other areas of the concha, which are thicker and irregular. Cymba cartilage is more curved, however, which makes it particularly suitable for this technique because, in contrast to the perichondrium/cartilage island flap technique, it does not require one large, flat piece of cartilage. The cartilage is cut into several slices that are subsequently pieced together, similar to a jigsaw puzzle, to reconstruct the tympanic membrane (Figs. 10-5 and 10-6). A large area of conchal eminence can be exposed by elevating the subcutaneous tissue and postauricular muscle from the conchal perichondrium. The cymba cartilage is the prominent bulge at the superior aspect of the concha (Fig. 10-7). A circumferential cut the size of the anticipated graft is made through the perichondrium and cartilage, but not through the anterior skin. The perichondrium is removed from the postauricular side, and the cartilage, with the perichondrium on the anterior
nique (right ear). TM, tympanic membrane. (From Dornhoffer JL: Cartilage tympanoplasty: Indications, techniques, and outcomes in a 1000-patient series. Laryngoscope 113:1844-1856, 2003.)
aspect, is dissected from the skin. This technique is also used for harvesting cartilage for canal wall reconstruction when the retrograde mastoidectomy technique is used for cholesteatoma surgery. The technique described here differs from the palisade tympanoplasty of Heermann and colleagues.23 Instead of placing rectangular strips of cartilage side to side, an attempt is made to cut one major piece of cartilage in a semilunar fashion, which is placed directly against the malleus on top of the prosthesis (Fig. 10-8A and B). This piece of cartilage acts to reconstruct a major portion of the posterior half of the tympanic membrane, and serves as a foundation for the rest of the cartilage pieces. A second semilunar piece is placed between this first piece and the canal wall to reconstruct the scutum precisely (Fig. 10-8C). Any spaces that result between this cartilage and the canal wall or scutum are filled in with small slivers of cartilage to prevent prosthesis extrusion and recurrent retraction (Fig. 10-8D). The reconstruction is covered
Chapter 10 • Cartilage Tympanoplasty
135
technique allows direct visualization and contact of the notched prosthesis to the manubrium handle, which has been shown to provide superior hearing results.24 The prosthesis acts as a scaffolding on which the cartilage is placed, which serves to reconstruct the tympanic membrane and prevent prosthesis extrusion. It likewise allows a precise and watertight fit between the reconstructed tympanic membrane and the canal wall in the posterior area, where recurrent cholesteatoma most frequently occurs. Typically, in these situations, the anterior half of the tympanic membrane is not altered or is grafted with conventional materials to allow cholesteatoma surveillance and possible intubation in the postoperative period if necessary.
1.
POSTOPERATIVE CARE 1. Cymba cartilage graft (donor site)
FIGURE 10-6. Schematic illustrating location of cymba cartilage (left ear). (From Dornhoffer JL: Cartilage tympanoplasty: Indications, techniques, and outcomes in a 1000-patient series. Laryngoscope 113:1844-1856, 2003.)
1. 3.
2. 1. Cartilage 2. Perichondrium 3. Post-auricular skin incision
FIGURE 10-7. Harvesting of cymba cartilage with postauricular incision (right ear). (From Dornhoffer JL: Cartilage tympanoplasty: Indi cations, techniques, and outcomes in a 1000-patient series. Laryngoscope 113:1844-1856, 2003.)
with the previously harvested perichondrium draped over the posterior canal wall (Fig. 10-9). Although this technique can be used for tympanic membrane reconstruction without ossicular reconstruction, it is favored when ossiculoplasty is performed in a situation with the malleus present, and is especially suitable for cholesteatoma surgery. Because the prosthesis is placed before the cartilage reconstruction, the palisade
At 1 to 2 weeks after surgery, the packing material of Gelfoam and antibiotic ointment is completely suctioned from the external canal. Antibiotic/steroid-containing drops are used for an additional 2 weeks to clear the ear of residual ointment and Gelfoam, the latter of which can lead to granulation and fibrous tissue formation if inadequately removed from the tympanic membrane. Adult patients are instructed to begin the Valsalva maneuver, and children are instructed to use the Otovent, three times a day beginning 2 to 3 weeks after the surgery. A postoperative audiogram is obtained 6 to 8 weeks after surgery, at which time the tympanic membrane is examined. Impedance tympanometry is unreliable after cartilage tympanoplasty, and generally yields a lowvolume, type B tympanogram, despite normal hearing, because of the noncompliant nature of the graft. It is necessary to check air and bone conduction after the surgery and to use hearing levels to determine whether an effusion is present. If the hearing result is good, and the tympanic membrane is clear, the ear is examined at 6 months and at 1 year from the date of surgery. If an effusion is present based on observation or conductive hearing loss, nasal steroids are added, the Valsalva maneuver (or Otovent) is encouraged, and the ear is examined at 3 months. In a case of postoperative conductive hearing loss, if it is unclear whether an effusion is present because of the opacity of the tympanic membrane, a computed tomography (CT) scan is sometimes necessary to assess the status of the middle ear. If an effusion is present at that time, the ear is intubated.
PROBLEMS AND PITFALLS Intraoperative Complications Intraoperative difficulties center around poor graft fit or difficult graft placement. If the perichondrium/cartilage island graft is inadvertently made too small, the problem can be rectified in one of two ways. The most effective
136
OTOLOGIC SURGERY
A
B
C
D
FIGURE 10-8. Sequence of palisade reconstruction (left ear). A, Total ossicular replacement prosthesis in place. B, Initial cartilage placement. C, Reconstruction of the scutum. D, Reconstruction of remainder of posterior tympanic membrane. (From Dornhoffer JL: Cartilage tympanoplasty: Indications, techniques, and outcomes in a 1000-patient series. Laryngoscope 113:1844-1856, 2003.)
FIGURE 10-9. Postoperative appearance of tympanic membrane after palisade reconstruction (right ear). (From Dornhoffer JL: Cartilage tympanoplasty: Indications, techniques, and outcomes in a 1000-patient series. Laryngoscope 113:1844-1856, 2003.)
is to piece leftover cartilage slivers between the cartilage island and annulus using the palisade technique combined with the island technique. This approach works particularly well when the posterior island is too small. The second way to rectify a graft that is too small is to cover the entire graft with the perichondrium that was
previously harvested from one side and tuck it under the annulus as an underlay graft. Although this approach results in two layers of perichondrium over the cartilage, no detrimental effect on hearing has been noticed with this slightly thicker graft, and it is used occasionally. Poor fit more commonly results when the island flap is made too large, especially the anterior island. Because the bony annulus is blunt and thicker anteriorly and superiorly, if the anterior island is too large, it is pushed medially by the ridge of bone, making a step-off just anterior to the malleus neck. Although at first it may seem that this problem could be rectified by additional Gelfoam packing anteriorly, grasping the graft with cup forceps and pulling it laterally through and into the defect would reveal that it is actually being held medially by the anterior bony annulus. When this problem is identified, it is easily corrected by making the graft smaller and ensuring a precise fit to the bony annulus. A problem with placement can occur with an intact chain exhibiting a medially rotated malleus, which is usually encountered with the atelectatic ear. This problem can be overcome in one of two ways. The first is to remove 1 mm of the manubrium at the umbo with the malleus head amputator (malleus “nippers”). This does not affect hearing and allows medial placement of the graft. Attempting to lateralize the malleus with an intact chain
Chapter 10 • Cartilage Tympanoplasty
137
T-tube
FIGURE
10-10. Technique of cartilage tympanoplasty with intraoperative placement of T-tube. (From Dornhoffer JL: Cartilage tympanoplasty: Indications, techniques, and outcomes in a 1000-patient series. Laryngoscope 113:1844-1856, 2003.)
should be discouraged because of the possibility of acoustic trauma. The second way to resolve this problem is to remove a slightly wider strip of cartilage (2 mm) to facilitate the malleus handle. The more medial malleus can indent further into the perichondrium, allowing the cartilage plates to move more laterally in the reconstruction and avoiding contact with the promontory. This solution also allows the anterior island of cartilage more flexibility in positioning, which is necessary to make good contact with the anterior annulus. It has been the author’s experience that the reconstructed tympanic membrane frequently pulls the malleus laterally during healing, even when the graft is making contact with the promontory at the level of the malleus manubrium after graft placement. The other pitfall specific to the atelectatic ear concerns management of the atrophic tympanic membrane. After elevating the atrophic tympanic membrane off the promontory, it is tempting to insert the cartilage medial to the intact tympanic membrane. It is important, however, to remove at least a portion of the atrophic tympanic membrane anterior and posterior to the malleus to ensure that the cartilage flap is incorporated into the reconstructed tympanic membrane.
Postoperative Complications The most significant complication seen in the postoperative period is persistent effusion with conductive hearing loss, requiring intubation of the reconstructed eardrum. This effusion is seen in about 7% to 10% of cases and can be problematic in cases where the entire tympanic membrane is reconstructed with the cartilage/ perichondrium island flap. (As mentioned earlier, the tympanogram, generally low-volume type B, often is unreliable in the postoperative situation because of the noncompliant graft. The hearing result is the best indicator of middle ear function, and, in some cases,
especially when the conductive loss is intermediate or difficult to ascertain, a CT scan may be needed to assess the middle ear.) Although the tympanic membrane remains relatively insensate after cartilage reconstruction, it is often necessary to take the patient to the operating room for eardrum intubation because tube placement can be difficult. If enough room exists between the anterior annulus and the cartilage graft, a traditional myringotomy can be made, with more depth frequently found in the anterosuperior quadrant. It is more often necessary, however, to remove an ellipse of cartilage with a sickle knife or laser for adequate tube fit. Because of the thickness of the cartilage graft, grommet tubes are not applicable to this situation, and silicone elastomer (Silastic) T-tubes are more frequently used. More recently, a titanium tube (Razrbac Tube; Grace Medical Inc, Memphis, TN) has been designed specifically for use in ears reconstructed with cartilage tympanoplasty. The tube extends through the cartilage, but has more rigidity than a Silastic tube to prevent compression by the rigid graft. In cases of more pervasive eustachian tube dysfunction, such as in patients with craniofacial abnormalities (including Down syndrome), previous head and neck cancer involving the nasopharynx, or a history of multiple ear surgeries, a T-tube is inserted in the cartilage graft during the initial surgery. The perichondrium/ cartilage island graft is harvested and prepared as previously described. Using a round knife, a window that is large enough to allow placement of a Xomed Modified Goode T-tube (Xomed Surgical Products, Jacksonville, FL) is cut into the anterior cartilage island. A straight pick is placed into the cartilage window to dilate the perichondrium to allow tube placement. Before insetting the graft, the tube is placed into the cartilage window and brought out through the perichondrial surface. If the malleus is present, the end of the tube is first
138
OTOLOGIC SURGERY
angled under the manubrium with small alligator forceps. After hooking the tube under the malleus, the graft is slid forward into place. If the malleus is absent, the graft/tube complex is slid directly into its final position (Fig. 10-10). If a T-tube is removed either accidentally or purposely from an ear reconstructed with cartilage, the tympanic membrane heals with a monomeric membrane; however, the cartilage defect remains. If tube reinsertion is needed, a myringotomy is first performed through the monomeric membrane. The tube is reinserted at the original site by grasping the end of the T-tube with alligator forceps and pushing the flanges through the cartilage defect. Because the tympanic membrane is rigid from the cartilage reconstruction, it does not medialize when pressure is placed laterally. This increased rigidity greatly facilitates secondary tube insertion. Another problem that may occur postoperatively concerns a patient with reconstruction after cholesteatoma removal. One serious disadvantage of using cartilage for reconstruction in cholesteatoma surgery is that it creates an opaque tympanic membrane posteriorly, which could potentially hide residual disease. This is a problem that should be recognized, and surgical discretion should be used. If major disruption of the cholesteatoma sac occurs at extirpation, one must consider the advisability of performing a second-look surgery at a later date. This consideration applies to cholesteatoma surgery in general, not only in cases where cartilage is used in the reconstruction. One must also recognize the fact that most residual disease occurs in the epitympanum, an area that is hidden by the bony canal wall and scutum when canal wall-up surgery of any type is performed.25 Although posterior cartilage tympanic membrane reconstruction can delay the diagnosis of residual cholesteatoma, the disease becomes manifest either anteriorly or as a recurrence of a conductive hearing loss, and there should be no major complications because of this delay in diagnosis.26,27 Revision surgery after cartilage tympanoplasty is generally straightforward. The tympanomeatal flap is elevated in the usual fashion, and the fibrous annulus is identified. The mucosa is incised, the middle ear is entered, and the posterior portion of the eardrum is lifted up. This technique is similar to elevating a tympanic membrane with posterior myringosclerosis as long as an adequate strip of cartilage was removed to facilitate the malleus in the initial surgery; this allows the graft to hinge at the malleus during revision surgery. Removing the perichondrium from the medial surface of the graft before placement allows the smooth surface of the cartilage to become mucosalized, and prevents adhesions similar to Silastic sheeting, so scarring is generally less than seen with other graft materials, especially in cases with mucosal disruption on the promontory.
RESULTS Successful graft take can be anticipated in greater than 95% of patients undergoing cartilage tympanoplasty. Shortterm (<3 months) graft failures are exceedingly rare, even in an infected environment, where the increased durability of the cartilage is a distinct advantage. If long-term failures occur, they are usually seen in the nongrafted portion of the tympanic membrane, between the cartilage graft and the normal tympanic membrane, or between the cartilage graft and the bony annulus. If revision is necessary, a small piece of cartilage can be placed using the palisade technique, after freshening the edges of the perforation. Although ossification of the cartilage graft has not proved to be a problem, the graft does soften over time. Usually, this softening is inconsequential, and the graft maintains its shape and integrity with little remodeling. When the ossiculoplasty is performed with cartilage palisades, however, problems can occur with a poorly fitted prosthesis. If significant pressure is exerted on a focal area, as with a sharp edge of a prosthesis that is tilted or too long, the cartilage can focally absorb at the pressure point, leading to prosthesis exposure. Although its rigidity would seem to afford a degree of protection, cartilage should not be considered to be a safety net for an ill-fitting prosthesis. Because of the good anatomic results, attention has more recently been given to the acoustic properties of cartilage compared with more traditional grafting materials. In a retrospective comparison between perichondrium and cartilage in type I tympanoplasties, we reported no significant difference in hearing between groups.28 Our larger series of more than 1000 cases continued to show encouraging results, which have been supported by other authors who have shown excellent closure of the air-bone gap with cartilage tympanoplasty techniques.14,19,20,29 Because of the postoperative appearance of the tympanic membrane after cartilage tympanoplasty, it is surprising that no hearing loss is incurred. There is no satisfactory explanation for this phenomenon, other than that the dictum of “form follows function” seems not to apply to cartilage tympanoplasty.
CONCLUSION Cartilage is proving to be a very effective material for the reconstruction of the tympanic membrane in cases of advanced middle ear pathology. It is particularly useful for the management of an atelectatic ear, cholesteatoma, and high-risk perforation, and for reinforcement of the tympanic membrane in conjunction with ossiculoplasty. Although postoperative tube insertion is rarely needed, it can prove to be difficult when the entire tympanic membrane is reconstructed with cartilage, emphasizing the need to optimize eustachian tube function and continue research to predict outcome better based on preoperative parameters.
Chapter 10 • Cartilage Tympanoplasty
REFERENCES 1. Sheehy J L : Surgery of chronic otitis media. In English G (ed): Otolaryngology. Philadelphia, Harper & Row, 1985, pp 1–86. 2. Glasscock M E III, Hart M J: Surgical treatment of the atelectatic ear. In Friedman M (ed): Operative Techniques in Otolaryngology–Head and Neck Surgery. Philadelphia, Saunders, 1992, pp 15–20. 3. Levinson R M : Cartilage-perichondrial composite graft tympanoplasty in the treatment of posterior marginal and attic retraction pockets. Laryngoscope 97:1069-1074, 1987. 4. Eviatar A : Tragal perichondrium and cartilage in reconstructive ear surgery. Laryngoscope 88(Suppl 11):11-23, 1978. 5. Adkins WY: Composite autograft for tympanoplasty and tympanomastoid surgery. Laryngoscope 100:244-247, 1990. 6. Milewski C : Composite graft tympanoplasty in the treatment of ears with advanced middle ear pathology. Laryngoscope 103:1352-1356, 1993. 7. Amedee RG, Mann WJ, Riechelmann H : Cartilage palisade tympanoplasty. Am J Otol 10:447-450, 1989. 8. Duckert LG, Muller J, Makielski K H, Helms J: Composite autograft “shield” reconstruction of remnant tympanic membranes. Am J Otol 16:21-26, 1995. 9. Loeb L : Autotransplantation and homotransplantation of cartilage in the guinea pig. Am J Pathol 2:111-122, 1926. 10. Peer L A : The fate of living and dead cartilage transplanted in humans. Surg Gynecol Obstet 68:603-610, 1939. 11. Kerr AG, Byrne J E, Smyth G D: Cartilage homografts in the middle ear: A long-term histological study. J Laryngol Otol 87:1193-1199, 1973. 12. Don A, Linthicum FH Jr: The fate of cartilage grafts for ossicular reconstruction in tympanoplasty. Ann Otol Rhinol Laryngol 84:187-191, 1975. 13. Yamamoto E, Iwanaga M, Fukumoto M : Histologic study of homograft cartilages implanted in the middle ear. Otolaryngol Head Neck Surg 98:546-551, 1988. 14. Hamed M, Samir M, El Bigermy M : Fate of cartilage material used in middle ear surgery light and electron microscopy study. Auris Nasus Larynx 26:257-262, 1999.
139
15. Black B : Ossiculoplasty prognosis: The spite method of assessment. Am J Otol 13:544-551, 1992. 16. Goldenberg R A : Hydroxylapatite ossicular replacement prostheses: Preliminary results. Laryngoscope 100:693700, 1990. 17. Buchwach K A, Birck HG: Serous otitis media and type 1 tympanoplasties in children: A retrospective study. Ann Otol Rhinol Laryngol Suppl 89:324-325, 1980. 18. Strong M S : The eustachian tube: Basic considerations. Otolaryngol Clin North Am 5:19-27, 1972. 19. Bailey H A Jr: Symposium: Contraindications to tympanoplasty, I: Absolute and relative contraindications. L aryngoscope 86:67-69, 1976. 20. Raine C H, Singh S D: Tympanoplasty in children: A review of 114 cases. J Laryngol Otol 97:217-221, 1983. 21. Dornhoffer J L : Hearing results with cartilage tympanoplasty. Laryngoscope 107:1094-1099, 1997. 22. Zahnert T, Huttenbrink K B, Murbe D, Bornitz M : Experimental investigations of the use of cartilage in tympanic membrane reconstruction. Am J Otol 21:322328, 2000. 23. Heermann J Jr, Heermann H, Kopstein E : Fascia and cartilage palisade tympanoplasty: Nine years’ experience. Arch Otolaryngol 91:228-241, 1970. 24. Dornhoffer J L , Gardner E : Prognostic factors in ossiculoplasty: A statistical staging system. Otol Neurotol 22:299-304, 2001. 25. Smyth G D: Cholesteatoma surgery: The influence of the canal wall. Laryngoscope 95:92-96, 1985. 26. Parisier S, Hanson M : Pediatric cholesteatoma: Results of individualized single surgery management. In Sanna M (ed): Fifth International Conference on Cholesteatoma and Mastoid Surgery. Alghero-Sardinia, Italy, CIC, Edizioni Internazionali, 1997, pp 375–385. 27. Hirsch B E, Kamerer D B, Doshi S: Single-stage management of cholesteatoma. Otolaryngol Head Neck Surg 106:351-354, 1992. 28. Dornhoffer J L : Hearing results with the Dornhoffer ossicular replacement prostheses. Laryngoscope 108:531536, 1998. 29. Dornhoffer J L : Surgical modification of the difficult mastoid cavity. Otolaryngol Head Neck Surg 120:361367, 1999.
11
Tympanoplasty—Undersurface Graft Technique Transcanal Approach M. Coyle Shea, Jr. and Gale Gardner Videos corresponding to this chapter are available online at www.expertconsult.com.
Many otologic surgeons prefer placing connective tissue grafts medial to the tympanic membrane remnant. This graft placement can be accomplished through either a transcanal or a postauricular approach. This chapter describes in detail the transcanal technique and discusses our rationale for favoring it.
RATIONALE AND REQUIREMENTS Based on our experience, the transcanal approach and medial graft placement provide significant advantages, including the following: 1. The transcanal approach is more direct and quicker. 2. We believe it results in less surgical trauma, and reduces the likelihood of healing problems, such as adhesions, narrowing, and stenosis of the ear canal. 3. It results in less postoperative discomfort for the patient. 4. We favor medial graft placement because of reduced risk of tympanic membrane blunting and burying of squamous epithelium beneath the graft. Several basic requirements for successful use of this approach are as follows: 1. The surgeon must be prepared to use ear instruments skillfully, with both hands, while working through a speculum. The technique that we advocate emphasizes the advantages of a speculum holder. 2. Very delicate ear instruments and suction tips are more likely to produce rapid and trouble-free healing compared with large, trauma-producing instruments. 3. The surgeon must be able to enlarge a narrow ear canal to provide adequate working space. 4. The surgeon must also have mastered the various techniques for controlling bleeding from the ear canal, which can otherwise impede the technique that we describe.
PREOPERATIVE EVALUATION AND PATIENT SELECTION If the canal is of adequate size, this technique is suitable for any tympanic membrane perforation in which tympanoplasty is indicated, regardless of size or location. If the canal is not large enough to accommodate a 5 mm or larger ear speculum, however, this procedure may not be technically feasible. The usual indications for closure of tympanic membrane perforations are to reduce the incidence of middle ear infections and to improve hearing. Not every perforation needs to be or should be closed, however. Each patient must be evaluated on the basis of what would be best for that individual. An elderly or debilitated patient with an asymptomatic perforation or a patient for whom the ear under evaluation is the only hearing ear is usually not a good surgical candidate. In the case of a young child who developed a perforation from a ventilation tube that was initially inserted because the child could not ventilate the ear, it would be unwise to repair the tympanic membrane until it is apparent that eustachian tube function has significantly improved, to avoid the pathologic process repeating itself. There is no infallible test of tubal function, but it is reasonable to assume that if the patient can autoinflate by the Valsalva maneuver preoperatively, he or she would be able to ventilate the ear by this method, if necessary, postoperatively. Consequently, the patient is instructed in this procedure before surgery.
GRAFT SELECTION In transcanal tympanoplasty using the underlay technique, the grafting material may be any type of autogenous connective tissue, such as vein, fascia, or perichondrium. In 1957, Shea,1 using vein, was the first to use the underlay grafting technique. Tabb,2 Austin and Shea,3 and others soon recognized the superiority of this method over onlay skin grafting and followed Shea’s lead. The use of 141
142
OTOLOGIC SURGERY
fascia as an underlay graft was first reported by Storrs.4 Tragal perichondrium was first used in tympanoplasty as an onlay graft by Goodhill and associates,5 and it is the material we prefer. It is in the immediate surgical field, is extremely durable, and is very easy to handle when pressed. Vein also is easy to position and, if large enough (as from the antecubital fossa), can be used to repair perforations of any size. Our primary objection to the use of vein grafts is that in the event of a serious future illness, the large vein could be an important means of administering parenteral medications. Temporalis fascia, its overlying areolar tissue, or even scar tissue from the vicinity of a previous postauricular incision can be used with the transcanal undersurface technique, but it is not as easy to handle as vein or perichondrium. Pressing the fascia in a vein or fascia press makes it much more manageable by eliminating the tenacious loose strands, and at the same time preventing the stiffness that occurs with drying. We do not recommend pressing the vein for tympanoplasty (as done in stapedectomy) because it results in excessive thinning. Instead, we trim away the adventitia and stretch the vessel between the blades of a vein scissors before opening it.
SURGICAL TECHNIQUE Preparation Perioperative antibiotics are rarely used. The hair and skin surrounding the ear are cleansed with 70% alcohol. The hair is not shaved unless fascia is to be taken, but it is combed away from the ear and sprayed with liquid spray bandage. The head is secured with tape in a standard foam headrest in the position most conducive to good visibility for the surgeon; this usually involves tilting the head back and the chin up slightly (Fig. 11-1). For this reason, the surgeon, rather than a nursing assistant, should prepare and position the patient. The ear canal, auricle, and surrounding skin are cleansed with povidone-iodine scrub and painted with povidone-iodine solution. The field is draped in a sterile manner.
Surgical Evaluation Before beginning the procedure, the surgeon should examine the ear carefully with the operating microscope. The size of the canal and the size and location of the perforation should be established. Any evidence of cholesteatoma should be identified. The condition of the ossicles also should be evaluated, depending on their visibility, and the preoperative audiometry should be correlated with this. The extent of the anterior canal wall projection and the integrity of the scutum should be noted. The presence or absence of infection and drainage should be noted. All of these factors may influence the surgeon’s surgical decisions.
Injection Local anesthesia using 1% or 2% lidocaine with 1:100,000 epinephrine is employed in combination with either a general endotracheal anesthetic or intravenous sedation. The local anesthetic is administered at the time of the immediate preoperative preparation of the surgical area to ensure adequate vasoconstriction by the time the actual surgical procedure is begun.
Visibility: Canal Enlargement and Speculum Placement The sine qua non for successful transcanal surgery is adequate exposure. Inadequate visibility is probably the primary objection to this technique, but there are several moves that can significantly improve it. The external auditory meatus can be enlarged by making a small slit in the superior aspect of its lateral end with a No. 15 scalpel blade and stretching it with a nasal speculum. The largest sized ear speculum that can be atraumatically inserted is used and secured with a speculum holder (Fig. 11-2). The speculum holder is essential because it allows the use of both hands and serves as a support for the surgeon’s fingers. The holder is quite mobile, and allows the speculum to be placed and secured in the optimal position. It should be repositioned as needed throughout the procedure. The microscope head should also be moved about frequently to provide an unobstructed view of the operative field. Generally, the less complex the microscope, the more mobile it is, and the easier it is to use. Another item that is extremely helpful in improving visibility is the hydraulic chair. Because it can be lowered or raised in a matter of seconds, it rapidly allows the surgeon to change position in relation to the patient’s ear without the need for a circulating nurse to tilt the operating table back and forth. In the case of anterior and large central perforations, the anterior margin frequently cannot be seen because of a bulging anterior canal wall (Fig. 11-3). This problem can be remedied easily in 99% of cases by removing the hump. Sometimes the removal of a “dog-house” segment of anterior canal wall skin is all that is necessary. Frequently, the bony hump must also be removed, and this can be done quickly with a curette or small cutting burr (Fig. 11-4). In the removal of this bony hump, one must be aware of the proximity of the temporomandibular joint and avoid penetration into it. It is important to leave a 2 to 3 mm strip of skin intact between the annulus and the medial end of the resected skin. The excised skin is preserved in physiologic solution until the end of the procedure, at which time it is replaced. If the anterior perforation is marginal, the Austin “reverse elevator” (Fig. 11-5) is used to elevate the annulus and the adjacent 1 to 2 mm of the canal wall skin to provide a larger raw surface area for graft attachment. This elevated area gradually retracts into its normal
Chapter 11 • Tympanoplasty—Undersurface Graft Technique
143
144
OTOLOGIC SURGERY
osition as the ear heals. It is a crucial maneuver in the p successful repair of the anterior marginal perforation.
Perforation Preparation When the margin of the perforation can be adequately visualized, it is prepared by incising the edge with a sharp, slightly angled pick (Fig. 11-6), removing the rim and about 1 or 2 mm of the mucosa with a cup forceps (Fig. 11-7).
Canal Incisions and Middle Ear Exposure A posterior tympanomeatal incision is made with the superior limb beginning 2 to 3 mm anterior to the malleus neck (Fig. 11-8). This makes it possible to elevate the drumhead remnant completely off of the malleus handle (Fig. 11-9). Removal of the drumhead remnant from the malleus handle is performed in most situations except those in which the perforation is in an inferior or extremely posterior position. Two possible major advantages result from this. First, removal of any squamous epithelial ingrowth along the medial aspect of the malleus handle is facilitated. And second, ossicular chain reconstruction may be benefited since placement of the graft lateral to the handle of the malleus allows a malleus to stapes prosthesis to be palced directly against the bare malleus, without an intervening graft. This second advantage becomes moot however if a PORP or TORP is used.
Perichondrial Graft The tragal perichondrial graft is taken. If the perforation involves 60% of the surface of the drumhead or less, the graft usually may be obtained from the posterior aspect of the tragus through an incision immediately posterior to the free border without removal of the cartilage itself (Fig. 11-10). If the perforation involves more than 60% of the drumhead, the entire tragus with its perichondrium is removed through an incision along the free border. The excess soft tissue is removed from the perichondrium overlying the anterior surface of the tragus, and the entire perichondrium is removed from both sides of the cartilage and over the free border with a duckbill or Freer elevator and thumb forceps (Fig. 11-11). The cartilage is reinserted into the wound in its normal position to preserve the tragal contour, and the incision is closed with fine absorbable suture. The perichondrium is pressed with a vein or fascia press (Fig. 11-12). This process thins the graft for easier handling and enlarges it so that any size perforation can be closed.
Middle Ear Preparation If there is extreme medial retraction of the malleus handle so that the umbo is touching the promontory, the handle can be slowly elevated laterally with a right angle pick to release the accompanying contracture of the tensor tympani tendon. When the mucosa is badly diseased or
eroded, the medial wall of the middle ear can be lined with absorbable gelatin film (Gelfilm) cut to the desired size and shape. Because of the potential for delayed reaction, we do not use silicone elastomer (Silastic) sheeting unless a second stage is planned, at which time the Silastic sheeting is removed. The middle ear is filled with absorbable gelatin sponge (Gelfoam) soaked in a physiologic solution, such as lactated Ringer solution or Tis-U-Sol irrigating solution (Fig. 11-13). It is imperative that the tympanic cavity be filled with the Gelfoam, especially in the anterior part at the eustachian tube orifice, to prevent medial displacement of the graft. The central part of the middle ear should not be filled lateral to the umbo, however. If it is overfilled at this point, lateralization and loss of the conical contour of the drumhead may result.
Graft Placement and Stabilization For medium-sized or large perforations, the tympanomeatal flap is lifted, and the graft is placed over the posterior middle ear space, and then advanced over the malleus handle to the anteriormost extent of the perforation. The tympanomeatal flap is laid back down over the graft, and the graft edges are tucked under the margins of the flap (Fig. 11-14). With smaller posterior and inferior perforations, the graft may simply be inserted through the perforation and smoothed out posteriorly by elevating the tympanomeatal flap. When using a perichondrial graft, the surface that was in contact with the cartilage should be positioned toward the middle ear. With a vein graft, the intimal surface should be medially placed. After the graft has been roughly positioned, the edges of the perforation are carefully everted bimanually with a 20 gauge suction tip and 90 degree pick to prevent the ingrowth of squamous epithelium. If the graft does not appear to be adequately supported by Gelfoam, it can be reflected back, and additional Gelfoam can be inserted. Some surgeons prefer to place the graft medial to the malleus handle.6 As we pointed out earlier, we prefer graft placement onto the surface of the malleus since this allows a malleus to stapes prosthesis to be approximated directly to the bony surface of the malleus without an intervening graft. Adhesions between the umbo and promontory are also less likely to occur if the drum is placed lateral to the malleus handle.
External Canal Packing If the anterior canal wall skin has been removed, it should now be replaced. Small pledgets of moist Gelfoam are used to overlap the junction of rim and graft circumferentially and over the canal incision for further stabilization (Figs. 11-15 and 11-16). Finally, the external canal is filled with an antibiotic ointment, such as polymyxin. In the event of antibiotic sensitivity, povidone-iodine ointment may be substituted see video.
Chapter 11 • Tympanoplasty—Undersurface Graft Technique
POSTOPERATIVE CARE The only dressing that is used is a small, sterile, cotton ball that is loosely placed in the conchal cavity to absorb drainage. When the drainage stops, the cotton is discontinued, and the ear is allowed to ventilate. In the event
145
that purulent discharge occurs, antibiotic otic drops are started and continued until the first postoperative visit. The patient is instructed to avoid getting water in the affected ear or blowing the nose until the first postoperative visit at 3 weeks. At that time, the ear is cleaned using the operating microscope, and the graft is inspected. Approximately
146
OTOLOGIC SURGERY
Chapter 11 • Tympanoplasty—Undersurface Graft Technique
95% of the time, the graft will have taken, and the drumhead will be intact. Autoinflation is now begun using the Valsalva maneuver. If the patient is unable to ventilate the middle ear within 1 week to 10 days after the first postoperative visit, a small ventilation tube is inserted in the drumhead. If the graft is intact but not completely epithelialized at the time of the first postoperative visit, antimicrobial drops or a vinegar/alcohol solution should be used for 1 to 3 weeks to promote healing.
147
Although frank graft failure is a rarity, a small area of residual perforation occasionally is found. If this occurs, the edges can be cauterized with trichloroacetic acid and covered with a cigarette paper patch impregnated with povidone-iodine solution. The area is re-examined in 2 to 3 weeks, at which time it is usually healed. If revision surgery is necessary, it should be delayed for at least 3 months to allow for resolution of postoperative inflammatory changes.
148
OTOLOGIC SURGERY
REFERENCES 1. Shea JJ: Jr: Vein graft closure of eardrum perforations. J Laryngol Otol 74:358, 1960. 2. Tabb HG: Closure of perforations for the tympanic membrane by vein grafts: A preliminary report of twenty cases. Laryngoscope 70:271, 1960. 3. Austin D F, Shea JJ Jr: A new system of tympanoplasty using vein graft. Laryngoscope 71:596, 1961.
4. Storrs L A : Myringoplasty with the use of fascia grafts. Arch Otolaryngol 74:45, 1961. 5. Goodhill V, Harris I, Brockman S F: Tympanoplasty with perichondrial graft. Arch Otolaryngol 79:131, 1964. 6. Hough JVD: Tympanoplasty with the interior fascial graft technique and ossicular reconstruction. Laryngoscope 80:1385, 1970.
12
Tympanoplasty—Undersurface Graft Technique Postauricular Approach C. Gary Jackson, David M. Kaylie, Michael E. Glasscock III, and Barry Strasnick Videos corresponding to this chapter are available online at www.expertconsult.com.
Since the fundamental principles of tympanoplasty were first introduced by Wullstein1 and Zollner,2 there has been great diversity in the accepted surgical techniques used for repair of the tympanic membrane. The multitude of graft materials employed is a testimony to the difficulty of middle ear reconstruction. With advanced microsurgical techniques, the state of the art has now developed to the extent that graft success rates of 90% to 97% are to be expected.3-5 Two basic grafting techniques have evolved based on where the graft material is placed in relation to the drum remnant (overlay versus underlay techniques). This chapter presents a method of undersurface grafting. Detailed surgical techniques and appropriate preoperative and postoperative care are presented.
HISTORICAL ASPECTS Modern middle ear reconstructive surgery represents a culmination of more than a century of contributions by numerous dedicated and innovative otologic surgeons. The term tympanoplasty was originally defined in 1964 by what was then known as the American Academy of Ophthalmology and Otolaryngology’s Committee on Conservation of Hearing as “an operation to eradicate disease in the middle ear and to reconstruct the hearing mechanism without mastoid surgery, with or without tympanic membrane grafting.”6 If a mastoid procedure is included, the term tympanoplasty with mastoidectomy is used. The era of surgical repair of the tympanic membrane dates as far back as the 19th century. In 1853, Toynbee7 described closure of a perforation of the tympanic membrane using a small rubber disk attached to a silver wire. Ten years later, Yearsley8 advocated placing a cotton ball over the perforation; in 1887, Blake9 introduced the concept of placing a thin paper patch over the membrane. The use of cautery to promote spontaneous healing of tympanic membrane perforations was introduced by Roosa in 187610; he used silver nitrate. Later, Joynt,11 Linn,12 and Derlacki13 described modifications of this
technique using various forms of cautery and patches. Closure of tympanic membrane perforations was considered appropriate only for dry central perforations, however. At this point, no one advocated the use of drum closure for the chronically draining ear. It was not until 1952 that Wullstein1 and Zollner2 revolutionized middle ear surgery by advocating reconstructive grafting of the chronically diseased ear through the use of full-thickness or split-thickness skin grafts. House and Sheehy14 and Plester15 later used canal skin, believing that it more closely resembled the squamous layer of the tympanic membrane. The overall poor success rates of these grafts and the development of iatrogenic cholesteatomas prompted the search for alternative grafting materials. Shea16 and Tabb,17 working independently, described the use of autogenous vein to close the tympanic membrane. Goodhill18 advocated tragal perichondrium in the mid-1960s, and tympanic membrane homografts became popular a few years later. Glasscock and House19 reported the first sizable series of homograft tympanic membrane transplants in 1968. Interest in homografts has waned, however, largely because of the fear of transmission of infectious diseases. Storrs20 performed the first fascia graft in the United States. Although vein, perichondrium, and homografts still have their advocates, autogenous fascia has now become the standard by which all other grafting materials are measured. The use of skin grafts required that the tympanic membrane perforation be repaired by laying the graft on top of the denuded drum remnant. This method of repair eventually became known as the overlay technique and was carried over to other forms of grafting material. With the use of connective tissue grafts, the graft material could be placed medial to the tympanic membrane remnant. The success of this approach eventually gave rise to the underlay technique of tympanic membrane grafting; Austin and Shea3 reported a large series. Proponents of the underlay procedure submit that it eliminates many of the problems associated with overlay grafts, such as anterior blunting, epithelial pearl formation, and lateralization of the new drum. 149
150
OTOLOGIC SURGERY
In 1973, Glasscock4 described an underlay grafting technique that relied on a postauricular approach. With minor modifications, this approach continues to be the preferred method of dealing with disorders of the tympanic membrane and the middle ear.
PREOPERATIVE FUNDAMENTAL PRINCIPLES Regardless of the grafting technique chosen, the preoperative evaluation and management of a patient with a tympanic membrane perforation remain the same. A complete clinical history is obtained, and a comprehensive head and neck examination is performed. Particular attention is addressed to the nasopharynx. Otoscopic examination is performed with the aid of an operating microscope if necessary. All findings are diagrammed on the patient’s chart. All patients receive a pure tone air and bone conduction audiogram along with speech discrimination testing. Tuning fork tests should be done on all patients to confirm the audiologic findings.
FUNDAMENTALS Traditional Objectives As with any surgical procedure, successful outcomes result from mastery of understanding and execution of each component of the process as a whole. It is useful to distill the task of tympanoplasty into its important fundamentals. The traditional objectives of tympanoplasty have not changed in 50 years and remain:
• Eradication of disease • Closure of the ear by grafting • Hearing rehabilitation
with its significant effects on the nasopharynx in the short term, simultaneously with tympanic membrane grafting. After the adenoidectomy is performed, the tympanic membrane grafting is scheduled as a separate procedure, approximately 4 weeks later. Tonsillectomy is performed as an independently indicated consideration. Its effect on tympanoplasty is negligible. The clinical setting in which this occurs is generally in children younger than 10 years.
Nasal or Sinus Condition Similar to nasopharyngeal disease, significant nasal septal deformity, polyposis, or acute sinusitis should be managed before grafting an ear. Acute sinusitis would warrant cancellation of tympanoplasty. Lesser degrees of nasal obstruction or chronic sinusitis are addressed as logically dictated by the clinical circumstances before or at some time after tympanoplasty.
Allergy Allergic disease is an inexorable detriment to the longterm success of tympanic membrane grafting. In endemic areas, it should not be disregarded. At some time in the perioperative period, the tympanoplasty patient is referred for comprehensive allergy diagnosis and management. Immunotherapy, when indicated, can have a tremendous influence on long-term outcomes. Acute exacerbations, often seasonal, are managed pharmacologically with antihistamines and nasal steroids.
Rare Disorders
The goals, in this order of priority, allow for patient counseling and reasonable expectations.
Particularly in recidivistic disease, the presence of rare associated diseases must be kept in mind. Tuberculosis, sarcoidosis, diabetes mellitus, hematologic disorders, histiocytoses, immunodeficiency syndromes, and, in recurrent adult disease, neoplasms should not be disregarded.
Predisposing Conditions
PREOPERATIVE PREPARATION
Tympanoplasty is usually successful in achieving the above-listed goals in the short term. To ensure long-term success, conditions predisposing to failure must be prospectively managed. Generally, the status of the upper respiratory tract influences eustachian tube function and, consequently, the long-term success of tympanoplasty.
Otorrhea
Adenoidal Hypertrophy and Adenoidectomy Excessive adenoidal hypertrophy is regarded, by consensus, to influence the success of tympanic membrane grafting. Adenoidectomy should be done prospectively. It is generally not advised to perform adenoidectomy,
Every attempt is made to operate on dry ears. Preoperative infection control in the involved ear is useful, but not essential. At the initial evaluation, the draining ear is otomicroscopically evacuated. Instructions are given to begin the instillation of steroid-containing antibiotic drops. Irrigating the ear with sterile 1.5% acetic acid solution can be helpful in eradicating recalcitrant infections. Instillation of drops in the infected ear affords minimal ototoxicity risk. Pain on administration constitutes an end point. Associated disorders or unusually significant infection may rarely warrant oral antibiotics. In the absence of any immunocompromising accompaniment, cultures are
Chapter 12 • Tympanoplasty—Undersurface Graft Technique
not routinely done. The physician must be vigilant for signs and symptoms of intratemporal or extratemporal complications. Surgery is scheduled, and the ear is operated on, draining or not.
Eustachian Tubal Tests No clinical test exists for eustachian tubal physiology. Eustachian tubal patency is testable via methods such as the Valsalva maneuver and the Toynbee test, but it is not important in the grand schematic of tympanoplasty. Eustachian tubal physiology tests exist (e.g., the Flisberg test), but are clinically impractical and are not done. A statement attributed to Sheehy is true: “Sometimes the best test of eustachian tubal function is a tympanoplasty.” Eustachian tube function is nonetheless important to tympanic membrane grafting success. Status of the contralateral ear often predicts the eustachian tubal capacity of the involved ear. Apparent current eustachian tubal dysfunction may be a consequence of active infections unilaterally or the aftermath of a lifetime of chronic otitis media. Tympanoplasty is not contraindicated. Postoperatively, when the ear is restored to a more normal state, its eustachian tubal function may also be restored. In the difficult situation of tracheostomy in which eustachian tube function is compromised or when effusion or retraction affects the successful graft, the ear can be ventilated in the office. Ventilation tubes should not be placed in tympanic membrane grafts because they promptly extrude. Tube placement can be performed in the first month after the procedure in the office because the tympanic membrane is still anesthetic. An atelectatic ear should not preclude tympanoplasty. Rather, it is a perfect indication for cartilage tympanoplasty.
Imaging The imaging standard for chronic otitis media is now highresolution temporal bone computed tomography (CT). Contrast medium is also employed to evaluate clinically silent epidural, intracranial, or lateral sinus abscess. Plain mastoid radiographs have been abandoned, as has polytomography. All ears for grafting are not imaged. Only hearing ears and disease in long-standing adult chronic otitis media are ideal candidates. It is prudent to obtain CT scans for revision surgery, particularly if the initial surgery was done by another surgeon. Selected cholesteatomas may be studied, although imaging is unnecessary for all cholesteatomas. Patients with a cholesteatoma in which an inner ear fistula or tegmen defect with encephalocele is suspected are good candidates for imaging. The current state of magnetic resonance imaging (MRI) precludes its routine use in ear imaging of chronic otitis media. Magnetic resonance angiography is useful in venous phase to assess the lateral venous sinus.
151
Complications Polypharmacy and accessible medical care have changed the face of diagnosis of temporal bone and intracranial complications in chronic otitis media. Pain, focal neurology, headache, vertigo, particularly noisome (anaerobic) otorrhea, cephalgia, and sensorineural hearing loss all should elicit the applicable cliché: “high index of suspicion.”
Informed Consent Preoperative counseling regarding the nature and extent of the problem, treatment options, surgical details, surgical staging, reasonable expectations, and risks and complications is comprehensive. This discussion is interpersonal and reviewed in professionally prepared videotapes. The consent and preoperative and postoperative instructions are provided in written form, and the patient is asked to sign the informed consent in the presence of a neutral witness. Common complications discussed include, but are not limited to, hearing loss, infection, graft failure, and facial nerve paralysis. Disorder-focused brochures are also given to the patient.
BASIC TECHNICAL PRINCIPLES The basic surgical principles we all learn as surgical interns are as important here as ever when the operating microscope is introduced as a surgical tool.
Infection Control Chronic ear surgery is clean-contaminated or contaminated; 90% of otologic wounds are colonized at the time of surgery. The general surgical principle of infection control seeks to minimize colony counts so that host defense mechanisms are not overwhelmed. Whether or not surgeons can accomplish this in ear surgery is highly debated. Otologists seeking higher graft take rates and fewer complications often resort to prophylactic antibiotics as a “protective umbrella.” In the only study of statistical power on this subject, Jackson23 concluded that prophylactic antibiotics are harmless, but useless. In common uncomplicated tympanoplasty, antimicrobial prophylaxis is unwarranted. An indication for prophylaxis exists in the draining ear, which, intuitively, has a high postinfection rate with graft failure. Despite this fact, no protocol exists to prevent such an outcome. This is an ideal indication for intraoperative irrigation, yet ototoxicity and medicolegal concerns have impeded human study design to address this issue. There are indications for antimicrobial prophylaxis in ear surgery. Violation of the dural integrity with or without cerebrospinal fluid leakage, violation of the
152
OTOLOGIC SURGERY
labyrinth, acknowledged aseptic technique breaks, only hearing ears, and implantation of indwelling devices such as cochlear implants all are valid indications. “The secret to pollution is dilution.” Aggressive irrigation throughout the procedure theoretically clears devitalized debris and clots, and is thought to reduce colony counts. Normal saline is used. The surgical preparation is described in the section on technique.
Hemostasis Prior injection of the surgical field with 2% lidocaine (Xylocaine) and 1:100,000 epinephrine helps reduce bleeding in the surgical field, which is essential to microsurgery. These solutions are contained in dental carpules so that loose solutions with their attendant risks are not on the surgical table to be confused. Electromicrobipolar cautery, adapted from neurotology, is useful in eliminating bleeding from the lateral venous sinus, dura, facial nerve, and delicate external auditory canal (EAC) flaps. Excessive bleeding from infected areas granulating can be difficult. Absorbable gelatin sponge (Gelfoam) soaked in 1:100,000 epinephrine placed on the involved site while surgical attention is directed elsewhere is efficient hemostasis. Aggressive irrigation is thought to help control blood loss and affect hemostasis. Graft placement of any kind cannot be efficient under water. A bloodless field to identify all margins is essential. While preparing the graft, the operative field is filled with epinephrine-soaked Gelfoam until graft placement is imminent.
Grafting Techniques and Exposure Two approaches to tympanic membrane grafting have evolved over the years. The overlay technique previously was popular because of its high success rate and reproducibility.24 Experience has recognized recurring downsides, however. Blunting of the anterior sulcus, when significant, can result in conductive hearing loss by malleus fixation, as can lateralization of the graft away from the malleus. Inability to denude the drum of epithelium completely could and has resulted in epithelial pearls or cholesteatoma or both. Because the EAC skin is removed and replaced, delays in healing occur. The undersurface graft placement technique has been propelled by the adoption of connective tissue as a grafting material. Because this surgery was commonly performed with a transcanal approach through a speculum, underlay grafting was regarded as more technically difficult. Irregular graft placement—hence failure—often resulted. Variations in size and contour of the EAC and operator experience using a speculum impaired visualization of the entire tympanic membrane remnant and anterior sulcus, making graft placement particularly
c hallenging. Adequate exposure of the middle ear and eustachian tubal orifice was rare. The postauricular approach obviated all of these problems using the vascular strip access. Anterior EAC wall bulges could be easily managed, and no speculum was needed. Overall exposure was enhanced, allowing luxurious middle ear exposure and precise graft placement. All the complications of the overlay strategy are avoided. For the experienced ear surgeon, either placement strategy produces the desired outcome in greater than 90% of grafts. For some ear surgeons, the postauricular undersurface technique affords higher take rates with fewer complications.
Mastoidectomy Tympanic membrane perforation with cholesteatoma virtually mandates performing a mastoidectomy. Data do not exist to guide the ear surgeon regarding performing mastoidectomy or not when tympanic membrane perforation exists. In uncomplicated tympanic membrane rupture, such as in trauma, mastoidectomy is rarely indicated. Tympanic membrane grafts fail most commonly because of infection in the middle ear or mastoid. We readily assess and manage middle ear infection at the time of tympanoplasty. In any suggestion of mastoid infection, recent otorrhea, refractory or recurrent disease, or difficult aeration occasions such as tracheostomy, if it seems appropriate, mastoidectomy is executed.
Ossicular Reconstruction The key to successful ossicular reconstruction is understanding the healing concept of changing anatomic relationships. When anatomic relationships in the middle ear are minimal or stabilized, ossicular reconstruction is unstaged. When anatomic relationships are expected to change, such as in the most diseased ears because of complicated tympanic membrane pathology, extensive mucosal disease, extensive ossicular pathology, a fixed footplate, or extensive tympanosclerosis or healing biology, ossicular reconstruction is staged. Autogenous cartilage is always interposed between platforms and tympanic membrane graft undersurfaces. Cartilage is not interposed for hydroxyapatite. If possible, autogenous material is preferred.
Facial Nerve Monitoring Neural integrity monitoring of the facial nerve is used for all otologic cases. Paralyzing agents are not used in ear surgery anesthesia. Nerve monitoring informs technique relative to the facial nerve. Monitoring is no substitute for knowledge of the anatomy, sound tissue technique, or surgical instinct.
Chapter 12 • Tympanoplasty—Undersurface Graft Technique
SURGICAL TECHNIQUE Surgical Preparation A 2-cm wide area of hair is shaved above and behind the auricle, and Mastisol is placed. Nonsterile 3M (No. 1010) plastic towel drapes are placed on the skin to cover the hair. Electrode insertion for intraoperative facial nerve monitoring occurs before the surgical preparation. An iodine soap solution is used to wash the auricle and the skin around it, which is blotted dry with a sterile towel. The ear itself is bathed in an iodine preparation solution for 3 minutes. The solution is allowed to enter the EAC.
153
Finally, the area around the ear is prepared with an alcohol-based solution (DuraPrep Surgical Solution). The postauricular region and the tragus are injected with a 2% lidocaine and 1:100,000 epinephrine solution. After injection, the patient is draped with two layers of sterile sheets. The first layer consists of a 3M plastic ear drape (No. 1030). This waterproof sheet effectively isolates the patient and prevents contamination if the paper drapes become soaked with irrigation solution. Finally, a custom-designed paper otologic drape and a plastic drainage bag are applied (Fig. 12-1). Suction-irrigation tubing and cautery lines are secured to the field. The scrub nurse attaches a compartmentalized plastic pouch to the Mayo stand to hold suction tips and electrocauteries.
Anesthesia
FIGURE 12-1. The ear and postauricular area are prepared and draped in a sterile fashion.
All patients undergo general endotracheal anesthesia with an epinephrine-compatible anesthetic agent. Long circuits are used on the anesthesia machine to enable the anesthesiologist to be seated at the foot of the table, opposite the surgeon (Fig. 12-2). The scrub nurse is positioned at the head of the table across from the surgeon. The vascular strip and four quadrants of the ear canal are injected with the 2% lidocaine and 1:100,000 epinephrine solution. The local anesthetic reduces pain and allows the patient to be kept relatively “light” during the procedure. No limitations on the use of nitrous oxide during graft placement are required because the undersurface placement of the graft, along with packing of the eustachian tube orifice, precludes the graft being elevated off the drum remnant. The nitrous oxide bubbles escape harmlessly up the posterior canal wall. Every attempt is made to extubate the patient deeply to avoid straining and potentially detrimental Valsalva efforts. Intravenous antiemetics are given to reduce postsurgical nausea and vomiting.
FIGURE 12-2. Operating room arrangement. The scrub nurse is located directly across from the surgeon, and the anesthesiologist (not shown) is seated at the foot of the table.
154
OTOLOGIC SURGERY
Incisions The vascular strip is outlined by making incisions at the tympanosquamous and tympanomastoid suture lines using a No. 67 Beaver knife blade. In addition, small inferiorly and superiorly based flaps are created by making right angle incisions to the vascular strip incisions. The medial end of the vascular strip is formed by connecting the two primary incisions with a No. 72 Beaver blade approximately 2 mm lateral to the annulus (Fig. 12-3). A postauricular incision is made approximately 5 mm behind the postauricular crease (Fig. 12-4). The surgeon firmly grasps the auricle in the left hand and forcefully pulls forward and outward. Constant tension allows identification of the loose areolar tissue overlying the temporalis fascia and creates a bloodless surgical plane. Incisional bleeding is controlled with electric cautery.
Harvesting Fascia When hemostasis is achieved, a small Weitlaner retractor is positioned to hold the auricle forward. The scrub nurse places a small Senn retractor under the upper part of the incision and pulls laterally, exposing the temporalis fascia. The loose areolar tissue overlying the temporalis fascia is ballooned up with a local anesthetic to facilitate its removal. An incision is made at the level of the linea temporalis, and the areolar tissue is dissected free from the temporalis fascia using Metzenbaum scissors (Fig. 12-5). This tissue is pressed onto a polytef (Teflon) block and placed on the back table under a gooseneck lamp to dehydrate it.
Exposing the Middle Ear The retractor is removed, and an incision is made along the linea temporalis extending anterior and superior to the EAC. A T-shaped incision is created by dropping a vertical limb from the midpoint of the linea temporalis to the mastoid tip (see Fig. 12-5). A Lempert elevator is used to mobilize the periosteum to the level of the ear canal. The vascular strip is identified from posteriorly, grasped with Adson forceps, and held forward in the blade of a Weitlaner retractor along with the auricle (see Fig. 12-5). A second Weitlaner retractor is placed between the temporalis muscle and the mastoid tip at right angles to the first retractor. The ear canal is copiously irrigated with a physiologic saline solution to remove blood debris. With a 20 gauge needle suction in the surgeon’s left hand and a House No. 2 lancet knife in his or her right hand, the skin of the inferior ear canal is elevated down to the fibrous annulus (Fig. 12-6), creating an inferiorly based flap. Next, a House No. 1 sickle knife is used to develop a superior flap just above the short process of the malleus. The fibrous annulus is mobilized out of its sulcus anterior to the malleus.
If the operating table is rotated away from the surgeon, the anterior drum remnant and annulus are easily seen. If a bony overhang obscures complete vision, the canal skin can be reflected laterally, the bone removed with a small diamond burr, and the skin reflected downward. Care must be taken to protect the anterior annulus and adjacent canal skin.
Eradication of Disease With the middle ear now well exposed, the primary disease process can be addressed logically. Cholesteatoma, granulation tissue, or polypoid disease can be removed through the middle ear itself or in conjunction with a mastoidectomy if indicated. For better exposure of the middle ear, the facial recess or posterior tympanotomy is opened, leaving the posterior canal wall intact.
Preparation of the Tympanic Membrane Remnant When the disease of the middle ear and mastoid has been eradicated, the drum remnant is prepared for grafting. Small attic, marginal, or central perforations are prepared to preserve the normal drum remnant. In the case of an extensive perforation of a severely diseased membrane, the entire drum remnant is removed to the level of the annulus. The manubrium of the malleus is denuded, preserving the fibrous annulus. The mucosa of the undersurface of the drum remnant or annulus is abraded through the use of a House No. 1 sickle knife and cup forceps to ensure further an adequate recipient surface for the graft (see Fig. 12-6).
Placement of the Graft It is imperative that excellent hemostasis be achieved before graft placement. Gelfoam saturated in 1:1000 epinephrine is packed into the middle ear space while the graft is being fashioned. A dried areolar tissue graft is removed from the Teflon block and trimmed to size (approximately 2.5 × 1.5 cm). A slit is made toward the superior aspect of the graft to accommodate placement medial to the malleus handle. If mucosa has been removed from the middle ear, a sheet of absorbable gelatin film (Gelfilm) is trimmed and placed onto the promontory to prevent adhesions. The epinephrine-soaked Gelfoam is removed, and the middle ear is packed with saline-moistened Gelfoam starting from the eustachian tube and working posteriorly. The graft is grasped using cup forceps, rehydrated in a physiologic saline solution, such as Tis-U-Sol irrigating solution, and placed in the middle ear. With a 22 gauge suction in the surgeon’s left hand and a right angle hook in his or her right hand, the graft is slid under the manubrium of the malleus onto the lateral
Chapter 12 • Tympanoplasty—Undersurface Graft Technique
155
FIGURE 12-3. Vascular strip is outlined with No. 67 and No. 72 Beaver blade. FIGURE 12-4. Standard postauricular incision is fashioned approximately 5 mm behind the postauricular crease. FIGURE 12-5. A, The loose areolar tissue overlying the temporalis fascia is harvested as a free graft. Note T-incision through the mastoid periosteum used to expose the external auditory canal. B, The vascular strip is held forward along with the auricle.
FIGURE 12-6. A, ���������������������������������������������������������������������������� The inferior canal flap is elevated to the level of the fibrous annulus.����
156
OTOLOGIC SURGERY
FIGURE 12-6. B, The malleus is denuded, and the undersurface of the drum remnant or annulus is abraded, while diseased mucosa is removed. FIGURE 12-7. A, The fascia graft is directed under the fibrous annulus and malleus handle. B, The superior and inferior canal flaps are replaced over the fascia graft.
attic wall. A House annulus elevator is used to tuck the fascia under the drum remnant anteriorly and inferiorly (Fig. 12-7). With this technique, there is a point in the inferior canal at approximately the 6 o’clock position where the graft makes a transition from lying medial to the annulus to being lateral to it. The remaining graft is draped along the posterior canal wall, and the inferior canal flap with attached annulus is repositioned over the graft. Similarly, the superior flap is placed, covering the fascia lying anterior to the malleus (see Fig. 12-7). A House annulus elevator is used to even all edges of the annulus and smooth out the graft. The surgeon must ensure that flap margins are flat because buried skin may result in pearl formation. Polymyxin B and bacitracin (Polysporin) ointment is placed over the fascia graft filling the anterior sulcus. The retractors are removed, and the vascular strip is carefully replaced to its original position. The mastoid periosteum
incision is closed with 3-0 polyglactin 910 (Vicryl) suture in interrupted fashion. The postauricular incision is closed with the same 4-0 suture in a subcuticular fashion. No skin sutures are used, obviating later removal. Proper position of the vascular strip is confirmed again under direct vision through an ear speculum, and the remainder of the ear canal is filled with antibiotic ointment. A cotton ball is placed in the meatus, and a sterile, prepackaged plastic mastoid bubble dressing is applied (Fig. 12-8).
POSTOPERATIVE CARE Before discharge, the mastoid dressing is removed. A sterile, prepackaged postauricular dressing is applied behind the ear, and a fresh cotton ball is placed in the meatus of the ear canal to absorb the ointment as it
Chapter 12 • Tympanoplasty—Undersurface Graft Technique
FIGURE 12-8. A sterile prepackaged mastoid dressing is applied.
liquefies. Patients are instructed to change the cotton ball at least three times per day and whenever it becomes soiled. A 3-week follow-up appointment is scheduled. Each patient is counseled about the warning signs of infection and instructed to contact the office immediately if these occur. A prescription for analgesics is given. Patients are asked to keep the ear dry and to avoid nose blowing. By the first postoperative visit, the postauricular wound should be well healed, the ear canal should be free of ointment, and the tympanic membrane should be epithelialized. All tympanoplasty patients undergo a pure tone audiogram at this visit. By 6 weeks, the grafted eardrum has thinned considerably and assumes an appearance of a normal tympanic membrane. Follow-up visits are arranged at 6 and 12 months and yearly thereafter.
RESULTS In 1982, Glasscock and associates5 reported their expe rience in 1556 cases of tympanic membrane repair using the postauricular undersurface grafting technique described here. Of these cases, 663 were simple tympanoplasties, 687 involved a mastoidectomy, 54 were performed to repair a graft failure at a second stage, 38 involved a tympanoplasty with mastoid revision, and 114 were canal wall down mastoidectomies in which the tympanic membrane was grafted. Of ears, 463 (34%) had undergone at least one previous surgery. All tympanic membranes were repaired using areolar tissue, temporalis fascia, or tragal perichondrium. Of the 1556 ears, there were 110 failures, for an overall graft success rate of 93%. Of the failures, 19 occurred within 3 weeks of surgery, 31 occurred at 3 months, 19 occurred at 6 months, and 41 occurred after 1 year. Successful grafting occurred in 91.5% of patients younger than 12 years compared with 93.3% of patients older
157
than 12 years. The graft success rate was 92.7% in draining ears and 93.1% in dry ears. The presence of cholesteatoma had no apparent effect on success. Success rate in ears with cholesteatoma was 92% and in ears without cholesteatoma was 93.2%. Complications were minimal in this series. Postoperative otorrhea occurred in 6%, varying from a mild otitis externa to a severe middle ear infection promoting loss of the graft. True wound infections were seen in less than 0.5%. Sensorineural hearing loss occurred in less than 1% of the cases. There were five cases of delayed facial paralysis, all of which recovered completely within 2 to 3 weeks. Serous otitis media occurred in 2%, whereas perichondritis, stenosis of the EAC, and epithelial pearls occurred in less than 0.5%. A more recent study examined results from revision chronic ear surgery.25 The authors found that control of recurrent cholesteatoma was achieved in 93% of cases. Hearing also can be significantly improved by closure of recurrent perforations or reconstruction of the ossicular chain. These authors found that hearing results after ossicular chain reconstruction were not as good for revision surgery as seen for primary surgery.25 Another study by the same authors compared surgical outcomes for chronic ear surgery between smokers, former smokers, and nonsmokers.26 This study showed that smokers have significantly more cholesteatomas, more canal wall down mastoidectomies, and overall worse hearing results than nonsmokers. Smokers also had significantly more postoperative complications than nonsmokers. Patients who quit smoking within 5 years of their surgery had the same risks as current smokers. In patients who quit smoking more than 5 years before surgery, the results of chronic ear surgery were no different than in patients who never smoked.
CARTILAGE GRAFT TYMPANOPLASTY In cases of severe atelectasis of the tympanic membrane, a cartilage tympanoplasty is often indicated. Cartilage autografts have long been used in repair of canal wall defects and ossiculoplasty.27-30 In 1982, Glasscock and associates5 first described the successful use of cartilage/ perichondrial autografts for severe atelectasis, attic cholesteatoma, and posterior retraction pockets.5 Since that time, other authors have reported their results with this technique.31,32 The goal of cartilage tympanoplasty is to prevent recurrent retraction along with the long-term sequelae, including cholesteatoma formation, ossicular erosion, and progressive hearing loss. Incorporation of cartilage in the repair of the eardrum provides sufficient structural integrity to resist recurrent retraction, yet imparts minimal impedance. This technique is ideally suited for patients who have persistent eustachian tubal dysfunction, including tracheostomy patients or patients with recurrent atelectasis after standard fascia graft tympanoplasty.
158
OTOLOGIC SURGERY
Chapter 12 • Tympanoplasty—Undersurface Graft Technique
159
FIGURE 12-9. Standard vascular strip incisions are made. A House sickle knife opens the middle ear and elevates the atelectatic drum. FIGURE 12-10. A, Tragal cartilage perichondrial graft is harvested by means of an incision on the posteromedial aspect of the tragus. B, The tragal perichondrium is elevated off the surface of the cartilage. C, The cartilage is trimmed to the desired size. FIGURE 12-11. A, The cartilage perichondrial graft is positioned medial to the manubrium and fibrous annulus. The areolar tissue graft is positioned lateral to the cartilage perichondrial graft, but medial to the annulus and manubrium. B, Lateral view shows cartilage perichondrial graft in place. Ossiculoplasty is typically performed at a second stage.
Surgical Techniques After exposure of the middle ear is obtained, the next step is to excise all diseased and atelectatic tympanic membrane. The posterior fibrous annulus is elevated from its bony sulcus with a House No. 2 lancet knife. Careful dissection is required to avoid tearing the atelectatic drum to ensure that no epithelium is left in the middle ear. To verify complete removal of an attic retraction, it is often necessary to perform mastoidectomy. Posterosuperior retractions must be elevated in continuity with the remainder of the tympanic membrane. When elevating the drum off the lenticular proc ess and the stapes suprastructure, applying force in the posterior-to-anterior direction allows the stapedius tendon to provide countertraction, preventing inadvertent stapes subluxation. A House No. 1 sickle knife is used to elevate the diseased membrane off the manubrium and lateral process of the malleus. Fibrous adhesions are lysed, and diseased middle ear mucosa is removed with cup forceps (Fig. 12-9). To harvest the cartilage perichondrial graft, an incision is made on the posteromedial surface of the tragus (Fig. 12-10). This incision is carried through the tragal cartilage, preserving the dome of the tragal cartilage for cosmesis. A House No. 2 lancet elevator is used to elevate the perichondrium from one surface of the cartilage, leaving it hinged on the other side, similar to a book cover. The cartilage is trimmed to the proper dimensions, depending on the degree of disease present. A posterosuperior quadrant retraction often requires a cartilage graft of approximately 4 mm in diameter. For cases of atelectasis of the entire tympanic membrane, the cartilage can be incorporated into the entire pars tensa. In this situation, a wedge-shaped area of cartilage is accessed to accommodate the manubrium, if present. When using large cartilage grafts not likely to move in healing, perichondrium is not needed and is detached. When the middle ear has been packed with moistened Gelfoam, the cartilage/perichondrial graft is placed with the perichondrium side facing laterally. The perichondrium is tucked under the manubrium and draped over the ear canal posteriorly. The cartilage should not overlap the posterior canal wall. The areolar tissue graft is trimmed to size and placed lateral to the cartilage perichondrial graft and medial to the fibrous annulus and manubrium (Fig. 12-11). This areolar graft serves to cover any remaining defects in the tympanic membrane or
exposed bone in the external canal. When perichondrium is not used, the cartilage graft is placed atop the middle ear Gelfoam and ������������������������������������������ medial������������������������������������ to the tympanic membrane graft. The cartilage is ultimately enveloped by the neo–tympanic membrane. The superior-based and inferior-based canal flaps are returned to their original positions, covering the grafts as they extend onto the posterior canal wall. The external canal is filled with Polysporin ointment, and closure proceeds in the manner previously described.
Results Results in 100 ears using cartilage tympanoplasty were reported more recently.34 The retracted portion of the tympanic membrane was in the posterosuperior quadrant in 37%, the entire drum in 34%, the posterior half in 18%, the pars flaccida alone in 16%, the pars flaccida and pars tensa combined in 4%, and the anterior half in 1%. Cholesteatoma was present in 33 ears. Ossicular reconstruction was required in 52 ears; 31 of these procedures were performed simultaneously with the cartilage tympanoplasty, and 21 were staged. Hearing results were reported in 79 ears by calculating the postoperative average air-bone gap for the speech frequencies (500 Hz, 1000 Hz, and 2000 Hz) in 10 dB increments. Of these 79 cases, 38% achieved closure of the air-bone gap to within 10 dB, 39% closed within 20 dB, and 18% closed within 30 dB. Five percent of the patients failed to show closure to less than 30 dB. Of the 100 ears, there were 4 failures (4%) in terms of the technique itself. There were two recurrent cholesteatomas, one perforation, and one recurrent retraction. Six patients (6%) developed serous otitis media postoperatively. Of these six patients, three eventually required ventilation tubes. Fifteen patients had mild retractions that were easily controlled with modified Valsalva exercises. There were two wound infections and six cases of EAC infections. Perichondritis and EAC stenosis did not occur in this series.
SUMMARY The major objectives of tympanoplasty may be prioritized as follows: (1) control of infection, (2) creation of an air-containing middle ear space, and (3) hearing rehabilitation. To accomplish these goals, it is imperative that
160
OTOLOGIC SURGERY
the otologic surgeon exercise sound clinical judgment in terms of selection of patients and surgical approach. The fundamental components of the surgery must be understood and assiduously executed. Success should be achievable in greater than 90% of patients. For the average surgeon, the postauricular undersurface graft technique of myringoplasty yields consistently superior functional results with fewer complications than seen with classic transcanal undersurface procedures or overlay techniques.
REFERENCES 1. Wullstein H : Funktionelle Operationen im Mittelokr mit Hilfe des Freven Spalthappen-Transplantes. Arch Ohr Nas Kehlhopfheilk 161:422, 1952. 2. Zollner F: The principles of plastic surgery of the soundconducting apparatus. J Laryngol Otol 69:637, 1955. 3. Austin D F, Shea JJ: A new system of tympanoplasty using vein graft. Laryngoscope 71:596, 1961. 4. Glasscock M E : Tympanic membrane grafting with fascia: Overlay versus underlay technique. Laryngoscope 5:754, 1973. 5. Glasscock M E, Jackson C J, Nissen A J, et al: Postauricular undersurface tympanic membrane grafting: A followup report. Laryngoscope 92:718, 1982. 6. Committee on Conservation of Hearing of the American Academy of Ophthalmology and Otolaryngology: Standard Classification for Surgery of Chronic Ear Infection. Arch Otolaryngol Head Neck Surg 81:204, 1964. 7. Toynbee J: On the Use of an Artificial Membrane Tympanic in Cases of Deafness Dependent Upon Perforations or Destruction of the Natural Organ. London, J Churchill & Sons, 1853. 8. Yearsley J: Deafness, Practically Illustrated. London, J Churchill & Sons, 1863, Ecl 6. 9. Blake C J: Transactions of the First Congress of the I nternational Otological Society. New York, D Appleton, 1887. 10. Roosa D B : St. J: Disease of the Ear. 3rd ed. New York, William Wood, 1876. 11. Joynt J A : Repair of the drum. J Iowa Med Soc 9:51, 1919. 12. Linn EG: Closure of tympanic membrane perforations. Arch Otolaryngol Head Neck Surg 58:405, 1953. 13. Derlacki E L : Repair of central perforations of the tympanic membrane. Arch Otolaryngol Head Neck Surg 58:405, 1953. 14. House WF, Sheehy J L : Myringoplasty. Arch Otolaryngol 73:407, 1961. 15. Plester D: Myringoplasty methods. Arch Otolaryngol 78:310, 1963.
16. Shea JJ: Vein graft closure of eardrum perforations. J Otolaryngol 74:358, 1960. 17. Tabb HG: Closure of perforations of the tympanic membrane by vein grafts: A preliminary report of 20 cases. Laryngoscope 70:271, 1960. 18. Goodhill V: Tragal perichondrium and cartilage in tympanoplasty. Arch Otolaryngol 85:480, 1967. 19. Glasscock M E, House WF: Homograft reconstruction of the middle ear. Laryngoscope 78:1219, 1968. 20. Storrs L A : Myringoplasty with the use of fascia grafts. Arch Otolaryngol 74:65, 1961. 21. Gates G A, Avery C A, Prihoda TJ, Cooper JC Jr: Effectiveness of adenoidectomy and tympanostomy tubes in treatment of chronic otitis media with effusion. N Engl J Med 317:1444-1451, 1987. 22. Fry TL , Pillsbury HC : The implications of controlled studies of tonsillectomy and adenoidectomy. Otolaryngol Clin North Am 20:409-413, 1987. 23. Jackson CG: Antimicrobial prophylaxis in ear surgery. Laryngoscope 98:1116-1123, 1988. 24. Sheehy J L , Glasscock M E : Tympanic membrane grafting with temporalis fascia. Arch Otolaryngol 86:391, 1967. 25. Kaylie D M, Gardner E K, Jackson CG: Revision chronic ear surgery. Otolaryngol Head Neck Surg 134:443-450, 2006. 26. Kaylie D M, Bennett M L , Davis B M, Jackson CG: Chronic ear surgical outcomes in smokers and non-smokers. Presented at American Academy of Otolaryngology–Head and Neck Surgery Annual Meeting. Washington, DC, September 16-19, 2007. 27. Donald FJ, McCabe B F, Loevy S S, et al: Atticotomy: A neglected otosurgical technique. Ann Otol Rhinol Laryngol 83:652, 1974. 28. McCleve D E : Repair of bony canal wall defects in tympanomastoid surgery. Am J Otol 6:76, 1985. 29. McCleve D E : Tragal reconstruction of the auditory canal. Arch Otolaryngol 90:35, 1969. 30. Linda R E : The cartilage-perichondrium graft in the treatment of posterior tympanic membrane retraction pockets. Laryngoscope 83:747, 1973. 31. Schwaber M K : Postauricular undersurface tympanic membrane grafting: Some modifications of the “swinging door” technique. Arch Otolaryngol Head Neck Surg 95:182, 1986. 32. Levenson R M : Cartilage-perichondrial composite graft tympanoplasty in the treatment of posterior marginal and attic retraction pockets. Laryngoscope 97:1069, 1987. 33. Adkins W: Composite autograft for tympanoplasty and tympanomastoid surgery. Laryngoscope 100:244, 1990. 34. Glasscock M E, Hart M J: Surgical treatment of the atelectatic ear. Otolaryngol Head Neck Surg 3:15, 1992.
13
Ossicular Reconstruction John T. McElveen Jr., Calhoun D. Cunningham III, and James L. Sheehy Videos corresponding to this chapter are available online at www.expertconsult.com.
Reconstruction of the middle ear sound transformer mechanism has continued to evolve since the pioneering efforts of Wullstein and Zollner in the 1950s. Despite improvements in materials and refinements in surgical technique, complete closure of the air-bone gap on a consistent basis has proven elusive. The excellent hearing results consistently obtained with stapedectomy procedures are not as readily achieved when reconstructing the other middle ear ossicles. In addition, problems with extrusion may jeopardize early hearing gains. This chapter addresses some of the factors responsible for this disparity, and discusses techniques and materials that may assist the surgeon in optimizing the postoperative hearing results, while minimizing the likelihood of extrusion.
HISTORICAL PERSPECTIVE Since Matte’s1 report of a myringostapediopexy in 1901, numerous methods have been attempted to bridge the gap between the tympanic membrane and the inner ear fluids. The modern era of reconstructive middle ear surgery began with reports by Zollner in 19552 and Wullstein in 1956.3 These early attempts focused on creating a sound pressure differential between the oval and round window by adapting the operation to the ossicular problem encountered. If the incus was missing, the graft was placed on the stapes capitulum (type III or columellar tympanoplasty). If the incus and the stapes crura were missing, the graft was laid on the promontory, leaving a mobile footplate exposed (type IV, oval window, or cavum minor tympanoplasty), producing sound protection fort the round window. These techniques usually altered the volume of the middle ear by creating an open mastoid cavity. In 1957, Hall and Rytzner4 described the use of an autogenous incus or malleus to reconnect the mobilized footplate to the tympanic membrane in patients with otosclerosis. Hearing results with this technique were very promising, and the importance of a closed mastoid cavity with a normal middle ear space was quickly realized. Soon thereafter, the search for the ideal material to reconstruct
the sound conduction mechanism was begun. In the late 1950s and early 1960s, much attention was focused on the use of autogenous and alloplastic materials. The first reported use of an artificial material to reconstruct the ossicular chain was by Wullstein in 1952,3 when he used a vinyl-acryl plastic known as palavit to connect the tympanic membrane to the stapes footplate. In 1958, Shea5 described the use of polyethylene tubing placed on the capitulum of the stapes and wedged under the tympanic membrane. His efforts were soon followed by others using various polyethylene prostheses and other inert materials, such as polytef (Teflon) and silicone elastomer (Silastic). Despite many excellent short-term and long-term hearing results, these early alloplastic materials often resulted in extrusion, significant middle ear reactivity, or, worse, penetration of the inner ear. As a result, many surgeons turned to autogenous prostheses that would be more compatible with the middle ear. Following the early work of Hall and Rytzner,4 several otologists, including Farrior,6 Sheehy,7 and Guilford,8 began reporting on the success of using autografts for ossicular reconstruction. The most commonly used autograft material was the body of the incus; however, cartilage and cortical bone were also used. These natural materials were well tolerated in the middle ear and provided reliable hearing results. The disadvantages that soon became apparent were the prolonged time required to sculpt the prosthesis and the lack of availability in chronically diseased ears. In an effort to circumvent some of these issues, in 1966, House and colleagues9 first reported the use of homografts in middle ear reconstruction. Other reports soon followed describing the use of irradiated ossicles, cartilage, and even homograft tympanic membranes with en bloc ossicles.10,11 Homografts had hearing results and biocompatibility similar to autografts; however, concerns regarding the risk of transmission of human immuno deficiency virus and prions (i.e., Creutzfeldt-Jakob disease) ultimately led to their decline in use. In a continued effort to find a safe, reliable, and easily available prosthesis, Shea in 197612 reported on the use of high-density polyethylene sponge (Plastipore) for middle 161
162
OTOLOGIC SURGERY
ear reconstruction. Made of porous polyethylene, this alloplast had nonreactive properties and sufficient porosity to allow ingrowth of tissue. It was readily available commercially and could be easily trimmed with a knife. A similar porous polyethylene that is thermal-fused (Polycel) was developed later and allowed the prosthesis to be coupled to other materials, such as stainless steel. This capability allowed it to be modified to various prosthesis designs. Early reports of porous polyethylene implants revealed a high incidence of extrusion when placed in direct contact with the tympanic membrane. This problem was significantly reduced by placing a disc of cartilage between the head of the prosthesis and the tympanic membrane, as advocated by Coyle Shea and reported by Brackmann and Sheehy.13 As a result, Plastipore and Polycel total ossicular replacement prostheses (TORPs) and partial ossicular replacement prostheses (PORPs) continue to be used with good long-term success today. In an effort to improve extrusion rates associated with porous polyethylene, in 1979 various ceramics were recommended for use in ossicular reconstruction. These alloplastic materials were termed either bioinert or bioactive. Bioinert implants, such as dense aluminum oxide ceramic, did not react with surrounding tissues and were popular in Germany and Japan. Bioactive implants, such as glass ceramic (Ceravital), were biocompatible and reacted with surrounding soft tissue and adjacent bone allowing a coupling between the implant and the ossicle in contact.14 The advantage of ceramic implants was that they could be placed directly under the tympanic membrane without interposing cartilage; however, they were difficult to handle and shape because of their glass nature. In 1984, Grote15 introduced the use of the calcium phosphate ceramic, hydroxyapatite, for tympanoplasty surgery. Subsequently, great interest developed in this material for middle ear prostheses. Hydroxyapatite, which is the mineral matrix of living bone, was known to be a bioactive material achieving integration with surrounding bone and tissue. In 1985, Wehrs16 developed an incus prosthesis and an incus/stapes prosthesis made of hydroxyapatite and reported successful hearing results with a low extrusion rate 4 years later. Since that time, this material has been adapted to various uses and prosthesis designs. The advantage of this material is that it is quite rigid and has a good sound transfer function. The disadvantages of hydroxyapatite are that it has a large mass creating a high input impedance, and it is solid, potentially obstructing the surgeon’s view. Because of its brittleness, it is often combined with other materials to create a prosthesis shaft that is more malleable and easier to shape. More recently, hydroxyapatite has been combined with polyethylene (HAPEX) to create an allograft material that approaches the mechanical strength of bone, but is soft enough to cut with a knife. In an attempt to find a prosthesis that had the rigidity and biocompatibility of hydroxyapatite, but not the mass, titanium prostheses were developed. The specific density
of titanium is low, less than 57% that of stainless steel, yet it is extremely rigid. In addition, it is nonmagnetic, has excellent biocompatibility, and lends itself to being manufactured into various shapes and sizes. Most of the titanium prostheses possess an open head, allowing better visualization during placement. First used for ossicular reconstruction in 1993 in Germany,17 the popularity of titanium prostheses has grown rapidly. Cartilage must still be used, however, between the platform and tympanic membrane to prevent extrusion. Several authors to date have published favorable hearing results with titanium prostheses, and compared with hydroxyapatite, titanium may provide improved hearing responses at higher frequencies because of its low mass.18-21 Despite the improvements in middle ear prostheses, there remained a need for an adhesive or bone cement to stabilize the prostheses or, in some cases, replace them altogether. In the 1980s, ionomeric cements were used for cranioplasties and ossicular chain reconstruction. Although effective in the middle ear,22 aluminum toxicity issues associated with cranioplasties resulted in ionomeric cements being largely supplanted by hydroxyapatite phosphate cements, which are free of aluminum. Two of these cements currently are being produced for otologic applications—Hydroset (Stryker) and OtoMimix (Gyrus). The cements seem to be well tolerated, and have been particularly useful in reconstructing the long process of the incus and stabilizing prostheses. Although numerous techniques and materials have been used for ossicular reconstruction, the quest for the ideal prosthesis is ongoing. Today, autogenous and alloplastic prostheses are used with equally good outcomes, and the surgeon should use what is comfortable and provides consistent, favorable results. This chapter primarily focuses on the use of alloplastic material in ossicular chain reconstruction, and principles that improve outcomes in attempts to reconstruct the middle ear sound transformer mechanism.
PATIENT SELECTION AND EVALUATION The selection of patients undergoing ossicular chain reconstruction largely depends on the problem at hand. Any patient with chronic otitis media may ultimately be a candidate for ossicular chain reconstruction. These patients often present because of the hearing loss associated with their infection or chronic middle ear disease. In other cases, initial hearing may be normal, but become compromised secondary to surgery to remove cholesteatoma. Still other patients present with hearing loss resulting from ossicular problems associated with trauma or congenital ear malformations. Regardless of the cause, the initial evaluation requires a thorough microscopic examination of the ear to identify the problem accurately. In cases of chronic ear disease, ossicular discontinuity may be diagnosed through a perforation or with severe
Chapter 13 • Ossicular Reconstruction
atelectasis of the tympanic membrane. Most commonly, incus erosion is the cause of conductive hearing loss; however, the presence or absence of all three ossicles should be noted. Cholesteatoma involvement of the ossicles is usually identified on examination. In the event of active infection and drainage, it is important to obtain a dry ear with meticulous cleaning and topical antimicrobial preparations before surgery. If a dry ear cannot be obtained, ossicular reconstruction is often staged after tympanoplasty so that ventilation and the middle ear mucosa can normalize. When there is a conductive hearing loss with an intact eardrum, the history should focus on any head or ear trauma, prior ear surgery, family history of otosclerosis, Tullio phenomenon, or unusual congenital abnormalities. Careful evaluation of the symptoms and findings guides the surgeon in the need for surgery, type of procedure, its urgency, and the anticipated result, based on the type of reconstruction required. Audiometric studies are essential in patients with suspected conductive hearing loss and should include pure tone air and bone conduction with masking, speech discrimination, and tympanometry. Severe eustachian tube dysfunction may necessitate concomitant ventilation tube placement or, in rare cases, preclude ossicular chain reconstruction. Acoustic reflex testing is helpful in distinguishing hearing losses resulting from otosclerosis versus an inner ear conductive hearing loss associated with dehiscence of the superior semicircular canal. In the latter, the reflex is present. Conductive hearing losses greater than 25 dB usually signify an ossicular problem. Audiometric results should always be confirmed with the Weber and Rinne tuning fork examination using 512 Hz and 1024 Hz tuning forks. Before operating on the involved ear, the contralateral ear must be assessed. In cases of a better or only hearing ear, alternatives such as a hearing aid or a canal wall down procedure in the setting of chronic otitis media with cholesteatoma may be more suitable. Imaging studies are rarely obtained to evaluate the middle ear and ossicles. Thin-section computed tomography (CT) of the temporal bone may provide useful information, however, in cases of extensive cholesteatoma, malleus fixation, incus dislocation, superior canal dehiscence, or suspected congenital ossicular abnormalities. After a thorough work-up, the physician is prepared to give realistic expectations regarding the chance of hearing restoration and the need for additional surgery. Although ossicular reconstruction may be performed at the time of initial repair or cholesteatoma removal, extensive cases often require reconstruction to be done during a second stage at a later date. In adult patients, 9 to 12 months are allowed to pass before performing a second look procedure to rule out residual cholesteatoma and reconstruct the ossicular chain; in pediatric patients, 6 months is an appropriate time interval. This staging period provides adequate time for small residual cholesteatoma to become apparent, and provides time for diseased mucosa to
163
ormalize. Typically, absorbable gelatin film (Gelfilm) is n placed over the promontory, and the middle ear is packed with middle ear packing soaked in Merogel a nonototoxic antibiotic solution. Enzymatic type packing in conjunction with the Gelfilm should decrease postoperative middle ear adhesions. Staging in this manner helps to ensure a disease-free ear, and improves hearing outcomes. The selection of pediatric patients for ossicular reconstruction depends largely on the status of the ear. In the setting of chronic otitis media, the most important goal is to provide a safe, dry ear regardless of age. When a conductive hearing loss is discovered without evidence of chronic ear disease, it is best to wait until the child is 5 to 7 years old to allow time for eustachian tube maturation. If hearing loss is bilateral, hearing aids are a suitable option until that time. When surgery is contemplated, the options, including the use of hearing aids, must be discussed with the parents, who will ultimately make the decision. When discussing outcomes with patients, it is important to provide a realistic expectation of hearing results. Successful hearing results in ossicular reconstruction are based on the postoperative air-bone gap and stratified as excellent (<10 dB), good (11 to 20 dB), and fair (21 to 30 dB). This success also depends on several factors including the presence or absence of a mobile stapes superstructure, intact canal wall with normal middle ear volume, and adequate eustachian tube function. Although outcomes vary slightly depending on the type of prosthesis used, successful improvement in hearing is generally broken down according to the use of a PORP versus a TORP. In patients undergoing ossicular reconstruction with a PORP, two thirds of patients should achieve hearing outcomes within 15 dB of their bone scores, whereas two thirds of patients with a TORP should achieve hearing outcomes within 25 dB of their bone scores. The disparity in outcomes between PORPs and TORPs is apparent in Brackmann and Sheehy’s review of 1042 cases in which successful hearing with an air-bone gap less than 15 dB was achieved in 63% of PORPs and only 42% of TORPs.23 Regardless of the technique or type of prosthesis, surgeons should use what they are most comfortable with and provides consistent good hearing results.
SURGICAL CONSIDERATIONS The surgeon may encounter ossicular abnormalities when intraoperatively examining the middle ear. During second look procedures after cholesteatoma surgery, the surgeon is aware of the ossicular defect before elevating the tympanomeatal flap, and can decide preoperatively what type of ossicular prosthesis is required. In other cases, the surgeon must make an intraoperative assessment and decision. Consequently, an assortment of prostheses must be available to the surgeon at the time of the procedure.
164
OTOLOGIC SURGERY
FIGURE 13-3. Use of the incus bridge prosthesis with bone cement to bridge a large gap between the foreshortened incus and capitulum of the stapes. FIGURE 13-1. Partial ossicular replacement prosthesis bridging the gap between the mobile stapes and mobile malleus.
FIGURE 13-2. Reconstruction of the ossicular chain using bone cement to augment the foreshortened incus.
The prostheses and ancillary equipment must be able to address the following ossicular abnormalities: 1. Mobile stapes and mobile malleus, but absent incus: In this situation, a standard PORP can be used to bridge the gap between the tympanic membrane and malleus and stapes (Fig. 13-1). 2. Mobile stapes and fixed malleus, but absent incus: With this scenario, the head of the malleus can be amputated with a malleus nipper, and a PORP can be used to connect the handle of the malleus/tympanic membrane with the stapes. 3. Incus necrosis and mobile stapes and malleus: The management of incus necrosis depends on the gap remaining between the capitulum and the incus remnant. If the gap is quite small, or if there is merely disarticulation of the incus, a small amount of bone cement can be used
FIGURE 13-4. Absent stapes superstructure with mobile footplate and mobile incus and malleus.
to restore continuity (Fig. 13-2). If the gap is more substantial (i.e., >1 to 2 mm), bone cement alone may be structurally inadequate to bridge the gap. In these cases, the incus/bridge prosthesis can be used in conjunction with bone cement (Fig. 13-3). Other options include the use of specialized prostheses, such as the Plester or Applebaum prosthesis. If the gap is too great, the incus is discarded and bypassed with a PORP. 4. Fixed footplate and mobile malleus and incus: A standard stapedectomy or stapedotomy is performed. 5. Absent stapes superstructure, but mobile footplate, incus, and malleus (Fig. 13-4): In this situation, a stapes prosthesis can be crimped to the incus and placed on the footplate. The length of the prosthesis should be shortened an additional 0.25 mm to account for the intact footplate. It is important to have adequate tension on the footplate. This technique is preferable to bypassing the incus with a TORP.
Chapter 13 • Ossicular Reconstruction
165
FIGURE 13-5. Foreshortened incus in ear with prior stapedotomy. FIGURE 13-6. Head of malleus removed and prosthesis clipped to malleus handle and extending to fenestra in fixed footplate.
6. Foreshortened incus and a fixed footplate or prior stape dectomy (Fig. 13-5): In the past, the incus would have been discarded, the footplate removed or fenestrated, and a TORP placed. Other options are also currently available, including placement of a Winkle prosthesis that attaches to the foreshortened incus and extends into a small fenestra in the footplate. One can also reconstruct the incus with the incus/ bridge prosthesis and cement, and attach a stapes prosthesis to the reconstructed incus. It may be difficult to crimp the prosthesis to the reconstructed incus. In these situations, a self-crimping stapes piston may be useful. 7. Fixed stapes and fixed or mobile malleus, but absent incus: In the 1960s and 1970s, the stapes would have been removed, the oval window covered with tissue, an incus replacement prosthesis wrapped around the manubrium of the malleus, and the malleus head amputated. Because of the technical difficulty associated with the incus replacement prosthesis, many surgeons would merely use a TORP. More recently, a titanium prosthesis with a ball joint has been developed. The prosthesis affixes to the manubrium with a special clip, and the piston extends medially into a fenestra created in the footplate (Fig. 13-6). 8. Total fixation of the ossicular chain: In this situation, the incus is discarded, a fenestra created in the footplate, and the head of the malleus amputated after the prosthesis has been clipped to the manubrium. In the future, this may be supplanted with a round window vibroplasty, using the floating mass transducer (Fig. 13-7). Although it is preferable to have an intact tympanic membrane when reconstructing the ossicular chain, it is usually not essential. If the reconstruction necessitates a stapedotomy or stapedectomy, however, the tympanic membrane must be intact, and the ear must be free of infection.
FIGURE 13-7. Floating mass transducer situated in round window niche (vibroplasty).
PROSTHESIS SELECTION The ideal prosthesis for reconstructing the ossicular chain is one that is biocompatible, easy to place, and stable over the long term with good sound transmission qualities. When selecting a prosthesis, the surgeon should consider not only the anatomic configuration to be repaired, but also a material that is comfortable to work with during surgery. The multitude of prostheses and techniques that continue to be used today is a testament to the fact that no prosthesis is perfect, and the search for the ideal reconstruction method continues. This section describes some common materials used for ossicular reconstruction today.
166
OTOLOGIC SURGERY
FIGURE 13-8. Autologous incus interposition graft.
Autografts, which have been used since the earliest reconstruction efforts, continue to be used by many surgeons with good results. These are usually autologous incus grafts that can be shaped intraoperatively with a notch for the malleus handle and a cup for the stapes capitulum when present (Fig. 13-8), or a straight shaft to be positioned on the footplate when the stapes is absent. Studies have shown good long-term viability of autologous ossicular grafts, although some concern remains about eventual reabsorption if blood supply is not maintained.24 The time required to sculpt these grafts and the unavailability in some cases has prompted many surgeons to turn to alloplastic materials. Since the 1970s, alloplastic prostheses have achieved significant popularity with the introduction of Plastipore. TORPs and PORPs are biocompatible with the middle ear and provide a reliable sound conduction mechanism. Use of PORPs and TORPs requires interposition of a thin disk of cartilage between the platform and the tympanic membrane to prevent extrusion. PORPs are used in the presence of an intact mobile stapes, and TORPs are used in the absence of the stapes superstructure with a mobile footplate. The malleus is not required for successful use of a PORP or TORP. Polycel, a similar material, has the advantage of being malleable and being coupled to other materials such as stainless steel or hydroxyapatite. Hydroxyapatite has become popular because of its excellent biocompatibility and multiple applications in prosthesis design. Hydroxyapatite is a calcium ceramic that chemically attaches to bone and is osteoconductive. With time, the implant gradually becomes covered with an epithelial layer resembling normal middle ear mucosa. Cartilage does not have to be used with hydroxyapatite, although some authors still recommend it. Early
rostheses were made entirely of dense hydroxyapatite p and noted to be brittle. More recent innovations have involved attaching a hydroxyapatite-reinforced polyethylene composite (HAPEX) shaft or cuff to a dense hydroxyapatite body. This allows the cuff or shaft to be easily trimmed to length with a knife. Various HAPEX prostheses have been produced and usually possess one or two notches in the hydroxyapatite body for placement under the malleus handle (Fig. 13-9). More recently, titanium has been favored as an alloplastic material because of its excellent biocompatibility, rigid strength, and light weight. Titanium prostheses are designed as PORPs or TORPs, and require interposition of cartilage between the prosthesis and tympanic membrane to prevent extrusion. The platform of the prosthesis has an open-top design facilitating placement (see Fig. 13-9). The prostheses come as presized implants of variable length that can be sized and adjusted intraoperatively. Hearing results to date with titanium implants are comparable to hydroxyapatite and polyethylene sponge, and may improve hearing at higher frequencies.17-21 In addition to conventional prostheses, bone cements have been used with success in ossicular reconstruction. These cements are currently hydroxyapatite-based cements that are mixed intraoperatively and harden within 4 to 6 minutes. They are useful in the setting of incus necrosis, when the gap is not too large, to span the incus and stapes. When a significant amount of incus necrosis is found, a titanium incus/bridge prosthesis can be used to set up a framework or carrier for the bone cement to be applied between the incus and stapes (see Fig. 13-3). Care must be taken to place Gelfoam around the oval window before application to decrease the risk of bone cement getting around the footplate and causing fixation.
Chapter 13 • Ossicular Reconstruction
167
gap between the incus and stapes capitulum. The malleus/oval window prosthesis has a ball joint that allows the shaft of the prosthesis to be rotated into the oval window after the prosthesis is clipped onto the malleus handle. Occasionally, one may encounter patients with hearing loss not amenable to standard reconstruction of the ossicular chain or who have failed multiple attempts at ossicular reconstruction. More recent attempts at improving hearing in such situations have focused on introducing sound through the round window using a vibratory transducer. The Vibrant Soundbridge (Med-El, Innsbruck, Austria) comprises an implanted receiver module placed behind the ear and is connected via a small cable to a floating mass transducer that is placed in the round window niche. Electric current from the receiver module drives the magnet within the transducer, transmitting vibratory stimulation to the inner ear. Early hearing results with this novel technique are promising.27
SURGICAL TECHNIQUE
FIGURE 13-9. Partial and total ossicular replacement prostheses currently available for implantation.
More recent reports of using bone cement in ossicular reconstruction have shown good hearing results (air-bone gap ≤20 dB) in 90% of patients.22,25,26 A few scenarios encountered during middle ear surgery may prompt the surgeon to use a specific prosthesis design. Three prostheses that warrant mention are the Plester (Kurz GmbH, Dusslingen, Germany) incudostapedial joint prostheses, the incus/bridge prosthesis, and the malleus/oval window prosthesis. The first two prostheses are used in the setting of incus necrosis to span the gap between the incus and stapes. The Plester angular prosthesis, made of titanium, has two clips that attach to the incus long process remnant with a cup portion that rests on the stapes capitulum. The cup of the prosthesis is first placed on the stapes capitulum, and then the clips are crimped onto the incus remnant. The shaft of the prosthesis comes in different lengths to accommodate differing amounts of incus necrosis. The incus/bridge prosthesis is designed to be used in conjunction with a bone cement. The prosthesis is clipped onto the incus remnant and extends to the capitulum of the stapes. Using the prosthesis as a scaffold, the bone cement is applied to bridge the
Although the specific surgical technique for reconstructing the ossicular chain may vary from surgeon to surgeon, there are common aspects inherent to all ossicular reconstructions. As with stapedectomies, the ossicular chain reconstruction can be performed under a straight local anesthetic, monitored anesthesia care (MAC), or general anesthesia. This decision is based on the age of the patient (i.e., child versus adult), the patient’s “fear factor,” whether a mastoidectomy is being performed in conjunction with the ossicular chain reconstruction, the patient’s other medical conditions (e.g., Parkinson’s disease), and the comfort level of the surgeon. Advantages of performing ossicular chain reconstruction under local anesthesia include the following: surgeon and patient ability to assess hearing intraoperatively, absence of gases accumulating in the middle ear space, altering the position of the tympanic membrane with respect to the prosthesis, and the cheaper cost compared with general anesthesia. Many patients prefer to be “asleep” during any surgical procedure, however. With the use of general anesthesia, the patient is unaware of any discomfort, the surgeon is not pressured from a time perspective, and the middle ear gas issue is largely circumvented by not using nitrous oxide.
Exposure and Assessment Regardless of the type of anesthesia, if one is performing an ossicular chain reconstruction without a concomitant mastoidectomy, the procedure usually is performed with a transcanal approach. If the canal is quite narrow, an endaural or postauricular approach can be employed. When using a transcanal approach, the ear canal is injected with 1% lidocaine (Xylocaine) and 1:40,000 epinephrine. If the patient is older than 65 years, a 1:60,000 solution is
168
OTOLOGIC SURGERY
FIGURE 13-10. Injection of lateral external auditory canal.
FIGURE 13-11. Vascular strip injection.
FIGURE 13-12. Canal incisions for tympanomeatal flap.
used. Using a 3 mL glass syringe, the solution is injected just inside the meatus, and not at the bony-cartilaginous junction where the skin is more adherent (Fig. 13-10). The ear canal is dilated with specula gradually increasing in size; this forces the local anesthetic solution medially. The vascular strip is injected last (Fig. 13-11); this is probably the most important injection. Vertical incisions are made at 6 o’clock and 12 o’clock positions, beginning 1 mm lateral to the annulus, and connected by a horizontal incision 10 to 12 mm lateral to the annulus (Fig. 13-12). Although adequate surgical exposure is important, less exposure may facilitate obtaining the proper tension on the prosthesis because the tympanic membrane is stabilized at the bony annulus.
Although the canal incisions are the same in patients with prior intact canal wall mastoidectomies, the technique is altered in patients with canal wall down mastoidectomies. The 6 o’clock position is shifted anteriorly to 4 o’clock, and the incision at the 12 o’clock position is avoided altogether. The horizontal incision begins anteriorly and extends toward the mastoid cavity. Using this technique, the exposure is more limited, but two point fixation is maintained. This facilitates slight tension on the prosthesis with respect to the tympanic membrane. Particular care should be taken with dissection in the area of the facial ridge, where the vertical segment of the facial nerve may not be protected by a bony coverage.
Chapter 13 • Ossicular Reconstruction
When adequate exposure is achieved, the status of the middle ear is assessed. In performing a second look procedure, it is essential to check carefully for any residual or recurrent cholesteatoma before considering ossicular chain reconstruction. In addition, the presence of serous fluid may necessitate concomitant placement of a ventilation tube, whereas a dehiscent facial nerve obscuring the oval window may preclude ossicular chain reconstruction using conventional techniques. Vibroplasty techniques may be useful in such cases and in cases with mixed hearing loss, allowing the surgeon to address the conductive and the sensorineural components.27 Although the surgeon’s options may be limited in most situations to a PORP or a TORP, with the advent of bone cements the surgeon must consider alternative techniques. Before indiscriminately removing a foreshortened incus, the surgeon should consider ways to conserve the middle ear transformer mechanism. In addition, it is particularly important to assess accurately the malleus and the stapes for fixation. Left undetected, fixation of either ossicle would compromise the postoperative hearing results.
Cartilage Preparation In partial and total ossicular chain reconstructions, cartilage is routinely used to interface between the prosthesis and the overlying tympanic membrane. Although the cartilage can be harvested from the tragus, the conchal cartilage is preferred. The conchal cartilage has the proper convexity and is cosmetically acceptable. If a postauricular incision is used, the cartilage is readily accessible without the need for an additional incision. When a transcanal approach is employed, however, a separate small posterior conchal incision is used. Optimally, one should attempt to maintain perichondrium on one surface of the cartilage; however, this is not essential. The cartilage should be beveled on its edges, adequately cover the platform of the prosthesis, and be quite thin. The thinning is typically performed using a standard scalpel, but devices are available, such as the “precise cartilage knife set” (Kurz), to assist the surgeon in thinning the cartilage to the proper thickness. Interposing the cartilage between the prosthesis and the tympanic membrane can be challenging. The prosthesis is tilted posteroinferiorly, the cartilage is placed over the anterior edge of the platform, and both are gently rocked back into position, maintaining slight tension on the tympanic membrane. Some surgeons suture the cartilage to the platform to circumvent this maneuver. Regardless of technique, it is important not to put too much downward pressure on the prosthesis and risk subluxation of the stapes or footplate into the vestibule when positioning the prosthesis. In some cases of ossicular chain reconstruction, a cartilage interface is not required. These cases typically involve the augmentation of the incus with a prosthesis or bone cement or both, or the use of a prosthesis that attaches to the manubrium of the malleus.
169
FIGURE 13-13. Cartilage shoe in oval window niche with prosthesis in position on mobile footplate.
Placement of Prosthesis Regardless of whether the surgeon is placing a PORP or a TORP, it is essential that the prosthesis be under slight tension and at a favorable angle. Various measuring devices are available to assist the surgeon in making the determination. In measuring, one must consider the additional length caused by the cartilage placed between the prosthesis and the overlying tympanic membrane. The prosthesis should fit perfectly without tension before placement of the cartilage. If the prosthesis is slightly short, the length may be corrected later by adding an additional piece of cartilage. Before placing the prosthesis, the middle ear is partially filled with a middle ear packing material that has been soaked in an antibiotic solution. The sponge facilitates placement of the prosthesis and cartilage, stabilizes them temporarily, and acts as counterpacking for the packing that is to be placed in the canal after replacement of the tympanomeatal flap. Despite perfect alignment of the PORP or TORP intraoperatively, the prosthesis can shift during the healing process. Adequate middle ear packing may minimize this; a small amount of bone cement placed at the junction of the PORP with the capitulum of the stapes gives the best stability. Bone cement cannot be used on the footplate, however. Instead, a cartilage punch (Kurz) is used to create a “cartilage shoe” that stabilizes the shaft of the TORP and minimizes slippage on the flat footplate surface (Fig. 13-13). A small piece of perichondrium placed on the footplate may limit ������������������������ further slippage ���������������� on the footplate. Bone cements are increasingly being used not only to stabilize the prosthesis, but also to augment the incus.
170
OTOLOGIC SURGERY
If the gap between the incus remnant and the stapes capitulum is quite small, the cement can be used alone. If the gap is more substantial, however, a prosthesis such as the incus/bridge prosthesis is placed and secured with bone cement. When using the cement, it is important that packing is placed over the footplate, to protect it from the cement. In addition, the mucosa should be carefully stripped off the portion of the incus and stapes where the cement is being applied. This facilitates a firm bond between the bone and the cement. If one is attaching a stapes prosthesis to a reconstructed incus, a self-crimping type of stapes prosthesis may be preferable. The most challenging ossicular chain reconstruction is when the incus is absent, and the footplate is fixed. One can use a TORP, but great care is needed to ensure that the prosthesis does not sublux into the vestibule. Several prostheses attach to the malleus handle to limit the incursion of the prosthesis into the vestibule. A slit is made in the periosteum of the malleus, and the prosthesis is secured laterally; the shaft of the prosthesis is then rotated into the fenestra created in the footplate (see Fig. 13-6). If performing a vibroplasty, the floating mass transducer can be placed in various locations. It is most frequently placed on the footplate or on the round window. When placed in the round window niche, the transducer should be reinforced with a piece of cartilage to prevent displacement of the transducer away from the round window during activation (see Fig. 13-7).
In patients undergoing ossicular chain reconstruction using either a transcanal or a postauricular approach, perioperative systemic antibiotics have been administered and a mastoid dressing has been placed. This dressing is removed on the patient’s first postoperative day. The patient is instructed to keep the ear dry. If a postauricular incision has been made, the patient is seen 7 to 10 days postoperatively for suture or Steri-Strip removal. The patient’s middle ear and external ear have typically been packed with Merogel and Gelfoam, respectively. The patient continues to take water precautions with respect to the external ear canal, but is allowed to get water behind the ear. Four weeks postoperatively, the patient is instructed to instill antibiotic ear drops in the external auditory canal, 3 drops three times daily for 2 weeks. The patient returns 3 months postoperatively to have the ear examined and the hearing assessed.
the surgeon and patient must have realistic expectations of hearing outcomes. A patient with poor eustachian tube function or a patient with a very limited middle ear space after a canal wall down mastoidectomy is not as likely to have the same hearing results as a patient with a posttraumatic incudostapedial discontinuity in an otherwise normal middle ear. In patients with poor eustachian tube function, the surgeon should use a slightly shorter prosthesis, anticipating some retraction medially; this optimizes hearing without risking extrusion. In some patients with severe eustachian tube dysfunction, a concomitant ventilation tube may be required. Despite careful prosthesis selection and placement, extrusion continues to be a cause of failure in ossicular reconstruction. Extrusion rates have ranged from 5% to 39% in the literature, and current rates of extrusion have been reported to be 1% for titanium prostheses and 3.3% for hydroxyapatite. Whether the middle ear space is narrowed or not, the prosthesis should always be under slight tension. This tension is achieved by minimally elevating the tympanomeatal flap and accurately sizing the prosthesis and cartilage. The cartilage should completely cover the platform of the prosthesis without touching the side walls. The middle ear is packed with a packing that breaks down adhesions and yet stabilizes the prosthesis. The ear canal is packed as well to stabilize the tympanic membrane and prosthesis further. Although PORPs are less likely to slip than TORPs, a small amount of bone cement placed on the interface between the prosthesis and the capitulum of the stapes alleviates the problem altogether. To minimize slippage of a TORP, the cartilage shoe prosthesis is placed on the footplate, and the shaft of the prosthesis extends through the fenestra created in the cartilage. If one attempts to place a TORP in a patient without a footplate, it is essential that a large piece of connective tissue be placed over the oval window and downward pressure on the prosthesis minimized. Otherwise, the prosthesis may extend into the vestibule, risking a sensorineural hearing loss. Staging the operation in patients with concomitant tympanic membrane perforations or in patients undergoing cholesteatoma surgery may make the difference between success and failure when reconstructing the ossicular chain. Although Silastic was previously used to cover the denuded promontory and maintain a middle ear space, Gelfilm works equally well, and does not create the thick mucosal sac that occasionally envelops the Silastic sheeting. In addition, the Gelfilm resorbs over time and does not obscure the middle ear structures during the second stage procedure.
FACTORS THAT MAKE A DIFFERENCE
SUMMARY
To achieve optimal hearing outcomes on a consistent basis, and to minimize the potential for extrusion, the surgeon must take into consideration various factors. First,
Since the 1950s, much progress has been made in reconstructing the middle ear sound transformer mechanism. Improvements in technique and materials have been
POSTOPERATIVE CARE
Chapter 13 • Ossicular Reconstruction
seen. In the future, further developments are anticipated in “powered prostheses,” which would allow the surgeon to address the conductive and the sensorineural components in a patient with a mixed hearing loss. Even with these innovations, however, the surgeon must continue to assess each patient on an individual basis, alter the surgical technique based on those findings, and adhere to the basic principles of chronic ear surgery.
REFERENCES 1. Matte: Uber Versuche mit Ainheilung des Trommelfells an das Kopfchen des Steigbugels nach operative Behandlung chronischer Mittelohreiter-ungen. Arch Ohren Nasen Kehlkopfheilkd 53:96, 1901. 2. Zollner F: Principles of plastic surgery of the sound conduction apparatus. J Laryngol Otol 69:637, 1955. 3. Wullstein H : Theory and practice of tympanoplasty. Laryngoscope 66:1076, 1956. 4. Hall A, Rytzner C : Stapedectomy and autotransplantation of ossicles. Acta Otolaryngol 47:318, 1957. 5. Shea JJ: Fenestration of the oval window. Ann Otol Rhinol Laryngol 67:932, 1958. 6. Farrior J B : Ossicular repositioning and ossicular prosthesis in tympanoplasty. Arch Otolaryngol 71:443-449, 1960. 7. Sheehy J L : Ossicular problems in tympanoplasty. Arch Otolaryngol 81:115-122, 1965. 8. Guilford F: Repositioning of the incus. Laryngoscope 75:236-242, 1965. 9. House WF, Patterson M E, Linthicum FH : Incus homografts in chronic ear surgery. Arch Otolaryngol 84:148153, 1966. 10. Wehrs R E : The borrowed incus in tympanoplasty. Arch Otolaryngol 85:371-379, 1967. 11. House WF, Glasscock M E, Sheehy J L : Homograft transplants of the middle ear. Trans Am Acad Ophthalmol Otolaryngol 873:836-841, 1969. 12. Shea JJ: Plastipore total ossicular replacement prosthesis. Laryngoscope 86:239, 1976. 13. Brackmann D E, Sheehy J L : Tympanoplasty: TORPs and PORPs. Laryngoscope 89:108-114, 1979.
171
14. Niparko J K, Kemink J L , Graham M D, et al: Bioactive glass ceramic in ossicular reconstruction: A preliminary report. Laryngoscope 95:249-258, 1985. 15. Grote JJ: Tympanoplasty with calcium phosphate. Arch Otolaryngol 110:197-199, 1984. 16. Wehrs R E : Incus replacement prosthesis of hydroxyapatite in middle ear reconstruction. Am J Otol 10:181-182, 1989. 17. Dalchow CV, Grun D, Stupp H F: Reconstruction of the ossicular chain with titanium implants. Otolaryngol Head Neck Surg 125:628-630, 2001. 18. Krueger WO, Feghali JG, Shelton C : Preliminary ossiculoplasty results using the kurz titanium prostheses. Otol Neurotol 24:836-839, 2002. 19. Gardner E K, Jackson CG, Kaylie D M : Results with titanium ossicular reconstruction prostheses. Laryngoscope 114:65-70, 2004. 20. Martin A D, Harner SG: Ossicular reconstruction with titanium prosthesis. Laryngoscope 114:61-64, 2004. 21. Zenner H P, Stegmaier A, Lehner R , et al: Open tubingen titanium prostheses for ossiculoplasty: A prospective clinical trial. Otol Neurotol 22:582-589, 2001. 22. Feghali J, Barrs D M, Beatty C, et al: Bone cement reconstruction of the ossicular chain: A preliminary report. Laryngoscope 108:829-836, 1998. 23. Brackmann D E, Sheehy J L : TORPs and PORPs in tympanoplasty: A review of 1042 operations. Otolaryngol Head Neck Surg 92:32-37, 1984. 24. Merchant S N, Nadol J B Jr: Histopathology of ossicular implants. Otolaryngol Clin North Am 27:813-833, 1994. 25. Babu S, Seidman M : Ossicular reconstruction using bone cement. Otol Neurotol 25:98-101, 2004. 26. Goebel J A, Jacob A : Use of Mimix hydroxyapatite bone cement for difficult ossicular reconstruction. Otolaryngol Head Neck Surg 132:127-134, 2005. 27. Colletti V, Soli S D, Carner M, et al: Treatment of mixed hearing losses via implantation of vibratory transducer on the round window. Int J Audiol 45:600-608, 2006. 28. Gardner E K, Jackson CG, Kaylie D M : Results with titanium ossicular reconstruction prostheses. Laryngoscope 114:65-70, 2004.
14
Canal Wall Reconstruction Tympanomastoidectomy Bruce J. Gantz, Samuel P. Gubbels, and Eric P. Wilkinson Videos corresponding to this chapter are available online at www.expertconsult.com.
The primary goal in the surgical management of chronic otitis media with cholesteatoma is the creation of a dry, safe ear through removal of disease and the alteration of anatomy to prevent recurrence. This goal can be accomplished effectively with preservation (canal wall up) or removal (canal wall down) of the posterior canal wall, both of which are described in other chapters 16 and 17. This chapter describes the technique of canal wall reconstruction tympanomastoidectomy with mastoid obliteration. Canal wall reconstruction tympanomastoidectomy combines elements of the canal wall up and canal wall down procedures to optimize surgical exposure for removal of disease, and creates a blockage of the attic that prevents recurrence of retraction pockets and recurrence of cholesteatoma. Many authors have described reconstruction of the posterior canal wall and mastoid obliteration using various methods, including composite osteoperiosteal flaps,1 composite cartilage/titanium grafts,2 ceramic alloplasts,3-7 bone pâté,8-10 costal cartilage,11 and bone cements.12-14 Canal wall reconstruction tympanomastoidectomy with mastoid obliteration was originally described by Mercke in 1987.15 Important modifications to the Mercke technique have been have been published by the senior author16 and are described in detail in this chapter.
ADVANTAGES AND DISADVANTAGES OF CANAL WALL UP AND CANAL WALL DOWN TECHNIQUES Canal wall down tympanomastoidectomy is the gold standard for surgical management of cholesteatoma.17,18 The enhanced exposure to the attic, antrum, and middle ear afforded by removal of the posterior canal wall provides for optimal visualization and removal of disease in cases of extensive cholesteatoma. Removal of the canal wall and lateral attic also prevents retraction and recurrent cholesteatoma formation. In
a ddition, all nitrogen-absorbing mucosa of the mastoid cavity and epitympanum is removed, and ultimately is replaced by stratified squamous epithelium after healing has occurred. When performed properly, the canal wall down procedure can result in a recidivism rate of 2%.17 Disadvantages of the canal wall down technique include the accumulation of debris in the mastoid cavity necessitating periodic débridement and potentially the need for ongoing water restrictions to help avoid bowl infections. The middle ear space that results after the canal wall down technique is narrower than that after the canal wall up technique, which can make ossicular reconstruction more difficult. The wide meatoplasty, which is a crucial component in the canal wall down technique, can present difficulties in the placement of a hearing aid, and may have an unacceptable esthetic appearance to some patients. Preservation of the posterior canal wall in cholesteatoma surgery has many advantages, including the elimination of the need for periodic cleaning and avoiding the need for water restrictions. The recidivism rate has been reported to be 36% in adults and 67% in children,19 however, higher by many reports than the incidence of recurrent disease seen with canal wall down procedures.18,20 The high rate of recidivism seen in canal wall up procedures can be due to numerous factors. First, exposure of the attic, antrum, and facial recess is more limited in canal wall up procedures compared with canal wall down approaches, which may lead to difficulty in complete removal of all involved air cell tracts and elimination of cholesteatoma at the initial procedure. Second, the epitympanum and mastoid cavity are ultimately relined with nitrogen-absorbing cuboidal mucosal epithelium after canal wall up procedures. The presence of this nitrogen-absorbing mucosa is thought to lead to negative middle ear and mastoid pressures, especially when continued inflammation with associated hypervascularity affects the mucosal layer.21 This large surface area of nitrogen-absorbing epithelium 173
174
OTOLOGIC SURGERY
along with underlying eustachian tube dysfunction can lead to progressive retraction of the tympanic membrane postoperatively, and ultimately to recurrence of cholesteatoma.16 Eustachian tube dysfunction exacerbates this scenario in children.
RATIONALE FOR CANAL WALL RECONSTRUCTION TYMPANOMASTOIDECTOMY WITH MASTOID OBLITERATION Canal wall reconstruction tympanomastoidectomy with mastoid obliteration is a technique in which the posterior canal wall is removed en bloc using a microsagittal saw to provide canal wall down–like exposure and ensure optimal visualization, aiding complete elimination of disease. The canal wall is replaced after the disease has been removed. The attic and mastoid cavity are isolated from the middle ear using pieces of cortical bone and are obliterated using bone pâté. Replacement of the posterior canal wall allows for preservation of near-normal middle ear anatomy, which can aid with ossicular reconstruction and avoid the need for routine cavity cleanings. Blocking the attic and mastoid cavities prevents re-retraction of the tympanic membrane by eliminating the negative pressure associated with a nitrogen-absorbing mucosa lining the cavity and physically limiting tympanic membrane retraction into the middle ear only.
PATIENT PREPARATION The patient is placed in the supine position on the operating room table with the head turned away from the side to be operated on. Facial nerve monitoring electrodes are applied to the orbicularis oris and orbicularis oculi muscles. Hair is clipped from the postauricular area, and a C-shaped incision 3 to 4 cm behind the pinna and within the hairline is marked. A solution of lidocaine 1% with epinephrine 1:100,000 is injected, and povidone-iodine solution is applied to the area, including the external auditory canal (EAC). Sterile towels are placed around the prepared area, being careful to include the face on the involved side in the field. A large sterile adhesive drape is applied over the field and surrounding towels followed by application of a split drape to the torso and surrounding area. A sterile irrigation/drainage bag is applied after trimming the adhesive drape around the auricle and postauricular area.
Posterior superior Cut Anterior superior cut Canal skin Inferior compound minor cut.
Canal wall reconstruction is indicated in patients of all ages with congenital or acquired cholesteatoma of the middle ear with extension to the attic, antrum, or mastoid not amenable to simple tympanoplasty or atticotomy. Canal wall reconstruction can be performed in patients who have had prior mastoid surgery, unless the posterior canal wall has been removed, or the lateral portion of the posterior canal wall drilled away in a prior canal wall up procedure. The procedure is not contraindicated in patients with facial paralysis, labyrinthine fistula, or an encephalocele. Canal wall reconstruction can be performed in patients with extensive scutum erosion or destruction of the medial aspect of the posterior canal wall from cholesteatoma or prior surgery, although there are additional reconstructive considerations in these circumstances. Canal wall reconstruction is contraindicated in patients with severe mastoid cholesteatosis that cannot be cleared from all mastoid air cell tracts, as sometimes occurs in young children. Care should be taken in the use of the canal wall reconstruction technique in patients with an extensive infectious component, although infection in general is not a contraindication to the procedure.
MFD
PATIENT SELECTION
. J.B
Fn
S. S.
Incus removed
FIGURE 14-1. Canal wall up mastoidectomy with facial recess. Dashed lines show proposed bone cuts. MFD-Middle Fossa Dural Plate; Fn- Facial nerve; SS-Sigmoid Sinus; JB-Jugular Bulb.
Chapter 14 • Canal Wall Reconstruction Tympanomastoidectomy
SURGICAL TECHNIQUE Mastoidectomy: Special Considerations A postauricular incision is made approximately 2 cm behind the hairline and carried to the level of the temporalis fascia superiorly. Dissection of the skin flap is carried anteriorly to within 2 to 3 mm of the EAC. A wide (3 to 4 cm) anteriorly based musculoperiosteal (Palva) flap is incised, elevated, and retracted anteriorly with the pinna. Bone pâté is collected from the mastoid cortex and squamous temporal bone using a large cutting burr and suction using a bone pâté collector (Otomed Corporation, Lake Havasu City, AZ, or Anspach, Palm Beach Gardens, FL). Bone pâté collection is stopped if an air cell becomes exposed. After collection, the bone pâté is soaked in bacitracin solution placed aside until the reconstruction. Calvarial bone slices approximately 4 × 10 × 0.5 mm in size are harvested from the mastoid tip or retrosigmoid area using an osteotome and are placed aside. An intact canal wall mastoidectomy and facial recess are performed as described in Chapter 16. Care is taken to maintain adequate posterior canal wall bone thickness and to avoid lowering (medializing) the lateral portion of the bony canal, both of which aid in the canal wall reconstruction to come. Cholesteatoma that is encountered in the mastoid or facial recess is removed as exenteration of air cells in these areas progresses. Meticulous removal of
175
all air cells in the sinodural angle, mastoid tip, and retrofacial areas is crucial. The facial recess is opened widely and extended inferior to the floor of the EAC. The incus is removed, as is the head of the malleus. The skin of the posterior, inferior, and superior EAC is elevated off the underlying bone without any prior canal skin incisions (Fig. 14-1). The annulus is elevated, and the tensor tympani tendon is cut allowing the tympanic membrane and canal skin to be reflected anteriorly as a sleeve. The tympanic membrane remnant is separated from the neck of the cholesteatoma sac in this process. A microsagittal saw (Anspach, Palm Beach Gardens, FL, or Jed Med Bien Air, St Louis, MO) is used under continuous irrigation to create superior and inferior cuts in the posterior canal wall. The cuts are performed in a locking fashion to help prevent collapse of the canal into the mastoid after reconstruction (see Fig. 14-1). Superiorly, two cuts are made. The first is parallel to the temporal lobe in a posterior-to-anterior direction (Fig. 14-2A). The second cut is perpendicular to the first, made from the superior external canal to meet the posterosuperior cut at a right angle (Fig. 14-2B). The inferior cut extends from the inferior facial recess laterally, widening and beveling as a compound miter (Fig. 14-2C). It is important to bevel the cut from medial to lateral and posterior to anterior so that when the canal wall is replaced, it does not fall back into the mastoid when the canal wall skin is packed back into place.
First bone cut
Second bone cut
A
First bone cut
B C FIGURE 14-2. A, Posterosuperior bone cut being made with microsagittal saw. B, Second bone cut made superiorly in external auditory canal at right angle to first cut. C, Inferior cut starting at inferior part of facial recess, widening and beveling laterally in a compound miter fashion. Note beveling of cut in medial to lateral direction and posterior to anterior direction to prevent collapse of posterior canal wall later in procedure.
176
OTOLOGIC SURGERY
PCW removed
4. 1.
5.
2.
3.
Fn
Cholesteatoma in: 1. 2. 3. 4. 5.
Antrum Epitympanum Mesotympanum Malleus-head clipped off Stapes
FIGURE 14-3. View of cholesteatoma after posterior canal wall (PCW) has been removed. Medial aspect of PCW is inspected for adherent squamous epithelium. Fn-Facial Nerve.
The posterior canal wall segment is removed, and the medial portion is inspected (Fig. 14-3). Any adherent squamous epithelium is removed with a diamond drill or gauze sponge. The posterior canal wall is placed aside for future reconstruction. The cholesteatoma is carefully dissected from the tympanic cavity, epitympanum, antrum, and mastoid (see Fig. 14-3). The cog is drilled, and the anterior epitympanic space is cleared of any residual disease, as is the sinus tympani. The zygomatic root air cells are drilled away similar to a canal wall down procedure. The wall down exposure provides undivided access to the mastoid, tympanic cavity, and antrum allowing for optimal exposure to the cholesteatoma to aid with complete removal (Fig. 14-4).
Reconstruction Reconstruction begins with the placement of a teardropshaped piece of reinforced silicone elastomer (Silastic) (0.04 mm thick) into the middle ear with the neck extending into the eustachian tube. The Silastic graft rests on top of the stapes superstructure (if present) and extends into the eustachian tube, but does not extend posteriorly
into the facial recess (Fig. 14-5A and B). In patients without an intact stapes superstructure, an additional wedgeshaped piece of Silastic is placed onto the footplate, resting underneath the larger teardrop-shaped Silastic graft (Fig. 14-5C); this holds a space for reconstruction during a second look procedure. A temporalis fascia graft is then harvested. This graft must be large enough to extend beyond the margins of the posterior canal wall cuts. The temporalis fascia is placed medial to the tympanic membrane remnant, extending laterally to the bony-cartilaginous junction. Absorbable gelatin sponge (Gelfoam) disks are placed between the Silastic graft and the underlay fascia graft to hold it in place (see Fig. 14-5A and B). A large calvarial bone slice is trimmed to extend from the zygomatic root to the margin of the posterior canal wall and placed in a near-axial orientation into the epitympanum to prevent future retraction of the grafted drum. The attic bone chip extends from the anterior epitympanum posteriorly to the posterior aspect of the lateral semicircular canal, and is wide enough to span from the tympanic facial nerve in a lateral direction to beyond the scutum (Fig. 14-6). The bony posterior canal wall is replaced, and an additional calvarial bone graft is placed in a near-coronal orientation into the facial recess (Fig. 14-7). The attic,
Chapter 14 • Canal Wall Reconstruction Tympanomastoidectomy
177
Cholesteatoma removed
1. 2.
A.
5.
3. 4.
7. 8.
C LS
MFD
6.
Fn
S.
D
PF
. S.S
Inf.
2.
1. 7. A.
3.
P.
5. 8. 6.
Fn
MFD
C
Cochleariform process Eustachian tube orifice Promontory Round window Stapes Pyramidal eminence (process) Cog removed Fn on floor of epitympanum
LS
1. 2. 3. 4. 5. 6. 7. 8.
4. Beveled (inferior cut) of PCW
FIGURE 14-4. Wall down view provided by canal wall reconstruction tympanomastoidectomy with mastoid obliteration approach with uninterrupted access to the mesotympanum, epitympanum, antrum, and mastoid to aid in complete removal of cholesteatoma. MFD-Middle Fossa Dural Plate; Fn-Facial Nerve; SS-Sigmoid ���������������������������������������������������������������������������������������������������������������� Sinus; LSC-Lateral ���������������������������������������������������������������������������������������������� Semicircular Canal; PFD-Posterior Fossa Dural Plate; PCW-Posterior Canal wall.
antrum, and mastoid complex is filled with bone pâté, and dry gauze is used to compress the pâté in place and wick off excess bacitracin solution from the pâté (Fig. 14-8). Attention is turned to the ear canal where a thin Cottle speculum is used along with the operating microscope to displace the posterior canal skin onto the bony posterior canal wall and visualize the tympanic membrane remnant. The underlay fascial graft is inspected through the perforation in the tympanic membrane remnant to verify that there is complete coverage of the perforation, and to ensure that no slippage of the graft has occurred during the reconstruction of the canal wall and obliteration of the mastoid cavity. Gelfoam is placed lateral to the grafted tympanic membrane, and the EAC is packed with ¼����������������������������������������� �� ���������������������������������������� inch iodoform strip gauze that has been soaked in bacitracin ointment (Fig. 14-9). The Palva flap is closed with absorbable sutures. A ¼ inch Penrose drain is placed between the Palva flap and the postauricular
flap and trimmed to size before closure of the postauricular tissues. The drapes are removed, and a sterile mastoid dressing is applied.
POSTOPERATIVE CARE Perioperatively, patients are given antibiotics, typically piperacillin-tazobactam and levofloxacin in adults and piperacillin-tazobactam alone in children younger than 18 years. Penicillin-allergic patients are typically administered clindamycin with levofloxacin. Postoperatively, intravenous antibiotics are continued for 48 hours. The Penrose drain is removed on the second postoperative day. The mastoid dressing is changed daily until discharge, at which time the patient is instructed to remove it in 2 to 3 days. Patients are discharged from the hospital with oral levofloxacin (in adults), amoxicillin/clavulanic
178
OTOLOGIC SURGERY
3A. 3B.
3B.
8.
5.
Fascia graft
3A.
2.
4.
1.
Fn
7. 5. 6.
A
1. 2. 3A. 3B. 4. 5. 6. 7. 8. 9.
Tympanic membrane TM defect Fascia graft Canal skin Eustachian tube Gelfoam pledgets Silastic sheet Facial recess Mastoid cavity Oval window Silastic wedge used when stapes superstructure missing 10. Stapes
B
6.
7. 5.
VII. 6. 9.
C
10.
FIGURE 14-5. A and B, Surgeon’s view (A) and horizontal cross section (B) of the temporal bone showing Silastic graft in mesotympanum extending into eustachian tube. The temporalis fascial graft has been placed medial to tympanic membrane (TM) remnant and extends laterally to bony-cartilaginous junction. Gelfoam is placed between Silastic and fascial graft. C, Horizontal cross section showing additional Silastic wedge in oval window resting on stapes footplate in patients without a stapes superstructure. Fn-Facial Nerve.
acid (in adults allergic to levofloxacin and patients <18 years), or clindamycin (in patients allergic to penicillin), to complete a 14-day course. The iodoform-bacitracin pack is removed from the external canal 1 week postoperatively, and the skin sutures are removed. A second look tympanoplasty with ossiculoplasty is performed, typically 6 months after the initial
t ympanomastoidectomy. During this second surgery, the status of the middle ear and tympanic membrane graft is assessed, the middle ear is examined for the presence of residual cholesteatoma, and ossicular reconstruction is performed. Ossiculoplasty usually involves placement of a thin tragal cartilage graft over a titanium ossicular prosthesis.
179
Chapter 14 • Canal Wall Reconstruction Tympanomastoidectomy Attic bone chip
Bone pâté.
FIGURE 14-8. Mastoid cavity and antrum obliterated with bone pâté.
FIGURE 14-6. Large calvarial bone chip placed into attic in near-axial orientation to prevent future tympanic membrane re-retraction. Chip extends from zygomatic root to posterior aspect of lateral semicircular canal and from tympanic facial nerve to well beyond scutum.
Anterior extension of bone chip 1.
A
S
2.
P
1.
2.
B A
I
Bone chips 1. Attic 2. Facial recess
FIGURE 14-7. A and B, Facial recess bone chip placed in near-coronal orientation after replacement of posterior canal wall.
180
OTOLOGIC SURGERY
Cottle speculum
10.
9.
8.
12.
5.
3A. 3B.
11.
2. 11.
7.
1.
6.
4.
1. Tympanic membrane 2. TM defect 3A. Fascia graft 3B. Canal skin 4. Eustachian tube 5. Packing 6. Silastic 7. Bone chip facial recess 8. Bone pâté in mastoid cavity 9. PCW bone 10. Palva flap 11. Gelfoam pledgets 12. Gauze
FIGURE 14-9. Horizontal cross section showing Cottle speculum being used to replace posterior canal wall skin onto reconstructed posterior bony canal wall. Iodoform strip gauze is used to pack the external auditory canal and fix the posterior bony canal wall in place during the healing process. PCW, posterior canal wall; TM, tympanic membrane.
POTENTIAL PITFALLS Complications of canal wall up procedures, such as tympanic membrane perforation and extrusion of ossiculoplasty prostheses, can occur with the canal wall reconstruction technique, but their incidence can be
minimized by meticulous attention to the placement of the fascial underlay graft (see Figs. 14-5B and 14-9), and the coverage of the titanium ossiculoplasty prosthesis with a cartilage graft. A greater concern with the canal wall reconstruction procedure is the possibility of wound infection,
Chapter 14 • Canal Wall Reconstruction Tympanomastoidectomy
which theoretically could progress to involve the mastoid bone pâté and posterior canal wall. Careful attention to numerous details in the canal wall reconstruction technique can effectively reduce the incidence of infection.16 First, perioperative antibiotics, with double coverage of Pseudomonas aeruginosa, are an important measure in preventing wound infections. Typically, patients are given piperacillin-tazobactam and levofloxacin intravenously before surgery and for 48 hours postoperatively. Patients are discharged on a 2-week course of oral levofloxacin. Pediatric patients are given piperacillin-tazobactam perioperatively and discharged on amoxicillin/clavulanic acid for 2 weeks. Penicillinallergic patients are given clindamycin and piperacillin-tazobactam perioperatively and complete a 2-week course of clindamycin postoperatively. Second, bone pâté is collected from nondiseased cortical bone and is soaked in bacitracin solution before placement in the mastoid. Pâté collection is stopped if air cells become exposed during the drilling process. Third, placement of a Penrose drain between the postauricular skin flap and the Palva flap for 48 hours postoperatively helps to prevent the formation of a hematoma, which may subsequently become infected. Re-retraction of the tympanic membrane can occur with any intact canal wall procedure (canal wall up or canal wall reconstruction), and can lead to recidivistic cholesteatoma. In the canal wall reconstruction procedure, the placement of large bone chips in the attic and facial recess prevents potential re-retraction. It is important that a large bone chip be carefully placed in a near-axial orientation into the attic, extending from the anteriormost part of the epitympanum to the posterior aspect of the lateral semicircular canal (see Fig. 14-6) and from the tympanic facial nerve well lateral to the medialmost extent of the posterior bony canal wall, which may have been eroded by cholesteatoma (see Fig. 14-7). The bone chip in the facial recess is placed in a near-coronal orientation and completely blocks this space, effectively separating the mastoid from the tympanic cavity (see Fig. 14-7). As with any mastoid obliteration procedure, complete removal of the cholesteatoma and all diseased mastoid air cell tracts is crucial in the canal wall reconstruction technique. Wide saucerization, thorough removal of the retrofacial cells, and careful sharpening of the sinodural angle are crucial steps in this process. Additionally, the cog is removed, providing wide exposure of the anterior epitympanum to allow for complete eradication of cholesteatoma.
HEARING RESULTS The primary goal of the canal wall reconstruction tympanomastoidectomy is the creation of a dry, safe ear. Hearing reconstruction is a secondary, although important, consideration. Ongoing eustachian tube dysfunction can
181
predispose patients with chronic otitis media to ossiculoplasty prosthesis extrusion and secretory otitis media, both of which may contribute to conductive hearing loss. When secretory otitis media occurs in patients who have undergone canal wall reconstruction, the authors do not offer patients ventilation tube placement because of the high risk of chronic otorrhea. Instead, the authors tolerate the conductive hearing loss that may result from secretory otitis media and prefer hearing aid placement for rehabilitation, if needed.
SUMMARY Canal wall reconstruction tympanomastoidectomy combines components of canal wall up and canal wall down techniques to allow wide visualization for removal of primary and recurrent cholesteatomas with optimal preservation of near-normal EAC and middle ear anatomy. Obliteration of the mastoid with blockage of the attic and facial recess in canal wall reconstruction tympanomastoidectomy helps to prevent recidivistic disease by removing nitrogen-absorbing mucosa and physically obstructing tympanic membrane re-retraction. Careful attention to detail in removing all mastoid and epitympanic air cells, reconstruction of the posterior canal wall and tympanic membrane, and obliteration of the mastoid and attic are crucial for success with the canal wall reconstruction technique.
REFERENCES 1. Ucar C : Canal wall reconstruction and mastoid obliteration with composite multi-fractured osteoperiosteal flap. Eur Arch Otorhinolaryngol 263:1082-1086, 2006. 2. Sudhoff H, Brors D, Al-Lawati A, et al: Posterior canal wall reconstruction with a composite cartilage titanium mesh graft in canal wall down tympanoplasty and revision surgery for radical cavities. J Laryngol Otol 120:832-836, 2006. 3. Reck R , Helms J: The bioactive glass ceramic Ceravital in ear surgery: Five years’ experience. Am J Otol 6:280-283, 1985. 4. Bagot d’Arc M, Daculsi G, Emam N: Biphasic ceramics and fibrin sealant for bone reconstruction in ear surgery. Ann Otol Rhinol Laryngol 113:711-720, 2004. 5. Leatherman B D, Dornhoffer J L : The use of demineralized bone matrix for mastoid cavity obliteration. Otol Neurotol 25:22-25, 2004. 6. Estrem S A, Highfill G: Hydroxyapatite canal wall reconstruction/mastoid obliteration. Otolaryngol Head Neck Surg 120:345-349, 1999. 7. Shea JJ Jr, Malenbaum BT, Moretz WH Jr: Reconstruction of the posterior canal wall with Proplast. Otolaryngol Head Neck Surg 92:329-333, 1984. 8. Roberson J B Jr, Mason TP, Stidham K R : Mastoid obliteration: Autogenous cranial bone pate reconstruction. Otol Neurotol 24:132-140, 2003.
182
OTOLOGIC SURGERY
9. Zini C, Quaranta N, Piazza F: Posterior canal wall reconstruction with titanium micro-mesh and bone pate. Laryngoscope 112:753-756, 2002. 10. Moffat DA, Gray RF, Irving RM: Mastoid obliteration using bone pate. Clin Otolaryngol Allied Sci 19:149-157, 1994. 11. Bacciu S, Pasanisi E, Vega Feijoo S, et al: [Total reconstruction of the posterior wall of the external ear canal with allogenic rib cartilage: Technique and results]. Acta Otorrinolaringol Esp 48:599-604, 1997. 12. Geyer G, Dazert S, Helms J: Performance of ionomeric cement (Ionocem) in the reconstruction of the posterior meatal wall after curative middle-ear surgery. J Laryngol Otol 111:1130-1136, 1997. 13. Lenis A: Hydroxylapatite canal wall reconstruction in patients with otologic dilemmas. Am J Otol 11:411-414, 1990. 14. Grote JJ, van Blitterswijk CA: Reconstruction of the posterior auditory canal wall with a hydroxyapatite prosthesis. Ann Otol Rhinol Laryngol Suppl 123:6-9, 1986. 15. Mercke U: The cholesteatomatous ear one year after surgery with obliteration technique. Am J Otol 8:534-536, 1987.
16. Gantz B J, Wilkinson E P, Hansen M R : Canal wall reconstruction tympanomastoidectomy with mastoid obliteration. Laryngoscope 115:1734-1740, 2005. 17. Palva T: Surgical treatment of chronic middle ear disease, II: Canal wall up and canal wall down procedures. Acta Otolaryngol 104:487-494, 1987. 18. Abramson M : Open or closed tympanomastoidectomy for cholesteatoma in children. Am J Otol 6:167-169, 1985. 19. Shohet J A, de Jong A L : The management of pediatric cholesteatoma. Otolaryngol Clin North Am 35:841-851, 2002. 20. Rosenfeld R M, Moura R L , Bluestone C D: Predictors of residual-recurrent cholesteatoma in children. Arch Otolaryngol Head Neck Surg 118:384-391, 1992. 21. Ars B, Wuyts F, Van de Heyning P, et al: Histomorphometric study of the normal middle ear mucosa: Preliminary results supporting the gas-exchange function in the postero-superior part of the middle ear cleft. Acta Otolaryngol 117:704-707, 1997.
15
Surgery of Acute Infections and Their Complications J. Gail Neely
DEFINITION AND CLINICAL SIGNIFICANCE Complications of acute or chronic suppurative ear disease manifest acutely and are medical and surgical emergencies. These complications are defined as a spread of infection beyond the confines of the pneumatized spaces of the temporal bone and the attendant mucosa. Complications are classified into two groups: aural (intratemporal) and intracranial. Aural complications include (1) mastoiditis, (2) petrositis, (3) labyrinthitis, and (4) facial paralysis. Intracranial complications include (1) extradural abscess or granulation tissue, (2) dural venous sinus thrombophlebitis, (3) brain abscess, (4) otitic hydrocephalus, (5) subdural abscess, and (6) meningitis (Fig. 15-1).1 Because of the significant reduction in absolute numbers of complications, individual clinicians do not have extensive experience in treating patients with complications of suppurative ear disease. This limited experience contributes to decreased familiarity and recognition of otogenic complications.2,3 The combination of lack of awareness and masking of early signs and symptoms leads to delay in diagnosis and subsequent treatment. Physician delay has been noted to be the most significant factor in late diagnosis and treatment of otogenic complications. Delay in diagnosis and treatment of complications of suppurative ear disease is associated with worsening morbidity and mortality.2,4
ETIOLOGY AND PATHOGENESIS The organisms responsible for otogenic complications in the acute setting are Streptococcus pneumoniae, Streptococcus pyogenes, Staphylococcus aureus, and Haemophilus influenzae.5 Organisms that cause complications in chronic otitis media are frequently gram-negative or anaerobic, or both; the presence of anaerobes is significantly associated with complications.6-9 Pseudomonas aeruginosa, Enterobacteriaceae, S. aureus, and anaerobic bacteria are most common in chronic mastoiditis; anaerobes tend
to predominate in chronic mastoiditis when techniques allow their discovery.5 In patients with infratemporal complications, isolates from cultures of middle ear effusions, otorrhea, and mastoid have shown S. pneumoniae to be the most common organism,10 followed by P. aeruginosa, S. pyogenes,11 S. aureus, and H. influenzae.12,13 P. aeruginosa was the most common organism found in a review of 134 patients with acute mastoiditis.14 S. pneumoniae15 and H. influenzae type B are the most common causes of bacterial meningitis.16 Proteus mirabilis, P. aeruginosa, and staphylococcal organisms are common pathogens isolated in patients with intracranial complications.17 Gram-negative isolates have been major organisms in other series.18,19 Polymicrobial cultures are common in brain abscess.8 These studies are important for a general reference; however, organisms change over time, and culturespecific treatment is important, as illustrated by the fact that a more recent study determined that methicillinresistant S. aureus was the second most common organism in chronic suppurative otitis media.20 Obstruction of the aditus ad antrum, congenitally preformed pathways through the oval or round window; obstruction of acquired pathways from fractures or chronic erosive infection, granulation tissue, or cholesteatoma, especially virulent organisms such as H. influenzae type B; and synergistic pathogenicity resulting from anaerobic organism microenvironmental changes all may play a role in the pathogenesis of complications (Fig. 15-2).1,21 Clinically, several important observations can be made that help alert the physician to the possible occurrence of a complication and may reflect some of the pathobiology. Signs and symptoms of possible impending complications are (1) persistent acute infection for 2 weeks; (2) recurrent symptoms of infection within 2 weeks; (3) acute, fetid exacerbation of chronic infection; (4) fetid discharge during treatment; (5) H. influenzae type B or anaerobes cultured from the ear; and (6) fever in the presence of a chronically perforated tympanic membrane, with or without cholesteatoma. 183
184
OTOLOGIC SURGERY
INTRACRANIAL COMPLICATIONS
(Subdural abscess) Brain abscess Brain indented and irritated by abscess
(4) Thrombophlebitis of cerebral vessel
Abscess (1) Mastoid granulations Arachnoid
(2) Extradural granulations
Bone
(3) Mural thrombus in sigmoid sinus
Monometer Elevated CSF
Otitic hydrocephalus
Middle ear Meningitis (hematogenously disseminated)
(CSF) Heart
FIGURE 15-1. Intracranial complications. The predictable pattern of associated complications is numbered in order of progression with the ultimate outcome being brain abscess or otitic hydrocephalus or both. CSF, cerebrospinal fluid.
Chapter 15 • Surgery of Acute Infections and Their Complications
185
Granulation from superior middle ear to aditus, then mastoid antrum
FIGURE 15-2. Obstruction of the aditus ad antrum, key to the development of complications.
CLINICAL PRESENTATION When the possibility exists that the patient has a complication from suppurative ear disease, the complete list of the 10 complications and the fact that more than one complication is likely can be overwhelming. The complications tend to manifest in obvious or predictable clinical patterns, however.
Obvious Complications The four most obvious complications are facial paralysis, labyrinthitis, meningitis, and mastoiditis.
Facial Paralysis Facial paralysis is obvious, and if it occurs as a result of acute infection, it is usually the only complication. If facial paralysis occurs as a result of cholesteatoma, a horizontal canal fistula may also exist.
Labyrinthitis Labyrinthitis manifests as ipsilateral sensorineural hearing loss, nystagmus toward the contralateral side, and vertigo. It is classified according to what enters the perilymphatic space: serous labyrinthitis (toxins), suppurative labyrinthitis (bacteria), or chronic labyrinthitis (soft tissue, such as cholesteatoma).22 Suppurative labyrinthitis destroys all the hearing and may rapidly progress to meningitis. Labyrinthitis with some hearing, resulting from acute infection, is usually serous and isolated without other complications. Labyrinthitis with hearing, resulting from a cholesteatoma, may be associated
with a labyrinthine fistula of the horizontal canal and a dehiscence of the fallopian canal, with or without facial paralysis.
Meningitis Meningitis associated with acute otitis media almost always is the result of hematogenous dissemination, and other complications are rare. Meningitis associated with chronic suppurative otitis media is usually the result of a dehiscence in the dura that allows continuity between an extradural abscess and the cerebrospinal fluid.
Mastoiditis Mastoiditis may be obvious if a subperiosteal abscess is present. When the mastoid infection spreads along the vessels laterally through the outer cortex of the mastoid at McEwen’s triangle, notably without cortex erosion, and purulent debris accumulates under the periosteum, a subperiosteal abscess results, and the classic picture of mastoiditis results. Patients with classic manifestations of mastoiditis have mastoid tenderness, with persistent pain and postauricular swelling, with or without fluctuation, edema and sagging of the posterosuperior canal wall skin, and anteroinferior displacement of the ear. The reasons for the direction of pinna displacement are (1) the major mass of mastoid air cells is immediately medial to the concha, and (2) the auricular cartilage of the pinna is anchored tightly to the osseous canal only by the large tragal cartilage anteriorly and inferiorly (this cartilage forms the “cartilaginous pointer” used so often to identify the facial nerve in parotid surgery).
186
OTOLOGIC SURGERY
Less Obvious Complications Classic mastoiditis is easily recognized; however, without such obvious signs, masked mastoiditis (defined as radiographic evidence of bone destruction) may manifest as mild discomfort in the ear, with or without mastoid tenderness. The pathophysiologic and diagnostic key to occult complications is mastoiditis. Without careful medical care, the classic presentation of acute coalescent mastoiditis with subperiosteal abscess predominates and is quickly diagnosed. Masked mastoiditis may occur in patients with seemingly adequate care, however, and is much harder to diagnose promptly, and consequently can be much more devastating and is more often associated with intracranial complications.23,24 Patients with any of the signs and symptoms of impending complication, particularly with a recurrence of deep, although not severe, pain, may have masked mastoiditis. If bone destruction is present, the diagnosis is made. Masked mastoiditis may exist, however, without early evidence of radiographic bone destruction. If pain persists despite seemingly adequate culture-guided antibiotic administration, surgical mastoid exenteration with facial recess approach to the middle ear creating a posterior tympanotomy, deep mastoid cultures with sensitivities for aerobes and anaerobes, and aggressive medical treatment may be indicated. Rarer subperiosteal or soft tissue abscesses can occur in the deep neck (Bezold’s abscess) or zygoma (Luc’s abscess). If purulent debris escapes through eroded bone and along vessels from the medial tip cells (medial to the digastric ridge and muscle) and diploetic bone and enters the neck through the incisura digastrica (the digastric groove on the inferior surface of the temporal bone), a Bezold’s abscess is formed. Purulent debris deep to the fascial planes of the sternocleidomastoid and trapezius muscles is difficult to localize by palpation.25 If pus tracks along the external auditory canal and accumulates under the temporalis muscle, a rare Luc’s abscess is formed.26 Other, less obvious complications follow. Subdural abscesses are extremely rare, usually obvious by coma and focal neurologic signs and easily seen by magnetic resonance imaging (MRI). If there is any doubt that a catastrophic intracranial lesion exists, the chance of a subdural abscess being present is remote. Otitic hydrocephalus characteristically manifests with headache, some degree of lethargy, and severe papilledema. Almost without exception, otitic hydrocephalus is associated with occlusive sigmoid sinus thrombophlebitis and extradural abscess or granulation tissue. Petrositis manifests with retro-orbital pain; however, the patient may not volunteer this symptom. It is crucial to ask about retro-orbital pain to be assured of its absence. Petrositis is rarely, if ever, present without mastoiditis. Intracranial complications are more frequent with petrositis. The full classic triad of Gradenigo (ear infection,
ipsilateral retro-orbital pain, and abducens palsy), often thought to be pathognomonic of petrositis, is rarely present in petrositis unless an extradural intracranial medial petrous abscess is present that compresses the abducens nerve in Dorello’s dural canal. Four silent, but extremely serious complications occur together in predictable patterns: (1) mastoiditis (classic or masked), (2) extradural abscess or granulation tissue, (3) dural venous sinus thrombophlebitis, and (4) brain abscess. Mastoiditis may occur alone, but often results in a silent accumulation of extradural abscess or granulation tissue. The extradural infection may occur, in the case of cholesteatoma, in the middle fossa at the tegmen, but it characteristically occurs along the extraluminal surface of the lateral wall of the sigmoid sinus, creating an often silent, nonoccluding phlebitis of the sinus wall and an induced mural thrombus, sigmoid sinus thrombophlebitis. In every case of suspected or operated mastoiditis, extradural granulation tissue and sigmoid sinus thrombophlebitis should be sought, preoperatively and intraoperatively. Brain abscess occurs as a result of retrograde thrombophlebitis of cerebral or cerebellar veins that are tributaries to the inflamed sigmoid sinus or other adjacent dural sinuses. Brain abscesses have four stages, as follows: 1. Invasion: The initial onset of cerebritis. This stage creates vague symptoms of mild headache, lethargy, and malaise that last several days and then resolve. 2. Localization: The stage of quiescence and latency. This stage is totally silent for weeks. 3. Enlargement: The stage in which most abscesses manifest with seizures or focal neurologic signs. 4. Termination: The stage in which the abscess catastrophically ruptures into the ventricle or subarachnoid space. Brain abscesses are silent; take weeks from onset to be detectable, even with imaging; and can be devastating. It is prudent to look for brain abscesses in cases of mastoiditis initially and again 3 to 4 weeks later (Fig. 15-3).
DIAGNOSIS The most powerful, rapid, efficient, and useful diagnostic tools are the expertly performed history and physical examination. Diagnoses are made from (1) the history, (2) the physical examination, and (3) surgical exploration. Careful surgical observations, through thin bone, of the dura at the tegmen, the sigmoid sinus, and the facial nerve are often crucial for complete diagnosis (Fig. 15-4). Imaging is important to observe brain and subdural abscesses; however, reliance on imaging modalities alone, at the expense of above-mentioned three elements, carefully done, leads to serious errors.
Chapter 15 • Surgery of Acute Infections and Their Complications Extradural granulation and abscess
187
Extradural granulation medial to bone
A
Abscess with connecting vessels
Mastoid granulations lateral to bone (AP Coronal section)
Eroded bone
B
Epidural abscess lateral to sigmoid with granulations compressing sigmoid medially
Inflamed wall of sigmoid sinus
Mural thrombus Granulation
FIGURE 15-3. A and B, Mechanism of development of brain abscess by retrograde thrombophlebitis of cerebral vessels, usually above the tentorium, from inflamed sigmoid and transverse sinus secondary to mastoid and extradural granulations.
In the history, look for the following:
In the physical examination, look for the following:
1. Symptoms suggesting impending complications 2. Symptoms of retro-orbital or deep, boring head pain 3. Symptoms of lethargy, headache, or both, currently and within the past 2 months
1. Signs suggesting impending complications 2. Signs diagnostic of obvious complications 3. Signs of catastrophic neurologic disease 4. Funduscopic examination for papilledema
188
OTOLOGIC SURGERY
FIGURE 15-4. Exploration of dura of the sigmoid sinus, middle fossa, posterior fossa, and sheath of the facial nerve as inspected through thin bone intraoperatively. It is unnecessary to thin all the bone, and unnecessary to transverse the bone to the structure, unless granulations or pus is found.
r easonable look at the intracranial structures.28 If bone is eroded over the sigmoid sinus or at the tegmen, MRI is indicated to get a better appreciation of possible extradural, intradural, subdural, and intracerebral granulation, edema, or abscess.8 MRI is more sensitive than CT in detecting early cerebritis and cerebral edema.6 MRI may be useful in imaging lesions at the petrous apex in Gradenigo’s syndrome.29 MRI is the most sensitive diagnostic tool in identifying the site and size of epidural, subdural, and brain abscess, and is more sensitive than CT in detecting extraparenchymal spread to the subarachnoid space or ventricle.30 If mastoiditis and particularly sigmoid sinus phlebitis are surgically proven, repeat MRI is indicated 3 to 4 weeks postoperatively to detect subsequent development of an occult brain abscess. In cases of suspected meningitis or otitic hydrocephalus, a thorough funduscopic examination and CT scan to determine the presence of a simultaneous intracranial abscess should precede lumbar puncture for fear of brain herniation during or after puncture. Typical findings on lumbar puncture for meningitis are high protein and low glucose levels and the presence of microorganisms on Gram stain. In otitic hydrocephalus, typical findings on lumbar puncture are normal protein and glucose levels, negative Gram stain, and elevated opening pressure.
TREATMENT Investigators reporting more recent series have found the following early findings to be associated with impending or established infratemporal or intracranial complications: 1. Pain and fever lasting more than 4 days despite appropriate treatment for acute otitis media10 2. Persisting fever and headache2 3. Radiographic evidence of a lytic lesion, presence of anaerobes, or excessive granulation tissue at surgery associated with chronic suppurative otitis media7 4. Chronic suppurative ear disease with fever, headache, ear pain, or vertigo27 5. Increasing otorrhea, meningeal signs, or impairment of consciousness17 6. Headache with vomiting in patients with chronic otitis media8 The most important error in diagnosis is a poor or nonexpert history with little attention paid to the details of previous symptoms and the associated time courses. Another error is to assume that the organisms cultured from the ear canal drainage accurately reflect the organism causing the complication; aspiration and intraoperative cultures from the complication and the involved tissues of the ear best reflect the responsible pathogens. When a complication is suspected, high-resolution computed tomography (CT) of the temporal bone and associated brain, with and without contrast infusion, is indicated. CT allows a good look at the bone and a
The treatment of all complications involves admission to the hospital, use of antibiotics that are guided by culture of the ear and the complication, and surgical intervention as outlined in this section.
Mastoiditis The treatment of acute coalescent mastoiditis and masked mastoiditis, with or without subperiosteal abscess, is wide myringotomy, complete canal wall up mastoidectomy, and wide facial recess approach from the mastoid into the middle ear, creating a posterior tympanotomy. The aditus ad antrum obstruction cannot be adequately maintained patent without the additional facial recess window. Chronic mastoiditis is a diagnostic dilemma; the primary way to establish the diagnosis is to identify radiographically or intraoperatively granulation tissue– induced erosion of bone, usually over the sigmoid sinus. The treatment of most chronic mastoiditis is the same as for cases of chronic suppurative otitis media, with or without cholesteatoma. Most of these cases improve with removal of disease and reconstruction of the middle ear at the same setting, with canal wall up or canal wall down procedures. In all patients undergoing mastoidectomy for suppurative disease, inspection of the dura of the tegmen, the sigmoid sinus, and the facial nerve through thin bone is important. Otherwise, granulation tissue in the middle
Chapter 15 • Surgery of Acute Infections and Their Complications
189
infection, is a middle fossa approach for the total exenteration of disease in the petrous apex and in the peri labyrinthine cells. This dissection is combined with the previous lateral approach, and all pneumatized spaces in the temporal bone are removed, sparing the vital structures of the inner ear, facial nerve, carotid artery, jugular bulb, and sigmoid and the contents of the internal auditory canal.33 The cavity is left open (Fig. 15-6). Malleus incus
TM remnant
FIGURE 15-5. Neely’s modification of a radical mastoidectomy in which the drum and the posterior and superior canal wall are removed, but the ossicles are left intact. Tympanoplasty may be performed later. TM, tympanic membrane.
and posterior fossae and along the facial nerve may go unrecognized and untreated.
Petrositis Sometimes the granulation tissue is severe in the middle ear and mastoid, and symptoms suggest petrositis. In such cases, total resection of the tympanic membrane with careful maintenance of the ossicles in situ, complete mastoidectomy, resection of the posterior and superior canal wall, and following the five tracts toward the petrous apex—posterosuperior tract (along the sinodural angle and superior semicircular canal), posteromedial tract (along the posterior fossa plate, posterior semicircular canal, and endolymphatic sac), subarcuate tract (through the arch of the superior semicircular canal, Frenckner’s approach), perilabyrinthine tracts (infralabyrinthine tracts inferior to the cochlea and supralabyrinthine tracts anterior to the geniculate ganglion, posterior to the carotid canal, and superior to the cochlea, removing the tegmen and cochleariform processes, Ramadier-Lempert approach), and peritubal tracts (removing the root of the zygoma, exposing and elevating a wide segment of dura to expose the petrous apex, Thornval’s approach)—usually allows adequate drainage with the need for a middle fossa petrosectomy.31,32 Reconstruction of the tympanic membrane can be done later in a healed ear (Fig. 15-5). The treatment of petrositis may require two stages, but the second stage is usually unnecessary. The first stage is a radical mastoidectomy, or the modification of the radical procedure mentioned in the preceding paragraph. The second stage, if the first does not resolve the
Labyrinthitis The surgical treatment of labyrinthitis in acute infection is confined to myringotomy. The treatment of labyrinthitis in chronic otitis media, with or without cholesteatoma, is tympanoplasty and mastoidectomy following the surgeon’s usual preference. If a fistula is identified in the cochlea, it is usually better to leave the cholesteatoma matrix on the fistula, close the ear, and return to remove it when the ear is well healed and sterile. This is a good technique for wide, deep fistulas in the vestibular labyrinth as well. For narrow, shallow fistulas in the semicircular canals, the cholesteatoma matrix may be carefully removed and the fascia placed over the fistula; if the matrix appears to be attached to the membranous labyrinth, dissection should cease and the matrix left to be removed later. An important point to re-emphasize is that one cannot be sure prospectively if the labyrinthitis is toxic (serous) or suppurative; nor can one be comfortable that a serous labyrinthitis may not become suppurative. Suppurative labyrinthitis may be soon associated with meningitis. Thus, hospitalization and intravenous antibiotics need to be continued until the infection has been eradicated.
Facial Paralysis The surgical treatment of facial paralysis from acute infection is myringotomy. If subacute or chronic infections are present, mastoidectomy to eradicate disease and to explore the fallopian canal for invasive granulation tissue is indicated; this exploration can be done by thinning the bone of the canal to allow observation of the contents. Decompression may not be worthwhile. Opening the sheath and exposing the nerve in the face of infection are probably contraindicated. If invasive granulation tissue is found within the osseous canal, the canal should be opened for at least the length of the extent of the granulation. External epineurial sheath granulations may be removed, but no attempt should be made to remove granulations from within the sheath; granulation tissue tends to infiltrate between fibers, and attempts to remove it would result in fiber destruction.
Extradural Granulation Tissue or Abscess The treatment of preoperatively or intraoperatively discovered extradural granulation tissue or abscess is the wide exposure of the abnormal dura. Careful attempts to
190
OTOLOGIC SURGERY
Greater superficial petrosal n. Eustachian tube Geniculate ganglion
I.C.A
E.A.C Incus Facial n. Bill’s bar Arcuate eminence
Geniculate ganglion
A
Tracts to petrous apex
Cochlea Cochlear n. Facial n. S.V.N I.V.N
B
I.C.A Eustachian tube Greater superficial petrosal n.
Arcuate eminence
Cochlea
IAC (covered by bone)
FIGURE 15-6. Two phases of management of petrositis. A, The first phase requires a radical mastoidectomy and following the five cellular tracts to the petrous apex. B, The second phase, if the first fails to resolve the infection, is the middle fossa approach to the complete petrosectomy. EAC, external auditory canal; IAC, internal auditory canal; ICA, internal carotid artery; IVN, inferior vestibular nerve, SVN, superior vestibular nerve.
remove excess granulation tissues bluntly are appropriate; it is unnecessary to remove all granulations if doing so would perforate the dura. Abscesses should be completely drained into the mastoid. If an abscess is encountered, the canal wall may be taken down so that complete drainage is ensured (Fig. 15-7). Good drainage usually can be achieved, however, with the addition of a large facial recess approach or extended facial recess approach taking the chorda tympani nerve.
Dural Venous Thrombophlebitis
Granulation tissue removed
FIGURE 15-7. Management of granulations on the sigmoid sinus thrombophlebitis. A conservative approach is advisable.
Dural venous thrombophlebitis rarely requires additional surgical treatment beyond a complete mastoidectomy and management of the extradural granulations or abscess; opening the sinus is not usually required. Medical therapy includes antibiotic therapy and measures to reduce any intracranial pressure using hyperventilation, therapeutic lumbar puncture, mannitol, and dexamethasone.7 In the rare case in which classic preoperative spiking fever and chills precede the discovery of a completely solidified sinus, careful opening of the sinus to explore for an intraluminal abscess may be indicated. Usually, a fibrotic, nonabscessing mural thrombus is found, for which no further intraluminal work should be done; if an easily identifiable intraluminal abscess is found, it should be drained into the mastoid.9,34,35 If bleeding occurs with sinus opening, it can be controlled with an extraluminal piece of absorbable knitted
Chapter 15 • Surgery of Acute Infections and Their Complications
191
If wall springs back, sinus is patent
B
Open sinus to evacuate abscess
A
Total obstruction
FIGURE 15-8. A-C, Exploration of the sigmoid sinus in search of an intraluminal abscess. This is very unusual and should not be performed without complete capability and familiarity with neurotologic surgery.
Open sinus to evacuate abscess
C
fabric (Surgicel), which is left in place at the conclusion of the procedure; a piece of fascia may be additionally placed lateral to this patch. Ligation of the internal jugular vein or use of anticoagulants and thrombolytics is not usually required; their use is still controversial. Some investigators avoid anticoagulants for fear of creating septic emboli and causing hemorrhagic complications. Continuing sepsis, extension of thrombus, or pulmonary complication with continuous spiking fevers may be an indication for internal jugular vein ligation, however. Anticoagulation is considered in cases of thrombosis spreading to the cavernous sinus.9 If anticoagulants and thrombolytics have a role, it is in cases associated with otitic hydrocephalus (Fig. 15-8). In cases with progressive neurologic deterioration with evolution of thrombus, transvenous, direct intrasinus thrombolytic treatment with urokinase and streptokinase has been successful.36-38
Brain Abscess The treatment of brain abscess is under the guidance of neurosurgery.39 Empirical treatment usually includes penicillin or β-lactam antibiotic, chloramphenicol, metronidazole, and intravenous dexamethasone. Additionally,
REMOVAL OF DEEP BRAIN ABSCESS Drain, then irrigate
Drill
Abscess
FIGURE 15-9. Burr hole over a brain abscess through a separate sterile approach, and the aspiration, with or without irrigation with antibiotics. Aspiration, irrigation, or resection of brain abscesses may not be required.
192
OTOLOGIC SURGERY
FIGURE 15-10. A, Severe Mondini malformation with cerebrospinal fluid (CSF) entering the vestibule through a dehiscence in the lateral wall of the internal auditory canal (IAC), and then entering the middle ear through dehiscences in congenitally malformed stapes footplate or round window. B, Treatment is fascia obliteration of the vestibule, taking care not to compress the facial nerve; a piece of cartilage is wedged through the oval window to create a self-retaining seal.
increases in intracranial pressure may require mannitol, hyperventilation, or dexamethasone preoperatively.6 In the past, the goal of traditional treatment was to remove surgically the intact, unruptured abscess capsule. Burr hole aspiration and stereotactic drainage is useful, however. Some authors advocate delay of mastoid surgery until after neurologic symptoms are stabilized.8,17 In a study of 36 patients with otogenic intracranial
abscess, Kurien and associates19 found that concurrent craniotomy and mastoidectomy avoids reinfection while the patient is waiting for definitive surgery; removes the source of infection at the same time complications are being treated; and results in a single, shorter hospital stay. CT-guided needle aspiration may be useful for aspiration for culture through a sterile field.6 Aspiration for
Chapter 15 • Surgery of Acute Infections and Their Complications
culture through a sterile field distant to the ear with or without intra-abscess instillation of antibiotics is not always done; systemic antibiotics may be the only treatment required to resolve the brain abscess, particularly in patients with multiple brain abscesses (Fig. 15-9). Except for aspiration, these abscesses are usually not openly drained or resected, unless they fail to resolve. Repeated aspirations are preferred to complete excision in cases with multiple abscesses, in cases with abscesses in a deep or dominant location, in cases with concomitant meningitis, in cases with early response to antibiotics, and in cases with abscesses less than 3 cm.6 Mastoidectomy can be done before or under the same anesthetic and after neurosurgical drainage of the abscess. The brain abscess rarely is in direct continuity with the ear; if it is, and it spontaneously drains into the mastoid, it can be drained through the ear (see Fig. 15-9).
Otitic Hydrocephalus The surgical treatment of otitic hydrocephalus is the management of mastoiditis, extradural granulations, and sigmoid sinus thrombophlebitis.34 An ophthalmologist should follow the patient’s vision. Long-term care of intracranial hypertension by a neurologist may be necessary. Most cases of otitic hydrocephalus resolve spontaneously over months, and neurosurgical intervention (lumboperitoneal shunt) should be reserved for patients with deterioration of vision or disabling pulsatile tinnitus. Many cases can be effectively treated with acetazolamide and furosemide or systemic steroids. Techniques of management of intracranial thrombophlebitis and chronic intracranial hypertension are controversial and evolving; optimal treatment is provided by a local team of expert consultants involved with the case. Ventricular shunting and optic nerve decompression may be required.40 The main point of treatment is careful management of the ear and chronic intracranial hypertension. Blindness and brain herniation are serious complications of this disease.41
Subdural Abscess The treatment of subdural abscess occurs under the guidance of neurosurgery. Medical treatment includes parenteral antibiotics, anticonvulsants, and corticosteroids. Surgical control of the mastoiditis, extradural granulations or abscess, and sigmoid sinus thrombophlebitis is crucial for recovery. Usually myringotomy for acute otitis media and complete mastoidectomy for coalescent mastoiditis or nonresolving otitis media are involved, and diagnostic subdural tap omission here. If the subdural space contains purulent material, the empyema must be removed by craniotomy or burr holes, and the subdural space must be irrigated.6 Craniotomy with abscess excision is sometimes the neurosurgical treatment of choice.9 Depending on the stability of the patient, the two surgical procedures (neurosurgical drainage with
193
astoidectomy) may be done together19 or sequentially. m Subdural abscesses are extremely rare from suppurative ear disease, tend to be quite severe, and may have direct connections through the infected dura. Experience with these lesions has led this author and neurosurgical colleagues to favor an aggressive approach.
Meningitis The treatment of meningitis from acute ear infection is predominantly medical except for myringotomy. With aggressive use of appropriate intravenous antibiotics, dexamethasone has been shown to reduce the incidence of neurologic sequelae, including deafness, and should be considered.9 If severe Mondini’s deformity with dehiscence of the medial wall of the vestibule is present on the infected side, intralabyrinthine obliteration is required after recovery of the meningitis (Fig. 15-10).21 Meningitis from chronic ear infection is a surgical and medical emergency because of the probability of a dehiscence in the dura and an ever-increasing flow of pus directly from the ear into the subarachnoid space. Radical mastoidectomy with exploration of all dural surfaces directly or through thin bone is necessary. When the extradural abscess is evacuated, care should be taken to identify the dural defect. When the defect is found, it may be repaired with fascia placed intradurally and extradurally. Wedging or suturing the graft is usually possible. Postoperative cerebrospinal fluid leak may occur, but tends to resolve as the meningitis and excessive production of cerebrospinal fluid resolve.
SUMMARY The most powerful tools of diagnosis are an expertly performed, detailed, and time-specific history and physical examination. Imaging is the only way to detect brain abscess and subdural abscess. A complete diagnosis may ultimately depend on carefully done intraoperative observations. Complications of suppurative ear disease manifest in obvious or predictable patterns that help guide preoperative diagnosis and intraoperative discovery and treatment. Treatment of complications requires hospitalization, culture-specific intravenous antibiotics, expedient surgical exenteration of the ear disease, and a specific tailored approach to the complication.
REFERENCES 1. Neely J, Arts H: Intratemporal and intracranial complications of otitis media. In Bailey B, Johnson J, Newlands S (eds): Head and Neck Surgery–Otolaryngology. 4th ed. Philadelphia, Lippincott Williams & Wilkins, 2006, pp 2041-2056. 2. Albers F: Complications of otitis media: The importance of early recognition. Am J Otol 20:9-12, 1999.
194
OTOLOGIC SURGERY
3. Friedberg J, Gordon D: Acute otitis media: The evolution of surgical management. J Otolaryngol 27:2-8, 1998. 4. Wang N, Burg J: Mastoiditis: A case-based review. Pediatr Emerg Care 14:290-292, 1998. 5. Brook I : The role of anaerobic bacteria in acute and chronic mastoiditis. Anaerobe 11:252-257, 2005. 6. Brook I : Brain abscess in children: Microbiology and management. J Child Neurol 10:283-288, 1995. 7. Panda N, Sreedharan S, Mann S, Sharma S : Prognostic factors in complicated and uncomplicated chronic otitis media. Am J Otolaryngol 17:391-396, 1995. 8. Yen P, Chan S, Huang T: Brain abscess: With special reference to otolaryngologic sources of infection. Otolaryngol Head Neck Surg 113:15-22, 1995. 9. Youngs R : Complications of suppurative otitis media. In Ludman H, Wright T (eds): Diseases of the Ear. London, Arnold, 1998, pp 398-415. 10. Harley E, Sdralis T, Berkowitz R : Acute mastoiditis in children: A 12-year retrospective study. Otolaryngol Head Neck Surg 116:26-30, 1997. 11. Goldstein N, Casselbrant M, Bluestone C, Kurs-Lasky M : Intratemporal complications of acute otitis media in infants and children. Otolaryngol Head Neck Surg 119:444-454, 1998. 12. Gliklich R , Eavey R , Iannuzzi R , Camacho A : A contemporary analysis of acute mastoiditis. Arch Otolaryngol Head Neck Surg 122:135-139, 1996. 13. Rosen A, Ophir D, Marshak G: Acute mastoiditis: A review of 69 cases. Ann Otol Rhinol Laryngol 95:222-224, 1986. 14. Khafif A, Halperin D, Hochman I, et al: Acute mastoiditis: A 10-year review. Am J Otolaryngol 19:170-173, 1998. 15. Grigoriadis E, Gold W: Pyogenic brain abscess caused by Streptococcus pneumoniae: Case report and review. Clin Infect Dis 25:1108-1112, 1997. 16. Haddad J: Treatment of acute otitis media and its complications. Otolaryngol Clin North Am 27:431-441, 1994. 17. Kangsanarak J, Navacharoen N, Fooanant S, Ruckphaopunt K : Intracranial complications of suppurative otitis media: 13 years’ experience. Am J Otol 16:104-109, 1995. 18. Hlavin M, Kaminski H, Fenstermaker R , White R : I ntracranial suppuration: A modern decade of postoperative subdural empyema and epidural abscess. Neurosurgery 34:974-981, 1994. 19. Kurien M, Job A, Mathew J, Chandy M : Otogenic intracranial abscess. Arch Otolaryngol Head Neck Surg 124:1353-1356, 1998. 20. Yeo S, Park D, Hong S, et al: Bacteriology of chronic suppurative otitis media—a multicenter study. Acta Otolaryngol 8:1-6, 2007. 21. Neely J: Classification of spontaneous cerebrospinal fluid middle ear effusion: Review of 49 cases. Otolaryngol Head Neck Surg 93:625-634, 1985. 22. Schuknecht H F: Pathology of the Ear. Philadelphia, Lea & Febiger, 1993. 23. Holt R , Gates G: Masked mastoiditis. Laryngoscope 93:1034-1037, 1983.
24. Koitschev A , Simon C, Löwenheim H, et al: Delayed otogenic hydrocephalus after acute otitis media in pediatric patients: The changing presentation of a serious otologic complication. Acta Otolaryngol 125:12301235, 2005. 25. Castillo M, Albernaz V, Mukherji S, et al: Imaging of Bezold’s abscess. AJR Am J Roentgenol 171:1491-1495, 1998. 26. Spiegel J, Lustig L , Lee K, et al: Contemporary presentation and management of a spectrum of mastoid abscesses. Laryngoscope 108:822-828, 1998. 27. Schwaber M, Pensak M, Bartels L : The early signs and symptoms of neurotologic complications of chronic suppurative otitis media. Laryngoscope 99:373-375, 1989. 28. Migirov L : Computed tomographic versus surgical findings in complicated acute otomastoiditis. Ann Otol R hinol Laryngol 112:675-677, 2003. 29. Murakami T, Tsubaki J, Tahara Y, Nagashima T: Gradenigo’s syndrome: CT and MRI findings. Pediatr Radiol 26:684-685, 1996. 30. Nissen A, Bui H : Complications of chronic otitis media. Ear Nose Throat J 75:284-292, 1996. 31. Gulya A J, Schuknecht H F: Anatomy of the Temporal Bone with Surgical Implications. 2nd ed. New York, Parthenon, 1995. 32. Mawson S R : Diseases of the Ear. Baltimore, Williams & Wilkins, 1974. 33. Visosky A, Isaacson B, Oghalai J: Circumferential petrosectomy for petrous apicitis and cranial base osteomyelitis. Otol Neurotol 27:1003-1013, 2006. 34. Commins D, Koay B, Milford C, Renowden S : Otitic hydrocephalus. J Otolaryngol 26:210-212, 1997. 35. Garcia R , Baker A, Cunningham M, Weber A : Lateral sinus thrombosis associated with otitis media and mastoiditis in children. Pediatr Infect Dis J 14: 617-623, 1995. 36. Barnwell S, Higashida R , Halbach V, et al: Direct endovascular thrombolytic therapy for dural sinus thrombosis. Neurosurgery 28:135-142, 1991. 37. Kermode A, Ives F, Taylor B, et al: Progressive dural venous sinus thrombosis treated with local streptokinase infusion. J Neurol Neurosurg Psychiatry 58:107-108, 1995. 38. Tsai F, Higahida R , Matovich V, Alfieri K : Acute thrombosis of the intracranial dural sinus: Direct thrombolytic treatment. AJNR Am J Neuroradiol 13:1137-1142, 1992. 39. Carpenter J, Stapleton S, Holliman R : Retrospective analysis of 49 cases of brain abscess and review of the literature. Eur J Clin Microbiol Infect Dis 26: 1-11, 2007. 40. Horton J, Seiff S, Pitts L : Decompression of the optic nerve sheath for vision-threatening papilledema caused by dural sinus occlusion. Neurosurgery 31:203-211, 1992. 41. Gower D, Baker A, Bell W, Ball M : Contraindications to lumbar puncture as defined by computed cranial tomography. J Neurol Neurosurg Psychiatry 50:1071-1074, 1987.
16
Mastoidectomy—Intact Canal Wall Procedure Charles A. Syms III, Mark J. Syms, and James L. Sheehy Videos corresponding to this chapter are available online at www.expertconsult.com.
Mastoidectomy can be performed in two ways in cases of cholesteatoma. The canal wall down technique is discussed in Chapter 17. This chapter addresses the canal wall up, or intact canal wall, technique. When the senior author would teach residents and fellows, he would emphasize that the call wall is intact, rather than “up.” The canal wall can be taken down surgically or left intact. The intact canal wall procedure is commonly referred to as canal wall up; however, in this chapter, the nomenclature is intact canal wall. In addition to the technique for intact canal wall, this chapter discusses the evolution of the technique, controversies in regard to intact canal wall versus canal wall down techniques, indications for canal wall down procedures, facial nerve in surgery for chronic ear disease, and management of the labyrinthine fistula.
DEFINITIONS The common mastoid operations performed for chronic ear infections are defined in this section. Technical surgical variations peculiar to each surgeon do not alter the fundamental classification. The basic classifications have remained unchanged since 1974.1
Modified Radical Mastoidectomy Modified radical mastoidectomy is performed to eradicate mastoid disease, in which the epitympanum, mastoid antrum, and external auditory canal are converted into a common cavity exteriorized through the external meatus. This technique differs from the radical operation in that the tympanic membrane, or remnants thereof, and ossicular remnants are retained to preserve hearing. (This operation does not involve any reconstructive procedure.) This is an infrequently performed operation in the junior authors’ clinical experience.
Tympanoplasty with Mastoidectomy Tympanoplasty with mastoidectomy is performed to eradicate disease in the middle ear and mastoid and to reconstruct the hearing mechanism, with or without tympanic membrane grafting. There are three variations of this operation; the classic procedure involves permanent exteriorization of the epitympanum and mastoid, a canal wall down procedure. A different approach is to perform a canal wall down procedure and obliterate the cavity or reconstruct the external auditory canal. The third variation is the intact canal wall procedure, the intact canal wall tympanoplasty with mastoidectomy, which is the subject of this chapter.
Radical Mastoidectomy Radical mastoidectomy is performed to eradicate middle ear and mastoid disease in which the mastoid antrum, tympanum, and external auditory canal are converted into a common cavity exteriorized through the external meatus. This operation involves removal of the tympanic membrane and ossicular remnants, with the exception of the stapes, and does not involve any reconstructive or grafting procedure. Frequently, the surgeon places a plug of soft tissue in the tubotympanum or may lay soft tissue over the middle ear to assist in healing, but this does not alter the name of the procedure. This is an uncommon procedure in the junior authors’ clinical experience.
EVOLUTION OF TECHNIQUE Before the mid-1950s, there were two operations for chronic otitis media with cholesteatoma: radical mastoidectomy and modified radical mastoidectomy. These are classic operations that are still indicated, but they are performed infrequently. Their objects are to create a safe ear by exteriorizing the disease and to preserve hearing, if possible. When tympanoplasty was first introduced by Wullstein2 and Zollner,3 exenteration of the mastoid was the rule. Two problems eventually became apparent. Moisture in the cavity had a deleterious effect on the fullthickness skin used to graft the tympanic membrane, and 195
196
OTOLOGIC SURGERY
the narrowed middle ear space created in the classic types III and IV tympanoplasty was prone to collapse, nullifying any hearing improvement (see Chapter 13). It became apparent that if satisfactory hearing results were to be obtained, some method of avoiding a narrow middle ear space would be necessary. Many investigators thought that the best way of solving this problem was by not creating an exteriorized cavity, but by reconstructing the tympanic membrane in a normal position, and then inserting some type of tissue or prosthetic device to re-establish the sound pressure transfer mechanism (see Chapter 13). Although this concept led to better hearing results, many problems developed over the years, some of which are still seen. The physicians at the House Ear Clinic began performing intact canal wall tympanoplasty with mastoidectomy in 1958, under the direction of William House.4 By 1961, more than half of all cholesteatoma cases were managed at the House Ear Clinic with an intact canal wall technique. Many revision operations were required for correction of recurrence of cholesteatoma resulting from retraction pockets. As a result, many physicians at the House Ear Clinic reverted to taking the canal wall down and then obliterating the cavity with muscle, based on a procedure suggested by Rambo.5 In 1963, 50% of cholesteatoma cases were managed this way. By 1964, it was realized that the technique of obliteration did not eliminate the cavity and the problems involved. In addition, the routine use of plastic sheeting through the facial recess in the intact canal wall procedure was reducing the number of cases that had to be revised because of retraction pockets (recurrent cholesteatoma). From that point on, the percentage of cases managed by a canal wall down technique gradually decreased to 10% in 1970. Since then, although there have been fluctuations, on average 15% to 30% of chronic ear surgery is managed by canal wall down procedures.6
CONTROVERSY The controversy over intact canal wall versus canal wall down centers mostly on safety—safety of the operative procedure and safety over the ensuing years.7 The technical ability of the surgeon also should be taken into consideration. In the surgery of aural cholesteatoma, be it intact canal wall or canal wall down, judgment and technical ability are major factors in the outcome.8 Let us assume that the technical ability and judgment are superior in the two groups. Why is there a difference in opinion as to what is best for the patient? Are hearing results a factor? Experienced otologic surgeons do not find much difference in hearing results. Surgeons are very careful not to narrow the middle ear space (see Chapter 13) and stage the operation almost as frequently as in an intact canal wall operation (see Chapter 18).
Is there a difference in the healing? Intact canal wall procedures, with lateral surface grafting (see Chapter 9), may take 6 to 8 weeks to heal. Open cavities frequently require 3 to 4 months and occasionally 6 to 8 months, and there is a small percentage that are never free of minor moisture problems. What about residual and recurrent disease? Some surgeons would argue that a primary canal wall down approach results in one operation, not two. Most experienced surgeons who use intact canal wall and canal wall down procedures find little difference in the incidence of middle ear residual disease, or disease left behind. They also find little difference in the incidence of staging the operation (see Chapter 18). Recurrent cholesteatoma is a different matter. Recurrent cholesteatoma characteristically results from a posterosuperior retraction pocket,9,10 which occurs only in intact canal wall procedures. Surgeons who have reported a 20% to 40% incidence of recurrent cholesteatoma have failed, with rare exceptions, to stage the operation when indicated (75% of the time), and have failed to use plastic sheeting through the facial recess, even when the operation was being performed in one stage. Advocates of the intact canal wall procedure (surgeons who have had extensive experience) have less than a 5% incidence of recurrent cholesteatoma. When a cavity is created, it is usually necessary to clean (remove dead skin) every 6 to 12 months for the rest of the patient’s life. Patients who have undergone the intact canal wall procedure need to be seen by the physician only once every year. Precautions relative to not getting water in the ear are necessary 50% or more of the time in canal wall down cases, depending on whether the cavity is healed, how large it is, whether an adequately sized meatus was created, and whether the cavity is round instead of bean-shaped. Finally, an adequate-sized meatus is relevant. If one creates a meatus large enough to have a trouble-free ear and allow water in the ear, the size can pose a problem with fitting a hearing aid, if and when there is a need for the aid in the future. The problem consists of getting a secure fit and preventing feedback. The behind-the-ear aid usually solves this problem and is the best aid for use in an ear that may have some drainage from a cavity.
INDICATIONS FOR MASTOIDECTOMY Mastoidectomy may be indicated in tympanoplasty s urgery to eliminate disease, to explore the mastoid to ensure that there is no disease, to enlarge the aircontaining middle ear–antral space, or occasionally to create temporary postauricular drainage (with a catheter) in patients with compromised eustachian tube problems or uncontrolled mucosal infection.11 The most common indication is the treatment of cholesteatoma and the associated infection.
Chapter 16 • Mastoidectomy—Intact Canal Wall Procedure
What about physicians who recommend at least a cortical (“simple”) mastoidectomy in all tympanoplasties? The rationale seems to be that it is “good practice,” and that “it’s better to be safe than sorry.” There are also arguments, mentioned earlier, that this practice can increase the middle ear cleft space, and that this is a good idea if there is compromised eustachian tube function. The indication for mastoidectomy is based on the clinical history and the appearance of the ear in the physician’s office. The final decision is made during surgery. Radiographs and imaging studies play little part in making the diagnosis or the decision to perform the surgery.
INDICATIONS FOR EXTERIORIZED MASTOID CAVITY The House Ear Clinic physicians prefer not to create a cavity, but they may do so sometimes. That decision may be made preoperatively, but more often than not, the operation is begun as an intact canal wall procedure, and the decision to exteriorize the mastoid is made intraoperatively.12
Preoperative Decisions The decision to perform a canal wall down procedure is made preoperatively in some cases. This decision is based on the consideration of the hearing in the involved ear, the status of the opposite ear, the preoperative complications, the degree of posterior canal wall destruction by disease, and the age and health of the patient. With rare exceptions, a cholesteatoma requiring mastoid surgery in an only hearing ear is managed with a canal wall down technique. Usually, the procedure is a classic modified radical mastoidectomy, leaving the middle ear and hearing the way they are. A classic modified radical mastoidectomy may be used in cases in which the affected ear has serviceable hearing, and the opposite ear has a severe uncorrectable impairment. One does not wish to jeopardize the only serviceable ear. In labyrinthine fistula cases, one may decide before the operation to use a canal wall down operation if the mastoid is small, or if the opposite ear has a cholesteatoma that would require surgery. If the hearing is serviceable, one would probably perform a classic modified radical procedure, particularly in patients in poor health or in elderly patients.12-14 A canal wall down operation may be decided on preoperatively if it can be seen that the cholesteatoma has destroyed a significant portion of the posterior canal wall. If the opposite ear already has a cavity, one may elect to create a cavity in the other ear at the time or surgery. In elderly patients or patients in poor health, we are more likely to use a classic modified radical mastoidectomy— the less done, the better.
197
Intraoperative Decisions Advocates of the intact canal wall procedure generally start the operation in this manner unless the decision has been made preoperatively. If one encounters a very contracted mastoid, particularly with an ear canal slanting up and forward, or if one encounters an unsuspected canal wall destruction, one would not hesitate to take the canal wall. Intraoperative decisions normally (at House Ear Clinic) account for two thirds of the decisions for canal wall down procedures.
PREOPERATIVE EVALUATION AND TREATMENT How does one make the diagnosis of cholesteatoma? What tests are necessary? How vigorously does one treat a chronically draining ear preoperatively? The diagnosis of cholesteatoma is based on a welltaken history by the physician and a careful examination of the ear under an operating microscope to confirm ingrowth of skin into the middle ear, the epitympanum, or both. The only routine testing is the hearing test. This test is not related to making the diagnosis, but is done to allow proper counseling. Occasionally, the hearing test results in a change of approach to the surgery, as noted in the preceding section. Radiographs and imaging studies play little part in making the diagnosis or directing the surgical approach. These tests are usually obtained if there is a complication, or if one is considered likely (e.g., semicircular canal fistula, facial paralysis, meningitis, or other intracranial complications).14 Under these circumstances, imaging studies rarely make a difference in the overall surgical approach, but should allow the surgeon to predict any complications or sequelae. The patient and family can be counseled properly and forewarned of problems.
Treatment of a Chronically Draining Ear How much, if any, treatment of the draining ear is indicated before surgery? Must the ear be dry before the surgery? If so, for how long? How vigorous should treatment be? Are cultures of the drainage indicated? Experienced otologic surgeons rarely take cultures of draining ears, unless there seems to be a subacute mastoiditis or a suspected complication. A culture may be indicated in such a situation. In an ear without cholesteatoma (the benign central perforation), local treatment is indicated to obtain a dry ear before surgery (see Chapter 8). It is preferable to have the ear dry for 3 or 4 weeks before tympanic membrane grafting. If the ear is draining at the time of surgery, it is probably best to perform at least an antrostomy through the mastoid cortex to ensure drainage while the tympanic membrane graft is healing.11 It is more likely
198
OTOLOGIC SURGERY
that a mucosal problem could dictate staging the operation if the ear is still draining at the time of surgery (see Chapter 18). Many ears with cholesteatoma have only intermittent discharge, if any. These ears usually respond quickly to various local medications. If the cholesteatomatous ear has a history of almost continual drainage, and there are no associated symptoms requiring treatment, it may be best to leave the ear alone. If local treatment is started, it should be effective quickly, if it is going to help at all. The exception to this course is an actively draining ear with polyps; these should be treated with local medication. Resolution of discharge despite the history is common.
PREOPERATIVE COUNSELING Physicians at the House Ear Clinic use a Chronic Ear Patient Discussion Booklet to explain to the patient how a normal ear functions, and how skin grows into the ear and forms a cholesteatoma. Then the procedure is described to the patient, as follows: Cholesteatoma is an ingrowth of skin into the mastoid. This forms a skin-lined cyst that we call a cholesteatoma. It is not a tumor or truly a growth. But it does tend to grow larger as time goes on if the ear continues to drain. There are three reasons why the ear should have surgery at some time. First, it is potentially dangerous. If the drainage continues, about 20% of patients eventually develop severe dizziness because the cholesteatoma breaks into the balance canal. There is about a 1% chance that it may break into the nerve to the face or break into the covering of the brain. Second, the longer the problem goes on, the more damage may be done to the hearing. If the cholesteatoma breaks into a balance canal, the patient might lose all hearing permanently. Third, there is the matter of the drainage. The objectives of surgery are to obtain a safe, dry, and hearing ear. Obtaining a safe, dry, healed ear is almost certain. To obtain a good hearing ear, it is frequently necessary to do the operation in two stages. There is a 60% to 70% chance of helping the hearing with the second operation. There is nothing urgent about having the surgery. It can be done in 1 month or 3 months, but it is inadvisable to put it off indefinitely. It is like sitting on a keg of dynamite: It is not a very good place to sit because you are not quite sure whether the dynamite will ever explode or cause a serious problem. In the back of the Chronic Ear Patient Discussion Booklet, there is a section called Risks and Complications of Surgery. The only serious complication that occurs with any regularity is a total, 100% loss of hearing in the ear operated on. The likelihood of this happening is 1% to 2%; all of the other things listed in the booklet
are either very remote or are temporary. (See Appendix 1 in Chapter 9 for the complete Risk and Complications sheet.) If the patient has a labyrinthine fistula, as 5% to 10% of patients with cholesteatoma at the House Ear Clinic do, the patient is told there is a 10% chance of a total loss of hearing and prolonged dizziness, which would eventually clear up.
PREOPERATIVE PREPARATION OF THE PATIENT Preoperative preparation of the patient differs little from what was described for tympanic membrane grafting (see Chapter 9). The operation is done under general anesthesia, unless the patient requests otherwise. Preoperative antibiotics are not routinely used.
SURGICAL TECHNIQUE Just as the preparation before and in surgery is the same as that described for tympanic membrane grafting (see Chapter 9), the initial steps are also the same: making canal incisions, elevating the vascular strip, turning the ear forward, removing and dehydrating the temporalis fascia, removing canal skin, enlarging the ear canal by removing the overhanging bone anteriorly and inferiorly, and ensuring that the remnant is de-epithelialized.15
Removal of Middle Ear Disease Cholesteatoma should be dissected in continuity to ensure total removal. All diseased tissue is dissected from the bone or mucosa, beginning in the anterosuperior quadrant, and proceeding inferiorly, posteriorly, and superiorly until the superior edge of the promontory, the lower edge of the oval window, is reached. Normal mucosa should not be sacrificed. No attempt should be made at this time to remove a cholesteatoma that surrounds the stapes or is in the oval window. Manipulations in this area should be postponed until the mastoidectomy has been completed, and the facial recess is open. Removal of oval window disease should always be deferred until the end of the procedure, so that if a fistula develops inadvertently, the operation may be terminated expeditiously. Before proceeding with the mastoidectomy, one must determine the status of the incudostapedial joint. If there is an intact chain, the incudostapedial joint should be separated at this time. This facilitates removal of the incus after the facial recess is opened, and prevents possible trauma to the inner ear, which could occur if the drill inadvertently touches the incus when the epitympanum or facial recess is being opened.
Chapter 16 • Mastoidectomy—Intact Canal Wall Procedure
Mastoid Exenteration The mastoid is exenterated, under the microscope, using a drill with various-sized round cutting burrs. Continuous suction-irrigation during drilling is used to cool the bone, to keep the field clean at all times, and to prevent clogging of the burr by bone dust. The initial burr cut is made along the linea temporalis. This marks the lowest point of the middle fossa dura in most cases. The second burr cut is along a line perpendicular to the one just described and tangent to the posterior margin of the ear canal (Fig. 16-1). These two burr cuts outline a triangular area posterior to the ear canal cut, the apex of which is at the spine of Henle. Projected into the mastoid, parallel to the direction of the ear canal, the apex of this triangle is directly over the lateral semicircular canal. The only structure of importance lying within this triangle as one proceeds with the exenteration is the lateral (sigmoid) sinus. The deepest mastoid penetration is always at the apex of this triangle. This ensures that the antrum is entered and the lateral canal identified before deeper penetration in other areas. The dural plate is skeletonized superiorly, and the lateral sinus is skeletonized posteroinferiorly as the dissection proceeds. Uncovering the middle fossa or sigmoid sinus dura is unnecessary, but should not result in any problem. It is not considered a complication. After the lateral semicircular canal has been identified and the cortical mastoidectomy has been completed, the zygomatic root is exenterated to allow access to epitympanum. The posterior bony canal wall is thinned at this time (Fig. 16-2). There are certain “rules of thumb” used in teaching in regard to the initial approach to the mastoid, as follows:
• Always keep the deepest area of penetration into the
mastoid at the apex of the two initial burr cuts. The direction in which one proceeds is not perpendicular to the bone; it should be parallel to the ear canal. Following parallel to the ear canal leads to the mastoid antrum. • Do not dig a “deep dark hole.” One must remember to saucerize the margins, to open the exposure as one proceeds deeper. In this way, it is possible to see what one is doing and to use the suction-irrigation with the drill. • Always use the largest burr possible. If one should inadvertently uncover the middle fossa dura, the facial nerve, or the sigmoid sinus, one is less likely to do serious damage with a large burr as opposed to a very small burr. If there is a problem, one will be able to see what was done. • Be mindful of the “dark side of the burr”—the anterior edge of the cutting burr, which is frequently not the focus of the surgeon’s attention. Many beginning otologic surgeons fenestrate the ear canal or middle cranial fossa plate by not “minding the dark side.”
199
• “If lost on the way to the antrum, go high and for-
ward.” By going high, one identifies the middle fossa dural plate. By going forward, one squeezes into the angle between the dural plate and the ear canal. Going medial in that direction leads to the epitympanum and avoids the matter of fenestrating the lateral semicircular canal or getting into other troubles should the mastoid antrum be filled with bone.
Opening the Facial Recess The facial recess is one of the posterior recesses of the middle ear. It is bordered laterally by the chorda tympani, medially by the upper mastoid segment of the facial nerve, and superiorly by bone of the fossa incudis (Fig. 16-3). This recess is frequently the seat of a cholesteatoma, particularly when the cholesteatoma is associated with a perforation below the posterior malleal fold. Bone in this area may be cellular even in poorly developed mastoid. The landmark for opening into the facial recess is the fossa incudis. One visualizes the triangular area that is inferior to the fossa and is bordered by bone of the fossa incudis superiorly, the upper mastoid segment of the facial nerve medially, and the chorda tympani laterally (Fig. 16-4). The bone is saucerized in this area with a large cutting burr. When the bony canal wall lateral to the recess has been thinned satisfactorily (care being taken not to perforate into the external canal), bone removal is continued with a smaller cutting burr. One should always stroke the burr parallel to the direction of the facial nerve, never allowing the burr to pass the bone of the fossa incudis superiorly (Fig. 16-5). The facial recess is opened from the mastoid to remove disease in the area, to gain additional access to the posterior middle ear (oval and round windows), to gain a better view of the tympanic segment of the facial nerve, and to facilitate postoperative aeration of the mastoid. Opening into the middle ear through the facial recess is a key step in performing an intact canal wall tympanoplasty with mastoidectomy; with rare exceptions, it should not be omitted (Fig. 16-6). There are two things one may notice when approaching the facial nerve in this area. Frequently, bleeding is encountered from one of the vessels intimately associated with, but lying outside, the bony canal of the nerve. Before actually uncovering the nerve, one may note its white sheath showing through the thin bone. Often this sheath is highlighted by one of the vessels on the sheath being visible through the bone. Identification of the facial nerve provides an additional landmark for opening into the facial recess. A small cutting burr is used to enter into the middle ear, just lateral to the facial nerve, and then the opening is enlarged to the extent possible with diamond or roughcut diamond burr. It is usually possible to obtain at least a 2 mm opening.
200
OTOLOGIC SURGERY
Chapter 16 • Mastoidectomy—Intact Canal Wall Procedure
201
FIGURE 16-1. Burr cuts to begin mastoidectomy. Note lateral semicircular canal ghosted in. FIGURE 16-2. Exenteration of the mastoid is completed. FIGURE 16-3���. Horizontal cross section of temporal bone, looking at the upper segment from below. Note facial recess (FR) and tympanic recess (TR).
FIGURE 16-4���. Facial recess area outlined by triangle: borders are the bone of the fossa incudis, upper mastoid segment of the facial nerve, and chorda tympani.
It is unnecessary to expose the nerve with this approach, but there is no harm in doing so. The facial nerve is quite resistant to gentle trauma.16 One can use the cutting burrs when approaching the nerve in the region of the facial recess. Bone removal is quicker than with the diamond stone, and it is easier to determine when the nerve is exposed. To make this differentiation, the exposed area can be probed with a Rosen needle. If the exposed area is nerve sheath, it rebounds immediately after release of the probe; it “bounces back.” If what has been uncovered is mucosa or cholesteatoma, one may note that it will “come back to you,” but that it does not bounce back or rebound. After the recess has been opened, the incus, if present, is removed along with bone of the fossa incudis if involved with cholesteatoma. It is then possible to see the pyramidal eminence, the oval and round windows, and the part of the tympanic portion of the facial nerve lying posterior to the cochleariform process.
Elimination of Disease As disease is encountered in the mastoid, it is removed by dissecting it from a posterior to anterior direction. It is important to remove all cholesteatoma matrix in continuity so that no remnant of epithelium remains. The mastoid is exenterated to the extent indicated by the disease process and to the extent necessary to obtain adequate exposure. It is unnecessary to remove normal-appearing cells, to exenterate all cells, as one would do in a canal wall down procedure. From the mastoid approach, all mastoid and facial recess disease may be removed by elevating it and dissecting it toward the epitympanum and middle ear. Unless there is an unusually narrow angle between the tegmen and the superior wall of the ear canal, it should be possible to remove all epitympanic disease. When the cholesteatoma has contacted the malleus head, as it frequently has, the entire malleus should be removed. Removal of the malleus exposes the opening into the supratubal recess and facilitates dissection of disease from behind and through the ear canal simultaneously. Posterosuperior middle ear disease is removed at this time through the ear canal and mastoid (via the facial recess). The cholesteatoma matrix is dissected in continuity, if possible. The areas that are most difficult to see with this or any other approach, even radical mastoidectomy, are the posterior middle ear recess: infrapyramidal and tympanic.17 These are the areas posterior to and between
the oval and round windows (Fig. 16-7). They extend for a variable distance medial to the pyramidal process and facial nerve (and lateral to the posterior semicircular canal). They are often the seat of cholesteatoma, particularly in cases associated with perforations below the posterior malleal ligament. The area must be cleaned with a right angle dissector. Removal of the pyramidal process and adjacent bone with a diamond burr may be necessary sometimes to facilitate the cleaning, but can be done safely only in a case where the stapes superstructure and tendon are missing. If the tympanic recess is deep, and there is disease in it, it can be approached in a well-developed mastoid from the mastoid side, medial to the facial nerve and lateral to the posterior semicircular canal.19 This approach is not usually feasible, or necessary, but should be kept in mind. It is an approach commonly used in connection with glomus tumor surgery. Infrequently, otologic endoscopes for viewing the posterior recesses from the opposite side of the usual surgical position of the operative ear may be helpful (viewing from the patient’s right side for a left ear or from the left side for the right ear).
Use of Plastic Sheeting Plastic sheeting is used routinely in the intact canal wall procedure, regardless of the status of the middle ear mucosa, to prevent adhesions between the raw undersurface of the tympanic membrane graft and the denuded bone of the epitympanum and facial recess area. (Other uses of plastic sheeting are discussed in Chapter 18.) Before it was realized that plastic sheeting should be used through the facial recess, there were many cases of recurrence of cholesteatoma because of retraction of the tympanic membrane into the facial recess and epitympanum (Fig. 16-8). If the operation is not being staged, thin silicone sheeting is used through the recess. An opening can be created in the plastic to allow for reconstruction to the stapes capitulum (Fig. 16-9). When staging of the operation is indicated, as it usually is in cholesteatoma cases, thick silicone sheeting is used (Fig. 16-10) (see Chapter 18).
Completion of the Operation The tympanic membrane is reconstructed with rehydrated fascia, the ear canal skin is replaced on the denuded bone, packing is inserted, the vascular strip is replaced, and closure postauricularly is with subcutaneous suture,
202
OTOLOGIC SURGERY
FIGURE 16-5. A-C, Horizontal cross section showing progressive saucerization and opening into the facial recess; compare with Figure 11-13. FIGURE 16-6���. Facial recess is open, and incus is removed. FIGURE 16-7���. View through the ear canal showing location of tympanic sinus, medial to the facial nerve.
all as described for tympanic membrane grafting (see Chapter 9). The mastoid is routinely irrigated with antibiotic solution, removing all bone dust and debris, and decreasing the likelihood of postoperative inflammation or infection. A superficial Penrose drain is occasionally
indicated when oozing has been a problem. The drain extends from the area of the fascia removal and exits at the lower part of the incision. This drain is usually attached to the dressing so that it is removed on the first postoperative day.
Chapter 16 • Mastoidectomy—Intact Canal Wall Procedure
Dressing and Postoperative Care There is no difference between the dressing or postoperative care for this procedure and the procedure for tympanic membrane grafting (see Chapter 9). The schedule and plan for office visits are the same. This is now most commonly performed as an outpatient procedure. The authors frequently use a Glasscock pressure dressing for up to 1 week after the surgery to prevent patient telephone calls regarding the “outstanding ear.”
FACIAL NERVE IN SURGERY OF CHRONIC OTITIS MEDIA One of the greatest fears of the inexperienced surgeon is that damage to the facial nerve may occur during a mastoidectomy.16 In the words of the senior author: “The facial nerve is exactly where it is supposed to be.” Fear may result in avoidance of the nerve rather than positive identification; this can result in inadvertent damage to the nerve that has not been identified. In canal wall down surgery, this fear results in the inadequacy of some of the surgical procedures, leaving the posterior bony canal wall (the facial ridge) high, creating a bean-shaped cavity. Familiarity with the facial nerve results in respect: respect for helping to guide one throughout the temporal bone and respect for its ability to withstand manipulations and minor trauma. There are three segments to the facial nerve: labyrinthine, tympanic, and mastoid. In surgery of chronic otitis media, concern is with the tympanic and mastoid segments.
Tympanic Segment The tympanic segment is the portion of the nerve extending from the geniculate ganglion to the second genu (adjacent to the pyramidal process). Landmarks for identification of the nerve in its tympanic course are the cochleariform process, the oval window, and the pyramidal process. From the mastoid approach, the lateral semicircular canal and the cog are useful. These are discussed later. The upper edge of the oval window is bordered by the facial nerve (Fig. 16-11). If this is not apparent, the cochleariform process may be identified. The facial nerve lies posterior and superior to the cochleariform. One may also identify the pyramidal process, and note that the facial nerve lies above and behind this structure. When none of these landmarks are apparent, the semicanal for the tensor tympani may be identified in the anterior middle ear and followed posteriorly. Its inferior border is continuous with the upper margin of the oval window, the facial nerve. Rarely, one may need to identify the vertical groove on the promontory for the tympanic nerve. This groove is followed superiorly to the cochleariform process or its
203
remnants. The facial nerve is posterior and superior to the cochleariform process. From the mastoid approach, there are two landmarks to the tympanic course of the facial nerve: the lateral semicircular canal and the cog. The posterior half of the tympanic segment is immediately inferior to the lateral canal. The relationship between the lateral canal and facial nerve is among the most consistent landmarks for the facial nerve. The nerve course is anteriorly, passing superior to the cochleariform process and anterior to the cog. The cog is a ridge of bone that extends inferiorly from the tegmen epitympani and partially separates the anterior tympanic compartment (supratubal recess) from the mesoepitympanum. The cog lies immediately superior to, and just slightly posterior to, the cochleariform process and anterior to the head of the malleus (Fig. 16-12). As the facial nerve runs between the cochleariform process and the geniculate ganglion, it courses under the base of the cog and anterior to it in the floor of the supratubal recess.
Mastoid Segment In mastoidectomy, the initial landmarks within the temporal bone are the mastoid antrum and the lateral semicircular canal. When the lateral canal is identified, the surgeon knows where the facial nerve is and is prepared to remove all the diseased tissue from the mastoid (Fig. 16-13). An intimate knowledge of this threedimensional anatomy and the relationship of surrounding structures to the facial nerve is the major reason a trainee in otology must spend sufficient time dissecting the temporal bone in the laboratory before operating on patients. The short crus of the incus is located inferior, and slightly lateral, to the anterior portion of the lateral canal bulge. The fossa incudis is at the tip of the short crus. The facial nerve lies medial to the fossa incudis and inferior to the lateral canal. As the nerve travels inferiorly in its course to the stylomastoid foramen, it travels in a slightly posterior direction, in most instances, and travels laterally. (The reader would be well served by reading the article by Litton and associates on the relationship of the fallopian canal to the tympanic annulus.18) The senior author routinely referred otology/neurotology fellows to this classic reference and supplied them for copies for review on multiple occasions. Two further landmarks to the mastoid segment of the facial nerve are the digastric groove and the posterior semicircular canal. Neither are usually involved (in facial nerve identification) in surgery of chronic otitis media because of the lack of cellular development in most cases of this type. The digastric groove leads to the stylomastoid foramen. The inferior portion of the posterior semicircular canal travels medial to the facial nerve.
204
OTOLOGIC SURGERY
Chapter 16 • Mastoidectomy—Intact Canal Wall Procedure
205
FIGURE 16-8. Retraction of new tympanic membrane into the facial recess, caused by scar tissue; plastic sheeting was not used. Arrows indicate development of recurrent cholesteatoma. FIGURE 16-9���. Thin plastic through the facial recess; capitulum is protruding through the opening to allow reconstruction. FIGURE 16-10���. Thick plastic sheeting extending from the mastoid into the middle ear through the facial recess. FIGURE 16-11���. Parasagittal section through the middle ear of the right temporal bone to show landmarks. Facial nerve lies superior to the oval window and posterior (also superior) to the cochleariform process. The nerve lies superior, and also posterior, to the pyramidal process, and anterior to the cog in the floor of the supratubal recess. The cog is a ridge of bone extending inferiorly from the tegmen epitympani, above the cochleariform process (see Fig. 16-12). Note the relationship of the semicanal for the tensor tympani and a groove in the promontory for the tympanic nerve of Jacobson. FIGURE 16-12���. View from the mastoid into the epitympanum (facial recess has been opened). Malleus head has been ghosted in to show the relationship to the cog (compare with Fig. 16-11). The cog is a ridge of bone extending inferiorly from the tegmen epitympani, anterior to the malleus head, above the cochleariform process. FIGURE 16-13���. Facial nerve (labyrinthine, tympanic, and mastoid segments) has been ghosted in to show the relationship to the ossicles, facial recess opening, and lateral semicircular canal.
MANAGEMENT OF THE LABYRINTHINE FISTULA There is no way of knowing for certain preoperatively that a patient with a cholesteatoma does not have a labyrinthine fistula; 10% of such patients do.13 Each mastoid must be approached as if a fistula existed. When the cholesteatomal sac is encountered in the mastoid, it should be opened, and the medial wall of the sac lying on the lateral semicircular canal should be palpated to detect any bony dehiscence. Bony erosion
FIGURE 16-14. Defect in lateral epitympanic wall. FIGURE 16-15���. Repair of lateral wall defect with cartilage.
may be obvious by flattening of the usual prominence of the lateral canal. If there appears to be a fistula, a decision has to be made regarding management of the mastoid: continue as an intact canal wall procedure, or create an open cavity? If an open cavity is to be created, the matrix may be left on the fistula permanently.12 If the operation is to continue as an intact canal wall procedure, the matrix around the fistula is carefully incised. The rest of the matrix may be removed without removing the segment covering the fistula. If a decision
206
OTOLOGIC SURGERY
has been made that the operation is to be performed in two stages (see Chapter 18), the matrix is left over the fistula, to be removed at the planned second stage when the ear is healed. If the operation is to be performed in one stage, the operation is completed, including grafting, ossicular chain reconstruction, and packing of the ear canal. If one suspects that the fistula is small, it is reasonable to remove the matrix at this time and cover the fistula immediately with fascia. If the fistula appears to be large, or the ear is infected, it is probably best to leave the matrix and come back into the mastoid in 6 months to remove it.
hole on the canal side before replacement of the ear canal skin.
REPAIR OF CANAL WALL DEFECTS
1. Committee on Conservation of Hearing of the American Academy of Ophthalmology and Otolaryngology: Standard classification for surgery of chronic ear infection. Arch Otolaryngol Head Neck Surg 81:204-205, 1965. 2. Wullstein H : Theory and practice of tympanoplasty. L aryngoscope 66:1076-1093, 1956. 3. Zollner F: Principles of plastic surgery of the soundconducting apparatus. J Laryngol Otol 69:637-652, 1955. 4. Sheehy JL, Patterson ME: Intact canal wall tympanoplasty with mastoidectomy. Laryngoscope 77:1502-1542, 1967. 5. Rambo J HT: Further experience with musculoplasty. Arch Otolaryngol Head Neck Surg 71:428-436, 1960. 6. Syms M J, Lexford WM : Management of cholesteatoma: Status of a canal wall. Laryngoscopy 113:443-448, 2003. 7. Sheehy J L : Intact canal wall tympanoplasty with mastoidectomy. In Snow J B (ed): Controversies in Otolaryngology. Philadelphia, Saunders, 1980. 8. Sheehy J L , Brackmann D E : Surgery of Chronic Ear Disease: What We Do and Why We Do It. In Instructional Courses, vol. 6, St Louis, Mosby, 1993. 9. Sheehy J L : Cholesteatoma surgery: Residual and recurrent disease. Ann Otol Rhinol Laryngol 86:451-462, 1977. 10. Sheehy J L , Robinson JV: Cholesteatoma surgery at the Otologic Medical Group: Residual and recurrent disease: A report on 307 revision operations. Am J Otol 3:209215, 1982. 11. Sheehy J L : Chronic tympanomastoiditis. In Gates G A (ed): Current Therapy in Otolaryngology–Head and Neck Surgery, vol. 4, Philadelphia, Decker, 1990, pp 19-22. 12. Sheehy J L : Cholesteatoma surgery: Canal wall down procedures. Ann Otol Rhinol Laryngol 97:30-35, 1988. 13. Sheehy J L , Brackmann D E : Cholesteatoma surgery: Management of the labyrinthine fistula. Laryngoscope 89:78-87, 1979. 14. Sheehy J L , Brackmann D E, Graham M D: Complications of cholesteatoma: A report on 1024 cases. In McCabe B, Sade J, Abramson M (eds): Cholesteatoma, First International Conference. Birmingham, AL, Aesculapius, 1977. 15. Sheehy J L , Brackmann D E : Surgery of chronic otitis media. In English G M (ed): Otolaryngology. Philadelphia, Lippincott, 1994. 16. Sheehy J L : Facial nerve in surgery of chronic otitis media. Otolaryngol Clin North Am 7:493-503, 1974. 17. Donaldson J A, Anson B J, Warpeha R L , et al: The surgical anatomy of the sinus tympani. Arch Otolaryngol Head Neck Surg 91:219-227, 1970.
Defects in the posterior or superior bony canal wall need to be repaired to prevent recurrence of cholesteatoma from retraction pockets. Defects may be the result of the disease or the surgery.
Defects Resulting from Disease It is not unusual for cholesteatoma to erode some of the lateral epitympanic wall, but it sometimes may destroy a larger portion of the posterior canal wall. If the canal wall destruction is extensive, it is probably wise not to repair the defect, but to change to a canal wall down procedure. In most cases, canal wall destruction is limited to the lateral epitympanic wall (Fig. 16-14). If a second-stage procedure is planned (see Chapter 18), it is wise not to reconstruct the defect; the thick silicone sheeting would prevent a retraction pocket between stages 1 and 2. At stage 2, after removing the plastic, one may see through the defect to detect any residual disease. The defect may be repaired at that time with ear cartilage. Smaller defects can be repaired with bone pâté. If a second-stage operation is not indicated, the defect may be repaired in a similar way after the graft has been tucked under the bony defect (Fig. 16-15).
Defects Resulting from Surgery If one is using an intact canal wall technique, it is unwise to perform atticotomy. Sometimes it is necessary to do so, however, particularly anteriorly, if the middle fossa dural plate is low, or the ear canal is angled in such a way that one cannot obtain adequate vision into the supratubal recess. The repair of these problems is the same as described earlier. Sometimes an inadvertent opening is made in the canal wall when the mastoid is drilled. To avoid this, the ear canal bone should be thinned as the final step in the procedure before the facial recess is opened. This procedure prevents inadvertently knocking a hole in the wall while drilling in the mastoid. If such a defect occurs, it may be repaired with a shaving of cartilage over the
Acknowledgments Many of the illustrations are modified from Sheehy JL, Brackmann DE: Surgery of chronic otitis media. In English GM (ed): Otolaryngology. Philadelphia, Lippincott, 1994.
REFERENCES
Chapter 16 • Mastoidectomy—Intact Canal Wall Procedure 18. Litton WB, Krause C J, Anson B A, et al: The relationship of the facial canal to the annular sulcus. Laryngoscope 79:1584-1604, 1969. 19. Pickett BW, Cail WS, Lambert PR: Sinus tympani: Anatomic considerations, computer tomography, and a discussion of the retrofacial approach for removal of disease.
207
SUGGESTED READING Sheehy J L : Management of cholesteatoma. In Pensak M (ed): Controversies in Otolaryngology. New York, Thieme, 2001, pp 214-218.
17
Mastoidectomy—Canal Wall Down Procedure Moisés A. Arriaga Videos corresponding to this chapter are available online at www.expertconsult.com.
There are two approaches to mastoidectomy in patients with cholesteatoma and chronic otitis media. The canal wall up (also known as intact canal wall) technique is discussed in detail in Chapter 16. This chapter describes the technique for canal wall down surgery. Related topics discussed in this chapter include the role of atticotomy, mastoid obliteration procedures, reconstruction of canal wall down cavities that have not been previously reconstructed, management of dural venous sinus injury during chronic ear surgery, and facial nerve monitoring in chronic ear surgery.
DEFINITIONS The surgeon may perform a series of mastoidectomy procedures that involve sacrificing a portion or all of the ear canal in continuity with the mastoidectomy.
Modified Radical Mastoidectomy (Bondy) In the classic modified radical mastoidectomy, the epitympanum and canal wall down mastoidectomy and external auditory canal (EAC) are converted into a common cavity. The tympanic membrane and middle ear are left undisturbed.
Radical Mastoidectomy The radical mastoidectomy eradicates middle ear and mastoid disease by converting the mastoid antrum, middle ear, and EAC into a common cavity. The tympanic membrane and ossicular chain are sacrificed. No effort is made at reconstructing a middle ear space; however, a tissue plug or graft is usually placed to seal the orifice of the eustachian tube.
Tympanoplasty with Canal Wall Down Mastoidectomy In tympanoplasty with canal wall down mastoidectomy, the mastoid air cells are exteriorized and form a common cavity with the EAC. The middle ear is reconstructed
by grafting the tympanic membrane and possibly reconstructing the ossicular chain. The terminology can sometimes be confusing. Some authors refer to this procedure as a modified radical mastoidectomy. To be accurate, that term should be applied to the Bondy modified radical mastoidectomy.
Atticotomy In an atticotomy, only a limited portion of the wall of the EAC is sacrificed. A small attic cholesteatoma is exteriorized by drilling the scutum to the limits of the cholesteatoma sac. The defect is reconstructed with a cartilage graft or autologous bone.
Mastoid Obliteration Procedure Mastoid obliteration procedure refers to a possible modification of the above-discussed mastoid procedures in which soft tissue, bone pâté, or biocompatible materials are used to fill the space of the mastoid cavity in an effort to limit postoperative mastoid cavity problems.
Mastoid Reconstruction Procedure Mastoid reconstruction is a two-stage procedure that involves creating an air-containing space with tympanoplasty and silicone elastomer (Silastic) sheeting in a previous radical mastoidectomy. A subsequent procedure is performed for ossicular reconstruction.
INDICATIONS FOR CANAL WALL DOWN MASTOIDECTOMY Mastoidectomy in chronic ear surgery is designed to eliminate mastoid disease in the face of suppurative otitis media and, more commonly, cholesteatoma of the middle ear or mastoid. Generally, canal wall up surgery is preferred to maintain the normal anatomic contours of the mastoid. Certain factors are strong indications for canal wall down surgery, including (1) extensive damage 209
210
OTOLOGIC SURGERY
by disease to the posterior canal wall, (2) severely contracted mastoid with low-lying tegmen and far forward sigmoid sinus preventing adequate visualization through a standard canal wall up approach, (3) cholesteatoma in an only hearing ear, and (4) labyrinthine fistula in an ear with extensive cholesteatoma. Some authors have argued that canal wall down surgery permits excision of cholesteatoma in the sinus tympani region.1 Anatomically, this is not full visualization because the depths of the sinus tympani are medial to the facial nerve. Nonetheless, Hulka and McElveen2 showed that canal wall down procedures do permit additional visualization in the anterior epitympanum and sinus tympani region. The latter is not fully visualized with any technique. Another relative indication for canal wall down mastoidectomy is failure of previous canal wall up procedures with recurrent cholesteatoma from epitympanic retraction pockets. The anatomy of a canal wall down mastoidectomy with full exteriorization of the epitympanum makes retraction pocket recurrences of cholesteatoma unlikely because the whole epitympanum has been exteriorized. Although the use of Silastic in canal wall up facial recess surgery and staging has been significant in reducing the incidence of recurrent cholesteatoma with canal wall up surgery, the presence of a scutal edge and a distinct epitympanum in cases with persistent eustachian tube dysfunction can produce recurrent cholesteatoma.
DECISION MAKING With the exception of the preoperative identification of an attic cholesteatoma in an only hearing ear, usually the decision to perform a canal wall down technique is made intraoperatively. Characteristics such as extensive canal wall destruction by cholesteatoma, a large labyrinthine fistula with an extensive cholesteatoma, and a severely contracted mastoid all are features identified intraoperatively. As the operation proceeds, the surgeon may discover that certain areas of the mastoid, such as the epitympanum, are poorly visualized and may elect to perform a canal wall down procedure for purposes of exposure. Similarly, as the operation is proceeding in a patient with multiple previous recurrences in the epitympanum through retraction pockets, the surgeon may elect to convert the mastoid into a canal wall down cavity, especially if no clear technical reason for failure of the previous procedures has been identified other than chronic eustachian tube dysfunction.
PREOPERATIVE EVALUATION A detailed microscopic examination of the ear is necessary preoperatively. Attic retraction pockets should be viewed with suspicion if the surgeon is unable to see the depths of the pocket. Using a right angle pick to feel the depths
of such a retraction is often helpful. The use of otoendoscopes has been quite helpful to permit a “fish-eye” view of these pockets. This technique permits wider visualization of the pocket. In these cases, computed tomography (CT) scan can be helpful to define whether what seems to be a small retraction represents the neck of a cholesteatoma or is merely a small retraction. Although the CT scan may underestimate the extent of disease in this area, a large cyst extending into the antrum would be identified with CT scan. An attic pocket in which the depths cannot be palpated that is beginning to retain debris is an indication for surgery. If the CT scan does not show a large cyst in the antrum, an atticotomy may be considered. Similarly, a distinct attic cholesteatoma with a positive fistula test and the subjective symptom of dizziness is also an indication for imaging. The possibility of a labyrinthine fistula must be considered. Especially if there is a large cholesteatoma, a canal wall down procedure should be considered. In a pre-existing canal wall down cavity that is draining or is retaining significant debris, the surgeon must evaluate four specific characteristics of the cavity: (1) adequacy of saucerization of the mastoid cortex margins, (2) adequate lowering of the facial ridge, (3) adequate management of the mastoid tip, and (4) adequacy of the meatus. Problems with any of these characteristics can contribute to cavity failures. In chronic drainage situations, areas of persistent mucosalization should be identified so that these are dealt with appropriately in the revision procedure. Audiometric studies are routinely obtained preoperatively. Generally, we repeat audiometric studies done elsewhere before the patient undergoes surgery. Studies particularly should be repeated if the “outside” audiograms do not coincide with tuning fork tests. Special attention must be given to the adequacy of masking with the audiometric studies performed.
PREOPERATIVE COUNSELING As with any otologic procedure, we counsel patients that there are three principal risks: (1) hearing loss that may be total in the operated ear (<1%); (2) dizziness that is usually temporary, but rarely can become permanent; and (3) facial nerve paralysis, which is quite rare, but is a distressing complication for the patient. Generally, patients who undergo canal wall down surgery also have facial nerve monitoring. Our indications and rationale for this are discussed later. Patients undergoing canal wall down surgery should be specifically advised that the ear canal will be larger postoperatively. Healing after a canal wall down procedure takes longer than in a canal wall up procedure. At the preoperative visit, patients are counseled that they should begin applying otologic antibiotic drops immediately after surgery to begin dissolving the packing. After
Chapter 17 • Mastoidectomy—Canal Wall Down Procedure
the initial postoperative visit at 2 weeks, half-strength vinegar irrigations are used to remove residual packing and limit granulation tissue formation. These cavities are usually healed by 2 to 3 months postoperatively.
SURGICAL TECHNIQUES
211
the chapters on tympanoplasty (undersurface graft Chapter 12 and lateral graft Chapter 9). The ear has been reflected forward, the middle ear work has been completed, the tympanic membrane remnant has been prepared for grafting, and the mastoidectomy has begun.
Basic Techniques
Patient Preparation The patient is placed supine on the operating table with the head turned away. Hair is shaved approximately 2 to 3 fingerbreadths behind the pinna. Adhesive plastic drapes are applied surrounding the edges of the shaved hair. Approximately 3 to 4 mL of lidocaine (Xylocaine) 1% with epinephrine 1:100,000 is injected in the postauricular region and in the posterosuperior aspect of the ear canal. Sterile preparation is accomplished with povidoneiodine solution, allowing solution to enter the ear canal as well. In cases in which facial nerve monitoring is used, electromyographic needle electrodes are placed in the orbicularis oculi and orbicularis oris muscles and in the forehead and ipsilateral shoulder to act as ground for monitoring and for stimulating. Sterile towels are placed around the prepared area. The area is dried with a sterile towel, and a large sterile adhesive drape is applied that overlies the sterile prepared area and holds the sterile towels in position. As mentioned previously, the decision to convert a mastoid operation from canal wall up to canal wall down is usually made intraoperatively. The usual steps for tympanoplasty with mastoidectomy surgery would have been accomplished, including canal incisions, as described in
The basic principle of canal wall down mastoidectomy surgery is to eliminate all of the disease and exteriorize the mastoid antrum in continuity with the EAC. In addition to removing all diseased air cells, the following four steps are basic to creating a trouble-free mastoid cavity.
Adequate Saucerization Removal of bone from the edges of the mastoid defect saucerizes or bevels the edges of the defect so that there is no overhanging bone obstructing the wider cavity below (Fig. 17-1). This saucerization ensures that there is no disease tissue laterally. Also, saucerization permits soft tissue surrounding the mastoid to slide into the defect. Paradoxically, removing additional bone around the margins of the mastoid cortex in this saucerization step actually makes a smaller cavity rather than a larger cavity.
Adequate Lowering of the Facial Ridge The boundary between the EAC and the mastoid cavity is defined by the height of the facial nerve. Leaving excessive bone overlying the facial nerve (facial ridge) between the EAC and the mastoid cavity creates a situation in which there is a deep trough on the mastoid side
Partial obliteration
Curved ledge
A
B
FIGURE 17-1. A and B, Saucerization. Removal of ledges and overhanging bone facilitates visualization and permits adjacent soft tissue to obliterate part of the cavity.
212
OTOLOGIC SURGERY Facial ridge lowered to canal floor
High facial ridge
B
A Bean-shaped cavity
Round cavity
High facial ridge Facial ridge lowered to level of ear canal
C D FIGURE 17-2. Facial ridge. A and B, Lowering the ridge permits easier access to the mastoid cavity. C and D, Bean-shaped (C) and round (D) cavities are created by lowering the facial ridge to the level of the floor of the ear canal.
(Fig. 17-2). This high ridge is sometimes referred to as a “beginner’s hump.” Occasionally, novice mastoid surgeons leave a large ridge of bone overlying the facial nerve for fear of injuring it. This ridge creates a difficult situation postoperatively with trapped mastoid spaces that are hard to clean in the office, and it prevents natural cleaning. The surgeon should lower the ridge of bone overlying the facial nerve so that the fallopian canal is barely visible with a thin amount of bone overlying the vertical segment of the facial nerve. Inferiorly, toward the stylomastoid foramen, the nerve takes a medial-to-lateral trajectory. In this area, it is helpful to bevel the bone of the EAC to parallel this more lateral track of the facial nerve inferiorly. Three adjuncts are used for appropriately identifying the level of the facial nerve. First, the nerve itself can be identified at the second genu where the short process of the incus points directly to the facial nerve. If a facial recess has already been accomplished, this level can be followed from superiorly toward inferiorly. The second adjunct is the periosteum and tendon of the digastric muscle in the digastric ridge inferiorly in the region of the mastoid tip. This also leads directly to the facial nerve. This periosteum surrounds the nerve as it exits the stylomastoid foramen. Finally, if facial nerve
monitoring is employed, soft tissue structures that parallel the course of the nerve can be stimulated directly with the facial nerve stimulator to confirm the anatomic position of the nerve. Facial nerve monitoring is more often useful in revision cases in which scar tissue may obscure the anatomy. The objective in adequately lowering the facial ridge is to make the cavity more rounded, rather than kidney-shaped, which facilitates postoperative hygiene.
Management of the Mastoid Tip If the mastoid tip is not pneumatized, it is incapable of retaining debris and suppurative mucosal air cells—no specific management is necessary (Fig. 17-3). Often in chronic draining situations, a pneumatized tip is encountered. This tip must be removed. There are two basic techniques. The mastoid air cells lateral to the digastric muscle can simply be drilled away, in this manner allowing the soft tissue of the skull base to obliterate the space. Alternatively, the digastric ridge can be followed from posterior to anterior. A Kocher clamp can be placed on the mastoid tip, and a curved Mayo scissors can be used to remove the muscular attachments from the mastoid tip to permit complete amputation of this structure. This
Chapter 17 • Mastoidectomy—Canal Wall Down Procedure
Exteriorize
A
213
Amputate
B
FIGURE 17-3. Mastoid tip. A, Mastoid tip is exteriorized with a burr. B, The tip is grasped with a Kocher clamp and amputated by cutting with curved Mayo scissors along the digastric ridge.
eliminates the tip and allows the soft tissue of the skull base to prolapse into the defect.
Adequate Meatoplasty An adequate meatus is essential to allow aeration, epithelialization, and adequate drainage of the cavity. Of the four steps for a trouble-free cavity, small errors in the first three can be compensated by an adequate meatus. The converse is usually not true. An adequate meatoplasty involves extending the vascular strip incisions laterally through the conchal cartilage (Fig. 17-4). The conchal cartilage is removed to permit this extended vascular strip complete mobility so that it can even be everted through the ear canal. The cartilage can be removed posteriorly by first freeing the soft tissue from the conchal cartilage. This soft tissue can remain pedicled medially and be used to obliterate partially a large mastoid cavity. Generally, we simply discard this tissue. At this point, the cartilage can be incised carefully so as not to transect the skin on the ear canal side. The wedge of cartilage can be removed. Alternatively, the vascular strip can be grasped from the ear canal side, and the level of the cartilage can be identified and peeled from the medial side. Additional conchal cartilage adjacent to the edges of the vascular strip should also be removed to allow this whole posterior aspect of the incision complete mobility to overlie the bony margins.
When the posterior meatoplasty has been accomplished, attention is focused anteriorly. At the superior edges of the tragal cartilage, there is usually a significant amount of soft tissue. By undermining the skin overlying the anterior canal and tragal cartilage, a portion of the tragal cartilage can be trimmed along with the adjacent soft tissue to permit the soft tissue mass from obliterating the meatus with either scar or contracture. When the soft tissue meatoplasty has been accomplished anteriorly and posteriorly, we have found that placement of sutures near the edge of cartilage removal is useful in the early postoperative period to maintain the patency of the meatus. In this manner, stenting is not needed to keep the meatus open during the initial healing phases. The four steps just described are the basic building blocks for an adequate canal wall down mastoidectomy procedure.3 These techniques as they apply to specific canal wall down procedures are described in the following sections.
BONDY MODIFIED RADICAL MASTOIDECTOMY The Bondy modified radical mastoidectomy is the original modified radical mastoid operation (Fig. 17-5). Before this procedure was developed, mastoid surgery entailed complete exenteration of the middle ear transducing
214
OTOLOGIC SURGERY
mechanism, including the tympanic membrane and ossicular chain. Despite its historical development, at this time this procedure is indicated for cholesteatoma that has spared the middle ear. An attic cholesteatoma that has extended through the attic into the epitympanum and mastoid antrum can be managed by this technique, which spares the tympanic membrane and middle ear.
Sigmoid sinus
FIGURE 17-4. Meatoplasty. Rosen needle marks the area of conchal cartilage removal.
This technique is particularly useful in cholesteatoma in an only hearing ear, in elderly patients in whom staging and multiple surgeries would not be advised, and in cases of lateral canal fistulas from extensive atticoantral cholesteatoma that has spared the middle ear. After the ear is reflected forward with standard vascular strip incisions, the canal wall down mastoidectomy is created in the usual manner as described in the techniques section. The edges of the cavity are saucerized. The facial ridge is brought down to the level of the nerve with management of the mastoid tip if it is pneumatized and adequate meatoplasty. The hallmark is that the tympanic membrane is left in position. Middle ear work such as ossicular reconstruction is not performed. Cholesteatoma matrix that is lining the epitympanum or overlying a semicircular canal fistula can be left in place because this begins the epithelialization process.
CANAL WALL DOWN MASTOIDECTOMY WITH TYMPANOPLASTY Canal wall down mastoidectomy with tympanoplasty is the procedure that has been renamed by some authors as a modified radical mastoidectomy (Fig. 17-6). The difference between this procedure and the procedure described in the previous section is that tympanic membrane grafting, staging, and ossicular reconstruction are performed with this procedure. Generally, it is preferable to accomplish the middle ear work first. We adhere to the strategy described in the previous chapter for addressing middle ear pathology before the mastoid pathology. Middle ear disease in the anteroinferior, anterosuperior,
Attic cholesteatoma that does not involve the middle ear
Thin layer of skin over ossicles Canal wall down mastoidectomy completed
Matrix over ossicles
A
B
FIGURE 17-5. A and B, Modified radical mastoidectomy. An attic cholesteatoma is exteriorized without entering the middle ear. Canal wall down mastoidectomy is completed, and the matrix covers the ossicle in the epitympanum.
Chapter 17 • Mastoidectomy—Canal Wall Down Procedure
215
Cartilage cap Partial ossicular prosthesis Stapes
A Canal wall down
B Cholesteatoma and diseased incus
FIGURE 17-6. A and B, Canal wall down mastoidectomy with tympanoplasty. In addition to the canal wall down cavity, disease is removed from the middle ear, and ossiculoplasty can be accomplished.
and posteroinferior quadrants is managed first. The posterosuperior quadrant is saved for the final manipulation in the event that an oval window fistula is encountered to permit expeditious termination of the procedure. The mastoidectomy procedure is then performed. Initially, the strategy would be to perform a canal wall up mastoidectomy; however, if any of the characteristics described earlier are encountered, which require changing over to a canal wall down procedure, the technical steps of saucerization, managing the facial ridge, managing the tip, and performing adequate meatoplasty are accomplished. When the technical maneuvers for the canal wall down procedure have been completed, the decision is made whether primary reconstruction can be performed. In the case of extensive cholesteatoma, the usual area for residual disease is not the mastoid, but the mesotympanum. If the cholesteatoma has not been removed in continuity, or if there is a question of complete cholesteatoma removal, or there is extensive mucosal disease, staging is performed with Silastic in the middle ear before tympanic membrane grafting. If ossicular reconstruction can be accomplished at the same stage, this is performed. We prefer porous polyethylene partial and total ossicular prostheses. These prostheses are covered with large cartilage grafts. In cases with a mobile stapes suprastructure, ossicular reconstruction in canal wall down surgery often requires only covering the capitulum with autologous cartilage because the middle ear space is narrow. The cartilage is usually left with an attached tail of perichondrium that serves to anchor the prosthesis/cartilage complex in position. Preservation of the canal wall has not been a prognostic factor
for hearing results in our cases or in those of others.4 If primary ossicular reconstruction is planned without staging in canal wall down surgery, medial grafting is performed if possible when a substantial anterior tympanic membrane remnant is present and the annulus is intact anteriorly. This technique has less risk of lateralization that would affect the ossicular reconstruction in a primarily reconstructed case.
ATTICOTOMY In cases of limited attic cholesteatomas, drilling the scutum around the epitympanic margins of the cholesteatoma can preserve a near-normal contour for the ear (Fig. 17-7). This procedure is limited to patients with cholesteatoma confined to the central epitympanic area. The surgeon must be prepared to convert to a complete canal wall down mastoidectomy if necessary. Preoperative CT has been useful to ascertain that a cyst is limited to the attic. Clinically, the clue that an attic pocket is transitioning into a cholesteatoma is when it is no longer self-cleaning and debris begins to accumulate. Even in a canal wall up procedure with staging and facial recess, this region of the scutum would require reconstruction with either cartilage or bone pâté. The procedure is approached such that conversion to a full canal wall down mastoidectomy with tympanoplasty is straightforward. Standard vascular strip incisions are performed followed by a postauricular incision and reflecting the ear anteriorly. The edge of the scutal defect is drilled with a small diamond burr and gradually widened.
216
OTOLOGIC SURGERY Cholesteatoma Scutum drilled to apex of retraction
M I High pocket (Limited cholesteatoma)
Tympanomeatal flap elevated
Stapes
A
C
Chorda tympani nerve
Inferiorly based tympanomeatal flap
Defect
B
Cartilage with perichondrium seals the defect and tympanomeatal flap is replaced
D
FIGURE 17-7. Atticotomy. A, Atticotomy is limited to small pockets in the epitympanum without extension into the antrum. B and C, Inferiorly based tympanomeatal flap is gradually elevated as the scutum is drilled to the apex of the retraction. D, Cartilage with attached perichondrium seals the defect.
The bone of the superior ear canal is thinned and saucerized to permit adequate visualization. When the margins of the cholesteatoma sac are identified, this is exteriorized. Often the cholesteatoma debris itself is excised, and the matrix can be reflected inferiorly and a small cartilage graft or bone chip can be placed in the scutal defect, and the inferiorly based tympanomeatal flap can be returned. The vascular strip is returned to position, and the ear is packed with absorbable gelatin sponge (Gelfoam). If the disease has been underestimated, and the cholesteatoma extends significantly toward the antrum, the surgeon at this point is usually committed to a full mastoidectomy with tympanoplasty.
RADICAL MASTOIDECTOMY In radical mastoidectomy, extensive disease does not permit ossicular reconstruction; instead, the middle ear and mastoid cavity are exteriorized (Fig. 17-8). Middle ear mucous membrane must be excised along with all dead mastoid air cells. The eustachian tube must be addressed specifically. If it is left open, nasopharyngeal reflux can create a moist cavity. Although the tube may be packed with pledgets of temporalis muscle, a small
temporalis fascia graft over the eustachian tube orifice works effectively.
MASTOID OBLITERATION PROCEDURE Patients with particularly large mastoid cavities occasionally have difficulty with drainage from retained moisture or difficulty being fitted with the hearing aid because of the volume of the cavity. In an attempt to eliminate this problem, soft tissue flaps, bone pâté, and biocompatible materials may be placed to obliterate the volume of the cavity. Rambo5 and Palva6 described a series of muscular flaps for mastoid obliteration. Rambo used inferior flaps, and Palva used more posteriorly fashioned flaps. More recent studies indicate that the long-term effect of mastoid obliteration is negligible because the obliterated and nonobliterated cavities ultimately heal with essentially the same mastoid volume.7 In portions of the cavity that are difficult to exteriorize, such as deep retrofacial air cells, fascia or muscle grafts can be used to seal these air cells to encourage their drainage into the eustachian tube through the air cell system. If the surgeon wishes to seal off a portion of the cavity, the soft tissue that is freed from the posterior aspect of the vascular strip before freeing the
Chapter 17 • Mastoidectomy—Canal Wall Down Procedure
217
Small graft over denuded mucosa seals eustachian tube orifice Oval window Labyrinth
Round window Facial ridge
FIGURE 17-8. Radical mastoidectomy. Fascia is used to seal the eustachian tube orifice after completing the canal wall down cavity and removing middle ear mucous membrane. No effort is made at tympanic membrane or ossicular reconstruction.
cartilage during meatoplasty can be left pedicled to the vascular strip and used as a vascularized flap to obliterate a portion of the cavity. Mastoid obliteration to reduce the volume of the cavity is an infrequently used technique in our practice. Wide saucerization to permit prolapse of adjacent soft tissue is a preferred method.
RECONSTRUCTION OF A RADICAL CAVITY In some patients who have undergone previous radical mastoidectomy, the ear itself becomes quiescent. Despite obliteration of the eustachian tube, recanalization occurs, and an air-containing space can be identified. This airspace is usually limited to the protympanum directly over the eustachian tube area. In the absence of weeping mucous membrane or disease in the rest of the cavity, these patients are candidates for reconstruction of the middle ear sound transforming mechanism. The patient must understand that this reconstruction requires a twostage procedure. Any form of revision surgery must be done with great caution, particularly in elevating the tympanomeatal flap over the level of the facial ridge. The initial surgeon may have lowered the ridge to the level of the nerve itself and not left bony cover. Alternatively, chronic osteitis may have dissolved the remaining thin bone over the facial nerve. Any previous labyrinthine fistula is in jeopardy. The tympanomeatal flap tends to shrink after elevation.
The incisions are usually made along the superior aspect of the lateral semicircular canal and on the posterior aspect of the facial ridge, leaving adequate room in the event of flap shrinkage so that the mesotympanum itself would be covered (Fig. 17-9). In these cases, we encourage the anesthesiologist to use nitrous oxide to inflate further whatever air-containing space exists within the middle ear. Facial nerve monitoring is useful in these cases to confirm the position and the adequacy of bony cover of the facial nerve. The tympanomeatal flap is elevated along with the epithelialized layer covering the promontory; often this elevates most of the mucosa over the promontory as well. When this is accomplished, the eustachian tube itself is examined carefully. We have encountered numerous patients whose eustachian tubes have been obliterated by previous surgeons using an ossicular remnant. This remnant should be removed with care because the petrous carotid artery is just medial to the eustachian tube in this area. When the eustachian tube patency is confirmed, Silastic is placed in the middle ear, the drum is regrafted if necessary with a medial technique with the fascia extending over the level of the facial ridge, and the tympanomeatal flap is returned. Six months later, the Silastic is removed, and the ossicular reconstruction can be accomplished. We have found the argon laser (similar to the method used for laser stapedotomy) to be a useful adjunct for excising fibrous tissue and defining the anatomy of the oval window and stapes footplate
218
OTOLOGIC SURGERY Eustachian tube orifice
Tympanic membrane
Elevate tympanomeatal flap
HSC Facial nerve Mastoid cavity HSC
A
B Incision line
Facial nerve Round window Oval window
FIGURE 17-9. A and B, Incisions for tympanomeatal flap. The tympanomeatal flap incisions are made along the superior edge of the horizontal semicircular canal (HSC).
itself. Reconstruction is usually accomplished with a total ossicular prosthesis and a cartilage graft that brings the level of the reconstruction barely above the level of the tympanic and vertical segments of the facial nerve.
REVISION CANAL WALL DOWN SURGERY Revision canal wall down surgery requires certain technical modifications. Specifically, it is difficult to accomplish vascular strip incision because the bone of the canal is missing. If the tegmen has been thinned, blindly cutting toward the tegmen could injure the dura or underlying structures. The principal technical modification in revision canal wall surgery is that no canal incisions are made at the outset. Instead, a postauricular incision is made, and the epithelialized covering of the contours of the mastoid is elevated in continuity. This tissue usually cannot be elevated to the level of the facial nerve, and when the cavity is entered medially, vertical incisions (the equivalent of incisions at 6 and 12 o’clock) can be fashioned from medial to lateral to permit reflection of the ear anteriorly. At this point, the tympanomeatal flap is elevated in a similar manner to that described earlier for reconstruction of a radical cavity. Care must be taken with regard to the facial nerve and possible labyrinthine fistulas. The middle ear pathology can be dealt with as described previously, and the specific problem with the cavity can be corrected, with the surgeon being methodical in performing each of the four steps described previously (saucerization, facial ridge, mastoid tip, and adequate meatoplasty). If areas of retained air cells with mucosal discharge have
been identified preoperatively, these are drilled. Areas that cannot be fully exteriorized, such as deep retrofacial air cells or perilabyrinthine cells extending toward the apex, are covered with a graft to encourage drainage toward the eustachian tube.
POSTOPERATIVE CARE We pack canal wall down cavities with moistened Gelfoam. Antibiotics containing potentially ototoxic medications are specifically avoided. Neomycin-containing otologic preparations are not used. Instead, cefazolin is usually placed in the irrigation fluid for drilling and for irrigating bone dust at the completion of drilling. The Gelfoam itself is moistened in saline and packed over the tympanic membrane and in the cavity. As described in the section on meatoplasty, the meatus is held open with two absorbable sutures (3-0 polyglactin 910 [Vicryl]) placed from the posterior aspect just at the edge of cartilage removal from the concha. The periosteum and subcutaneous tissue are reapproximated with interrupted 3-0 absorbable sutures. Steri-Strips are applied, and the remainder of the cavity is packed with large pieces of Gelfoam. The patient is instructed to begin otologic drops (usually a sulfa and steroid preparation) the day after surgery; this is continued until the 2 week postoperative visit. At that point, most of the packing is removed from the cavity in the office. The patient is instructed in using a dilute vinegar solution (acetic acid) to wash out the remainder of the packing and to minimize granulation formation. The patient irrigates twice a day and uses the antibiotic
Chapter 17 • Mastoidectomy—Canal Wall Down Procedure
219
(Canal wall down) Facial nerve on floor of epitympanum S
Supratubal recess Cochleariform process Eustachian tube orifice
VII
Petrous carotid Hypotympanum
Jugular bulb
I
P
A
FIGURE 17-10. Facial nerve on floor of epitympanum. Schematic view of labyrinthine segment of the facial nerve just medial to the epitympanum. A, anterior; I, inferior; P, posterior; S, superior.
drops at bedtime. The second postoperative visit is 6 weeks after the surgery. At that time, any residual packing is cleaned. Significant granulations are managed with judicious application of silver nitrate, vinegar irrigations at home, and antibiotic-steroid drops.
POTENTIAL PITFALLS Dural Venous Sinuses Although management of the dural venous sinus is not usually considered a component of chronic ear disease, these structures are intimately related to the surgical field, and the surgeon must be comfortable in managing any injuries to these structures because cholesteatoma itself can affect them. The principle outlined in Chapter 16 of using the largest possible burr is necessary for preserving the safety of the dural venous sinuses. A large cutting burr would bounce off the wall of the sigmoid sinus, whereas a small cutting burr with the same amount of force applied would lacerate the sinus, necessitating management. If the sigmoid sinus is entered, the immediate management is to remove the suction aspirator from the field and place a finger over the opening. This is venous bleeding, and usually pressure is adequate for its control. A large piece of Gelfoam over the surface of the opening with gentle pressure such as with a cottonoid is usually adequate for
coagulation and termination of the venous bleeding. If there is a surrounding bony cover, bone wax can be used effectively. If any material is placed intraluminally, care must be exercised to maintain a tail of the packing material extraluminally. Packing material that is placed within the lumen of the sinus continues through the jugular bulb into the central venous circulation and produces pulmonary emboli. An injury to the superior petrosal sinus can usually be repaired with bipolar cautery of the sinus or placement of bone wax. Injury of the jugular bulb is usually more problematic. Temporary occlusion with the surgeon’s finger or a cottonoid is necessary. In the meantime, a patch can be prepared combining absorbable knitted fabric (Surgicel) or Gelfoam with a large piece of bone wax to cover the injured area and apply pressure for coagulation to occur. If conservative methods are unsuccessful, the surgeon must be prepared to ligate the jugular vein in the neck and proceed with intraluminal packing.
FACIAL NERVE The floor of the epitympanum must be approached cautiously because the facial nerve labyrinthine segment is just medial to the geniculate ganglion, and aggressive dissection in this region can injure the nerve (Fig. 17-10).
220
OTOLOGIC SURGERY
FACIAL NERVE MONITORING Facial nerve monitoring in ear surgery has become a contentious topic. We do not believe that facial nerve monitoring is the current standard of care in surgery for chronic ear disease. Nonetheless, because our team is involved in many neurotologic cases as well, we monitor most otologic cases done under general anesthesia. The surgical team is comfortable with the use of a facial monitor; the anesthesiologist knows muscle relaxants are not used in otologic cases, and the quality of monitoring improves. With regard to monitoring in canal wall down surgery, revision canal wall down surgery (because of the risk to the facial nerve in the previously lowered facial ridge and in a large cholesteatoma) is a definite indication for facial nerve monitoring. Although the monitoring is not a substitute for anatomic knowledge, chronic ear disease can distort the anatomy, making monitoring useful. We have encountered cases in which edematous changes to the dehiscent tympanic segment of the nerve have increased its diameter threefold so that the nerve itself appeared to be a mound of granulation tissue. In large cholesteatomas when the fallopian canal has been eroded, monitoring permits separation of the cholesteatoma from the nerve with auditory feedback from the monitor in a similar manner to removal of an acoustic tumor from a facial nerve in cerebellopontine angle surgery. The advantages of facial monitoring include identification of unexpected anatomic variations, identification of anatomic distortions from chronic ear disease, and assistance with lowering the facial ridge to an appropriate level. Also, in cases of dehiscent nerves or significant dissection of granulation from the nerve, anatomic integrity of the facial nerve can be confirmed by stimulating proximal to the questionable area before the conclusion of the procedure.
OUTCOMES Adherence to the principles of canal wall down surgery outlined earlier results in more than 90% of cavities requiring yearly or biannual cleaning without drainage or
infection. The anatomy of the ear has been changed by canal wall down surgery. Failure to débride desquamated epithelium can produce infections, cholesteatomas, and serious complications. We counsel patients that cholesteatoma is a lifelong disease that is either active or inactive. Periodic follow-up is necessary to prevent reactivity. We prefer patients to have otologic drops available at home in the event of cavity drainage. Development of a severe infection is prevented by 3 drops, three times a day for 3 days at the onset of drainage. If drainage persists, office follow-up is advised.
REFERENCES 1. Edelstein D R , Parisier SC : Surgical technique and results in cholesteatoma. Otolaryngol Clin North Am 22:10291040, 1989. 2. Hulka G F, McElveen JT Jr: A randomized blinded study of canal wall up versus canal wall down mastoidectomy determining the differences in viewing middle ear anatomy and pathology. Am J Otol 19:574-578, 1998. 3. Sheehy J L : Surgery in chronic otitis media. In English G M (ed): Otolaryngology, vol. 1, Philadelphia, Lippincott, 1984. 4. Sheehy J L , Beneche J E : Middle ear reconstruction: Current status. Adv Otolaryngol Head Neck Surg 1:143-170, 1987. 5. Rambo J HT, Musculoplasty: Ann Otol Rhinol Laryngol 74:535, 1965. 6. Palva T: Meatally based musculoperiosteal flap in cavity obliteration. In Sade J (ed): Cholesteatoma and Mastoid Surgery: Proceedings, Second International Conference. Amsterdam, Kugler, 1982. 7. Toner J L , Smyth G D, Kerr AG: Realities in ossiculoplasty. J Laryngol Otol 105:529-533, 1991.
18
Tympanoplasty—Staging and Use of Plastic James L. Sheehy and Clough Shelton*
Elimination of disease and restoration of function are the two aims of tympanoplasty. In most teaching situations, one can separate the two aims, limiting the discussion to one or the other. The staging of the operation and the use of plastic in the middle ear require, however, that the discussion consider both objectives. Staging the operation involves disease and function, and it is not technique oriented; that is, staging does not vary significantly with the technique of tympanic membrane grafting or of restoring the sound pressure transfer mechanism, or even the management of the mastoid. This chapter discusses the indications for staging tympanoplasty and mastoidectomy, and techniques used in performing tympanoplasty in two stages. The controversies surrounding the procedure are discussed at the end of the chapter.
POSTOPERATIVE COLLAPSE OF TYMPANIC MEMBRANE Retraction and collapse of the tympanic membrane is a well-recognized postoperative problem. Many authors blame the collapse on continued poor eustachian tube function.1 Others blame the collapse on fibrous adhesions between the denuded middle ear surfaces and the tympanic membrane graft (see later). The introduction of a barrier material, such as silicone elastomer (Silastic) sheeting, between these two raw surfaces prevents the formation of fibrous adhesions, with subsequent retraction and collapse of the tympanic membrane. This alternative explanation is supported by observations of healing after staging. In a review of 400 planned two-stage tympanoplasty operations, 89% of *Editorial Note: James Sheehy was a skillful surgeon and a master teacher. For many of us, his name is synonymous with surgical management of chronic otitis media. Jim was the sole author of this chapter in our first two editions. It is difficult to improve on a chapter that is done so well. The current chapter represents an update of the chapter in the previous edition with minor revisions.
patients achieved an aerated middle ear, 5% required the placement of a ventilation tube, and the remainder developed collapse of the middle ear space.2 These results would indicate that continued eustachian tube dysfunction is an uncommon cause for postoperative tympanic membrane retraction.
INDICATIONS FOR STAGING There are two reasons for staging the operation in tympanoplasty: (1) obtaining a permanently disease-free ear and (2) obtaining permanent restoration of hearing.3,4 Whether one finds any indication for staging depends on how vigorously a good functional result is pursued in badly diseased ears. The decision whether or not to stage is made at the time of surgery. With experience, one usually can make this judgment preoperatively and alert the patient to the possible necessity of a two-stage procedure. The decision is based on three factors: (1) the extent of the mucous membrane problem, (2) the certainty (or lack thereof) of removal of cholesteatoma, and (3) the status of the ossicular chain. Taking these three factors into account, we stage about 75% of tympanoplasty and mastoidectomy procedures and about 15% of tympanoplasties not requiring mastoidectomy.
Mucosal Disease Factors There are frequently large areas of diseased or absent mucosa in the chronically infected middle ear. Groundwork is necessary to promote regrowth of normal mucosa. The first step is elimination of infection before surgery, if possible. The second step is removal of all squamous epithelium, granulations, and irreversibly diseased mucosa at the time of surgery. The middle ear is sealed with a graft to prevent squamous epithelium from migrating back into the middle ear. This sealed middle ear space fills with a blood clot, and this clot supports fibroblastic invasion with eventual formation of scar tissue or adhesions between 221
222
OTOLOGIC SURGERY
the denuded surfaces. To prevent these adhesions from forming, and to allow mucosa to migrate in, plastic sheeting is used over the denuded areas. A two-stage operation is indicated to obtain the best hearing results, and to prevent recurrence of cholesteatoma (retraction pocket) in patients with extensive mucous membrane destruction. The objective of the two-stage procedure is to obtain a well-healed ear with a mucosa-lined pneumatized middle ear cleft so that ossicular reconstruction may be performed later under ideal circumstances.
Ossicular Chain Factors An increase in the incidence of sensorineural hearing impairment has been observed in patients in whom the inner ear has been opened in the presence of actual or potential infection. Because of this result, a fixed stapes should not be removed at the time of tympanic membrane grafting. In cases of otosclerosis, a two-stage procedure is almost always indicated. When the fixation is due to tympanosclerosis, it may be possible to mobilize the stapes, depending on the area of fixation. In diffusely involved cases, a laser can be useful to remove the suprastructure and to char the tympanosclerosis, allowing its removal and resulting in a mobile footplate. If it is impossible to mobilize the footplate, a second stage procedure should be carried out. At that time, the tympanic membrane would be intact, and the middle ear would be without infection, allowing for removal of the fixed footplate in a sterile environment and completion of ossicular reconstruction.
Residual Cholesteatoma Factor It may seem illogical to leave behind epithelial disease, removing it at a planned second stage procedure, but this is exactly what is done under certain circumstances. Removal of cholesteatoma in the middle ear may be questionable sometimes in an acutely inflamed ear, in which differentiating between granulation tissue and matrix is difficult. Differentiating becomes a particular problem when granulations fill the oval and round windows. Excessive manipulation in these areas could result in an inner ear complication. Removal of matrix involving a mobile stapes with an intact suprastructure can be challenging. Sometimes it is impossible to be certain that every shred of cholesteatoma has been removed. In such cases, a laser can be used at the second stage to cut the crura off the mobile footplate and facilitate removal of residual disease. The surgeon may have torn the matrix when removing it from the tympanic recess and may be uncertain of complete removal, which presents a considerable problem under the pyramidal process, an area hidden from view regardless of the technique of surgery, whether it is an open or closed cavity technique. Removal of the
pyramidal process with a diamond burr may or may not resolve the problem. One third of patients with middle ear cholesteatoma at the first operation have residual disease at the second stage.2 It is much easier to be certain of cholesteatoma removal from the mastoid, especially in a small apneumatic one. Extensive cholesteatoma in a pneumatized mastoid poses a problem. In using the intact canal wall procedure, one should usually revise the mastoid in such cases within 1 to 2 years to be certain not to leave disease behind. The mastoid and epitympanum are often re-explored in patients in whom excessive bleeding occurred at surgery. Unexpected residual disease in the epitympanum has been noted in some cases of this type in the past.
Timing the Second Stage The second-stage operation may be performed in 6 to 9 months if the primary indication for staging was an ossicular or a mucous membrane problem. The middle ear should be well healed by that time. If the primary reason for staging is reinspection of the mastoid and epitympanum for possible residual cholesteatoma, it is best to wait 9 to 18 months. The delay allows time for any residual disease to have grown to a 1 or 2 mm cyst so that it may be identified with greater ease. The only exception to this rule is if this disorder occurs in a child, or if serous otitis media develops; a residuum may grow faster under these circumstances. What is the best strategy for managing a patient with bilateral cholesteatomas who needs staged procedures on both ears? The decision here is based on disease activity and hearing level. After the first stage, it is common for the ear to have a maximal conductive hearing loss until the second operation. This ear would not provide the patient with functionally useful hearing without a hearing aid. If only one ear has active disease or poor hearing, that ear is operated initially. After the second stage is completed, surgery begins on the second ear. If both ears have active disease, after healing has occurred from the first stage on the initial ear, the second ear can be operated. The patient typically requires a hearing aid on the first ear until the ossicular reconstruction is performed at the second stage. Fitting a behind-the-ear hearing aid provides the capability to switch the aid easily to the opposite ear as needed. Patients with chronic otitis media present the surgeon with many management complexities. Bilateral disease is common, and each ear may require two surgeries. It is helpful to record the plan for the second stage at the time the first stage is completed; this can be done through a handwritten chart note, a customized surgical data sheet, or prominent placement in the dictated operative report. Common included items are timing of the next stage, approach (transcanal or postauricular), type
Chapter 18 • Tympanoplasty—Staging and Use of Plastic
of prosthesis, location of possible residual cholesteatoma, need for special equipment (laser, facial nerve monitor), and areas of special caution (exposed dura, dehiscent facial nerve or jugular bulb). These details are best documented at the time of the first operation, and provide a clear basis for the planning of the second stage.
PREOPERATIVE EVALUATION AND COUNSELING The ability to predict the need for staging the operation depends on one’s experience and philosophy regarding the badly diseased ear. We discuss the possibility of needing a planned two-stage operation with the patient at the initial consultation and include it in printed patient educational materials. In some cases, it is clear at the initial evaluation that two stages are required; in others, the need for staging can be determined only intraoperatively. Typically, when the rationale for staging is explained to the patient, it is readily understood and accepted.
PLASTIC IN THE MIDDLE EAR The most common middle ear indication for staging the operation is a mucous membrane problem. To avoid confusion, the following discussion is limited to that problem. The techniques are the same when there are other indications.6
Plastic Sheeting Silicone sheeting is referred to here as “thin” and “thick.” The thin sheeting is 0.005 inch; the thick sheeting is 0.040 inch. Silicone sheeting is distributed by Invotec International, Inc., Jacksonville, Florida. An alternative to thin silicone sheeting is absorbable gelatin film (Gelfilm). Thin silicone sheeting is malleable. It adapts easily to the middle ear space and does not tend to curl when exposed to body temperature. Thick silicone sheeting is stiff and is not deformed by fibrous tissue that may develop in the middle ear. Although stiff, it is malleable enough to be withdrawn from the mastoid and middle ear through a tympanotomy exposure. Supramid Extra is the trade name for a medical-grade nylon 6. The sheeting thickness of Supramid Extra Foil is 0.3 mm. Because it is thinner than thick silicone sheeting, it is easier to insert. It is not malleable, however, and this quality prevents it from being extracted from the mastoid without a mastoid re-exploration. Most cases can be managed with Gelfilm for minimal mucosal disease and thick silicone sheeting for severe disease. Gelfilm is used for cases not requiring a second stage, and thick silicone sheeting is used for cases that are to be staged.
223
Mucous Membrane Indications When the mucous membrane problem is limited, there is no need for staging (Fig. 18-1). Adhesions form between denuded bone of the middle ear and the tympanic membrane graft, but usually do not pose a long-term problem. Nonetheless, Gelfilm is frequently placed over the denuded areas to prevent adhesions (Fig. 18-2). Alternatively, one may use thin silicone sheeting, if desired. The silicone sheeting may be left in place indefinitely. In more diseased ears, one may find that normal mucosa remains only in the tubotympanum, facial recess, and epitympanum (Fig. 18-3). Reconstruction of this ear without regard to the mucous membrane problem usually results in a fibrosed middle ear. When thin silicone sheeting was used in such cases in the early 1960s, the results were frequently disappointing. The silicone sheeting was rolled up or deformed by advancing fibrous tissue. If the silicone sheeting contacted the tympanic membrane, extrusion often occurred. To obtain the best hearing results in such cases, it was necessary to perform the reconstruction in two stages. At the initial operation, the tympanic membrane was grafted over a piece of thick silicone sheeting that filled the middle ear (Fig. 18-4). Six months later, the thick silicone sheeting was removed, and a suitable prosthesis was inserted. In badly infected ears, there may be no mucosa remaining in the middle ear cleft except for the tubotympanum (Fig. 18-5). Some physicians believe that nothing can be done to reconstruct such an ear; radical mastoidectomy has been advised. Ears of this type can be reconstructed by staging the tympanoplasty, as long as some mucosa remains in the tubotympanum, and the plastic sheeting can reach the area to prevent eustachian tube closure. In cases of extensive or total mucous membrane destruction, it may be wise to perform an intact canal wall mastoidectomy even though there may not be other indications for a mastoid exploration. The mastoid is opened into the middle ear through the facial recess for insertion of a sheet of Supramid or thick silicone sheeting (Fig. 18-6). This process ensures that the plastic would not roll up—hence, no fibrous tissue. During revision, the plastic is removed, and a suitable prosthesis may be placed between the stapes capitulum or footplate and the mobile tympanic membrane in a well-healed normal middle ear.
Canal Wall Down Procedures The principles and indications involved in staging the operation in canal wall down procedures are the same as in canal wall up procedures, but the technique is different. Many physicians who use both techniques in tympanoplasty have stated that their hearing results are less satisfactory in canal wall down cases. The probable
224
OTOLOGIC SURGERY
FIGURE 18-1. Absence of mucosa over promontory. FIGURE 18-2. Silicone sheeting has prevented adhesions between tympanic membrane graft and denuded promontory. FIGURE 18-3. More extensive mucous membrane destruction and absence of the stapes. Good mucosa remains in the recesses and the epitympanum. FIGURE 18-4. Stiff plastic sheeting holds its position and prevents adhesions between the tympanic membrane graft and the raw areas of bone. FIGURE 18-5. Mucous membrane destruction is extensive. Note a few islands here and there of mucosa and some in the tubotympanum. FIGURE 18-6. Thick silicone or Supramid sheeting is not displaced by fibrous tissue. Mucosa should regrow over all denuded bone and the undersurface of the graft.
Chapter 18 • Tympanoplasty—Staging and Use of Plastic
r eason for these results is the narrowing of the middle ear space, which should not be allowed to occur in canal wall down procedures. In staging the operation in canal wall down procedures, a wide middle ear space is obtained by extending the thick silicone sheeting onto the fallopian canal. It is important to bevel the superior edge of the plastic to help delay extrusion; the graft rests on that edge.
Plastic Extrusion There is no question that plastic material placed in the middle ear may extrude if an edge of the plastic comes in contact with the tympanic membrane. It is our experience, however, that extrusion occurs in less than 0.5% of the cases. Surgeons who have reported a major problem with extrusion are probably using thin silicone sheeting in situations in which they should have used thick plastic and staged the operation.
CONTROVERSY There is considerable difference of opinion among experienced otologists in regard to staging the operation.5 The major difference occurs over more diseased ears, usually with cholesteatoma, and is related not to the management of the mastoid (canal wall up or canal wall down), but instead to the surgeon’s philosophy.7 Staging proponents believe that the great divergence of opinion on staging that one encounters is related to two factors. How hard does the individual pursue a good functional result in the most severely diseased ear? What does one accept as a satisfactory functional result? In an ear with no remaining mucosa except in the tubotympanum, it is our experience that unless the operation is staged, one cannot usually obtain a satisfactory functional result.
225
Acknowledgments Many of the illustrations are modified from Sheehy JL, Brackmann DE: Surgery of chronic otitis media. In English GM (ed): Otolaryngology. Philadelphia, Lippincott, 1994.
REFERENCES 1. Sheehy J L : Testing eustachian tube function. Ann Otol 90:562-564, 1981. 2. Rambo J HT: Use of paraffin to create a middle ear space in musculoplasty. Laryngoscope 71:612-619, 1961. 3. Tabb HG: Surgical management of chronic ear disease with special reference to staged surgery. Laryngoscope 73:363-383, 1963. 4. House H P: Polyethylene in middle ear surgery. Arch Otolaryngol Head Neck Surg 71:926-931, 1960. 5. Austin D F: Types and indications of staging. Arch Otolaryngol Head Neck Surg 89:235-242, 1969. 6. Smyth G D L : Staged tympanoplasty. J Laryngol 84:757764, 1970. 7. Sheehy J L : Plastic sheeting in tympanoplasty. Laryngoscope 83:1144-1159, 1973. 8. Sheehy J L , Crabtree J A : Tympanoplasty: Staging the operation. Laryngoscope 83:1594-1621, 1973. 9. Sheehy J L , Shelton C : Tympanoplasty: To stage or not to stage. Otolaryngol Head Neck Surg 104:399-407, 1991. 10. Sheehy J L , Brackmann D E : Surgery of chronic otitis media. In English G M (ed): Otolaryngology. Philadelphia, JB Lippincott, 1994. 11. Sheehy J L , Brackmann D E : Surgery of Chronic Ear Disease: What We Do and Why We Do It. Instructional Courses, Vol. 6. St. Louis, CV Mosby, 1993.
19
Complications of Surgery for Chronic Otitis Media Richard J. Wiet, Steven A. Harvey, and Philip D. Littlefield Videos corresponding to this chapter are available online at www.expertconsult.com.
Chronic otitis media is a formidable challenge to the otologist. Competent management requires a solid understanding of temporal bone anatomy and the subtleties of the disease. It is ideal to avoid complications, so one should begin with a careful history and physical examination along with thoughtful preparation. Nevertheless, nobody has total control in life, and even a master surgeon eventually confronts a serious complication. When this happens, the surgeon first must recognize that there is a complication, and then handle it in a way that minimizes morbidity. The ability to address a serious complication well requires composure, knowledge, and judgment. Afterward, an unassuming review of the experience leads to surgical wisdom. This chapter discusses complications of surgery for chronic otitis media, including otic capsule fistulas caused by the surgeon or the underlying disease, because either kind is dangerous if improperly managed. Other topics discussed include sensorineural hearing loss from various causes, iatrogenic facial nerve trauma, ossiculoplasty complications, dural injury, and vascular injuries. We describe how to prevent or deal with each of these problems. The surgeon should not hesitate to consult an expert about a complication; this is especially true if the surgeon has not experienced many. Few surgeons have practice at treating unexpected facial paralysis after mastoid surgery. It is wise to consult whenever in doubt so that a bad situation is not exacerbated. Likewise the consultant should view the situation in an objective, nonjudgmental way because behavior to the contrary is counterproductive at best.
PREOPERATIVE COUNSELING Each patient should be informed of the indications, alternatives, and potential complications of surgery. We describe the diagnosis and procedure in layman’s terms. We explain the probability of success in simple ratios, such as “8 or 9 times out of 10 we close the perforation with a gain in hearing.” We then explain the potential risk of partial or total sensorineural hearing loss, dizziness,
facial weakness, and other complications of mastoid surgery. Our group uses a handout that follows the example outlined by Sheehy1 to facilitate the discussion. With practice, this can be done in a way that does not sound negative. To do so, the patient must understand the goals of the surgery and how they are to be accomplished. The principal goal of chronic otitis media surgery is to eradicate infection. Correction of hearing loss is secondary and often is delayed until a second procedure. Preoperative counseling depends on the findings of a proper history and physical examination. A history of vertigo or sensorineural hearing loss is important. The condition of the tympanic membrane should be noted because central perforations are rarely associated with cholesteatoma, whereas marginal perforations or retractions are more problematic. We carefully inspect retraction pockets of the pars flaccida or the posterosuperior pars tensa. A hearing loss of greater than 30 to 35 dB implies erosion of the ossicular chain, but hearing is often normal with posterosuperior quadrant cholesteatomas. The patient should be forewarned that sound transmission is occurring through the mass, and that hearing is likely to decrease after the initial surgery. The patient with an only hearing ear needs a conservative approach, and this must be conveyed as well.
LABYRINTHINE FISTULA SECONDARY TO CHRONIC OTITIS MEDIA Labyrinthine fistulas are well-known complications of chronic otitis media. They are an unusual entity—but by no means rare—and are a serious operative hazard. An inexperienced surgeon may not foresee this and spoil an otherwise good result. The incidence of labyrinthine fistulas secondary to chronic otitis media in the modern literature ranges from 3.6%, reported by Palva and colleagues,2 to 12.9%, reported by Sanna and coworkers.3 Although the incidence of infectious complications of chronic otitis media, such as meningitis, sigmoid sinus thrombosis, and intracranial abscess, has declined, 227
228
OTOLOGIC SURGERY
TABLE 19-1 Site of Labyrinthine Fistula and
Associated Eye Movements
Site of Fistula
Eye Movements
Lateral canal, postampulla Lateral canal, preampulla Vestibule
Horizontal, toward normal ear Horizontal, toward diseased ear Rotary, horizontal, toward diseased ear Rotary, toward normal ear Vertical, with an arc
Superior canal, ductal side Posterior canal
Adapted from McCabe BF: Labyrinthine fistula in chronic mastoiditis. Ann Otol Rhinol Laryngol 93(Suppl 112):138-141, 1983.
labyrinthine fistulas have had a remarkably stable prevalence of about 10%.4 The most common location of a labyrinthine fistula is the lateral semicircular canal in all published series. Most authors have reported isolated lateral semicircular canal fistulas in about 80% of cases, but this figure ranges from 57%5 to 90%.6 This preponderance shows that the lateral semicircular canal is more exposed to cholesteatoma than the rest of the labyrinth. The remainder of fistulas in most series involved the lateral canal plus one or more other sites. There is a difference between a fistula and a bone erosion, although both appear as a blue-gray line under the microscope. A fistula is an abnormal communication between a hollow organ and the outside. With erosion, the bone is thin, but the layer next to the endosteum is still intact. Consequently, the labyrinth behaves differently than with a fistula.5 With a fistula, only the endosteal membrane separates the cholesteatoma matrix from the perilymphatic space. Pressure changes are transmitted directly to the membranous labyrinth causing the findings described in most reports (Table 19-1). Many authors describe fistulas as small or large without quantifying these terms. Sanna and coworkers3 have classified fistulas as small (0.5 to 1 mm), medium (1 to 2 mm), and large (>2 mm). They found that most fistulas were large (74%), followed by medium (18.4%) and small (7.6%). Gacek5 also classified fistulas as either small or large with a cutoff of 2 mm in the greatest dimension. Although Gacek said that 2 mm was an arbitrary dividing point, he thought that the bony margin of the opening could support the cholesteatoma matrix in fistulas smaller than 2 mm. This forms a subtle separation from the endosteal membrane and allows safer removal. Although the surgeon should be prepared to deal with a fistula in any case of chronic otitis media, the length of symptoms may heighten suspicion. Sheehy and Brackmann,7 reporting on 97 cases of labyrinthine fistulas, noted that greater than 50% had a history of chronic otitis media for 20 years or longer. Ritter8 also noted that many of his patients had lifelong otorrhea, often dating back to childhood.
Vestibular symptoms increase the probability of a labyrinthine fistula. Sheehy and Brackmann7 noted that almost two thirds of patients were dizzy, and dizziness was constant in 12%. Ritter8 said that 76% of his patients complained of vertigo, which was usually brief, lasting seconds to minutes. Ostri and Bak-Pedersen,9 with 20 cases, and Gormley,10 with 35 cases, noted frank vertigo in 65% of patients. McCabe4 reported the highest prevalence of vestibular symptoms of any larger series: 90% of 79 cases. In addition, there was a high prevalence of positive fistula tests (72%). McCabe gave an outstanding description of the test along with the anticipated eye movements for different fistula locations (see Table 19-1). Dizziness is much less common in cases of chronic otitis media without fistulas. One review found the prevalence to be 15% in such cases, and more related to age than to duration of disease.11 With a positive fistula test of a postampullary lateral canal fistula, air compression into the ear causes conjugate deviation (not nystagmus) of the eyes to the opposite ear; this displaces the endolymph and cupula toward the vestibule. There is nystagmus toward the tested ear with sustained pressure. If the pressure is pulsed, there is still deviation to the opposite ear, but the eyes drift back to the midline with pressure release. If the fistula is located at the junction of the lateral canal ampulla and vestibule (preampullary region), positive pressure displaces the cupula away from the vestibule with deviation of the eyes toward the diseased ear. Positive pressure on a fistula of the vestibule moves the horizontal and superior canal ampullae, causing rotary-horizontal eye movements toward the diseased ear.4 Pressure changes are not the only way for a fistula to cause vestibular symptoms. Extension of infection into the perilymph is rare, but causes purulent labyrinthitis with complete loss of vestibular function. Serous labyrinthitis is more common and occurs when only inflammation and bacterial toxins enter the labyrinth.8 Sensorineural hearing loss is another clue that there might be a fistula. Sheehy and Brackmann7,11 found an impairment in more than half of their fistula cases compared with only 20% of nonfistula cases. Sensorineural levels were more often diminished in patients with extensive fistulas at sites other than the lateral canal. Preoperative anacusis existed in 12%.7 Ritter8 also found decreased sensorineural levels in his series—70% had decreased bone conduction, which averaged 26 dB. Speech discrimination was less than 80% in 20% of patients, and 30% were deaf. Farrior12 reported nonserviceable hearing in 13% of 31 cases, whereas Ostri and Bak-Pedersen9 noted anacusis in 15%. Preoperative facial weakness, although rare, is more common with fistulas than without fistulas: 4% versus 1%.7 All series report a high incidence of facial nerve dehiscence, including Gormley (27%),10 Ritter (36%),8 Sheehy and Brackmann (50%),7 and Ostri and BakPedersen (55%).9 There is a greater potential for intraoperative nerve injury with fistulas.
Chapter 19 • Complications of Surgery for Chronic Otitis Media
229
Fistula
FIGURE 19-1. Fistula of the lateral semicircular canal secondary to cholesteatoma. The facial recess is opened.
INTRAOPERATIVE MANAGEMENT OF LABYRINTHINE FISTULAS The suitability of leaving cholesteatoma matrix over a fistula versus removing it remains a passionate controversy. Walsh13 and Baron14 drew attention to this in 1953 during a symposium on the management of cholesteatomas. Walsh13 thought the cholesteatoma had a chemical osteolytic effect and advocated removal in all cases. Baron14 believed that bone erosion from cholesteatoma was mainly from pressure, so exteriorization of the sac was sufficient. Various techniques have been promoted since then. All authors support three concepts with labyrinthine fistula treatment, as follows: 1. Open and evacuate any cholesteatoma sac in the mastoid (Fig. 19-1), and carefully palpate the medial wall to detect any bony erosion—especially on the dome of the lateral canal. 2. Leave matrix over any fistula to protect it, even if planning removal, and focus on the rest of the ear disease. Any removal is done immediately before closing. When exposed, quickly cover the fistula with tissue such as fascia, vein, or perichondrium (Fig. 19-2). 3. Leave matrix alone whenever fistulas are extensive, multiple, or involve the vestibule or cochlea (Fig. 19-3).3,10 A promontory cochlear fistula is one of the few absolute indications for radical mastoidectomy because postoperative sensorineural hearing loss is likely. Sheehy and Brackmann7 had a 56% incidence of severe or total
FIGURE 19-2. Fascia placed over lateral semicircular canal fistula after removal of cholesteatoma matrix.
sensorineural hearing loss when they attempted matrix removal from extensive fistulas. Gacek5 reported a partial to profound hearing loss in three cases in which he removed matrix from cochlear fistulas. Gacek believed that the membranous semicircular canals and their ampullae have thicker, more rigid walls than Reissner’s membrane or the basilar membrane, and are less likely to rupture with manipulation. In addition, the cochlear duct is in the outermost portion of the bony cochlea, so it is more closely in contact with the overlying matrix than the membranous labyrinth is elsewhere. With regard to the isolated lateral canal fistula, some authors always leave the matrix alone because of the risk to the inner ear.8,16,17 Ritter8 recommended a canal wall down mastoidectomy and leaving the matrix over the fistula. He reported hearing loss in 47% of cases when he removed the matrix compared with 22% when it was undisturbed. He left 80% of fistulas uncovered after matrix removal; this increases the possibility of an inner ear injury. Nevertheless, the senior author (R.J.W.) agrees with this philosophy—especially when the fistula looks larger than 2 mm on computed tomography (CT) scan. Other surgeons perform a canal wall down tympanomastoidectomy and matrix removal in cases of isolated lateral canal fistulas. Farrior12 prefers a single-stage operation, but cautions against matrix removal from an acutely inflamed or better hearing ear. Palva and associates2 also use this technique. Their preference is due to cases of periodic otorrhea or persistent vertigo when the matrix was left in place. All cases in which the matrix was removed and the fistula was covered healed without further otorrhea or vestibular symptoms.
230
OTOLOGIC SURGERY
Squamous debris Cholesteatoma matrix
Swollen mucosa Bone
FIGURE 19-3. Cochlear promontory fistula secondary to cholesteatoma. The posterior external auditory canal is gone in this dissection.
According to Gacek,5 the decision to remove matrix is based on the following factors: 1. Ability and experience of the surgeon: The surgeon should not disturb a fistula if he or she is not confident about atraumatic removal. 2. Location and size of fistula: Matrix removal is safe with lateral canal fistulas less than 2 mm in diameter, but only an experienced surgeon should attempt removal when a fistula is greater than 2 mm. Dissection should be stopped if the matrix seems adherent to the membranous labyrinth. 3. Function of both ears: Keeping cholesteatoma out of the labyrinth takes precedence when there is no hearing. The matrix should be removed, and a labyrinthectomy should be considered. Only an experienced surgeon should attempt matrix removal in a better hearing ear, but the fistula must be smaller than 2 mm. 4. Mechanism of bone erosion by the cholesteatoma: When osteolysis seems to be from the pressure of an expanding cholesteatoma sac, one should decompress and exteriorize the sac, and leave the matrix down. This condition occurs with noninfected cholesteatomas in which the keratin debris does not liquefy and remains impacted within the sac. In contrast, it may be better to remove the cholesteatoma if it seems to be destroying bone through biochemical mechanisms. There usually is granulation tissue in these cases. The
enzyme collagenase, which can cause bone resorption, exists in cholesteatoma and in granulation tissue, but its activity increases when both are present.18-20 Sheehy and Brackmann7 have an eclectic approach to lateral canal fistulas. They consider all of the factors mentioned by Gacek5 except for the final one about mechanisms of bone erosion. They prefer the intact canal wall tympanomastoidectomy with removal of the matrix over the fistula in a second operation 4 to 6 months later, after the ear is healed, and any infection is gone. They remove the matrix at the first stage in selected cases. It must be a small lateral canal fistula in a noninfected ear with normal bone conduction; however, they do not define “small.” When there is a fistula in a much better hearing ear, Sheehy and Brackmann,7 similar to Farrior,12 do a canal wall down procedure to avoid hearing loss or a second operation. They leave the matrix over the fistula when doing this. Sanna and associates3 and Gormley,10 reporting on Smyth’s series of labyrinthine fistulas, have nearly the same approach as Sheehy and Brackmann.7 They prefer a staged intact canal wall tympanomastoidectomy with matrix removal from the fistula during the second surgery. Sanna and associates3 do a canal wall down mastoidectomy without matrix removal in patients with an only hearing ear, in patients with a large external auditory canal defect, in patients with multiple fistulas, or in
Chapter 19 • Complications of Surgery for Chronic Otitis Media
elderly patients. In contrast, Ostri and Bak-Pedersen9 encourage matrix removal, even with large fistulas, using a one-stage intact canal wall tympanomastoidectomy. Ostri and Bak-Pedersen report that they have hearing improvement in most of their cases, and that anacusis is uncommon. There are a variety of surgical procedures for labyrinthine fistulas. These procedures range from always exteriorizing the disease and leaving matrix over the fistula to removing the matrix over large fistulas during one closed mastoid operation.8,9 Other authors have an approach between these extremes.3,5,7,10,12,21 Several technical points warrant mention. As stated, matrix removal should be at the end of the surgery. One should see if the matrix is attached to the underlying membranous labyrinth, and if so, stop dissecting.5 Conversely, Bellucci22 does not think it can get attached, and that it always can be elevated under high magnification. Bellucci believes that damage to the labyrinth results directly from instruments such as the suction. Farrior12 agrees and believes that no instrument should touch the fistula. Instead, he recommends using a strip of cellulose sponge for this part of the dissection. It is soft, but also slightly abrasive, which helps with removal from the bone. Irrigation should be gentle, but it is still important to prevent contamination by infected debris. A new technique for fistula repair uses hydroxyapatite bone cement. The matrix is elevated off the fistula, while observing the previous technical points. One immediately places fascia and then some cement over the site. The hardened cement is covered with another layer of fascia to promote coverage with mucosa (if an intact canal wall mastoidectomy) or epithelium (in a canal wall down procedure). We have found that hydroxyapatite reconstruction makes the vestibular apparatus less susceptible to barometric changes, such as with sneezing or the Valsalva maneuver. Hearing preservation is an important goal of labyrinthine fistula treatment. Postoperative hearing results vary among authors, and direct comparisons are impossible because of differences in surgical approaches and disease extent. Gormley10 reported a 3.3% incidence of anacusis in ears that were functioning before surgery. Ostri and Bak-Pedersen9 had a 5% incidence, whereas Sanna and associates3 separated their results between open (9.5%) and closed (2.6%) mastoidectomy groups. Gacek5 had a 14% incidence of anacusis, all in ears in which he attempted matrix removal from a cochlear fistula. Sheehy and Brackmann7 noted severe or total loss of hearing in 8% of lateral canal fistulas, and in 56% of fistulas involving other sites—predominantly cochlear fistulas. Law and colleagues21 reported postoperative anacusis in 22% of their cases. Decreased postoperative sensorineural levels or discrimination scores without total hearing loss are reported as follows: Sanna and associates (4% with closed mastoidectomy, 6.4% with open mastoidectomy),3 Gacek (7%),5
231
Ostri and Bak-Pedersen (17%),9 and Ritter (37%).8 Sensorineural hearing loss from surgery is common, and this must be explained beforehand to any patient who may have a fistula. Finally, extensive labyrinthine destruction does not invariably cause hearing loss. Phelps23 described a patient with an extensive cholesteatoma involving the vestibule along with the basal and middle turns of the cochlea who still had hearing, but it was lost during surgery. We saw something similar in four patients with slow-growing cholesteatomas. Bumstead and coworkers24 reported four cases of extensive labyrinthine destruction in which vestibular function was gone, but the hearing remained, even after surgery. They suggested that the inflammatory response sealed the cochlea and protected it. There is at least one report of a labyrinthine fistula with preoperative anacusis and a large hearing gain after surgery.25 The hearing loss was thought to be from serous labyrinthitis that later resolved.
IATROGENIC LABYRINTHINE FISTULA Although iatrogenic labyrinthine injury and hearing loss is always possible, it is infrequent. The three sites at the most risk during chronic otitis media surgery are the lateral semicircular canal, promontory, and oval window.26 Similar to a cholesteatoma fistula, an iatrogenic breach of the membranous labyrinth is treacherous, but the hearing is not always lost. Multiple case reports from the fenestration era show this to be the case.27-30 That said, the surgeon must recognize and repair the injury. Palva and associates31 reported iatrogenic injury to the lateral canal in the absence of a cholesteatoma fistula in 0.1% of chronic otitis media cases (2 of 2192). One case had stable hearing, whereas the other had an approximately 20 dB bone conduction loss in the speech frequencies. Jahrsdoerfer and colleagues32 described two episodes of drilling into the lateral canal without hearing loss. A more recent review by Canalis and coworkers33 reported a 0.08% incidence of this mishap, similar to Palva and associates.31 They had several observations about this injury.33 There was an acute, moderately severe sensorineural loss in cases that ultimately had a favorable outcome—hearing returned close to preoperative levels after 3 to 6 weeks. A permanent mild high-frequency loss was common, but speech discrimination scores returned to normal. Tinnitus was rare. Vertigo and nystagmus occur as anticipated, but commonly last beyond the usual period of central compensation. Patients sometimes have unsteadiness, positional vertigo, and spontaneous nystagmus 1 to 2 years after surgery. This is consistent with a partial labyrinthine injury along with continued neurosensory activity. How is hearing preservation possible with labyrinthine injuries? There are theories, but no conclusions. Canalis and coworkers33 believe that there is a better
232
OTOLOGIC SURGERY 4.
3.
9.
6.
2.
I 10. 5. 1. 7. 1. 2. 3. 4. 5.
Scutum (partially removed) Promontory Malleus TM flap Facial n.
8. 6. 7. 8. 9. 10.
Chorda tympani (moved inferiorly) Annular lig. (stapes) Pyramidal eminence Cholesteatoma matrix Stapes tendon
FIGURE 19-4. Correct method of cholesteatoma removal from the stapes. Dissection should be from posterior to anterior, allowing the stapedius tendon to stabilize the stapes. TM, tympanic membrane.
outcome when the lateral canal is opened on its posterior limb where symptoms may be from hemorrhage and serous labyrinthitis. Injuries near the vestibule would be more likely to cause contamination of the perilymph by endolymph, which causes permanent damage. They suspect that their cases of cochlear preservation were due to local changes at the fistula. The walls of the membranous labyrinth might collapse from a loss of fibrovascular support within the perilymphatic space. This, along with a sudden loss of endolymph, collapses the membranous labyrinth and seals it. Jahrsdoerfer and colleagues32 have a different theory. They think there is a closure of the pars superior from the pars inferior near the utriculoendolymphatic valve. An acute loss of endolymph from the vestibule causes the utricular wall to collapse, closing the valve area and protecting the cochlea from decompression. They recommend sealing iatrogenic fistulas with a fascia plug. Bone wax34 and muscle plugs33 are also used. Accidental opening of the inner ear is more common at the oval window than at the lateral semicircular canal. Palva and associates2 had a 1.4% incidence of this iatro genic fistula, and of their 12 cases, 11 occurred at the oval window.2 Two had complete avulsion of the footplate, three had dislocation, and the rest were fractures without dislocation. There were no drastic hearing losses, despite preoperative infections, but three cases did have depressed bone conduction thresholds several months later. In a 5-year follow-up report, Palva and associates31
opened the oval window once in an additional 1362 cases, for an overall incidence of 0.5%. This is still higher than the rates of lateral semicircular canal injury mentioned earlier. The relative retention of hearing in most iatrogenic fistulas at the oval window supports the findings of an earlier report by Weichselbaumer.35 He had 27 cases of accidental fenestration of the oval window, but no dead ears. Only two had new sensorineural losses—each by 20 dB. Of his cases, there was complete footplate avulsion in 11, partial avulsion in 12, and footplate perforations in the remainder. Likewise, Sheehy and Brackmann,7 in their report on labyrinthine fistulas from cholesteatoma, accidentally opened the oval window in 11 cases. These events did not cause sensorineural impairment. We offer several guidelines to protect the oval window, as follows: 1. Remove cholesteatoma by dissection parallel to the stapedius tendon to steady the stapes (Fig. 19-4).12 Manipulations of the stapes in a superoinferior direction or depressing it can disrupt the annular ligament (Fig. 19-5). 2. Be cautious about removing any granulation tissue or inflamed mucosa from around the stapes. This area naturally heals when the rest of the middle ear is disease-free and sealed by tympanoplasty.36 When needed, a laser is useful for atraumatically removing tissue from the stapes and oval window.37,38
Chapter 19 • Complications of Surgery for Chronic Otitis Media
233
Correct
Stapes tendon Incorrect
A Stapes is dislocated out of oval window Annulator lig. (stapes) Oval window
B FIGURE 19-5. A and B, Comparison of correct and incorrect cholesteatoma removal from the stapes. With incorrect removal, the stapes is not stabilized by stapedius tendon and is at risk for dislocation.
3. If the dissection becomes difficult, leave the cholesteatoma, reconstruct the tympanic membrane, and perform a second stage operation in 6 to 9 months. Any inflammation should be gone, and there usually is an epithelial pearl that is easily removed. In addition, there is less risk of inner ear contamination and hearing loss if a fistula occurs during the second stage.12 4. Immediately cover any opening with fascia, avoiding any suction. Palva and associates31 recommend delaying ossicular reconstruction until the area heals.
SENSORINEURAL HEARING LOSS Opening the labyrinth, as discussed earlier, can cause hearing loss, but hearing loss also may result from excessive ossicular manipulation, drilling on the ossicles, noise, and other unidentified factors. Smyth39 reviewed his series of 3000 chronic ear operations and found sensorineural hearing loss in 2.5%. He defined significant sensorineural depression as a 10 dB decrease from 500 to 4000 Hz, or a 10% reduction in speech discrimination. In reviewing his tympanoplasties without mastoidectomy, he found sensorineural hearing loss to be twice as high with ossicular chain disarticulation (2.6%) than without it (1.3%).
Smyth39 thought this difference was due to intentional disarticulation when removing tympanosclerotic plaques. The incidence of sensorineural depression reversed in cases of intact canal wall tympanomastoidectomy with a facial recess approach. It was higher for intact chain cases (5.6%) than for disarticulated chain cases (2.5%). Hearing loss in the intact chain cases occurred-equally from drilling on the incus and excessive malleus manipulation of the malleus during cholesteatoma removal (Fig. 19-6). Smyth thought that most losses with a disarticulated ossicular chain were due to excessive manipulation of the stapes during disease removal or ossicular reconstruction. With canal wall down procedures, there was a high incidence of hearing loss (50%) when the ossicular chain stayed intact, but there were no losses when it was disarticulated. Palva and associates36 reported their incidence of sensorineural hearing loss after chronic ear surgery to be 4.5% of 1680 cases. Most of these were high frequency, in the range of 4000 to 8000 Hz, and only 8% were across all test frequencies. All ears had some recovery in the first 3 postoperative months, but there were no appreciable changes afterward. Discrimination scores remained greater than 80% if the hearing loss was at 2000 Hz or more, and varied between 50% and 80% if the low frequencies were affected. Palva and associates36 found that most losses
234
OTOLOGIC SURGERY
Vibration of burr is transmitted to incus
FIGURE 19-6. Inadvertent contact between the burr and an intact incus. This could cause dislocation of the incus, or vibratory transmission to the stapes if the incudostapedial joint is intact.
(81%) occurred in ears with intact ossicular chains, and when there was cholesteatoma in the epitympanum. These studies allow some generalizations about protecting hearing, as follows: 1. When removing cholesteatoma from the malleus handle, dissect parallel to it, and in a slow, deliberate manner. This moves the stapes in a longitudinal direction, decreasing inner ear fluid displacements.31 Slow dissection allows time for the perilymph to pass through the helicotrema toward the round window without damaging the organ of Corti.39 2. Dissect cholesteatoma off the lateral surface of the incus body and short process in an anteroposterior direction, but in a superoinferior (parallel) direction along the long process. This gives the incus stability from its posterior and lateral suspensory ligaments, and from its connections to the malleus head.12 If dissection in the epitympanum is required, it is safest to disarticulate the incudostapedial joint first.36 3. Vibratory transmission to the ossicular chain and dislocation of the incus are prevented by careful drilling near the aditus ad antrum. Raise and lower the irrigant level to identify the incus promptly by its reflection (the water sign).2 4. Remove disease from the stapes and oval window as described earlier. Additionally, be cautious about removing tympanosclerotic plaques in the oval window. Delaying stapedectomy until a second stage decreases inflammation and the risk of cochlear injury, but many of these patients are better served by a hearing aid.39
Using these guidelines, Palva and associates31 had no further episodes of high-frequency sensorineural hearing loss in a 3-year follow-up report of an additional 512 procedures for chronic otitis media. They did have a 0.2% incidence of unexplained anacusis. Noise from the drill and suction-irrigator might be a factor. Low-intensity noise can lead to spasm of the vessels in the zona arcuata of the basilar membrane in laboratory animals.40 Also the greater metabolic needs of hair cells in the presence of noise can cause ischemia and subsequent damage.39
FACIAL NERVE INJURY Iatrogenic facial paralysis is a crushing experience that is best avoided. The literature22,41-44 supports the precept that the best precaution is a thorough knowledge of temporal bone anatomy. The landmarks for the facial nerve and the possible variations in its course must be fixed in the surgeon’s mind. This requires an active pursuit of knowledge that is acquired only by habitual, meticulous dissections in the temporal bone laboratory, attentive surgery, and years of experience. We also have a low threshold for imaging the temporal bone, and then study the course of the facial nerve immediately before starting surgery. Without hesitation, we promote routine facial nerve monitoring during mastoid surgery. The incidence of iatrogenic facial paralysis from otologic surgery ranges from 0.6% to 3.6% and exceeds 5% in revision cases.26,43 There were only two cases of postoperative facial palsy in a series of 958 operations for chronic otitis media.42 Both had radical mastoidectomies.
Chapter 19 • Complications of Surgery for Chronic Otitis Media
235
Incus LSC VII skeletonized
FIGURE 19-7. Cutting burr injury to the mastoid segment of the facial nerve. Complete transection of the nerve is most common here. LSC, lateral semicircular canal.
Partial disruption
Complete disruption
VII
FIGURE 19-8. Partial disruption of the facial nerve with less than one
FIGURE 19-9. Complete transection of the facial nerve.
third of the diameter affected.
One patient had a dehiscent tympanic segment, but the other had normal anatomy. Both of these nerves regained full function several weeks later. The authors restated the need for a thorough knowledge of temporal bone anatomy, combined with careful dissection, because normal facial nerve landmarks can be distorted by disease.42 Similarly, near paranoia is required for revision cases because past alterations of the anatomy can be deceiving. The most frequent facial nerve injury is transection of the mastoid segment with a cutting burr (Figs. 19-7
to 19-9).41 A perfectly oriented surgeon can cause this injury because of a moment of careless technique. Drill control must be absolute, and one must move in the direction of the nerve rather than across it. In addition, one should drill under continuous irrigation to prevent thermal injury.43 Diamond burrs generate heat rapidly. Cool irrigation also is hemostatic to bleeding inflammatory tissue and improves visualization. If a facial nerve injury is recognized during surgery, the facial canal should be opened proximally and distally
236
OTOLOGIC SURGERY
Beveled edges of nerve graft
Optional tissue cover
Gr.
A
B
FIGURE 19-10. A and B, Great auricular nerve (Gr) interposition graft. The ends of the nerves are beveled to increase their surface area. No sutures are used.
to decompress and examine it. Paparella and colleagues44 suggest preserving the sheath, unless there is discontinuity, because extra dissection would provoke fibrous proliferation and compromise nerve regrowth. Other authors43 advise maximum decompression of the edematous segment by opening the sheath until normal-appearing nerve is visible proximal and distal to the injury. Several possibilities should be considered when a patient unexpectedly awakens with a facial paralysis. Any intramastoid packing that could put pressure on the nerve needs to be loosened.43 Local anesthetic can cause a temporary facial paralysis and needs time to wear off. Surgical exploration should be considered if there is no improvement.
FACIAL NERVE GRAFTING Facial nerve rehabilitation depends on the type of injury. Neurapraxia resolves in time, but a transection never heals without surgical repair. The surgeon must understand the situation through objective studies and historical information from any previous surgeon. Neurapraxia is most likely if the prior surgeon inspected the nerve and was confident that it was not transected. An expectant course based on electric studies may be followed. Operative examination should be considered if there are doubts concerning the trauma, or if there is electric silence 3 days beyond the event. Most authors believe that if the diameter of the disruption is one third of the nerve or greater, facial function is ultimately better with resection of the injured segment and anastomosis (Fig. 19-10).45-48 Much has been written— and sometimes refuted—about the timing and technique of nerve repair over the years. Controversies remain.
Most recommendations are based on basic science studies of nerve injury and regeneration. Multiple studies have shown that maximal axonal regenerative ability is present 21 days after injury.49-51 Previous authors advocated delaying repair for several weeks to allow maximum axonal regeneration to push the axons across the anastomotic site.52,53 More recent literature supports early repair within several days of injury, and certainly within 30 days.54-56 After deciding to resect the injured portion of the facial nerve, one must determine if rerouting or grafting is best suited for reconstruction. With rerouting, the labyrinthine and tympanic segments are removed from the fallopian canal to close the defect. This difficult dissection requires opening the labyrinthine segment without violating the labyrinth—and this often is impossible. In addition, the greater superficial petrosal nerve and vessels tether the facial nerve at the geniculate ganglion and must be cut. A maximum gap of 1 cm is the traditional cutoff point for rerouting and anastomosis,45 but anatomic studies show that a tensionless anastomosis is possible with defects of 17 mm.57 The senior author (R.J.W.) has limited success with this technique compared with grafting; however, some authors believe that the results are ultimately superior as long as there is a tensionless anastomosis.58 Rerouting disrupts the blood supply to a large segment of the nerve.45,59 Other authors think that interposition grafts have results equal to the results of end-to-end anastomosis, at least when the grafts used are greater than 1 cm.60 The ipsilateral greater auricular nerve is the donor site of choice for interposition grafting.43,45 It can make a 10 cm graft. The proximal-distal orientation of the graft is reversed to prevent regenerating axons from growing into endoneural tubes that exit before the end of the graft.61 The graft should be long enough to overlap the gap by 3 to 5 mm on each end.62
Chapter 19 • Complications of Surgery for Chronic Otitis Media
The use of epineural versus perineural coaptation is controversial, although perineural techniques are more common.63 The great auricular nerve has a fascicular pattern of axons, but the facial nerve changes from a monofascicular to a multifascicular pattern as it travels from the geniculate ganglion to the stylomastoid foramen.61 We find it extremely difficult, if not impossible, to see or align any fascicles of the nerve within its course through the temporal bone. We concede that a perineural repair might be possible distal to the stylomastoid foramen. Fisch and Lanser61 encourage removing 3 to 4 mm of epineurium from the ends of the graft and nerve stumps. This recommendation is based partly on the work of Berger and Millesi,64,65 who showed that connective tissue forms within several days of neural injury, and is capable of infiltrating distal endoneural tubules to the exclusion of regenerating axons. The main source of this fibrosis is not surrounding connective tissue, but the epineurium itself.64,65 Removing the epineurium also allows the surgeon to assess better the true cross section of the nerve. Nonetheless, we believe that the best way to prevent fibrosis is by avoiding any possible trauma to the nerve. We barely dissect the epineurium because the potential benefits seem offset by additional trauma. A clean, oblique cut should be made through ends of the facial nerve and cable graft. This cut increases the surface area of contact and allows more regenerating axons to cross the anastomosis.66,67 It is difficult to determine the best site to freshen the proximal nerve stump, and no adequate guidelines are available. In animals, axonal sprouting by the proximal facial nerve stump can be at least 10 mm proximal to the injury.54 Sutureless reapproximation is preferred within the temporal bone because no motion is present, and because it decreases trauma to the anastomotic sites.45,48 The fallopian canal naturally cradles the graft. This differs from extratemporal nerve grafting, where we place sutures through the epineurium to ensure close approximation. Some authors advocate using foreign materials, such as fibrin glue or bovine collagen membrane, to stabilize the anastomosis.45,61 Other materials that have been used to cover the anastomotic site include topical thrombin, absorbable gelatin sponge (Gelfoam), gold foil, and skin grafts.44,47 Brackmann46 recommended splinting the anastomosis with clotted blood without using any other supporting material. Others also rely on natural tissue adhesiveness for this.45,48 The senior author (R.J.W.) has a similar approach, believing that a close, tensionless approximation is possible in the fallopian canal without any special materials over the anastomosis—they are more likely to interfere with healing rather than help it. Even with the best reconstruction, only 30% to 50% of the original facial nerve fibers traverse the graft and innervate motor end plates.61,68 Fisch and Lanser61 stated that maximal restoration of facial movement after grafting does not exceed 75% of normal, which corresponds to a grade III on the House-Brackmann scale. Farrior12 also reported a 60%
237
to 80% recovery of facial function with meticulous repair. Some return of facial nerve function should be expected within 5 to 7 months after repair of the tympanic or mastoid segments.59 Synkinesis is always noticeable because the regenerated fibers intermix as they travel distally.
OSSICULOPLASTY COMPLICATIONS Ossicular reconstruction is discussed extensively in other chapters, and this discussion is limited to prosthesis extrusion. Various materials have been advocated for ossicular reconstruction, such as autologous tissue (bone or cartilage), polyethylene, polymaleinate ionomer, hydroxyapatite, and titanium.69-75 Multiple variables related to chronic otitis media cause extrusion, including underlying eustachian tube dysfunction, the extent and severity of mucosal inflammation or granulation tissue, the status of the native tympanic membrane, and the method of grafting employed. These variables are not completely controlled by the surgeon. Multiple techniques to inhibit extrusion have been proposed, including engaging the prosthesis beneath the malleus manubrium or using a cartilage platform over the head of the device, especially with polyethylene.73 Hydroxyapatite is currently the most popular allograft material. It has agreeable biocompatibility and does not need cartilage protection from the tympanic membrane, but we do use cartilage when the middle ear environment is hostile. The extrusion rate of this material is 4% to 7%, even with shortterm follow-up.72,74 Titanium prostheses are increasingly popular and are not immune to extrusion.75 An extruding prosthesis may be carefully observed in certain circumstances, including the dry, noninfected ear, without evidence of progressive sensorineural hearing loss or dizziness, as long as water precautions are observed. Infection that does not promptly resolve with antibiotics, granulation tissue around the prosthesis, or cholesteatoma recidivism requires intervention. This complication is addressed in the operating room; a partially extruded prosthesis should never be extracted in the office because of potential injury at the oval window or the facial nerve. The use of autologous tissue for ossicular reconstruction is not a new concept, but it is a good idea in a severely diseased ear. We sometimes use cartilage in various configurations, such as the double-cartilage block ossiculoplasty described by Luetje and Denninghoff.76 Good hearing results with essentially absent extrusion rates are reported.69,76,77 The incus and the malleus head are useful for similar purposes.
DURAL INJURY Normal dura is capable of supporting the cerebrum without bony support, so limited dural exposure usually does not require treatment.44 Dural injury is a more
238
OTOLOGIC SURGERY
Bony tegmen defect with exposed dura
FIGURE 19-11. Laceration of exposed middle fossa dura causing
serial lumbar punctures or a lumbar drain may be needed to reduce the intracranial pressure.22,44 Additional materials and techniques are used with large dural defects. Bellucci22 described using fascia, skin, and bone to repair the defect through a middle fossa craniotomy (Fig. 19-13). He used a tight intramastoid packing to keep the grafts firmly against the dura. Neely and Kuhn79 used cartilage/perichondrial grafts to repair middle fossa defects. Kamerer and Caparosa80 also recommended packing the mastoid cavity—for up to 3 weeks—to support large repairs. They used perioperative antibiotics when doing this. Extreme cases require rotational flaps, free flaps, or fat obliteration.80 Hydroxyapatite bone cement is a promising adjunct that simplifies cranial defect repairs.81,82 It is biocompatible, is easy to mix and work with, provides structural support, and forms a watertight seal. It must be used correctly to work. Cerebrospinal fluid egress must be temporarily stopped, or the cement will wash away before it sets. We use our usual tissue plugs for this, and dry the area with cotton pledgets. The cement is covered with fascia after it is set so that mucosa will grow over it (see Fig. 19-12). We now prefer this for most repairs, and rarely resort to craniotomies, flaps, or mastoid packings.
cerebrospinal fluid leak.
VASCULAR INJURY serious matter. This injury became less frequent with the microscope and refinements in operative techniques.78 It is sometimes caused by indiscriminate cauterization of dural vessels, and this is avoided by using only bipolar cautery on the dura. The drill or a misguided curette causes more direct damage.22 Cerebrospinal fluid is usually seen leaking from the opening (Fig. 19-11), but small injuries to the middle fossa dura are sometimes concealed by the abundant arachnoid tissue in this area. Any dural injury should be repaired as soon as it is noticed. With middle fossa dural injury, Paparella and colleagues44 recommend removing the surrounding 5 mm of bone to inspect the dura and the brain. A fascia graft may be placed between the normal dura and the surrounding bone. A linear defect sometimes can be closed with sutures. Macerated muscle gently placed into the defect stops most leaks (Fig. 19-12), and this is further secured with an overlaying suture through the edges of the dura. A limited injury to the temporal lobe is treated similarly, but we encourage neurosurgical consultation in all cases of brain injury because life-threatening complications are possible. Posterior fossa dural injuries are more problematic because the brain and arachnoid tissue less readily facilitate spontaneous repair. More profuse cerebrospinal fluid leaks are likely. Repair involves exposing the margins of normal dura to place a graft.44 The patient should have a few days of bed rest with the head elevated to reduce the pressure on the repair until it seals. If the leak persists,
The three major vascular structures at risk during chronic otitis media surgery are the sigmoid sinus, jugular bulb, and internal carotid artery. Iatrogenic injury is more likely with anatomic variations, a poorly aerated mastoid, and revision surgery. Venous injuries are the most common.22
Sigmoid Sinus The sigmoid sinus is normally encountered during mastoidectomy. It is intradural, as are all intracranial sinuses, and is formed by two layers of dura that split to envelop it. The outer layer is the thinner of the two, and is approximately half the thickness of the intracranial portion.83 The walls of the sinus lack smooth muscle, so it cannot collapse without removing the overlying bone and compressing it. It is best identified by a bluish coloration beneath an intact bony plate, or by an increase in the pitch of the burr as it touches the compact bone overlying the sinus. The most common anomaly of the sigmoid sinus is anterior displacement into the mastoid cavity.26 Injuring this anomaly can be avoided by identifying the sinus in the sinodural angle and following it inferiorly.26 The mastoid emissary vein enters the upper third of the sigmoid sinus along its posterolateral side. Multiple veins are occasionally found. Bleeding is prevented by elevating the soft tissue flap from superior to inferior and identifying any vein in the field. The surgeon cauterizes it with a bipolar, cuts it, and obliterates the intramastoid foramen with bone wax.84 Multiple veins also are encountered
239
Chapter 19 • Complications of Surgery for Chronic Otitis Media Fascia (over bone cement)
Bone Dura
Bone cement over muscle plug
Fascia Cement Muscle plug
CSF leak stopped with muscle plug
Bone cement used to repair tegmen defect. All bone cement is covered with fascia.
FIGURE 19-12. Cerebrospinal fluid (CSF) leak stopped with a muscle plug. Bone cement is used to repair the tegmen defect. All bone cement is covered with fascia.
Repair through middle cranial fossa approach. Retractor
Forceps Temporalis m. Fascia bone graft
E.A.C.
Site of mastoidectomy
Tegmen defect
Bony scutum of EAC
FIGURE 19-13. Repair of a large dura and tegmen defect with bone and
Sigmoid sinus laceration during mastoidectomy
fascia placed from a middle fossa approach. EAC, external auditory canal.
FIGURE 19-14. Large sigmoid sinus laceration.
uring retrosigmoid bone removal. If one is in the way, d the surgeon exposes it with a diamond burr, coagulates it with a bipolar, and divides it near its junction with the sigmoid sinus. The surgeon obliterates the proximal foramina with bone wax.84,85 Similar maneuvers are used to control bleeding from an avulsed vein, but the bleeding is immediately covered with bone wax before further exposure.
Bleeding from the sigmoid sinus is minimal to profuse depending on the injury (Figs. 19-14 and 19-15). Oozing may be from small dural veins overlying the sinus. Small lacerations of the sinus wall itself can be caused by carelessly removing overlying bone fragments. These usually
240
OTOLOGIC SURGERY
vein in the neck is ligated to prevent embolization. The sinus lumen is packed inferior to the opening while controlling bleeding from above with extraluminal compression.85 Another method requires wide exposure of the dura surrounding the injury. The dura is punctured, and large hemoclips or sutures are used to ligate the sinus.86
Superior Petrosal Sinus The superior petrosal sinus is occasionally lacerated when the posterosuperior cell tract is drilled out in a wellaerated mastoid. The bleeding usually stops with bipolar cautery. If not, control is gained by exposing the sinus medial and lateral to the injury. The medial portion is obliterated by intraluminal or extraluminal techniques.85
Jugular Bulb
Surgicel placed intraluminally
FIGURE 19-15. Intraluminal obliteration of the sigmoid sinus. A tail of Surgicel remains outside to prevent embolization.
have sharp edges, or they can tear the sinus if tugged forcefully. This kind of bleeding is typically controlled with bipolar cautery or thrombin-soaked Gelfoam. Long segments of the sinus are cauterized in a stripping fashion instead of a spotting one.85 Monopolar electrocautery should not be used because it often enlarges the opening—and the current can thrombose the sinus. Larger lacerations require packing. One circumferentially removes bone from around the injury first. The simplest method is to obliterate the sinus extraluminally by placing a large sheet of absorbable knitted fabric (Surgicel) between the bone and the wall of the sinus. This can sometimes be done without completely obstructing the lumen, but it also can push the entire sinus away from the overlaying bone before stopping the bleeding. The sinus may also be obliterated intraluminally. Either way, a large sheet of Surgicel (4 × 4 cm) is needed to prevent embolization (see Fig. 19-15).85 Also, a tail of Surgicel is left outside the opening to prevent further migration and embolization. Another technique is to place a bolus of Surgicel directly over the injury and to secure it with bone wax. It may also be held in place with sutures that go through the dura on both sides of the sinus.85 This technique avoids total obstruction of the lumen. Massive sinus bleeding is controlled by obliteration above and below the injury. Temporary control is obtained with direct pressure, and the internal jugular
The jugular bulb has great anatomic variation. Its average dimensions are 15 mm wide × 20 mm high, but the latter varies up to 10 mm.83,87 The right jugular bulb is slightly larger than the left one in most people. The dome of the bulb is generally covered by bone and resides in the hypotympanum, but the more anterior the sigmoid sinus is, the higher the dome lies.83 One study showed the jugular bulb lying above the level of the inferior tympanic annulus in 6% of cases.88 A different study found dehiscent bone over the bulb in 7%.83 Injury to an exposed jugular bulb can occur during dissection of cholesteatoma from either the hypotympanic or retrofacial air cells. In addition, massive hemorrhage from the hypotympanum can occur simply by elevation of a tympanomeatal flap.83,89,90 An exposed bulb is especially prone to injury because the vessel wall is thin in this area. The endothelium of the sigmoid sinus is covered by dura, but this ends as the sinus leaves the posterior fossa. Inferiorly, a heavy adventitial layer surrounds the jugular vein after it leaves its foramen. These extra layers of support are missing over the jugular dome.83 Highresolution CT is the best way to image this area and to detect a high bulb before surgery.91 Bleeding from the jugular bulb may be managed in several ways: 1. Occlude small openings with bone wax or Gelfoam that is pressed into place with a cotton-tipped applicator or a cottonoid.84 2. Control larger lacerations by packing Surgicel between the bony defect and the bulb.85 3. Keep it in place with bone wax if you need to drill more. 4. Do not overpack the bulb—this would put pressure on cranial nerves IX, X, and XI. 5. If this is not enough, ligate the internal jugular vein in the neck, and obliterate the bulb and the sigmoid sinus with Surgicel.
Chapter 19 • Complications of Surgery for Chronic Otitis Media
Minor bleeding from a high jugular bulb injury during tympanomeatal flap elevation stops when covered with Gelfoam. The surgery often can continue. If flap elevation causes a rapid hemorrhage, one packs the middle ear with Gelfoam, replaces the flap, and firmly packs the external auditory canal with petrolatum gauze.83 This packing is left in place for 1 week, and then is removed over several days to a week.
Carotid Artery Carotid artery injury is rare in chronic ear surgery, but the artery is closely related to several middle ear structures. It can be exposed either congenitally or secondary to disease. The vertical portion of the petrous carotid normally lies slightly medial and anterior to the basal turn of the cochlea. At the genu where the artery turns into the horizontal segment, it lies anteroinferior to the cochleariform process, medial to the eustachian tube orifice, and anterior to the cochlea.92 Usually the petrous carotid artery is well covered by bone; however, this can be less than 0.5 mm thick near the eustachian tube orifice, and it is dehiscent in 1% of cases.93 An exposed artery must be kept in mind whenever dissecting in the anterior mesotympanum or eustachian tube. The lack of pulsations in the petrous carotid artery makes positive identification difficult.22 The petrous carotid artery rarely can take an anomalous course through the temporal bone.94-96 The artery normally is found medial to the vestibular line, which is a vertical plane that runs through the lateral aspect of the vestibule and approximates the level of the promontory.95 The anomalous vessel can be lateral to this line,95 and then the carotid foramen opens into the posterior hypotympanum.95,96 Internal carotid artery bleeding is often from the vasa vasorum of the artery, rather than from the lumen. A true rupture of the carotid wall causes rapid exsanguination, unless there is immediate control. This control is very difficult with the limited exposures of tympanomastoid surgery. The surgeon slows the bleeding with direct pressure and begins resuscitation while an interventional radiologist does a temporary balloon occlusion under fluoroscopic guidance. Repair can be with Nos. 7 to 10 polypropylene (Prolene) suture if the injury is a small laceration.84 Otherwise, the patient is transported to an interventional radiology suite for more sophisticated angiography. Permanent endovascular occlusion is acceptable if angiography shows adequate collateral circulation via the circle of Willis.97,98 There are now reports of using covered stents to control intracranial internal carotid artery injuries without vessel sacrifice.99,100 The damaged area can also be resected and grafted by a vascular neurosurgeon if permanent occlusion or stenting is impossible, but first broad exposure of the artery within the skull base must be provided.
241
SUMMARY We have discussed specific complications of chronic otitis media surgery. These problems are infrequent, but there is an element of difficulty when they happen because subsequent maneuvers are unfamiliar. The diversity of chronic otitis media and temporal bone anatomy adds to this. Each case is unique and requires special attention. A command of the anatomy and proper surgical technique guides the surgeon to that goal first stated by Hippocrates, “Above all, do no harm.”
REFERENCES 1. Sheehy J L : Surgery of chronic otitis media. In English G M (ed): Otolaryngology vol. 1, Philadelphia, Harper & Row, 1985, pp 1-86. 2. Palva T, Kårjå J, Palva A : Opening of the labyrinth during chronic ear surgery. Arch Otolaryngol 93:75-78, 1971. 3. Sanna M, Zinni C, Gamoletti R , et al: Closed versus open technique in the management of labyrinthine fistula. Am J Otol 9:470-475, 1988. 4. McCabe B F: Labyrinthine fistula in chronic mastoiditis. Ann Otol Rhinol Laryngol 112(Suppl):138-141, 1984. 5. Gacek R R : The surgical management of labyrinthine fistulae in chronic otitis media with cholesteatoma. Ann Otol Rhinol Laryngol 83(Suppl 10):1-19, 1974. 6. Wayoff MR, Friot JM: Analysis of one hundred cases of fistulas of the external semicircular canal. In McCabe BF, Sadé J, Abramson M (eds): First International Conference on Cholesteatoma. Birmingham, AL, Aesculapius Press, 1977, pp 463-464. 7. Sheehy J L , Brackmann D E : Cholesteatoma surgery: Management of the labyrinthine fistula—a report of 97 cases. Laryngoscope 89:78-87, 1979. 8. Ritter FN: Chronic suppurative otitis media and the pathologic labyrinthine fistula. Laryngoscope 80:10251035, 1970. 9. Ostri B, Bak-Pedersen K : Surgical management of labyrinthine fistulae in chronic otitis media with cholesteatoma by a one-stage closed technique. ORL J Otorhinolaryngol Relat Spec 51:295-299, 1989. 10. Gormley PK : Surgical management of labyrinthine fistula with cholesteatoma. J Laryngol Otol 100:1115-1123, 1986. 11. Sheehy JL, Brackmann DE, Graham MD: Complications of cholesteatoma: A report of 1024 cases. In McCabe BF, Sadé J, Abramson M (eds): First International Conference on Cholesteatoma. Birmingham, AL, Aesculapius Press, 1977, pp 420-429. 12. Farrior J B : Surgery for cholesteatoma. In Wiet R J, Causse J B (eds): Complications in Otolaryngology– Head and Neck Surgery vol. 1, Philadelphia, Decker, 1986, pp 69-76. 13. Walsh TE : Why I remove the matrix. Symposium on the Surgical Management of Aural Cholesteatoma. Trans Am Acad Ophthalmol Otolaryngol 57:687-693, 1953.
242
OTOLOGIC SURGERY
14. Baron S H : Why and when I do not remove the matrix. Symposium on the Surgical Management of Aural Cholesteatoma. Trans Am Acad Ophthalmol Otolaryngol 57:694-706, 1953. 15. Glasscock M E : The open cavity mastoid operations. In Glasscock M E, Shambaugh G E Jr (eds): Surgery of the Ear. Philadelphia, WB Saunders, 1990, pp 228-247. 16. Freeman P: Fistula of the lateral semicircular canal. Clin Otolaryngol Allied Sci 3:315-321, 1978. 17. Ruedi L : Acquired cholesteatoma. Arch Otolaryngol Head Neck Surg 78:252-261, 1963. 18. Abramson M: Collagenolytic activity in middle ear cholesteatoma. Ann Otol Rhinol Laryngol 78:112-124, 1969. 19. Abramson M, Gross J: Further studies on a collagenase in middle ear cholesteatoma. Ann Otol Rhinol Laryngol 80:177-185, 1971. 20. Abramson M, Huang CC : Localization of collagenase in middle ear cholesteatoma. Laryngoscope 87:771-791, 1977. 21. Law K P, Smyth G D, Kerr AG: Fistulae of the labyrinth treated by staged combined-approach tympanoplasty. J Laryngol Otol 89:471-478, 1975. 22. Bellucci R : Iatrogenic surgical trauma in otology. J Laryngol Otol 8(Suppl):13-17, 1983. 23. Phelps P: Preservation of hearing in the labyrinth invaded by cholesteatoma. J Laryngol Otol 83:1111-1114, 1969. 24. Bumstead R M, Sadé J, Dolan K D, McCabe B F: Preservation of cochlear function after extensive labyrinthine destruction. Ann Otol Rhinol Laryngol 86:131137, 1977. 25. Sheehy J L : Dead ear? Not necessarily: A report of three cases of chronic otitis media. Am J Otol 4:238-239, 1983. 26. Wiet R J, Herzon G D: Surgery of the mastoid. In Wiet R J, Causse J B (eds): Complications in Otolaryngology– Head and Neck Surgery vol. 1, Philadelphia, Decker, 1986, pp 25-31. 27. Cawthorne T: The effect on hearing in man of removal of the membranous lateral semicircular canal. Acta Otolaryngol 78(Suppl):145-149, 1948. 28. Jonkees L BW: On the function of the labyrinth after destruction of the horizontal canal. Acta Otolaryngol (Stockh) 38:505-510, 1950. 29. Sadé-Sadowsky N: The damage to the membranous labyrinth during fenestration and its influence upon hearing. J Laryngol Otol 69:753-756, 1955. 30. Thomas R : Fenestration operation: Experience of first ninety-six (consecutive) cases. J Laryngol Otol 65:259274, 1951. 31. Palva T, Kårjå J, Palva A : Immediate and short-term complications of chronic ear surgery. Arch Otolaryngol 102:137-139, 1976. 32. Jahrsdoerfer R A, Johns M E, Cantrell RW: Labyrinthine trauma during ear surgery. Laryngoscope 88:15891595, 1978. 33. Canalis R F, Gussen R , Abemayor E, Andrews J: Surgical trauma to the lateral semicircular canal with preservation of hearing. Laryngoscope 97:575-581, 1987. 34. Cullen J R , Kerr AG: “How I do it.” Iatrogenic fenestration of a semicircular canal: A method of closure. L aryngoscope 96:1168-1169, 1986. 35. Weichselbaumer W: The opening of the vestibule during tympanoplasty. Z Laryngol Rhinol Otol 44:457-464, 1965.
36. Palva T, Kårjå J, Palva A : High-tone sensorineural losses following chronic ear surgery. Arch Otolaryngol 98:176-178, 1973. 37. McGee TM: The argon laser in surgery for chronic ear disease and otosclerosis. Laryngoscope 93:1177-1182, 1983. 38. Parkin J L : Lasers in tympanomastoid surgery. Otolaryngol Clin North Am 23:1-5, 1990. 39. Smyth G D: Sensorineural hearing loss in chronic ear surgery. Ann Otol Rhinol Laryngol 86:3-8, 1977. 40. Lawrence M : In vivo studies of the microcirculation (with 16-mm color motion picture). Adv Otorhinolaryngol 20:244-255, 1973. 41. May M, Wiet R J: Iatrogenic injury—prevention and management. In May M (ed): The Facial Nerve. New York, Thieme, 1986, pp 549-560. 42. Vartiainen E, Kårjå J: Immediate complications of chronic ear surgery. Am J Otol 7:417-419, 1986. 43. Wiet R J: Iatrogenic facial paralysis. Otolaryngol Clin North Am 15:773-780, 1982. 44. Paparella M M, Meyerhoff WL , Morris M S, DaCosta S S : Mastoidectomy and tympanoplasty. In Paparella M M, Shumrick D A, Gluckman J L , Meyerhoff WL (eds): 3rd ed Otolaryngology vol. 2, Philadelphia, Saunders, 1991, pp 1405-1439. 45. Adkins WY, Osguthorpe J D: Management of trauma of the facial nerve. Otolaryngol Clin North Am 24:587611, 1991. 46. Brackmann D: Otoneurosurgical procedures. In May M (ed): The Facial Nerve. New York, Thieme, 1986, pp 589-618. 47. Kamerer D B : Intratemporal facial nerve injuries. Otolaryngol Head Neck Surg 90:612-615, 1982. 48. McCabe B F: Symposium on trauma in otolaryngology, I: Injuries to the facial nerve. Laryngoscope 82:18911896, 1972. 49. Grafstein B : The nerve cell body response to axotomy. Exp Neurol 48(Pt II):32-51, 1975. 50. McCabe B F: Facial nerve grafting. Plast Reconstr Surg 45:70-75, 1970. 51. Sunderland S : Some anatomical and pathophysiological data relevant to facial nerve injury and repair. In Fisch U (ed): Facial Nerve Surgery. Birmingham, AL, Aesculapius Press, 1977, pp 47-61. 52. Ducker TB, Kauffman FC: Metabolic factors in surgery of the peripheral nerves. Clin Neurosurg 24:406-424, 1977. 53. McQuarrie IG, Grafstein B : Axon outgrowth enhanced by previous nerve injury. Arch Neurol 29:53-55, 1973. 54. Barrs D M : Facial nerve trauma: Optimal timing for repair. Laryngoscope 101:835-848, 1991. 55. May M : Facial reanimation after skull base trauma. In May M (ed): The Facial Nerve. New York, Thieme, 1986, pp 421-440. 56. Bascom D A, Schaitkin B M, May M, et al: Facial nerve repair: A retrospective review. Facial Plast Surg 16:309313, 2000. 57. Yarbrough WG, Brownlee R E, Pillsbury HC : Primary anastomosis of extensive facial nerve defects: An anatomic study. Am J Otol 14:238-246, 1993. 58. Malik TH, Kelly G, Ahmed A, et al: A comparison of surgical techniques used in dynamic reanimation of the paralyzed face. Otol Neurotol 26:284-291, 2005.
Chapter 19 • Complications of Surgery for Chronic Otitis Media 59. Johns M, Crumley R : Facial Nerve Injury: Repair and Rehabilitation (SIPac). 2nd ed. Alexandria, VA, American Academy of Otolaryngology, 1977, p 9. 60. Fisch U, Rouleau M : Facial nerve reconstruction. J Otolaryngol 9:487-492, 1980. 61. Fisch U, Lanser M J: Facial nerve grafting. Otolaryngol Clin North Am 24:691-708, 1991. 62. Fisch U: Facial nerve grafting. Otolaryngol Clin North Am 7:517-529, 1974. 63. Orgel MG: Epineurial versus perineurial repair of peripheral nerves. Clin Plast Surg 11:101-104, 1984. 64. Berger A, Millesi H : Nerve grafting. Clin Orthop Relat Res 133:49-55, 1978. 65. Millesi H : Healing of nerves. Clin Plast Surg 4:459-473, 1977. 66. Yamamoto E, Fisch U: Experiments on facial nerve suturing. ORL J Otorhinolaryngol Relat Spec 36:193204, 1974. 67. Yasargil MG, Fisch U: Unsere Ertahrungen in der mikcrochirurgischen Exstirpation der Akustikusneurinome. Arch Ohrenheilk 194:243, 1969. 68. Ashur H, Vilner Y, Finsterbush A, et al: Extent of fiber regeneration after peripheral nerve repair: Silicone splint vs. suture, gap repair vs. graft. Exp Neurol 97:365-374, 1987. 69. Altenau M M, Sheehy J L : Tympanoplasty: Cartilage prosthesis—a report of 564 cases. Laryngoscope 88:895904, 1978. 70. Slater PW, Rizer FM, Schuring AG, Lippy WH : Practical use of total and partial ossicular replacement prostheses in ossiculoplasty. Laryngoscope 107:1193-1198, 1997. 71. McElveen JT, Feghali JG, Barrs D M, et al: Ossiculoplasty with polymaleinate ionomeric prosthesis. Otolaryngol Head Neck Surg 113:420-426, 1995. 72. Goldenberg R A : Hydroxylapatite ossicular replacement prosthesis: A four-year experience. Otolaryngol Head Neck Surg 106:261-269, 1992. 73. Brackmann D E, Sheehy J L , Luxford WM : TORPs and PORPs in tympanoplasty: A review of 1042 operations. Otolaryngol Head Neck Surg 92:32-37, 1984. 74. Black B : A universal ossicular replacement prosthesis: Clinical trials of 152 cases. Otolaryngol Head Neck Surg 104:210-218, 1991. 75. Ho SY, Battista R A, Wiet R J: Early results with titanium ossicular implants. Otol Neurotol 24:149-152, 2003. 76. Luetje C M, Denninghoff J S : Perichondrial attached double-cartilage block: A better alternative to the PORP. Laryngoscope 97:1106-1108, 1987. 77. Harvey S A, Lin SY: Double-cartilage block (DCB) ossiculoplasty in chronic ear surgery. Laryngoscope 109:911-914, 1999. 78. Glasscock M E, Dickins J R , Jackson CG, et al: Surgical management of brain tissue herniation into the middle ear and mastoid. Laryngoscope 89:1743-1754, 1979. 79. Neely JG, Kuhn J R : Diagnosis and treatment of iatrogenic cerebrospinal fluid leak and brain herniation during or following mastoidectomy. Laryngoscope 95:1299-1300, 1985. 80. Kamerer DB, Caparosa RJ: Temporal bone encephalocele—diagnosis and treatment. Laryngoscope 92: 878-881, 1982.
243
81. Catalano PJ, Post K, Sen C, et al: Prevention of cerebrospinal fluid rhinorrhea in neurotologic surgery. Am J Otol 21:265-269, 2000. 82. Kveton J F, Goravalingappa R : Elimination of temporal bone cerebrospinal fluid otorrhea using hydroxyapatite. Laryngoscope 110:1655-1659, 2000. 83. Graham M D: The jugular bulb: Its anatomic and clinical considerations in contemporary otology. Laryngoscope 87:105-125, 1977. 84. Leonetti J P, Smith PG, Grubb R L : Control of bleeding in extended skull base surgery. Am J Otol 11:254-259, 1990. 85. Moloy PJ, Brackmann D E : “How I do it.” Control of venous bleeding in otologic surgery. Laryngoscope 96:580-582, 1986. 86. Hotaling A J, Rejowski J E, Kazan R F, Wiet R J: Control of sigmoid sinus in glomus jugulare tumor resection. L aryngoscope 95:481-482, 1985. 87. Glasscock M E, Dickins J R , Jackson CG, Wiet R J: Vascular anomalies of the middle ear. Laryngoscope 90:77-88, 1980. 88. Overton S B, Ritter FN: A high-placed jugular bulb in the middle ear: A clinical and temporal bone study. L aryngoscope 83:1986-1991, 1973. 89. Hough JV: Congenital malformations of the middle ear. Arch Otolaryngol 78:335-343, 1963. 90. West J M, Bandy BC, Jafek BW: Aberrant jugular bulb in the middle ear cavity. Arch Otolaryngol 100:370-372, 1974. 91. Caldemeyer K S, Mathews VP, Azzarelli B, Smith R R : The jugular foramen: A review of anatomy, masses, and imaging characteristics. Radiographics 17:1123-1139, 1997. 92. Leonetti J P, Smith PG, Linthicum FH : The petrous carotid artery: Anatomic relationships in skull base surgery. Otolaryngol Head Neck Surg 102:3-12, 1990. 93. Myerson M D, Ruben H, Gilbert JG: Anatomic studies of the petrous portion of the temporal bone. Arch Otolaryngol Head Neck Surg 20:195-210, 1934. 94. Goldman NC, Singleton GT, Holly E H : Aberrant internal carotid artery presenting as a mass in the middle ear. Arch Otolaryngol 94:269-273, 1971. 95. Valvassori G E, Buckingham R A : Middle ear masses mimicking glomus tumors: Radiographic and otoscopic recognition. Ann Otol Rhinol Laryngol 83:606-612, 1974. 96. Lapayowker M S : Presentation of the internal carotid artery as a tumor of the middle ear. Radiology 98:293297, 1971. 97. Andrews JC, Valavanis A, Fisch U: Management of the internal carotid artery in surgery of the skull base. L aryngoscope 99:1224-1229, 1989. 98. de Vries E J, Sekhar L N, Janecka I P, et al: Elective resection of the internal carotid artery without reconstruction. Laryngoscope 98:960-966, 1988. 99. Leung G K, Auyeung K M, Lui WM, Fan YW: Emergency placement of a self-expandable covered stent for carotid artery injury during trans-sphenoidal surgery. Br J Neurosurg 20:55-57, 2006. 100. Kocer N, Kizilkilic O, Albayram S, et al: Treatment of iatrogenic internal carotid artery laceration and carotid cavernous fistula with endovascular stent-graft placement. AJNR Am J Neuroradiol 23:442-446, 2002.
20
Dural Herniation and Cerebrospinal Fluid Leaks Douglas A. Chen Videos corresponding to this chapter are available online at www.expertconsult.com.
An encephalocele occurs when brain tissue herniates through a dural defect of the skull. Temporal bone encephaloceles manifest either as a mass or cerebrospinal fluid (CSF) in the middle ear or mastoid or both (Figs. 20-1 and 20-2). Encephaloceles have also been called dural herniations, brain herniations, brain prolapse, and meningoencephalocele. They have been documented since the early 1900s.1,2 They were associated with chronic suppurative otitis media and transmastoid surgery, but, with a decreasing incidence of otogenic intracranial complications in modern-day patients, they have become uncommon. Contemporary otologic literature describes a myriad of signs and symptoms that are frequently not associated with suppurative disease.3 Other causes include congenital cranial base defects, spontaneous hernias, and trauma. Iurato and associates4 reviewed 139 cases of temporal bone encephaloceles, and found that 59% occurred as a complication of mastoid surgery, 21% were spontaneous or idiopathic, 9% were a complication of chronic otitis media or chronic mastoiditis, and 9% resulted from trauma. More recently, Scurry and coworkers5 reported encephalocele formation associated with morbid obesity in eight patients. The middle cranial fossa is the most common site of occurrence; encephaloceles rarely originate from the posterior fossa.6
PATHOGENESIS Basic knowledge of the embryology of the temporal bone is helpful in the understanding of formation of dural herniation and CSF leaks. There are four ossification centers that form the temporal bone: squamous, tympanic, petrous, and mastoid. Pneumatization follows ossification and continues into adulthood and consists of a process of marrow resorption, mucosal advancement, and bone remodeling. Disturbances in the normal ossification or pneumatization process may lead to encephalocele formation. Portions of the petrous and squamous ossification centers of the temporal bone ultimately form the roof of the middle ear and mastoid, areas that are the most likely to form encephalocele.7
Many authors have reported multiple tegmen defects (Fig. 20-3). Ahern and Thulen8 noted a 6% incidence of multiple defects in the tegmen in 94 consecutive cadaveric specimens. Ferguson and colleagues6 noted a 22% incidence of multiple tegmen defects in 27 preserved dried temporal bones. Lang9 found tegmen defects in 20% of 70 temporal bones. Bilateral spontaneous encephaloceles could be predicted from such studies, and have been reported by Iurato and colleagues.10 The high incidence of tegmen defects is contrasted by the low incidence of spontaneous encephalocele and CSF leak formation. This observation lends credibility to the widely held belief that bony defects are necessary, but not solely responsible, for the formation of encephaloceles. The other factor that is necessary for encephalocele formation is a pathologic process at the dural level. Inflammatory process, chronic suppurative otitis media, and transmastoid surgery are the most common factors in the dura. Other factors, such as benign intracranial hypertension, obesity, and aging, also have been suggested.5,11 The association with chronic suppurative otitis media and transmastoid surgery is well documented in the literature.4,12,13 In the preantibiotic era, meningeal complications from chronic suppurative otitis media were common. The incidence of encephalocele during this period was increased by transmastoid trephination of peridural and brain abscesses. Encephalocele can occur in the face of chronic suppurative otitis media with or without surgery, with or without cholesteatoma.14,15 Paparella and coworkers16 reported 10 cases of encephalocele all associated with chronic otitis media, 6 of which had not had previous surgery, and 2 of which had cholesteatoma. These authors speculated that the inflammatory response led to destruction of the tegmen, followed by extension to adjacent dura and brain, resulting in herniation. Encephalocele formation can form when a dural tear occurs at the site of a bone dehiscence from a drill, scalpel, or cautery.12 Dural injuries may go unrecognized during the surgery; however, if recognized at the time of surgery, they should be repaired to prevent an immediate or delayed problem. Although chronic suppurative otitis media and transmastoid surgery are still the 245
246
OTOLOGIC SURGERY
FIGURE 20-1. Temporal lobe herniation into the mastoid cavity through tegmen defect.
most common causes for encephalocele and CSF leak, a decreasing incidence has been cited because of improved surgical techniques, including the operating microscope and the use of drills versus gouges and chisels.17 CSF leak secondary to head trauma is well documented. Dura in the skull base that is adherent to adjacent bone tears along fracture lines. CSF leak usually resolves spontaneously or with lumbar drainage within 1 to 2 weeks. Savva and coworkers18 managed 26 of 29 patients with CSF otorrhea secondary to head injury successfully with conservative treatment. In contrast, only 1 of 53 patients with surgically induced or spontaneous CSF leak was successfully managed conservatively. Bony dehiscence and dural injury from head trauma can lead to either immediate or delayed encephalocele formation. Spontaneous encephalocele and CSF leak has been attributed to multiple factors. Arachnoid granulations of the temporal bone may be a cause of spontaneous encephalocele and CSF fistula formation. Normally, arachnoid granulations protruding into the lumen of venous structures are involved in the resorption of CSF. When not associated with venous structures and surrounded by bone, they may enlarge and cause bone erosion as a result of intermittent subarachnoid pressure associated with age and physical activity. Gacek19 observed in temporal bones that in the tegmen tympani and mastoideum 22%
had tissue on the dural surface, and 9% had tissue in the posterior fossa. Rao and colleagues11 reported a series of 10 patients with spontaneous CSF leak with an average age of 58 and an increased incidence in women. They hypothesized that the increased incidence in women suggested an association with benign intracranial hypertension, although many of their patients did not have clinical findings. Scurry and colleagues5 reported an association of morbid obesity in eight patients with spontaneous CSF leak. They also posited a theory that obesity ultimately resulted in increased intracranial pressure and subsequent CSF leakage and encephalocele formation. Many of their patients did not have findings consistent with increased intracranial pressure, which questions the idea that it plays a significant role.
DIAGNOSIS Diagnosis of encephalocele and CSF leak is primarily clinical with supportive radiologic and laboratory information. Clinical findings may or may not include a history of chronic suppurative otitis media or trauma. Middle ear effusion with conductive hearing loss is the most common presenting symptom.19-21 A subsequent
Chapter 20 • Dural Herniation and Cerebrospinal Fluid Leaks
247
FIGURE 20-4. Coronal CT scan with bone defect in the tegmen (arrow).
FIGURE 20-2. Coronal cut showing encephalocele through tegmen.
FIGURE 20-3. View of middle cranial fossa floor showing frequent sites of bony defect.
myringotomy and tube results in a persistent clear, odorless, watery type of discharge.22-24 Leakage may vary with various maneuvers that increase intracranial pressure, such as supine positioning, Valsalva maneuver, and compression of the jugular vein.25 Leaning forward may elicit rhinorrhea. A mass in the ear canal or behind the tympanic membrane is a rarely described finding.26,27 A more common finding is a soft compressible mass in the mastoid cavity arising off the tegmen that may mimic a blue dome cyst or cholesterol granuloma.28,29 Less common presentations include meningitis and seizures.2,6,17,28,30-36 Tension pneumocephalus is another uncommon presentation of temporal bone encephalocele.37 Air enters the intracranial space through the bone defect when the
extracranial pressure exceeds the intracranial pressure. Air accumulates when a valve phenomenon allows air to enter the intradural space, but does not allow the air to escape. Tension pneumocephalus has been reported after insufflation of the external auditory canal.38,39 If enough fluid can be collected, various tests can help determine whether the fluid is consistent with CSF. A glucose value 60% of serum levels, a protein concentration of less than 200 mg/dL, and chloride level greater than normal serum levels are consistent with CSF. β2-Transferrin has replaced glucose, protein, and chloride as a more sensitive and specific test for CSF.40 β2-Transferrin is a protein specific to CSF that is produced by neuraminidase activity in the brain. Zapalac and coworkers41 retrospectively reviewed 52 cases of CSF leak and found the test to be 98% sensitive. The high sensitivity, specificity, and ability to detect small amounts in specimens contaminated with blood or mucus has made β2-transferrin a powerful diagnostic tool.42 Computed tomography (CT) is ideally suited to imaging the bony anatomy of the temporal bone. Highresolution axial and coronal CT scans are needed to define defects in the tegmen tympani or mastoideum (Fig. 20-4). CT is limited in its ability to distinguish the various soft tissue densities in the mastoid. CT with intrathecal injection of metrizamide can show CSF leak fistulas for diagnosis and localization. Other radiopaque agents used for intrathecal injection, such as fluorescein and indigo carmine, have been reported.43 The main limitation is that small or intermittent leaks can lead to false-negative testing. Magnetic resonance imaging (MRI) is superior to CT in differentiating encephalocele from other soft tissue densities in the temporal bone, such as effusion, cholesteatoma, and other lesions (Fig. 20-5).44 Bone detail is lacking, however. Shetty and colleagues45 prospectively studied the accuracy and sensitivity of high-resolution CT and MRI in localizing fistulas in patients with CSF rhinorrhea. CT revealed an accuracy of 93% and a sensitivity of 92%. MRI cisternography showed an accuracy of 89% and a sensitivity of 87%. Zapalac and coworkers41 retrospectively examined 52 patients with skull base CSF leak fistulas
248
OTOLOGIC SURGERY
and determined that the most cost-efficient algorithm for diagnosis is an initial β2-transferrin, followed by CT for preoperative localization. MRI cisternography was reserved for cases in which CT was nonlocalizing.
TREATMENT Most CSF leaks after head trauma resolve with conservative management.18 However, CSF leaks from encephaloceles need surgical intervention. A high incidence of
FIGURE 20-5. T1-weighted MR image shows soft tissue that is contiguous with temporal lobe herniating into the mastoid (arrow).
unsuspected encephaloceles was found in the series by Kamerer and Caparosa.12 They urged that all tissue adjacent to dural defects be submitted for histologic confirmation. Herniated brain is nonfunctional and can be amputated with bipolar cautery without neurologic sequelae. Histologically, it is consistent with neural tissue, scarring, and inflammation. A basic concept of encephalocele repair involves a multilayered closure including a soft tissue repair of the dura combined with a repair of the bone defect.46 Soft tissue repairs without bone repair result in recurrence in some cases.11 The dura can be repaired primarily with suture. Alternatively, fascia or dural substitutes can be used to repair the defect. Intradural repairs have the advantage of pressure gradients of CSF and brain pushing the graft against the dural defect. The disadvantage is increased intradural dissection and possible brain parenchyma or vascular injury. Extradural repairs have worked consistently and eliminate the intradural dissection and risk.47 Extradural graft repairs have the theoretical disadvantage of the graft being pushed away by CSF pressure; however, extradural graft migration has not been a source of failure, presumably because it is secured by being wedged between the bone repair and the expanding brain. Various materials have been used for the bone repair. Commonly, a piece of cortical bone is used to plug the bone defect.17,28,48 Some pliable materials, such as cartilage, can be insinuated intracranially through a transmastoid approach.49,50 Hydroxyapatite bone cement is a synthetic material that was approved by the U.S. Food and Drug Administration for repair of CSF leaks in 1996. Constantino and associates51 reported 21 patients with CSF leaks from various sites in the head and neck. Kveton and Goravalingappa52 reported on its use in the temporal bone, and specifically reported it being used successfully for encephalocele repair through
FIGURE 20-6. Tegmen defect repaired with fascia-bone-fascia technique via middle fossa craniotomy approach.
Chapter 20 • Dural Herniation and Cerebrospinal Fluid Leaks
249
FIGURE 20-7. Temporal lobe encephalocele reduced and repaired by fascia-bone-fascia technique.
FIGURE 20-8. Fascia “dumbbelled” through small dural defect.
a transmastoid approach. In addition, in the long-term 10-year follow-up study, Kveton and Coelho53 showed good stability with low infection and extrusion rates. The hydroxyapatite cement needs to be mixed with sodium phosphate or citric acid to begin the setting proc ess. When they are mixed thoroughly together over a 30 to 40 second period, the cement is in a semisolid state that can be shaped and molded to the defect. The cement sets up and hardens over 4 to 5 minutes even in wet environments. Care should be taken not to allow the cement to come into contact or drain onto the ossicular chain to prevent conductive hearing loss. Other synthetic materials have been used in the past, such as silicone elastomer (Silastic) sheeting, methyl methacrylate, and mesh, but may have a higher incidence of infection or extrusion.
Three surgical approaches have been described: transmastoid, middle fossa craniotomy, and combined transmastoid/middle fossa approach. The transmastoid approach is best suited for isolated tegmen mastoideum, tympani, and posterior fossa defects. Anterior medial sites are not accessible through this approach if hearing is to be preserved. A wide mastoidectomy is performed with large exposure of the tegmen and its dura. The presigmoid bone dura should also be exposed and examined. When the dura bony defect is isolated, the encephalocele is amputated with the bipolar cautery. The dural defect can be repaired with fascia that is insinuated through the bone defect intracranially but extradural.46 Cartilage or bone cement can be used to reconstruct the bone defect. In canal wall down mastoid cavity situations, alloplastic materials should not be used for reconstruction during
250
OTOLOGIC SURGERY
encephalocele repair because there is a higher incidence of infection and extrusion. The preferred approach by many neurotologists for many reasons is a combined mastoid and middle fossa craniotomy (Figs. 20-6 and 20-7). Surgical identification of the encephalocele is often easier from the mastoid side. Middle fossa dural dissection from the skull base floor can result in dural tears, making it difficult to distinguish the encephalocele process from iatrogenic dural tears. The middle fossa craniotomy allows a wide exposure for dural and bony defect repair, which is especially important with multiple defects that may be otherwise missed through a transmastoid approach. Lesions that herniate through the tegmen tympani into the epitympanum can be removed through a combined approach without disturbing the ossicular chain. The combined mastoid and middle fossa approach begins with a mastoidectomy. When the defect or defects are identified from the mastoid side, and the encephalocele is amputated, the skin incision can be extended onto the temporal area to expose the squamous surface of the temporal bone. A fascia graft from the temporalis muscle is obtained. A temporal craniotomy is performed. The temporal lobe is elevated, and the dual defect is identified. The bony defect is easily identified in the floor of the middle fossa, especially because it can now be seen communicating into the mastoid cavity. The defect is repaired with a dural repair (Fig. 20-8), bony defect repair, and another layer of soft tissue. Anterior medial petrous apex leaks can be approached with a middle fossa approach without mastoidectomy. The bone flap is replaced and secured with plating system. The skin incision is closed in layers. Lumbar drainage of CSF postoperatively is not routinely used.
REFERENCES 1. Caboche H: De la hernie cerebral dans les interventions intracraniennes dirigees contre les moyennes suppurees. Ann d Mal de Poriell du Laryngx 28:278-294, 1902. 2. Canfield R B : Some conditions associated with the loss of cerebrospinal fluid. Ann Otol Rhinol Laryngol 22:604, 1913. 3. Brown N E, Grundfast K M, Jabre A, et al: Diagnosis and management of spontaneous cerebrospinal fluid-middle ear effusion and otorrhea. Laryngoscope 114:800-805, 2004. 4. Iurato S, Ettorre GC, Selvini C : Brain herniation into the middle ear: Two idiopathic cases treated by a combined intracranial-mastoid approach. Laryngoscope 99:950954, 1989. 5. Scurry WC, Ort S A, Peterson WM : Idiopathic temporal bone encephaloceles in the obese patient. Otolaryngol Head Neck Surg 136:961-965, 2007. 6. Ferguson B J, Wilkins R H, Hudson W: Spontaneous CSF otorrhea from tegmen and posterior fossa defects. Laryngoscope 96:635-644, 1986.
7. Proctor B, Nielson E, Proctor C : Petrosquamosal suture and lamina. Otolaryngol Head Neck Surg 89:482-495, 1981. 8. Ahern C, Thulen C A : Lethal intracranial complication following inflation of the external auditory canal in treatment of serous otitis media and due to defects in the petrous bone. Acta Otolaryngol (Stockh) 60:407-421, 1965. 9. Lang DV: Macroscopic bony deficiency of the tegmen tympani in adult temporal bones. J Laryngol Otol 97:685688, 1983. 10. Iurato S, Colucci S, Davidson C, et al: Histopathology of spontaneous brain herniations into the middle ear. Acta Otolaryngol (Stockh) 112:328-333, 1992. 11. Rao A K, Merenda D M, Wetmore S J: Diagnosis and management of spontaneous cerebrospinal fluid otorrhea. Otol Neurotol 26:1171-1175, 2005. 12. Kamerer D B, Caparosa R J: Temporal bone encephalocele—diagnosis and treatment. Laryngoscope 92:878882, 1982. 13. Neely JG, Kuhn J R : Diagnosis and treatment of iatro genic cerebrospinal fluid leak and brain herniation during or following mastoidectomy. Laryngoscope 95: 1299-1300, 1985. 14. Alberti PWR M, Dawes J D K : Cerebrospinal otorrhea and chronic ear disease. J Laryngol Otol 75:123-135, 1961. 15. Dedo H H, Sooy FA : Endaural encephalocele and cerebrospinal fluid otorrhea: A review. Ann Otol Rhinol Laryngol 79:168-177, 1970. 16. Paparella M M, Meyerhoff WL , Oliviera C A : Mastoiditis and brain hernia (mastoiditis cerebri). Laryngoscope 88:1097-1106, 1978. 17. Glasscock M E, Dickins J R E, Jackson CG, et al: Surgical management of brain tissue herniation into the middle ear and mastoid. Laryngoscope 89:1743-1754, 1979. 18. Savva A, Taylor M J, Beatty CW: Management of cerebrospinal fluid leaks involving the temporal bone: Report on 92 patients. Laryngoscope 113:50-56, 2003. 19. Gacek R R : Evaluation and management of temporal bone arachnoid granulations. Arch Otolaryngol Head Neck Surg 118:327-332, 1992. 20. Gavilan J, Trujillo M, Gavilan C : Spontaneous encephalocele of the middle ear. Arch Otolaryngol Head Neck Surg 110:205-207, 1984. 21. Levy R A, Platt N, Aftalion B : Encephalocele of the middle ear. Laryngoscope 81:126-130, 1971. 22. Dysart B R : Spontaneous cerebrospinal fluid otorrhea—a report on a case with successful surgical repair. Trans Am Laryngol Rhinol Otol Soc 62:381-387, 1959. 23. Adams G L , McCoid G, Weisbeski D: Cerebrospinal fluid otorrhea presenting as serous otitis media. Minn Med 65:410-415, 1982. 24. Briant TD R , Bird R : Extracranial repair of cerebrospinal fluid fistula. J Otolaryngol 11:191-197, 1982. 25. Lalwani A K : Temporal bone encephalocele. In Jackler R K, Brackmann D E (eds): Neurotology. St Louis, Mosby, 1994, pp 1149-1155. 26. Lalwani A K, Jackler R J: Endaural encephalocele. Otolaryngol Head Neck Surg 106:309-310, 1992. 27. Jahrsdoerfer R A, Richtsmeier WJ, Cantrell RW: Spontaneous CSF otorrhea. Arch Otolaryngol 107:257-2261, 1981.
Chapter 20 • Dural Herniation and Cerebrospinal Fluid Leaks 28. Graham M D: Surgical management of dural and temporal lobe herniation into the radical mastoid cavity. Laryngoscope 92:329-331, 1982. 29. Baron S H : Herniation of the brain into the mastoid cavity: Post-surgical, postinfectional, or congenital. Arch Otolaryngol Head Neck Surg 90:127-133, 1969. 30. Kline O R : Spontaneous cerebrospinal otorrhea. Arch Otolaryngol Head Neck Surg 18:34-39, 1933. 31. Mealey J Jr: Chronic cerebrospinal fluid otorrhea—report of a case associated with chronic infection of the ear. Neurology 11:996-998, 1961. 32. Bowes A K, Wiet R J, Monsell E M, et al: Brain herniation and space occupying lesions eroding the tegmen tympani. Laryngoscope 97:1172-1175, 1987. 33. Feenstra L , Sanna M, Zini C, et al: Surgical treatment of brain herniation into the middle ear and mastoid. Am J Otol 6:311-315, 1985. 34. Moore G F, Nissen A J, Yonkers A J: Potential complications of unrecognized cerebrospinal fluid leaks secondary to mastoid surgery. Am J Otol 5:317-323, 1984. 35. Hyson M, Andermann F, Olivier A, Melenson D: Occult encephaloceles and temporal bone epilepsy: Developmental and acquired lesions in the middle fossa. Neurology 34:363-366, 1984. 36. Kemink J L , Graham M D, Kartush J M : Spontaneous encephalocele of the temporal bone. Arch Otolaryngol Head Neck Surg 112:558-561, 1986. 37. Cartwright M J, Eisenberg M B : Tension pneumocephalus associated with rupture of a middle fossa encephalocele. J Neurosurg 76:292-295, 1992. 38. Fairman H D, Brown NJ, Hallpike C S : Air embolism as a complication of inflation of the tympanum through the external auditory, meatus. Acta Otolaryngol 66:65-71, 1968. 39. Fisnes K A : Lethal intracranial complication following air insufflation with a pneumatic otoscope. Acta Otolaryngol 75:436-438, 1973. 40. Oberascher G: A modern concept of cerebrospinal fluid diagnosis in oto- and rhinorrhea. Rhinology 26:89-103, 1988.
251
41. Zapalac J S, Marple B F, Schwade N D: Skull base cerebrospinal fluid fistulas: A comprehensive diagnostic algorithm. Otolaryngol Head Neck Surg 126:669-676, 2002. 42. Rauch S D: Transferrin microheterogeneity in human perilymph. Laryngoscope 110:545-552, 2000. 43. Dedo H, Sooy F: Endaural encephalocele and cerebrospinal fluid otorrhea. Ann Otol Rhinol Laryngol 79:168177, 1970. 44. Kaseff LG, Seidenwarm DJ, Nieberding PH, et al: Magnetic resonance imaging of brain herniation into the middle ear. Am J Otol 13:74-77, 1992. 45. Shetty PG, Shroff M M, Sahani DV, et al: Evaluation of high-resolution CT and MR cisternography in the diagnosis of cerebrospinal fluid fistula. AJNR Am J Neuroradiol 19:633-639, 1998. 46. Lundy L B, Graham M D, Kartush J M, et al: Temporal bone encephalocele and cerebrospinal fluid leaks. Am J Otol 17:461-469, 1996. 47. Chen D A : Contemporary diagnosis and management of temporal bone encephalocele. In Arriaga M A, Day J D (eds): Neurosurgical Issues in Otolaryngology: Principles and Practice of Collaboration. Philadelphia, Lippincott Williams & Wilkins, 1999, pp 189-192. 48. Bartels L , Luk L J, Balis C : Endaural brain hernia: Repair using mastoid cortical bone. Am J Otol (Suppl):121-125, 1985. 49. Adkins WY, Osguthorpe J D: Mini-craniotomy for management of CSF otorrhea from tegmen defects. Laryngoscope 93:1038-1039, 1983. 50. Jahn A F: Endaural brain hernia: Repair using conchal cartilage. J Otolaryngol 10:471-475, 1981. 51. Constantino PD, Chaplin J M, Wolpoe M E, et al: Applications of fast setting hydroxyapatite cement: Cranioplasty. Otolaryngol Head Neck Surg 15:409-412, 2000. 52. Kveton J F, Goravalingappa R : Elimination of temporal bone cerebrospinal fluid otorrhea using hydroxyapatite cement. Laryngoscope 110(10 Pt 1):1655-1659, 2000. 53. Kveton J F, Coelho D H : Hydroxyapatite cement in temporal bone surgery: A 10 year experience. Laryngoscope 114:33-37, 2004.
21
Total Stapedectomy Howard P. House and Jed A. Kwartler
The goal of stapes surgery is to re-establish sound t ransmission through an ossicular chain stiffened because of otosclerosis. Various techniques have been used to accomplish this goal, including stapes mobilization, fragmentation, small fenestration, and partial and total stapes footplate removal. The history of surgery for otosclerosis began in the latter part of the 19th century. A group of pioneering surgeons, including Kessel,1 Boucheron,2 Miot,3 Faraci,4 and Passow,5 began to mobilize the stapes. At about this time, Jack6 reported on a series of cases in which he removed the stapes entirely. The unacceptably high rate of inner ear injury and infection led to the abandonment of stapes surgery. As stated by Goodhill,7 “it was probably Siebenmann8 along with Moure9 who closed the door on further stapes surgery at the turn of the century.” Surgery for otosclerosis was reactivated in 1923, when Holmgren10 bypassed the stapes area by creating a fenestra in the horizontal canal to stimulate inner ear fluids in response to sound, and the fenestration operation was reborn. In 1937, Sourdille11 presented a series of fenestration cases before the New York Academy of Medicine. In 1938, Lempert12 introduced his unique one-stage fenestration technique using his endaural approach and a dental drill to create the fenestra. Surgeons throughout the world clamored to learn his technique, which became the standard. Lempert will forever be known as the father of otosclerosis surgery. In 1952, Rosen13 reintroduced stapes mobilization for otosclerosis. For a brief time, this technique was widely used and threatened to replace the Lempert procedure. It was soon realized, however, that refixation of the footplate often occurred. Shea14 introduced his technique of total stapedectomy. After removing the total stapes, he covered the oval window with a vein graft and introduced an artificial stapes made of polytef (Teflon) by Treace to make the connection with the incus. This reactivation of stapedectomy by Shea replaced Lempert’s fenestration procedure and Rosen’s mobilization operation, and, with modification, is now used universally throughout the world. We are greatly indebted to Shea for his tremendous contribution to otosclerosis surgery.
PATIENT COUNSELING All patients should be told that even though their hearing loss is hereditary, their children or grandchildren will not necessarily have a similar problem. They should be assured they are not going to become totally deaf. The mechanics of the hearing loss should also be explained in detail, preferably by use of a suitable illustration. Patients who are suitable for stapes surgery should be told that they have the option of wearing hearing aids, and if there is any doubt about the decision to have surgery, they should be encouraged to have a trial period with hearing aids unless they are already wearing them. The expected hearing result and all possible risks, such as further or total hearing loss, taste disturbance, dizziness, the effect on tinnitus (if present), and the very remote possibility of a partial or total facial paralysis, should be clearly understood. Hearing improvement is in direct proportion to the preoperative bone conduction level, and the patient must understand the degree of improvement to be expected.
SELECTION OF PATIENTS FOR STAPES SURGERY All patients who are suitable for stapes surgery should be thoroughly informed of the advantages and the possible complications of the operation. For some patients, serviceable hearing can be restored with no need for a hearing aid. In others, the hearing can be improved so that they may need a hearing aid only for distant conversation. In still other patients, the hearing can be improved, and they may be able to convert their postauricular aid to an all-in-the-ear hearing aid or from an all-in-the-ear aid to an intracanal aid. Occasionally, patients have a totally blank audiogram and are still suitable for stapes surgery. This situation occurs when the bone conduction level exceeds the capability of the audiometer. There may be a 75 or 80 dB bone conduction level, but a 40 to 50 dB air-bone gap. Typically, when the patient is initially evaluated, he or 253
254
OTOLOGIC SURGERY
she is hearing surprisingly well with a powerful hearing aid and possesses excellent speech quality. On examination, one may note a positive Schwartze’s sign, but the 512 Hz tuning fork is not helpful. After surgery, these patients are grateful because they can now wear less powerful hearing aids with fewer feedback problems.
Indications for Stapes Surgery The patient should understand the details of the operation, including the operative procedure itself, and all admission and discharge procedures. The following principles also apply: 1. The patient should be in reasonably good health, especially if general anesthesia is contemplated. 2. The age of the patient is not a factor in the decision to perform surgery. The youngest in our series was 7 years old, and the oldest was 98. 3. The poorer hearing ear, based on the patient’s statement and not on the audiogram, should be chosen for surgery. In children, the poorer hearing ear should be operated on so that the hearing aid can be eliminated before they enter school. Surgery for the second ear should be delayed until they are old enough to make their own decision. 4. Tuning forks should be used to confirm the audiometric findings. If bone conduction is heard louder than air conduction with a 512 or a 1024 Hz tuning fork, with proper masking of the better hearing ear, the individual is a suitable candidate for surgery. If one reverses the 2048 Hz fork, he or she is an excellent candidate. The minimum air-bone gap should be 15 dB, as averaged in speech frequencies. 5. Speech discrimination is not a major factor in determining stapes suitability. If the patient understands sentences and answers questions correctly using a speaking tube while masking the opposite ear with a Bárány apparatus, he or she is suitable, provided that the earlier criteria are met. The improvement the patient receives with a cochlear implant substantiates the value of any sound that helps patients hear and react better to their environment. 6. Indications for stapes surgery are essentially the same whether the hearing loss is unilateral or bilateral. Surgery in the opposite ear can occur 6 months later, provided that it is then the poorer hearing ear.
Contraindications of Stapes Surgery Stapes surgery is contraindicated in patients (1) in poor physical health; (2) with a current balance problem, such as active Meniere’s disease or a fluctuating type of hearing loss; (3) with pre-existing tympanic membrane perforation; (4) with active external or middle ear infection; and (5) with an inadequate air-bone gap confirmed by an audiogram and the 512 Hz tuning fork.
SURGICAL TECHNIQUE Step 1 Stapes surgery may be performed with a local or a general anesthetic. We prefer local anesthesia with adequate sedation that may be supplemented during the procedure with intravenous midazolam (Versed) or diazepam (Valium), if necessary. We prefer local anesthesia because there is less bleeding, and the surgeon is alerted if any vertigo occurs while working on the footplate, inserting the prosthesis, or removing the prosthesis, in a revision case.
Step 2 During surgery, the patient’s vital signs are monitored by electrocardiography, blood pressure, and oxygen saturation. The auricle and the surrounding area are cleaned with povidone-iodine (Betadine) solution, and plastic drapes and folded towels are applied. A final head drape with an opening exposing the ear is placed over the head and rested on a metal support to help prevent the feeling of claustrophobia.
Step 3 The operating table is placed slightly in Trendelenburg’s position and rotated toward the surgeon so that he or she can see directly down the ear canal from the sitting position.
Step 4 The ear canal is washed with warm saline solution to remove the povidone-iodine, and local anesthesia containing 2% lidocaine in epinephrine 1:100,000 is infiltrated. The initial injections are made with a 30 gauge needle around the periphery of the entrance to the ear canal. Approximately 2.5 to 3 mL of this solution is injected, and 1 or 2 drops are placed in the vascular strip just external to the tympanic membrane. This helps reduce the bleeding at the time of the incision. The tissue to be used, whether vein, fascia, perichondrium, or fat, may be obtained before or after the canal surgery is started. Several sizes of oval and round specula should be on the tray, and the largest one that can be seated into the canal is used. The shafts of the instruments entering the speculum are in firm contact with the middle finger, which is stabilized against the speculum, and the speculum is stabilized by the other fingers against the head. A fixed speculum holder is not used. The advantage of not using a fixed speculum holder is flexibility of the speculum for viewing purposes and for allowing the patient to move his or her head, if desired. The specula and all instruments should be plain metal because black specula and instruments absorb much-needed light. The shafts of
Chapter 21 • Total Stapedectomy
the needles and hooks should be malleable so that they can be bent slightly to reach difficult areas. The inferior and superior vertical incisions are made at the 6:30 and 11:30 o’clock positions (Fig. 21-1). The point of the sickle knife is started 1 mm from the edge of the tympanic membrane to prevent a possible tear. It is extended externally approximately 8 mm, and this distance can be confirmed when the curve of the incision knife strikes the edge of the properly inserted speculum. If one extends the incisions further externally, the skin becomes thicker, and more bleeding occurs. Several sweeps are made to ensure that one cuts through the periosteum. The horizontal incision begins by elevation of the skin from the depth of the suture indentation, and continues inferiorly in short increments toward the inferior vertical incision to avoid tearing of the skin toward the eardrum. Several clean sweeps of the knife are again made to ensure that the periosteum that connects to the inferior vertical incision has been cut through. A similar superior incision is made in increments halfway to the superior vertical incision. Several sweeps are made on this partially completed incision. The knife is inserted beneath the remaining skin to elevate it. Scissors are used to connect with the end of the superior vertical incision. Scissors crush the vessels in the vascular strip and help reduce the bleeding. After these incisions, a broad separator is used to elevate the skin flap in a uniform manner toward the eardrum. Considerable pressure is applied on the instrument, especially inferiorly, to stay under the periosteum until it enters the middle ear area posterior to the ligament. A curved Rosen needle is used superiorly to elevate the eardrum and identify the position of the chorda tympani nerve (Fig. 21-2). When the nerve is identified, the needle is inserted superiorly to the nerve and carried forward to contact the malleus. This action provides the superior exposure. The needle is used inferior to the chorda tympani to identify the beginning of the tympanic membrane ligament. An elevator is used to lift the ligament inferiorly and to identify the round window. At this point, a cotton ball soaked in the lidocaine/epinephrine solution is placed on the raw surface of the skin flap to lessen the bleeding for a moment. The entire skin flap is elevated anteriorly, and a few drops of lidocaine/epinephrine are dropped onto the mucosa of the middle ear for anesthesia and to control any mucosal bleeding later as work is done in the stapes area. To visualize the footplate, bone of the posterior scutum is removed. At the upper limit of exposure superiorly, one should observe the lower half of the transverse portion of the fallopian canal (Fig. 21-3). If one can see the beginning of the curve of the body of the incus, a subsequent retraction pocket may develop. The posterior exposure is limited to observation of the stapedial tendon
255
and the attachment of the posterior crus of the stapes to the footplate. If more bone is removed, a posterior retraction pocket may develop. This exposure is usually done with curettes, but a diamond burr may be necessary to expose the posterior portion of the chorda tympani. On completion of this exposure, a square area is created posterosuperiorly. When the curette or the burr is used, the patient under local anesthesia should be forewarned of the noise that is created so there is no surprise to the patient. If the footplate area appears to be thin, a sharp needle is used to make a small perforating hole in the thinnest area. This small hole later provides an opening in which an obtuse hook can be inserted if the footplate is inadvertently mobilized at the time the crura are fractured toward the promontory. A small, round right angle knife is used now to separate the incudostapedial joint (Fig. 21-4). The intact stapedial tendon helps prevent an inadvertent mobilization from occurring. In this manipulation, a mild pressure and “jiggling” of the knife blade back and forth is used because if strong direct pressure is applied, dislocation of the incus may occur when the joint is suddenly separated. The necessary limits of exposure are the round window inferiorly, the lower half of the fallopian canal superiorly, the stapedial tendon pyramidal eminence and posterior crus posteriorly, and the malleus anteriorly (Fig. 21-5). The malleus is now checked to determine its mobility. One in 200 patients has a fixed malleus. If fixed, alternative techniques must be used (described later). The stapedial tendon is cut; then the patient is forewarned that a loud sound is forthcoming. The Rosen mobilizing needle is placed on the superior side of the stapes arch near the neck, and the superstructure is sharply fractured toward the promontory and removed (Fig. 21-6). The distance from the top of the incus to the thin, fixed footplate is measured (Fig. 21-7). Some surgeons measure from the inferior surface of the incus, and the prosthesis length is made accordingly. The prosthesis length should be checked before it is inserted. The measurement from the outer portion of the incus to the footplate is usually 4.5 mm, but may vary from 3.5 to 5.5 mm. The membrane is left intact over the footplate because it helps prevent bony chips from dropping into the vestibule. Obtuse, right angle hooks and finally the Hough hoe is used to remove the posterior and then the anterior portion of the footplate, effecting a total removal (Fig. 21-8). Great caution must be used to avoid suction of the perilymph when blood is suctioned from around the oval window. During footplate removal, bone chips or blood entering into the vestibule is left undisturbed. The previously prepared tissue of choice, measuring about 5 × 5 mm, is grasped with a nonserrated alligator forceps and slipped over the oval window so that it covers
256
OTOLOGIC SURGERY
FIGURE 21-1. Inferior and superior vertical incisions. FIGURE 21-2.� Elevating the skin flap. FIGURE 21-3. Middle ear exposure. FIGURE 21-4. Separating the incudostapedial joint.
Chapter 21 • Total Stapedectomy
all the edges and is positioned superiorly over a portion of the fallopian canal (Fig. 21-9). The prosthesis is inserted with a nonserrated forceps into the center area of the oval window and over the incus (Fig. 21-10). A crimper is used to close the loop on the incus, and the wire is moved toward the lenticular process. The skin flap is then placed back in its normal position, and a small gauze wick is inserted into the hypotympanic area. A cotton pledget is placed over the opening of the external ear canal, and a Band-Aid or two holds the cotton in position, completing the procedure. In our experience, whether one totally or partially removes the footplate or uses the small fenestra technique, or whether one uses the diamond burr or the laser to create the small fenestra, the end result is quite similar. More surgeons are using the small fenestra technique, and the postsurgery imbalance is less noticeable because perilymph disturbance is minimal. It is not the technique, the instruments, or a particular prosthesis that leads to a successful result, but rather the hands and mind behind the instruments. If one is closing the air-bone gap in 90% of the cases and encountering no more than a 1% severe sensorineural loss, one should stay with that technique.
POSTOPERATIVE CARE Some patients may have dizziness for a few hours after surgery. They are cautioned not to blow their nose hard. If patients sneeze, they should do so with their mouth open, and they should avoid excessive straining for 2 weeks. If operated on in the morning, the patient usually can go home that night and remove the Band-Aid, cotton, and gauze wick the following morning. Patients are allowed to travel by air 3 days after surgery and are instructed to use a nose spray and swallow frequently on descent. Most patients hear immediately after surgery, but their hearing level may decrease a few hours later. Barring complications, patients should have their first hearing test 3 weeks after surgery, at which time most have recovered their hearing. In others, the hearing improves over the next 3 months. The hearing they have at this time is what they will keep. We do not do revision surgery on any patient until 4 months have passed. All patients are routinely placed on a sodium fluoride and calcium carbonate supplement (Florical), 8 mg three times a day, and are followed up routinely on an annual basis.
INTRAOPERATIVE PITFALLS Chorda Tympani Nerve During the procedure, the chorda tympani nerve may be enlarged, or may be in a position to interfere with proper visualization of the stapes area (Fig. 21-11). The chorda tympani may be gently moved superiorly and inferiorly. If it is stretched to the point of a partial tear, it should be
257
severed rather than left partially functioning—the patient has less taste disturbance than when a partially functioning chorda is left intact. One should not sever the chorda if the opposite ear is operated on because a dry mouth and a severe taste disturbance may result.
Eardrum Perforation If a small perforation of the tympanic membrane occurs during surgery, it often heals if the edges are placed together. If a larger tear occurs, tissue is used with an underlay technique.
Malleus Fixation If the malleus is fixed, and the stapes is mobile, further stapes surgery is abandoned. The incudostapedial joint is carefully separated, and the incus is removed (Fig. 21-12). The head and neck of the malleus are exposed further anteriorly, and the Lempert snipper is used to sever the neck. Fixation usually results from tendon ossification, and tapping with a small chisel on the head of the malleus frees it, enabling removal. Continuity between the mobile stapes and the eardrum may be re-established by reconstruction. If the stapes is also fixed, an incus replacement prosthesis is used. Incisions are made to separate the periosteum from the middle one third of the malleus, and an incus replacement wire prosthesis is inserted through the opening. A right angle hook is used to rotate the loop portion and placed on the surface of the footplate to determine if the length is proper. The usual length is 5.5 mm. After this step, the loop is elevated away from the footplate, and the head and neck of the malleus are removed. The footplate is removed, and the loop of the prosthesis is placed into the oval window. The shaft of the prosthesis is grasped with a nonserrated alligator forceps to stabilize it, and with the right hand, it is tightened on the malleus by use of a right angle hook. Fat or absorbable gelatin sponge (Gelfoam) centers the loop in the oval window. Other types of prostheses, such as clamp-on plastic pistons, are available, and for some surgeons, these are more easily attached and inserted into the oval window. Another option is to use a total ossicular replacement prosthesis.
Facial Nerve Abnormalities The facial nerve is often dehiscent in its tympanic s egment, but the dehiscence is rarely seen because it often occurs on its undersurface. If the dehiscence is extensive, a prolapse covering half of the footplate area may occur. In this case, the facial nerve may be carefully elevated superiorly, and an opening may be made in the footplate for insertion of the prosthesis. If the shaft of the prosthesis needs to be bent, it usually does not provide a satisfactory hearing result. If the prosthesis rubs on the prolapsed facial nerve, it does not disturb facial
258
OTOLOGIC SURGERY
Chapter 21 • Total Stapedectomy
259
FIGURE 21-5. Adequate exposure. FIGURE 21-6. Fracturing the crura from footplate. FIGURE 21-7. Measuring for prosthesis. FIGURE 21-8. Removing the footplate. FIGURE 21-9. Placing tissue over the oval window. FIGURE 21-10. Placing the prosthesis.
nerve function. A marked prolapse covering the entire oval window requires a prosthesis from the malleus. In these cases, the incus is removed, and the facial nerve is moved slightly so that the inferior part of the footplate can be visualized and then fragmented for insertion of the incus replacement prosthesis with the loop into the fragmented area.
the prosthesis of choice. The edges are always jagged because there is no demarcation between the attachment of the footplate to the surrounding bone as there is in the solid footplate. These patients respond well to the small fenestra, the subtotal footplate, or the total removal technique. The floating footplate does not occur in obliterated cases.
Floating Footplate
Round Window Closure Owing to Otosclerosis
The most difficult complication in otosclerosis surgery is the solid floating footplate; this requires the most complicated and difficult procedure encountered in stapes surgery and should be managed only by highly experienced surgeons. In this problem, a small cutting burr is used posteroinferiorly on the promontory “side” of the oval window to gain room. The drilling stops 1 mm from the floating footplate. Straight and obtuse needles are used to enter the vestibule along the side of, but at no time touching even the edge of, the floating footplate. A right angle hook is inserted into the opening and then rotated to the undersurface of the footplate. It is elevated outward and anteriorly toward the fallopian canal. Blood adhesiveness allows the plate to remain there after the hook is removed and reinserted under the edge of the footplate to slip it up and over the fallopian canal, where it can be grasped. Only then can the surgeon take his or her first breath with a sigh of relief!
Solid or Obliterated Footplate By observation, the surgeon cannot determine whether the footplate is minimally fixed at the edges or is obliterated (Fig. 21-13). In each instance, it is necessary to use a cutting burr with a gentle paintbrush-type stroking motion anteroposteriorly. If the slightest give is felt with the drill, minimal fixation and a floating footplate exist. Perilymph escape is noted around the edges. The technique to be used is the same as for the solid floating plate.
Obliterated Footplate With an obliterated footplate, a cutting burr is used anteroposteriorly and around the edges to develop a thin blue area. The burr is used to enlarge the opening sufficiently to cover the area with a graft and to insert
A round window closure owing to otosclerosis is rare and is best left undisturbed. In these cases, there is always a fibrous band leading to the round window that is apparently sufficient to produce the necessary relationship between the oval window and the round window. Complete drilling out of the round window has not given the desired result and may make the hearing worse.
Perilymph Gusher When the cochlear aqueduct is widely patent, it may result in an excessive perilymph flow after the surgeon perforates the footplate. If it is a massive gusher (e.g., like a broken fireplug), one should not proceed with the removal of the stapes. It would be best to scrape the membrane of the footplate and the surrounding oval window area with suction in one hand, followed by the placement of small pieces of fascia or perichondrium in the arch of the stapes to help hold them in place. On one such occasion, we proceeded to remove the stapes and its footplate. The tissue placed over the oval window floated away, and it needed to be held in place with a suction tip in one hand until the prosthesis was inserted onto the incus. We packed the middle ear solid with Gelfoam and replaced the flap, only to note the flap being elevated by the flow of perilymph. The ear canal was tightly packed to hold the flap in place, and the patient was elevated to a sitting position. The mastoid dressing was changed repeatedly until the flow stopped.
Intraoperative Vertigo When patients are under local anesthesia, their response to sudden vertigo is immediately noted by the operating surgeon. Vertigo may occur more often in revision
260
OTOLOGIC SURGERY
FIGURE 21-11. A and B, Superior retraction of chorda tympani for visualization. FIGURE 21-12. A-F, Incus replacement prosthesis.
Chapter 21 • Total Stapedectomy
261
AUTHORS’ NOTE
FIGURE 21-13. Minimally fixed solid footplate.
Most patients with otosclerosis who opt for surgery today usually undergo a small fenestra-type procedure, with the use of either a microdrill or a laser. The evolutionary changes in stapes surgery have primarily occurred in the prosthesis itself, with the development of new self-crimping materials, and in the drill and laser instrumentation. The general concepts of stapes surgery still stand, however, as they did since the late 1950s. This chapter serves as a reminder of those general concepts and the importance of anticipating, and then dealing with, the unexpected in surgery, such as when the footplate mobilizes or fractures and the whole thing comes out. Knowing the steps to completing a total stapedectomy and having the confidence that the results can be consistently as good as a small fenestra approach are requirements for anyone contemplating performing stapes surgery.
REFERENCES rocedures if the covering tissue membrane is inserted p too deeply into the vestibule, or if the prosthesis is too long. If the cause is tissue, it should be carefully removed and replaced. If the prosthesis is too long, it should likewise be removed and replaced with a shorter one. In revision, the incidence of further sensorineural loss is slightly more common than in a primary case.
Deep Oval Window In some cases, the oval window niche is very narrow and very deep. If this condition results from otosclerotic encroachment on the sides of the oval window, it can be enlarged by use of a cutting burr. If there is another cause, a broad posterior crus located in the deep narrow niche may be difficult to fracture from the footplate. In this instance, after a small perforating hole has been made in the footplate, a tiny cutting burr or a laser may be used to cut through the posterior crus, making removal of the superstructure of the stapes possible.
SUMMARY Training surgeons is difficult because of the scarcity of suitable otosclerotic patients, and it is difficult for residents to obtain sufficient experience to ensure good hearing results in their patients. If a surgeon is obtaining the results being reported in the literature—90% closure of the air-bone gap within 10 dB and no more than 1% further sensorineural loss—there is no need to change technique. It is not the instruments or technique that ensures success, but rather the minds and the hands in control of the instruments.
1. Kessel J: Uber das Mobilisieren des Steigbugels durch Ausschneiden des Trommelfelles, Hammers und Amboss bei undurchgagikeit der Tuba. Arch Ohrenheilkd 13: 69-88, 1878. 2. Boucheron E : La mobilisation de l’etrier et son procede operatoire. Union Med Can 46:412-416, 1888. 3. Miot C : De la mobilisation de l’etrier. Rev Laryngol Otol Rhinol (Bord) 10:49-54, 1890. 4. Faraci G: Importanza acustica e funzionale della mobilizzazione della staffa: Risultati di una nuova serie di operazioni. Arch Ital Otol Rinol Laringol 9:209-221, 1899. 5. Passow K A : Operative anlegung einer offnung in die mediale paukenhohlenwand bei stapesankylose. Ver Dtsch Otol Ges Versamml 6:141, 1897. 6. Jack FL : Remarkable improvement of the hearing by removal of the stapes. Trans Am Otol Soc 284:474-489, 1893. 7. Goodhill V: Stapes Surgery for Otosclerosis. New York, Paul B. Hoeber, 1961. 8. Siebenmann F: Traitement chirurgical de la sclerose otique. Cong Int Med Sec Otol 13:170, 1900. 9. Moure E J: De la mobilisation de l’etrier. Rev Laryngol Otol Rhinol (Bord) 7:225, 1880. 10. Holmgren G: Some experiences in surgery of otosclerosis. Acta Otolaryngol (Stockh) 5:460, 1923. 11. Sourdille M : New technique in the surgical treatment of severe and progressive deafness from otosclerosis. Bull N Y Acad Med 13:673, 1937. 12. Lempert J: Improvement in hearing in cases of otosclerosis: A new, one-stage surgical technic. Arch Otolaryngol Head Neck Surg 28:42, 1938. 13. Rosen S : Restoration of hearing in otosclerosis by mobilization of the fixed stapedial footplate: An analysis of results. Laryngoscope 65:224-269, 1955. 14. Shea J Jr: Fenestration of the oval window. Ann Otol R hinol Laryngol 67:932-951, 1958.
22
Laser Stapedotomy Clough Shelton Videos corresponding to this chapter are available online at www.expertconsult.com.
The surgical treatment of otosclerosis has been evolutionary in nature. Multiple techniques are now available to accomplish the same goal, which is illustrated by the number of chapters devoted to this subject in this book. The small fenestra technique evolved as a less invasive method to accomplish stapedectomy.1 It allows a smaller opening into the inner ear with the potential for less surgical trauma. Many techniques have been used in the past to accomplish a small fenestra, including a manual pick method, a hand drill, a microdrill, and a laser technique.2 The laser technique was developed by Perkins3 in the early 1980s. Several different lasers have been used to perform the small fenestra, including argon, KTP, and CO2.4-6 A hand-held fiberoptic probe or a micromanipulator can be used to deliver the laser beam, and each method has its champions.7 The theoretical advantage of a laser stapedotomy over a mechanical stapedotomy is a “no touch” technique. Theoretically, there would be less mechanical energy imparted to the inner ear, which should result in better sensorineural hearing results.
PREOPERATIVE EVALUATION The surgical candidate should have a conductive hearing loss on audiogram and confirmed by tuning fork of at least 15 dB. Acoustic reflexes can be used as a simple screening tool for conductive hearing loss owing to superior canal dehiscence, and should be absent. The tympanic membrane should be intact, and there should be no evidence of ongoing infection in the ear. The patient should not have evidence of cochlear hydrops (see the section on pitfalls). The patient should have no medical contraindications to a short surgical procedure performed under local anesthesia. See Chapter 21 for indications and contraindications of stapes surgery.
PATIENT COUNSELING Before surgery, the patient is educated regarding the anatomy and physiology of the ear, and the effect of otosclerosis on hearing. The expected outcomes of the procedure
are described, including limitations of hearing improvement that might be imposed by a pre-existing sensorineural component of the hearing loss. The possible risks and complications of the procedure include worsened hearing or total hearing loss, worsened tinnitus, dizziness, taste disturbance, and facial paralysis. These risks are outlined in a patient education booklet that the patient receives on the initial office visit. The author performs stapedectomy under local anesthesia. The recovery seems easier for the patient, and there is less intraoperative bleeding. Disturbances of the inner ear related to footplate manipulation, such as dizziness, are also apparent intraoperatively. As part of the preparation of the patient for a procedure done under local anesthesia, explanation of what is to be expected in the operating room seems to alleviate patient anxiety and results in a calmer and more cooperative patient during the procedure. The preparation steps up to the start of surgery are described in detail to the patient, including the expectations regarding the level of sedation. During the procedure, the patient is sedated to the point of somnolence, but is still conscious and able to respond to questions in an appropriate manner. Too deep sedation can result in disinhibition of the patient and render him or her much less cooperative.
SURGERY Operating Room Setup and Preparation Stapedectomy surgery requires great finesse for good results. Thorough preoperative preparation results in smoother intraoperative execution and consistently good results. A surgical team that is familiar with the procedure and setup is of paramount importance. A scrub nurse who is knowledgeable about otologic instruments and can anticipate what is needed for the next step of the procedure is crucial. A microscope mounted closed circuit television camera with monitor allows the scrub nurse to see what the surgeon is doing and anticipate the next instrument that is needed. In a difficult case, with unfavorable anatomy or bleeding or both, the surgeon 263
264
OTOLOGIC SURGERY
must have the right instrument in a short amount of time to complete the operation effectively. A scrub nurse who is not trained to assist in a stapes surgery is unable to perform adequately in these situations. The anesthesiologist or local monitoring nurse is positioned at the foot of the table. The scrub nurse is positioned directly across the table from the surgeon. This positioning allows the nurse to hand instruments to both of the surgeon’s hands, greatly improving the efficiency. The microscope is positioned at the head of the table. Proper patient positioning can greatly facilitate the ease of this procedure. The patient’s head is turned away from the surgeon with the chin tucked toward the opposite shoulder. This maneuver aligns the external auditory canal with the view of the surgeon sitting in a neutral position. Failure to tuck the chin can result in the surgeon leaning over the patient’s chest and much less effective visualization of the field. The operating table is positioned in slight Trendelenburg. This head-down position counteracts the normal inferior angulation of the external auditory canal and results in improved visualization. Generally, the patient is positioned so the plane of the tympanic membrane is parallel with the plane of the floor in the operating room. A support bar (Fig. 22-1) is attached to the operating table to keep the drapes off the patient’s face. By attaching the bar directly to the table, the bar moves in conjunction with the table. Using a floor-mounted device, such as a Mayo stand, does not allow the coordination of movement between the two items and can result in limitations of table movement. The operative site is prepared by surrounding the ear with a plastic drape (1030; 3M) that keeps the hair out of the operative field. The ear and a margin of the plastic drape are prepared with povidoneiodine (Betadine) solution, and the povidone-iodine solution is allowed to fill the ear canal. Intravenous midazolam (Versed) is used for sedation. This medication is given in 0.5 mg increments until the patient is sleepy
and relaxed. Typically, the patient requires 4 to 5 mg. Intravenous meperidine (Demerol), given in 12.5 mg increments, can be used to supplement the midazolam. An antiemetic such as ondansetron (Zofran) is also administered during surgery. The amount of sedation is titrated to the point that the patient is somnolent, yet appropriately responsive. At this point, 3 mL of blood is drawn and transferred to the scrub nurse. A local field block is performed with 1% lidocaine with 1:100,000 epinephrine. A four quadrant block is performed in the meatus, and an additional injection is administered to the vascular strip and posteroinferiorly in the canal. Blanching of the canal skin is important for good hemostasis. Also important for good hemostasis is a snugly fitting speculum. The pressure provided by the speculum helps tamponade the vessels supplying the ear canal from the meatus. If one encounters troublesome bleeding during the procedure, a different speculum size can be tried, which typically results in improved hemostasis.
FIGURE 22-1. Drape support bar is attached directly to operating
FIGURE 22-2. Canal incisions are outlined for the tympanomeatal
table.
flap.
Surgical Technique A tympanomeatal flap is incised using disposable tympanoplasty blades (7200 and 7210 BD Beaver). These disposable blades result in a sharp incision and rapid healing of the ear canal. Two incisions are made tangential to the annulus, one beginning adjacent to the short process of the malleus and following the tympanosquamous suture line. The second is adjacent to the tympanic annulus starting at approximately the 6 o’clock position and extending posterosuperiorly in the canal. These two incisions (Fig. 22-2) are connected with a transverse incision approximately 5 mm posterior to the tympanic annulus. The flap length is approximately the width of two Sheehy weapons (3 mm round knife). An excessively long flap
Chapter 22 • Laser Stapedotomy
would result in more bleeding, and can be difficult to fold forward in a narrow ear canal because of the bulk of the flap. The flap is elevated with a Sheehy weapon to the edge of the bony tympanic annulus. During this elevation process, the tip of the weapon is kept in contact with the ear canal bone. A No. 3 Baron suction is used behind the weapon, avoiding suction on the skin flap. A Rosen needle is used to enter the middle ear in a posterosuperior location. The tip of the instrument is placed through the middle ear mucosa, and the mucosa is divided. It is imperative not to place the instrument too deeply into the middle ear during this maneuver, or incus dislocation can occur. When the middle ear space is identified, the dissection is carried inferiorly, and the tympanic annulus is directly visualized. The annulus is elevated with the Rosen needle out of its sulcus. A House annulus elevator is also useful at this point. The annulus elevator is a blunt instrument, and if used posterosuperiorly, it too can result in incus dislocation. The flap is elevated inferiorly to allow visualization of the round window (Fig. 22-3). It is elevated superiorly until the neck of the malleus is visualized. The chorda tympani nerve is adherent to the posterosuperior bony tympanic annulus. It is separated from this bone, and separated from the undersurface of the malleus. Separating the chorda tympani allows mobilization of the nerve inferiorly for good visualization of the stapes, and prevents injury during curetting. Posterior external auditory canal bone is removed with a curette to visualize the oval window. Initial curetting occurs several millimeters posterior to the edge of the tympanic annulus (Fig. 22-4). Initial curetting on the edge can result in displacement of the curette into the middle ear with dislocation of the incus. The posterior curetting allows for the bone to be weakened by creating a trough in this area. When the bone is weakened, the edge of the bone can be cracked off with the curette and little force is needed. The bone should be removed
265
osteriorly to allow visualization of the stapedial tendon p and pyramidal process (Fig. 22-5). For a right-handed surgeon working on a right ear, the bone should be removed superiorly so that half of the diameter of the horizontal fallopian canal can be visualized. For a righthanded surgeon working on a left ear, more curetting is necessary to introduce instruments in the middle ear, and bone should be removed so that the entire diameter of the horizontal fallopian canal can be visualized. At the conclusion of the curetting, all of the resultant bone chips should be removed from the ear canal to prevent postoperative healing problems.
Initial curette stroke
FIGURE 22-4. Initial curetting begins posterior to the edge of the scutum.
Separate chorda from manubrium
Reflected tympanomeatal flap
Incisions
FIGURE 22-3. Tympanomeatal flap is elevated exposing the neck of the malleus and the round window.
FIGURE 22-5. Completed curetting exposes stapedial tendon, pyramidal process, and horizontal fallopian canal.
266
OTOLOGIC SURGERY Incudostapedial joint sectioned
FIGURE 22-6.
Stapedial tendon vaporized with laser
FIGURE 22-7. FIGURE 22-6. Incudostapedial joint is separated with joint knife.
FIGURE 22-7. Stapedial tendon divided with
Posterior crus now visible
laser.
FIGURE 22-8.
The incudostapedial joint is separated with a joint knife (Fig. 22-6). The joint is typically 1 mm medial to the undersurface of the incus and can be identified by pushing the incus slightly forward and laterally. The joint is separated with a side-to-side motion by the knife. When the joint is separated, the malleus mobility is tested to rule out idiopathic malleus head fixation. The stapes is tested for motion, and the anterior oval window is inspected visually for evidence of otosclerosis. Otosclerosis can reliably be seen in this area and appears to be bright white bone with prominent vessels on it. The adjacent normal otic capsule bone looks slightly gray compared with the otosclerotic bone. If otosclerosis is not visualized, one must be quite confident that the stapes is not mobile before proceeding (see the section on pitfalls). In some cases, patients experience pain during manipulation of the middle ear structures. The local
FIGURE 22-8. Posterior crus of stapes is now exposed.
anesthesia can be supplemented by topical application of 1% lidocaine with 1:100,000 epinephrine. The solution can be squirted into the middle ear. It is important to remove the solution quickly, however, so that it does not diffuse through the round window and result in postoperative severe vertigo. Using the laser, the stapedial tendon and posterior crus are removed (Figs. 22-7 through 22-12). A typical setting for the KTP or argon laser is 1.5 W, at 1⁄10 of a second duration. The author prefers a fiberoptic probe for delivery of the laser. Any mucosal adhesions in the oval window area are also removed with the laser to prevent bleeding from the mucosa. Care should be taken not to use the laser directly on the tympanic facial nerve, particularly if dehiscent. In some cases, the posterior stapes crus has a three-dimensional shape like the letter C. It is important to char not only the posterior aspects of
Chapter 22 • Laser Stapedotomy
267
Remove char residue
Tympanosulcus Facial ridge
Bony scutum (bony annulus)
FIGURE 22-9.
FIGURE 22-10.
FIGURE 22-11.
FIGURE 22-9. Posterior crus is vaporized with laser. FIGURE 22-10. Charred posterior crus is picked away. FIGURE 22-11. Posterior crus is now divided.
the crus, but also the superior and inferior aspects. The charred area is picked through with a Rosen needle (see Fig. 22-10), and the superstructure is down fractured toward the promontory with a Rosen needle (see Fig. 22-12A). A quick sharp jerk on the superstructure is important to fracture the anterior crus (see Fig. 22-12B). A slow down fracture movement may result in mobilization of the footplate. The distance between the surface of the footplate and the lateral aspect of the incus is measured (Fig. 22-13). In most cases, this measurement is 4.5 mm. If it is shorter, one should suspect a thick footplate. Measurements can also be routinely taken to the undersurface of the incus. The surgeon should always measure in a consistent fashion, however, and know how the prosthesis he or she uses is sized to determine the correct prosthesis length (Fig. 22-14). Optimally, the prosthesis should extend 0.5 mm into the vestibule.8 The fenestra in the footplate is usually placed in the middle or posterior half. Using the laser, a footplate vessel is the initial target. With a visible wavelength laser, the energy is absorbed by pigment. The white footplate is a poor absorber of this energy; however, if an initial char is developed by using a laser on a red vessel, the char can be overlapped with each subsequent laser blast (Fig. 22-15) creating a rosette of chars. Perilymph seeps through the charred footplate and can be aspirated with a 24 gauge suction tip (Fig. 22-16). The charred area is created larger than the intended fenestra (Fig. 22-17). It is easier to remove
charred footplate than normal footplate if the fenestra needs to be enlarged. When a char of sufficient size has been fashioned, a sizing instrument “disk on a stick” (N1685-spec; Storz Instruments) is used to create the fenestra. The disk is 0.6 mm in diameter and creates a fenestra the same size as a 0.6 mm diameter piston. The disk is gently pushed through the char (Fig. 22-18) to create the hole. It can be used to rasp the hole, resulting in a hole slightly larger than the disk. When the fenestra is open, it is important to avoid suctioning directly into the oval window. A No. 24 suction is used at the margins of the oval window with the surgeon’s finger off the control hole. If a blood clot accumulates in the oval window obscuring the view, it can be lifted out using a Rosen needle rather than suction. As an alternative to the laser, a small fenestra can be easily created using a microdrill. To make a footplate opening, a 0.7 mm diamond burr is used, which allows sufficient clearance for a 0.6 mm piston prosthesis. After the superstructure has been removed, the burr is lightly placed on the footplate (Fig. 22-19). Pressure is not exerted against the footplate, and a light touch is used. The drill motor is activated, and when the fenestra is completed, the surgeon senses a subtle resistance change at his or her fingertips. The drilling can be quite noisy and can startle a patient who is under local anesthesia. In these cases, it is helpful to place the drill against the promontory and activate it, alerting the patient to the noise of the drill and avoiding unexpected patient movement. The microdrill provides a uniformly sized footplate opening.
268
OTOLOGIC SURGERY
A
B
Prosthesis length
FIGURE 22-12. A, Stapedial superstructure is down-fractured toward the promontory. B, A quick sharp motion toward the promontory results in fracture of the anterior crus.
FIGURE 22-13. Measurement is taken between the surface of the footplate and the lateral aspect of the incus.
The prosthesis is introduced into the oval window using nonserrated alligator forceps. After placement in the oval window, the prosthesis is maneuvered into position with a strut guide. If using a platinum wire prosthesis, the wire is quite malleable, and care must be taken not to bend the wire using excessive force while manipulating the prosthesis into position. In some cases, the diameter of the incus is wider than the opening of the shepherd’s crook. Gentle downward pressure on the shoulder of the piston, pushing it into the open fenestra, allows the crook opening to enlarge and slip over the incus. When the piston is in the fenestra, and the wire is around the incus, it is crimped with crimping forceps.
FIGURE 22-14. Piston prostheses are typically sized to the bottom of the shepherd’s crook.
Allowing both jaws of the forceps to contact the prosthesis simultaneously prevents displacement during this maneuver. The crimp should be firm and not oval in shape. An oval crimp can lead to excessive movement of the wire on the incus and ultimately lead to incus necrosis. The prosthesis is crimped over the narrowest portion of the incus (Fig. 22-20), and then slid inferiorly toward the lenticular process. The incus typically widens at this point, and this results in a very snug connection.
Chapter 22 • Laser Stapedotomy
269
Beam overlaps previous hole Rosette of holes creates fenestra
FIGURE 22-17. Char on the footplate is created larger than desired opening.
FIGURE 22-15. Char is created on the footplate with the laser. Subsequent chars overlap creating a rosette. Check diameter of fenestra
Aspirate perilymph
FIGURE 22-18. “Disk-on-a-stick” is used to make the opening in FIGURE 22-16. Perilymph seepage is removed with a 24 gauge suc-
footplate.
tion tip.
The new nitinol-polytef (Teflon) piston (SMart; Gyrus Medical) allows heat-activated crimping of the wire. The argon or KTP laser is set at 0.75 W, and the wire is lasered resulting in self-crimping of the prosthesis. In some cases, the wire is bent slightly when removing it from the packaging. The shaft of the prosthesis can also be lasered to straighten the wire into its native shape. The heat-activated crimping results in a very tight crimp. The length of the SMart piston may shorten on crimping, resulting in a shorter effective length than a manually crimped prosthesis. To correct for this, a longer prosthesis size is used, adding 0.25 mm to the usual length. In the author’s practice, the most common prosthesis length for the SMart piston is 4.5 mm compared with 4.25 mm for a manually crimped prosthesis design. Autologous venous blood is applied to the oval window to seal it. This is usually placed in a 3 mL syringe with a 20 gauge suction tip. Enough blood is applied to fill the oval window. The tympanic membrane is returned
to its anatomic location, and the ear canal is filled with an inert ointment packing (Kos-House Otic Cream; Otomed Inc). A cotton ball is placed in the ear canal, and a BandAid is placed over the cotton ball.
SPECIAL CONDITIONS Overhanging Facial Nerve The facial nerve may be dehiscent over the oval window, and in some cases prolapse inferiorly to obscure the view of the oval window. Obliterative otosclerosis on the promontory side can narrow the oval window and lead to a similar appearance as an overhanging facial nerve, and so this possibility must also be evaluated. The otosclerotic bone is very vascular and can be differentiated from the normal promontory When faced with a significantly overhanging facial nerve intraoperatively, the most conservative way to manage this situation is to close the ear and have the patient evaluated for a hearing aid.
270
OTOLOGIC SURGERY
FIGURE 22-19. Microdrill is used to make the small fenestra in the footplate.
B
A FIGURE 22-20. A, Wire is crimped around the incus at its narrowest portion. B, Piston protrudes slightly into vestibule.
Chapter 22 • Laser Stapedotomy
271
For an experienced stapes surgeon, however, it is usually possible to complete the operation in this situation. Typically, the facial nerve can be retracted superiorly with a No. 24 suction allowing creation of a fenestra in the footplate either with a fiberoptic laser probe or with a microdrill. Contact of the prosthesis with the facial nerve causes no facial nerve dysfunction. In some cases, removing part of the promontory bone with a microdrill allows additional access. Creating the fenestra on the inferior edge of the footplate can allow for a better angle of the prosthesis past the facial nerve. Bending the wire of the prosthesis around the facial nerve is generally unsuccessful.
failure to elevate the annulus in this location. It is possible to elevate the skin of the ear canal and tympanic membrane lateral to the fibrous layer of the drum, which results in perforation. Direct visualization of the annulus to allow its elevation is the best way to avoid this complication. If this complication occurs, the operation can be completed, and the perforation can be closed at the end of the procedure. A small piece of fat is harvested from the lobule, and a fat patch myringoplasty is performed by placing the fat through the perforation in a dumbbell fashion.
Obliterative Otosclerosis
The incus can be dislocated while entering the middle ear when raising the tympanomeatal flap or during the curetting process (see earlier). Typically, the incus remains in near-normal anatomic location, and the incudomalleal joint can heal and result in an intact ossicular chain. During the healing process, however, the incus tends to lateralize. In this situation, a slightly longer prosthesis is placed, 0.25 mm longer than would be used in the usual situation.
These cases have a thickened footplate, sometimes with loss of the normal oval window landmarks. A shallow measurement from the footplate to the incus may be the first clue that one is dealing with an obliterative case. Rather than seeing a thin blue footplate, an opaque white footplate is visible. It is possible to create a fenestra through an obliterative footplate with a laser; however, this is a slow and tedious process. The author prefers to use the laser initially to char the footplate to prevent bleeding of the otosclerotic bone. Next, a microdrill is used to create the fenestra. The fenestra is created gradually by excavating the footplate in layers. The goal is a funnelshaped opening, with the opening wider on the surface. This allows a wider angle of approach by the prosthesis into the fenestra, and prevents binding of the prosthesis against the edges of the footplate. When drilling the fenestra, serial measurements to the incus can be taken with a measuring device to monitor the progress toward the vestibule.
Floating Footplate In some instances, the footplate mobilizes during down fracture of the stapes superstructure. With the laser technique, it is possible to complete the operation with little difficulty. The fenestra in the footplate is created as previously described. Care in opening the fenestra should be observed to avoid fracture of the footplate and a resultant large opening. A large opening can be managed with a large-diameter prosthesis (0.8 mm). Also, the force used to place the prosthesis in the fenestra in a mobile footplate should be reduced.
INTRAOPERATIVE COMPLICATIONS Tympanic Membrane Perforation Perforation of the tympanic membrane may occur during elevation of the tympanomeatal flap. Most commonly, this occurs in a posteroinferior location and results from
Dislocated Incus
Chorda Tympani Nerve Injury Care should be taken to preserve the chorda tympani nerve, if possible. The nerve is particularly vulnerable to injury during curetting. In some cases, it is necessary to mobilize the nerve out of its bony canal in the posterior ear canal. Curetting directly over the course of the nerve unroofs the canal and allows the nerve to be moved. If the nerve is partially injured or stretched, the nerve should be completely transected in a sharp fashion with microscissors. Sharp transection leads to fewer postoperative symptoms of dysgeusia than does partial injury or stretching. The nerve can also be desiccated by the intense light of the microscope. Saline can be installed into the middle ear to rehydrate the nerve.
POSTOPERATIVE CARE The patient leaves the hospital several hours after the conclusion of the operation. A postoperative instruction sheet is provided describing the care of the ear and restrictions in activity. The patient is allowed air travel 3 days after surgery, and must keep water out of the ear, not blow the nose, and avoid heavy lifting (>30 lb) or vigorous physical activity for 3 weeks. The patient returns to work 3 to 5 days after the procedure. Postoperative prescriptions include acetaminophen with codeine (Tylenol with Codeine) and amoxicillin. Routine postoperative follow-up occurs at 1 month and at 4 months. Audiograms are obtained at both visits; the final postoperative hearing result is assessed at the 4 month visit. The audiogram at the 1 month visit is mainly to assess the sensorineural
272
OTOLOGIC SURGERY
hearing, and is performed while the packing is still in place. Before this visit, the patient is instructed to place mineral oil in the ear canal using an eye dropper; this facilitates removing the ear canal packing, which is done at this visit.
RESULTS Success Rate Small fenestra stapedectomy provides reliable hearing improvement for most patients. Reported closure of the air-bone gap to within 10 dB (when measured by comparing the postoperative bone conduction with the postoperative air conduction) ranges from 56% to 94%.9-12 Some authors have found an advantage of small fenestra stapedectomy performed with a laser compared with procedures done with a nonlaser technique; others have not found a difference. Sedgwick and coworkers13 compared small fenestra hearing results done by drill or laser, and found no difference for the closure of the air-bone gap within 10 dB or postoperative sensorineural hearing loss. When comparing small fenestra techniques with partial or total stapedectomy, there was better preservation of the high-frequency sensorineural hearing with a small fenestra technique, but the difference was small. Several prosthesis materials have been introduced more recently. Titanium has been advocated as a superior prosthesis material because of its light weight and ease of crimping. Some authors have reported better results14 with this prosthesis material, whereas others have found no advantage of it over a platinum and Teflon prosthesis.15 Nitinol is a bimetallic wire (nickel and titanium) that reforms to a preset shape on heating.16 It has been incorporated into a piston prosthesis (SMart) that allows for heat-activated crimping of the shepherd’s crook around the incus. Heat can be provided by laser or cautery. Early reports are encouraging, showing results comparable or superior to manually crimped prosthesis.17-19 The author has analyzed his own outcomes comparing the SMart piston with platinum-Teflon piston. Closure of the postoperative four-frequency air-bone gap to within 10 db occurred in 96% of the SMart prosthesis cases compared with 86% of the platinum-Teflon group. The SMart piston is now the author’s preferred prosthesis.
Postoperative Complications The management of the stapedectomy patient with postoperative dizziness or sensorineural hearing loss is controversial. Some surgeons believe that these symptoms represent either a perilymphatic fistula or a reparative granuloma and advocate immediate surgical re-exploration.20 Other surgeons believe that these symptoms are caused by postoperative serous labyrinthitis, and no surgical intervention is needed. Blood entering the
vestibule during surgery can irritate the labyrinth, particularly during its metabolism and degradation. These surgeons believe that early surgical exploration may be more harmful than helpful. Some authors have observed a transient decrease in the sensorineural hearing level after stapes surgery as a routine occurrence during the postoperative period.21 Reparative granuloma was a phenomenon that was encountered on the East Coast in the 1960s, but was rarely seen on the West Coast. Some have explained the observed geographic variation as due to differences in the composition of Gelfoam in various parts of the United States at that time (Linthicum FH, personal communication, 2006). It is theorized that ethylene oxide, used during sterilization, was retained by the lot of Gelfoam used on the East Coast for the Gelfoam-wire prosthesis technique, and this led to the cases of reparative granulomas. Many surgeons believe that reparative granuloma have historical significance only and does not occur with modern stapes prostheses. The author’s approach to postoperative dizziness is to examine the patient in the office to rule out evidence of infection. An audiogram is obtained, and if the sensorineural level has worsened compared with the preoperative levels, the patient is placed on a course of oral steroids for presumed serous labyrinthitis. This treatment has been extremely successful in managing these patients. Early postoperative exploration has not been needed.
Patient Selection Pitfalls Most patients with a conductive hearing loss and a normal tympanic membrane have otosclerosis. The term inner ear conductive hearing loss has been used to describe patients who had either a normal ossicular chain at the time of stapedectomy or a persistent conductive hearing loss after an uncomplicated stapedectomy.22 More recently, superior semicircular canal dehiscence23 and intralabyrinthine neuromas24 have been identified as possible causes of inner ear conductive hearing loss. In the case of an unchanged postoperative conductive loss after an uneventful stapedectomy, one should consider both entities, and further imaging may be indicated. Superior semicircular canal dehiscence typically has very good to supranormal bone thresholds and can be diagnosed on coronal CT scan. Intralabyrinthine neuromas have a mixed hearing loss and can be diagnosed on highresolution MRI. One must approach otosclerosis patients who have coexisting dizziness with caution. Most patients with otosclerotic inner ear syndrome25 have improvement in vertigo after stapedectomy, whereas patients with Meniere’s disease may have a profound postoperative sensorineural hearing loss. The dilated saccule caused by endolymphatic hydrops can be damaged during stapedectomy and lead to the sensorineural hearing loss. Issa and coworkers26 found stapedectomy to be safe in patients with Meniere’s
Chapter 22 • Laser Stapedotomy
disease if the bone conduction threshold at 500 Hz was 35 dB or better, and there was no high-frequency sensorineural hearing loss.
REFERENCES 1. Bailey H AT, Pappas JJ, Graham S S : Small fenestra stapedectomy technique: Reducing risk and improving hearing. Otolaryngol Head Neck Surg 91:516-520, 1983. 2. Rizer FM, Lippy WH : Evolution of techniques of stapedectomy from the total stapedectomy to the small fenestra stapedectomy. Otolaryngol Clin North Am 26:443-451, 1993. 3. Perkins RC : Laser stapedotomy for otosclerosis. Laryngoscope 91:228-241, 1980. 4. Hodgson R S, Wilson D F: Argon laser stapedotomy. L aryngoscope 101:230-233, 1991. 5. Bartels L J: KTP laser stapedotomy: Is it safe?. Otolaryngol Head Neck Surg 103:685-692, 1990. 6. Lesinski SG: Lasers for otosclerosis: CO2 versus argon and KTP-532. Laryngoscope 99(Suppl 46):1-8, 1989. 7. Gherini SG, Horn K L , Bowman C A, Griffin G: Small fenestra stapedotomy using a fiberoptic hand-held argon laser in obliterative otosclerosis. Laryngoscope 100:12761282, 1990. 8. Pauw B K H, Pollack A M, Fisch U: Utricle, saccule and cochlear duct in relation to stapedotomy. Ann Otol R hinol Laryngol 100:966-970, 1991. 9. Grolman W, Tange R A : First experience with a new stapes clip piston in stapedotomy. Otol Neurotol 26:595598, 2005. 10. Quaranta N, Besozzi G, Fallacara R A, Quaranta A : Air and bone conduction change after stapedotomy and partial stapedectomy for otosclerosis. Otolaryngol Head Neck Surg 133:116-120, 2005. 11. Jovanovic S, Schonfeld U, Scherer H : CO2 laser stapedotomy with the “one-shot” technique—clinical results. Otolaryngol Head Neck Surg 131:750-757, 2004. 12. Vincent R , Sperling N M, Oates J, Jindal M : Surgical findings and long-term hearing results in 3,050 stapedotomies for primary otosclerosis: a prospective study with the otology-neurotology database. Otol Neurotol 27(Suppl 2):S25-S47, 2006. 13. Sedgwick J D, Louden C L , Shelton C : Stapedectomy vs stapedotomy: Do you really need a laser? Arch Otolaryngol Head Neck Surg 123:177-180, 1997.
273
14. Zuur C L , de Bruijn A JG, Lindeboon R, Tange R A : Retrospective analysis of early postoperative hearing results obtained after stapedotomy with implantation of a new titanium prosthesis. Otol Neurotol 24:863-867, 2003. 15. Massey B L , Kennedy R J, Shelton C : Stapedectomy outcomes: Titanium vs. Teflon wire prosthesis. Laryngoscope 115:249-252, 2005. 16. Knox GW, Reitan H : Shape-memory stapes prosthesis for otosclerosis surgery. Laryngoscope 115:1340-1346, 2005. 17. Sorom A J, Driscoll C LW, Beatty CW, Lundy L : Retrospective analysis of outcomes after stapedectomy with implantation of a self-crimping Nitinol stapes prosthesis. Otolaryngol Head Neck Surg 137:65-69, 2007. 18. Harris J P, Gong S : Comparison of hearing results of Nitinol SMART stapes piston prosthesis with conventional piston prostheses: Postoperative results of the Nitinol stapes prosthesis. Otol Neurotol 28:692-695, 2007. 19. Rajan G P, Diaz J, Blackham R , et al: Eliminating the limitations of manual crimping in stapes surgery: Midterm results of 90 patients in the Nitinol stapes piston multicenter trial. Laryngoscope 117:1236-1239, 2007. 20. Gacek R R : The diagnosis and treatment of poststapedectomy granuloma. Ann Otol Rhinol Laryngol 79:970975, 1970. 21. Somers T, Vercruysse J P, Zarowski A, et al: Stapedotomy with microdrill or carbon dioxide laser: Influence on inner ear function. Ann Otol Rhinol Laryngol 115:880885, 2006. 22. House JW, Sheehy J L , Antunez JC : Stapedectomy in children. Laryngoscope 90:1804-1809, 1980. 23. Hillman TA, Kertez TR , Hadley K, Shelton C : Reversible peripheral neuropathy: The treatment of superior canal dehiscence. Otolaryngol Head Neck Surg 134:431436, 2006. 24. Kennedy R J, Shelton C, Saltzman K L , et al: Intralabyrinthine schwannomas: Diagnosis, management and a new classification system. Otol Neurotol 25:160-167, 2004. 25. Linthicum FH, Ghorayeb VY: Otosclerotic inner ear syndrome. Ann Otol Rhinol Laryngol 87:85-90, 1978. 26. Issa TK, Bagguette M A, Linthicum FH, House H P: The effect of stapedectomy on hearing of patients with otosclerosis and Ménière’s disease. Am J Otol 4:323-326, 1983.
23
Partial Stapedectomy Mendell Robinson
Many surgeons prefer a partial stapedectomy technique for the surgical treatment of otosclerosis. A partial stapedectomy should be performed, however, only when the surgeon follows certain basic principles in stapedial footplate surgery. This chapter presents those surgical principles and the manner in which they apply to the partial stapedectomy technique. This chapter also addresses the application of these principles to total stapedectomy and stapedotomy techniques.
HISTORY In 1958, Shea1 advocated total stapes footplate removal when using the polyethylene strut prosthesis and a vein graft to seal the oval window. Shortly thereafter, Schuknecht2 advocated the use of a wire and fat prosthesis, and House and Greenfield3 later advocated the use of a wire with absorbable gelatin sponge (Gelfoam) prosthesis. The use of the wire/Gelfoam or wire/fat prosthesis necessitated total removal of the footplate for the wire prosthesis to function satisfactorily. Other methods that could be considered a partial stapedectomy technique included the shattered footplate procedure with the use of a polyethylene strut, and the subluxated footplate procedure, both of which necessitated removal of the superstructure of the stapes.4 These two techniques were quickly rejected because of the associated sensorineural hearing loss resulting from either surgical trauma or a perilymphatic fistula. In 1961, the piston concept was introduced, in which a cup/piston prosthesis was used with a connective tissue graft of vein to seal the oval window.5 The introduction of the piston prosthesis no longer required a total removal of the stapes footplate. The concept evolved of “removing only that part of the footplate which comes out easily.”6 This newly introduced surgical technique consequently produced more successful hearing results and fewer inner ear complications. The same technique could be used for the thin blue footplate with minimal otosclerotic involvement, or for the obliterative footplate, in which only a stapedotomy opening (“drill-out”) could be created.
Measurement of the distance between the long process of the incus and stapes footplate was eliminated because a 4-mm prosthesis would protrude into the vestibule 0.2 to 0.3 mm in virtually every case. This slight protrusion would create its self-centering effect. By interposing a vein graft between the prosthesis and vestibule, there was rarely a regrowth of otosclerotic bone, including the obliterative type. Also, because of this protrusion, migration of the piston in subsequent years was virtually unknown. Because of the unique design of the cup/piston prosthesis, the surgeon was permitted a choice of a total stapedectomy or partial stapedectomy or stapedotomy. The classic cup/piston prosthesis was fabricated of 316L stainless steel, an alloy that is inert in living tissue and is nonmagnetic. Its design enhanced the self-centering effect because of the cup attachment to the lenticular process of the incus (the true physiologic point of attachment), and because of the axially placed stem protruding minimally through the footplate opening. The four holes in the cup and the hole on the distal end of the piston encouraged tissue and vascular ingrowth to secure the prosthesis, and to allow capillaries to vascularize the lenticular process, eliminating the avascular necrosis that so frequently occurred with the polyethylene strut. The 4 mm length is used in virtually every stapedectomy case whether it is a drill-out stapedotomy, a partial footplate removal, or a total footplate removal. Any connective tissue can be used with this prosthesis, but I prefer a vein graft that outlines the oval window opening and ensures a complete immediate seal of the oval window. Moon7 advocates the cup/piston prosthesis with areolar tissue. The cup/piston prosthesis is radiopaque and easy to localize and identify on routine mastoid radiographic views. Comparison of impedance studies of various stapes prostheses has shown the stainless steel cup/piston to be the closest in compliance to that of a normal mobile stapes.8 Hearing results reported by otologic surgeons who have used this technique have consistently confirmed a 96% air-bone gap closure to within 10 dB.9-11 Similarly, the complete closure (and overclosure) rate has been repeatedly confirmed at 80%, which the wire/tissue prosthesis and wire pistons have yet to attain. 275
276
OTOLOGIC SURGERY
Because the stapes does not increase in size with age, the 4 mm cup/piston prosthesis has been used in children 5 years old, adults, and elderly individuals. All patients who show an air-bone gap and normal results on otologic examination are candidates for stapedectomy surgery if stapedial fixation is found at the time of the middle ear exploration. Approximately 22% of patients with otosclerosis have a diminished cochlear reserve with a Shambaugh preoperative bone conduction classification of D or E. These patients have very severe mixed hearing losses, and in many, only a partial footplate removal can be obtained. These patients also most frequently show complete closure and overclosure of the air-bone gap. For these patients, the extra 10 dB of hearing gain is crucial because of their mixed hearing loss. Approximately 74% of the patients undergoing surgery have partial or total footplate removal that does not necessitate drilling. The remaining 26% require drilling to create a fenestra 0.8 mm in diameter or larger to accept the vein graft and cup/piston prosthesis.6 In 1980, Austin12 compared the results of total stapedectomy and partial stapedectomy, tissue seal and no tissue seal, and the small fenestra of Smyth. His statistical analysis, using chi-square tables for success or failure, sensorineural hearing loss, fistula, and complete air-bone gap closure, was computed. He concluded that a tissue seal provides a better success rate, a significantly lower risk of fistula, and a better hearing result in terms of complete closure or overclosure of the air-bone gap. Austin noted that in addition to the increased risk of sensorineural complications and fistulas, the small-diameter pistons used in stapedotomies did not provide as good a hearing result.
SURGICAL TECHNIQUE Stapedectomy footplate surgery follows a very basic principle—be as atraumatic as possible in removing the footplate of the stapes. This principle is contingent on removal of “only that part of the footplate which comes out easily.”6 Routine stapedectomy is divided into three stages: (1) exposure of the middle ear, incus, stapes, and footplate area; (2) removal of the stapes superstructure; and (3) removal of the stapes footplate. If the opening is 0.8 mm or larger, the piston can function satisfactorily. The new fenestra must be sealed with a tissue graft; a selfcentering 4 mm cup/piston bridges the gap from the oval window membrane to the incus (Figs. 23-1 to 23-5). Footplate surgery can be classified as follows: 1. Total footplate removal, in which a blue footplate with all oval window margins is visualized 2. Partial footplate removal, in which absent margins are in part of the circumference of the footplate, but where there is a blue area for perforating 3. Total footplate removal with drilling, in which a thick footplate with margins and without a blue center exists for perforating the footplate (biscuit footplate)
4. Partial footplate removal with drilling, in which absent margins are in part of the circumference of the footplate, and there is no blue area for perforating the footplate 5. Drill-out for obliterative otosclerosis, in which no margins of the oval window and thick-mounding otosclerotic bone exist The technique used in this method involves the use of the cup/piston stainless steel stapes prosthesis and a vein graft to seal the recreated oval window. The vein graft is usually taken from the dorsum of the opposite hand by an assistant surgeon at the same time the ear is being operated on. After the vein is opened, it is thinned, and as much of the adventitia is removed as possible. It is trimmed to a final size of 4 × 8 mm. The graft is placed in a small bowl of intravenous saline until it is ready for insertion. The vein is folded over the tip of a smooth-jawed microalligator forceps, umbrella style, and is inserted into the middle ear, avoiding contamination by not touching the ear speculum or the wall of the external auditory canal. The vein is placed over the oval window opening with its adventitial surface facing the vestibule. It is indented slightly into the vestibule with a fine, curved needle so that the exact location of the oval window is visualized. Measurement of the distance between the long process of the incus and the oval window is unnecessary because the 4 mm long prosthesis is used in virtually every case. If there is a drill-out, a 4.5 mm prosthesis optionally may be used to ensure the self-centering effect by protruding 0.7 mm into the vestibule; this also retards formation of new otosclerotic bone. The prosthesis is grasped at its cup end with a small, smooth alligator forceps and targeted into the dimpled portion of the vein graft. It is engaged on the lenticular process of the incus by slightly depressing the socket with the curved needle, and then allowing the prosthesis to rise up and lock onto the lenticular process. The wire loop is rotated over the long process of the incus. The wire is not crimped because the fit is precise. Mucosa grows over the wire loop and secures its position. Before the middle ear is closed, the malleus is gently palpated to ensure that there is mechanical transmission of vibration from the malleolar handle to the prosthesis. The vein graft is inspected to ensure that the edges of the graft overlap the entire oval window area, and that there is no inversion of any of the edges toward the vestibule; otherwise, this problem may predispose to a perilymphatic fistula. The tympanomeatal flap is replaced, and the ear canal is packed with a strip of silk and an expandable methylcellulose otowick and saturated with neomycin and polymyxin B (Neosporin) solution. The canal packing is removed in 1 week. Prophylactic antibiotics are used preoperatively and postoperatively (tetracycline), and steroid/antibiotic eardrops (Cortisporin otic suspension) are used in the ear canal for 1 week to maintain sterility of the ear canal, and to prevent the otowick from drying out.
Chapter 23 • Partial Stapedectomy
277
278
OTOLOGIC SURGERY
This prosthesis, because of its unique design, is a “smart” prosthesis—it can be fitted to the incus under many circumstances. It can fit an angled lenticular proc ess; an extra-large lenticular process; an extra-long long process; a short long process; or a fractured, atrophic, or necrotic long process. When a narrow oval window niche or a prolapsed facial nerve is present, the prosthesis easily self-centers into the vein graft. When there is only partial footplate removal or drill-out for obliterative otosclerosis, the vein graft and prosthesis align perfectly because of the self-centering feature.9
MODIFICATIONS OF THE CUP/PISTON STAPES PROSTHESIS Infrequently, an anatomic variation in the middle ear may result in a prosthesis that performs suboptimally. For these infrequent situations, a cup/piston prosthesis with an offset shaft compensates for a short long process of the incus, a prolapsed facial nerve, and an abnormally high rise of the promontory. When the lenticular process of the incus is absent, or there is erosion of the long process of the incus, the modified Robinson-Moon-Lippy stapes prosthesis compensates for the absent lenticular process. A cutout is in the cup of the prosthesis to accept the eroded long proc ess of the incus, and the shaft is offset in a way similar to the Moon modification. The length of this prosthesis should be 4.5 mm to compensate for the fact that the long process is in the socket rather than above it. The standard dimension of the cup is 0.875 mm, inside diameter, and 0.6 mm, stem diameter, with a 4 mm length. The cup/piston prosthesis is available with a large well (1 mm) and a narrow stem (0.4 mm) for a large lenticular process or narrow oval window niche.
INDICATIONS FOR CUP/PISTON PROSTHESIS The partial or total stapedectomy technique described in this chapter is used for stapes fixation owing to otosclerosis, tympanosclerosis, and osteogenesis imperfecta. It is also used for congenital stapes anomalies and stapedial injuries, including fracture or dislocation of the stapes.
POSTOPERATIVE RESULTS The postoperative hearing results with this partial-total stapedectomy technique have been consistent and repeatable since first reported in 1961. Subsequent reports by myself and others have repeatedly shown a success rate of 96% for closure to within 10 dB of the air-bone gap. A partial success rate of 3% exists in closure to within 20 dB, leaving only 1% unsuccessful.
Delayed hearing losses have been infrequent. In a 21-year period, the incidence of delayed conductive losses was 1.6%, and incidence of cochlear losses of more than 10 dB was 1.2%. The total continued longterm success rate is 93%. These data were derived from a 20-year study of 4815 stapedectomy cases.13 The causes of a delayed conductive loss and their incidence are as follows: otosclerotic regrowth, 3/1000; tympanofibrosis, 2/1000; malleus fixation, 1/1000; postoperative tympanic membrane perforation, 1/1000; incus necrosis, 1/1000; prosthesis migration, 1/3000; prosthesis extrusion (preoperatively healed perforation), 3/1000; and perilymphatic fistula (first vein graft too small), 1/1000. Delayed cochlear losses resulted primarily from perilymphatic fistula (vein graft too small), 1/2000; cochlear otosclerosis, 1/1000; presbycusis, 4/1000; and viral labyrinthitis, 2/1000.
DISCUSSION Comparison with Wire/Stapes Prosthesis With wire prostheses, design, fabrication, and measurement are not standardized. Wires attach to the incus on its long process and not at the physiologic site where the capitulum of the stapes is attached to the incus. Wires are at a disadvantage when there is a short incus, an eroded incus, or a prolapsed facial nerve. Over a long period, there is usually notching of the long process of the incus and occasionally a necrosis of the long process from the notching. Wire/tissue prostheses require total footplate removal, which can be traumatic, and wires are not selfcentering and frequently develop a delayed conductive hearing loss because of migration and impingement on the margin of the oval window. Wire/piston prostheses also exhibit some of the same disadvantages, including the lack of standardization and the incus attachment problems, mentioned earlier. The oval window opening must be precise because an exact fit is mandatory for the stapedotomy technique. The length must be precise: the piston must protrude 0.2 mm into the vestibule. When a stapedotomy opening is created, microfractures of the footplate may occur, increasing the risk of a perilymphatic fistula. Results with wire prostheses vary greatly; success rates range from 80% to 95%. With wire/piston stapedotomy procedures, there frequently is a residual, small air-bone gap in the low frequencies. Reports of complete closure are sparse, but the complete closure rate of the wire/fat prosthesis has been reported to be 15% to 20%, and the complete closure rate of the wire/ piston stapedotomy procedure has been reported to be 50% or less. The cup/piston complete closure rate has consistently been 80%, however. Impedance studies have shown that wires have a very high compliance, whereas cup/piston prostheses have a compliance similar to that of a normal mobile stapes. This high compliance
Chapter 23 • Partial Stapedectomy
of wire prostheses reflects the imperfect attachment to the incus because the wire is crimped over the long proc ess and eventually loosens slightly when notching of the incus occurs. Notching and loosening do not occur with the cup/piston prostheses.8
Stainless Steel Prostheses versus Polytef (Teflon) Prostheses Many materials have been used and advocated for the fabrication of stapes prostheses. Polytef (Teflon) has many advocates because of its inertness and lack of tissue reaction. The stainless steel cup/piston prosthesis, which is fabricated of 316L stainless steel, has also shown a lack of tissue reaction. The main difference between both materials is the weight. A Teflon cup/ piston prosthesis weighs 3.3 mg, and a stainless steel cup/piston prosthesis weighs 12.5 mg. An intact stapes freshly removed from the ear weighs 6 mg. Bluestone14 first reported that by loading a polyethylene stapes prosthesis with a steel core, he would obtain a 10 dB increase in hearing at 4000 Hz and 8000 Hz compared with that of the control group. I reported a study comparing the hearing results in patients with a Robinson stainless-steel cup/piston prosthesis and vein graft in the first ear and a Robinson Teflon cup/piston prosthesis and vein graft in the second ear.15 This study eliminated all variables except the difference in weight of the prostheses. The preoperative hearing level in each ear was the same, as was the footplate pathology in both ears. The design of the prosthesis was exactly the same; only the weight differed. The stainless steel prosthesis weighed almost four times that of the Teflon prosthesis. The results showed that the high-frequency gains in hearing were slightly better for the stainless steel ear (14 dB) than for the Teflon ear (11 dB). The rate of complete closure and overclosure of the air-bone gap was significantly greater in the stainless steel ear (80%) than in the Teflon ear (52%), even though the overall closure of the air-bone gap to within 10 dB was 97% in the stainless steel ear and 96% in the Teflon ear. Impedance studies of this group showed that the compliance of the stainless steel stapes prosthesis most resembled that of the normal mobile stapes, whereas the Teflon prosthesis showed a slightly reduced compliance (high impedance), but as a cup/piston, its compliance curves were more within the range of the mobile stapes than any of the wire prostheses. This study proved that the heavier prosthesis resulted in a much greater complete closure and overclosure of the air-bone gap, and that there was a small but significant hearing advantage in postoperative results when a metallic prosthesis was used (6 dB). In mixed-type hearing loss, this increased hearing can have a significant effect on the patient’s reaching a serviceable postoperative hearing level.15
279
Juvenile Otosclerosis Otosclerosis is usually considered to be a disease of young and middle-aged adults, but juvenile otosclerosis occurs in 15.1% of stapedectomy cases before age 18. I presented an in-depth study of 610 patients (of a total of 4014 patients) who developed clinical otosclerosis before age 18 and underwent stapedectomy.16 Of these 610 patients, 35 underwent surgery before age 18, and 574 underwent surgery after age 18, but their hearing loss had developed during their juvenile years. The youngest patient was 5 years old, and the average age of onset was 11.5 years. The surgical technique performed on all 610 patients consisted of a partial or total footplate removal, as described in this chapter. All patients younger than 18 received general anesthesia, whereas all patients older than 18 had local anesthesia (lidocaine [Xylocaine] 2% with 1:30,000 epinephrine). The surgical footplate pathology varied considerably between patients younger than 18 and patients older than 18. In the patients younger than 18, 66.7% had a thin blue footplate, which was totally removed, whereas in patients older than 18, only 45.1% had similar pathology. Partial footplate removal was performed in 5.5% of the patients younger than 18 and 9.7% of the patients older than 18. More significantly, drilling of the footplate because of diffuse otosclerotic involvement was necessary in 27.8% of the patients younger than 18 and in 45.2% of patients older than 18. Rate of obliterative otosclerosis, which required drill-outs, was 5.6% in the juvenile group and 12% in their older counterparts. This result is in contrast to 3% prevalence of obliterative otosclerosis for all stapedectomies. The hearing results were more favorable in patients who underwent surgery before age 18; 100% had an airbone gap closure of 10 dB or less, and 77.6% had complete closure or overclosure of the air-bone gap. Patients who underwent surgery after age 18 had a very successful closure rate, but only 93.6% had the air-bone gap closed to within 10 dB. Similarly, 77.3% had complete closure or overclosure. A postoperative delayed conductive hearing loss (20 dB) occurred in only one patient 5 years after the initial surgery in the group younger than 18. The postoperative hearing level in all patients younger than 18 (4000 Hz) remained the same or improved. This study also revealed a very high incidence of bilateral otosclerosis (92%) when the hearing loss occurred before age 18. Of patients younger than 18 with otosclerosis, 80% had excellent cochlear reserve and did not show a deterioration of sensorineural function after stapedectomy. The longer the hearing loss existed, the greater the degree of footplate pathology that occurred. The probability of requiring a drill-out for obliterative otosclerosis increased fourfold when surgery was deferred to after age 18. Deferring a stapedectomy procedure in a child may not be in the best interest of the patient because of the progressive nature of the footplate pathology and the increased necessity of drilling the footplate.16
280
OTOLOGIC SURGERY
SUMMARY Stapedectomy and stapedotomy for the correction of hearing loss resulting from otosclerotic bony fixation of the stapes are surgical procedures that have evolved almost entirely in the “golden era” of otologic surgery (1955-1985). This era has passed, and even the most popular of the nationally known surgeons who perform stapedectomy have experienced a dramatic decrease in the caseload for stapedectomy. This decrease, in addition to the dispersion of patients to a larger number of trained otolaryngologists, has to some degree compromised the position of the otologic surgeon not only in developing skills and judgment, but also in maintaining the acquired skill necessary to perform the more complicated and precise techniques. General otolaryngologists and otologists should adopt a technique that is uncomplicated, easy to perform, predictable, and safe for the patient. This technique should be standardized so that it can be performed in the same way for all stapedectomies. Reducing the variables and making one technique work for all cases of stapes fixation would allow the general otolaryngologist gradually to develop the high degree of skill necessary for this procedure. The technique adopted should have minimal risks and immediate or delayed complications, and, if possible, should have the lowest incidence of revision because revisions demand more skill, judgment, and knowledge than primary stapedectomies. More than 200,000 stapedectomies have been performed in the United States with the technique and prostheses described in this chapter. Despite the overall decrease in the number of stapedectomy procedures performed annually, the number of Robinson cup/piston stapes prostheses used each year continues to increase, indicating the popularity of this simpler and safer technique.
REFERENCES 1. Shea JJ Jr: Fenestration of the oval window. Ann Otol Rhinol Laryngol 57:932, 1958. 2. Schuknecht H : Stapedectomy and graft prosthesis operation. Acta Otolaryngol (Stockh) 51:241-243, 1960.
3. House H P, Greenfield EC : Five-year study of wire-loop absorbable gelatin sponge technique. Arch Otolaryngol Head Neck Surg 89:420-421, 1969. 4. Goodhill V: Stapes Surgery for Otosclerosis. St Louis, Paul B Hoeber, 1961, p 136. 5. Robinson M : The stainless-steel stapedial prosthesis: One year’s experience. Laryngoscope 73:514, 1962. 6. Robinson M : A four-year study of the stainless-steel stapes. Arch Otolaryngol Head Neck Surg 82:217-235, 1965. 7. Moon C : Stapedectomy connective tissue graft and the stainless-steel prosthesis. Laryngoscope 78:798-807, 1968. 8. Feldman A : Acoustic impedance measurement of poststapedectomized ears. Laryngoscope 79:1132-1155, 1969. 9. Schondorf J, Pilorget J, Graber S : [The influence of the stapes prosthesis on the long-term results of stapedectomy.]. Head Neck Otolaryngol 28:153-157, 1980. 10. Elonka D, Derlacki E, Harrison W: Stapes prosthesis comparison. Otolaryngol Head Neck Surg 90:263-265, 1982. 11. Girgis T: Stapedectomy: Robinson stapes prosthesis versus wire prosthesis. Presented at American Society of Otology, Rhinology, and Laryngology, Middle Section Meeting, Chicago, January 1985. 12. Austin D F: Stapedectomy with tissue seal. In Snow J B Jr (ed): Controversy in Otolaryngology. Philadelphia, Saunders, 1980. 13. Robinson M : Total footplate extraction in stapedectomy. Ann Otol Rhinol Laryngol 90:630-632, 1981. 14. Bluestone C D: Polyethylene stainless-steel core in middle ear surgery. Arch Otolaryngol Head Neck Surg 76:303, 1962. 15. Robinson M : Stapes prosthesis: Stainless steel versus Teflon. Laryngoscope 84:1982-1995, 1974. 16. Robinson M : Juvenile otosclerosis. Ann Otol Rhinol L aryngol 92:561-565, 1983.
24
Laser Revision Stapedectomy Larry B. Lundy Videos corresponding to this chapter are available online at www.expertconsult.com.
In recent years, the safety and efficacy of revision s tapedectomy have come under scrutiny. Experienced surgeons report that the results of revision stapedectomy are often worse than results of primary stapedectomy, and that the risks of sensorineural hearing loss, tinnitus, and vertigo are increased. With the application of laser technology to revision stapes surgery, less traumatic and more precise techniques can be applied, allowing better results and diminished risks compared with revision stapedectomy without lasers. This chapter reviews the clinically relevant principles of laser technology, compares results of revision stapedectomy with and without laser application, defines candidates for surgery, and reviews surgical technique.
LASER PHYSICS AND PRINCIPLES Laser energy is derived from the release of energy (photons) occurring when stimulated electrons return to their resting orbital. As proposed by Einstein in 1917, when photons of the appropriate wavelength strike excited atoms, a second additional photon is released as the electron returns to its ground state (Fig. 24-1). In this stimulated emission situation, the photon that is emitted from the excited atom has exactly the same frequency, direction, and phase as the incident photon, providing laser energy that is collimated, coherent, and monochromatic. Lasers are typically named by the active medium, or the source of atoms that are excited and undergo stimulated emission of photons. The active medium can be a liquid, solid, or gas. Common gas lasers include CO2, argon, and helium-neon. An example of solid state lasers are the neodymium:yttrium-aluminum-garnet (Nd:YAG) and the potassium titanyl phosphate crystal (KTP). The KTP laser is simply a Nd:YAG laser beam that passes through a KTP crystal, which halves the wavelength and doubles the frequency of the laser beam (Table 24-1). The wavelength of the emitted photons, or laser beam, has important characteristics for tissue interaction. Lasers whose wavelengths fall into visible (380 to
700 nm) and infrared (700 nm to 1 mm) portions of the electromagnetic spectrum are considered thermal lasers. Interaction of these lasers with normal biologic materials is mediated by a photothermal process. On contact with tissue, the laser energy is converted to thermal energy, resulting in a rapid increase in tissue temperature. The laser-tissue interaction depends as much on the tissue type and its composition (e.g., bone, muscle, cartilage, or nerve) as it does on the laser energy. Visible-spectrum laser (argon at 514.5 nm and KTP at 532 nm wavelengths) energy absorption by tissue depends partly on tissue color. For soft tissue work, chromophores of hemoglobin and melanin absorb most of the energy. Lighter color tissues reflect most of the laser energy. Energy absorption from the invisible CO2 laser (10,600 nm wavelength) is primarily by intracellular and extracellular water, which is instantaneously converted to steam. For any laser, the magnitude of the laser-tissue interaction can be regulated by the laser’s power output, the power density at the point of impact, and the energy fluence. Every surgeon who uses a laser should thoroughly understand these fundamental concepts. Power is the time rate at which energy is emitted, and is expressed as watts. The power output is directly adjusted by the control panel on the laser console. Laser energy is delivered through a focusing lens. Power density is a measure of the intensity, or concentration, of the laser beam spot size (Fig. 24-2). It is the ratio of power to surface area of the spot size, and is expressed in terms of watts per square centimeter: Power density =
power (W) area of spot size (cm2)
where area of spot size is πr2, and where r = spot size radius in centimeters. Power density is inversely proportional to the square of the radius of the spot size. Consequently, for any specific power output, changes in the spot size can have a tremendous effect on power density (Fig. 24-3). The third fundamental, practical concept is that of fluence, which is simply the power density × time, 281
282
OTOLOGIC SURGERY
FIGURE 24-1. According to the Bohr atomic model, an electron can absorb a photon, and the electron makes a transition to a higher energy level from its normal ground state. The electron eventually returns to the ground state by the spontaneous emission of a photon. In the condition of stimulated emission, a photon of appropriate energy interacts with the electron already in its excited state, causing the release of two photons. (From Weisberger EC: Lasers in Head and Neck Surgery. New York, Igaku-Shoin, 1991.)
TABLE 24-1 Characteristics of Laser Types LASER TYPE Characteristic
Argon
KTP-532
CO2
Medium Wavelength Color Smallest spot size Delivery Absorption
Gas 488-514 nm Blue-green 0.150 mm Handpiece or micromanipulator Pigment
Crystal 532 nm Green 0.150 mm Handpiece or micromanipulator Pigment
Gas 10,600 nm Invisible 0.150 mm Micromanipulator or optical fiber Water
expressed in joules (J). This is the total amount of energy delivered to the tissues: Fluence (J) =
power (W) × exposure time (sec) area of spot size (cm2)
As fluence increases, the volume of affected tissue also increases. The thermal energy having an impact on tissue increases dramatically as the time of exposure increases. If power density (W/cm2) is held constant, and exposure time is doubled, the energy delivered is doubled. The thermal effect of the tissue increases significantly, however, because the increase in temperature is continuous. For example, assume the laser power is set at 2 W, the spot size is 0.2 mm2, and exposure time is 0.2 second. Is this the same as delivering two separate impulses at 0.1 second each? Yes and no. Yes, it is the same in terms of energy delivered from the laser. But no, it is not the same in terms of thermal energy imparted to the tissue because during the time between the two separate 0.1 second pulses, no matter how brief, the tissue is cooling (Fig. 24-4). Other factors come into play, such as the absorption characteristics of the damaged tissue in the center of the laser spot and dissipation of heat, but
the important concept is that of the increase and decrease of the temperature.
HISTORY OF REVISION STAPEDECTOMY Background After the introduction of the stapedectomy procedure by Shea1 in 1958, the surgical treatment of otosclerosis was revolutionized. During the 1960s and 1970s, otologic surgeons performed hundreds of thousands of stapedectomies. An otologic surgeon commonly performed thousands of stapes procedures during his or her peak professional years. The accepted success rate (as defined by closure of the air-bone gap to ≤10 dB) was 90% or greater, with a 1% or less incidence of significant sensorineural hearing loss, including deafness. The most common technique during this time was the total stapedectomy, with removal of the entire stapes footplate. The small fenestra technique began gaining some acceptance, although not universal, in the early 1980s. As with any surgical procedure, the success rate of stapedectomy was not 100%; with even a small percentage of failures (i.e., air-bone gap closure of >10 dB)
Chapter 24 • Laser Revision Stapedectomy
283
FIGURE 24-2. Planes of focus with spot size and power density.
indow and vestibule, and would place a prosthesis on w the existing oval window membrane. By doing so, the prosthesis would often rest on a thick fibrous oval window membrane or perhaps a residual bony footplate or new bone growth. Although this technique protected against sensorineural hearing loss, the lower incidence of closure of the air-bone gap to within 10 dB remained. The stark contrast of results of revision stapedectomy compared with primary stapedectomy, plus technologic advances in hearing aids, diminished the enthusiasm for revision stapes surgery in all but the most experienced hands. FIGURE 24-3. Relationship between spot size and power density.
in such a large pool of patients, there was a significant number of patients who were candidates for revision stapes procedures. Stapes surgeons soon discovered two important facts: The success rate of revision stapedectomy was not nearly as high as that of primary stapedectomy, and the incidence of significant sensorineural hearing loss, including dead ears, was significantly higher than the incidence associated with primary stapedectomy. Prominent otologists obtained air-bone gap closure within 10 dB in 50% or less of revision cases.2-7 The incidence of significant postoperative sensorineural hearing loss ranged from 3% to 20%, with 14% having profound loss.2,3,6,8 Glasscock2 and Sheehy3 and their associates and Lippy and Schuring8,9 advocated leaving the oval window neomembrane intact and undisturbed, if possible, in revision cases to reduce the risk of severe sensorineural hearing loss, even though it may result in fewer patients with postoperative hearing improvement. Feldman and Schuknecht,4 Pearman and Dawes,10 and Derlacki5 reported opening the neomembrane to identify the vestibule and ensure correct prosthesis placement. To diminish the risk of significant sensorineural hearing loss, the surgeon often would not open the oval
Laser Concomitant with the realization that revision stapedectomy surgery was not as successful as primary stapedectomy was the introduction of lasers in temporal bone surgery. The use of the laser in temporal bone surgery was the object of experiments in 1967 by Sataloff11 and 1972 by Stahle and colleagues.12 The evolution of laser otologic surgery has been based on a mixture of clinical, experimental animal, and laboratory observations. In 1977, Wilpizeski13 examined argon and CO2 lasers on monkeys by performing myringotomy, ossicular amputation, stapes fenestrations, lysis of stapedial tendon, and crurotomy. He noted damage to the organ of Corti in monkeys after using “excessive power.” Escudero and associates14 were the first to use a laser in human otologic surgery in 1977. They used the argon laser with a fiberoptic handpiece to tack temporalis fascia to tympanic membrane perforations. In 1979, Perkins15 presented a preliminary report of argon laser stapedotomy with excellent initial results in 11 patients. In 1980, DiBartolomeo and Ellis16 expanded argon laser applications in 30 patients for middle ear and external ear soft tissue and bony problems. In 10 patients, otosclerosis was corrected, including one revision case. In 1983, McGee17 reported on the use of argon laser in
284
OTOLOGIC SURGERY
A
B FIGURE 24-4. A and B, Temperature and time of exposure relationship for a single pulse (A) versus two separate pulses (B).
more than 500 otologic cases, 100 of which were primary stapedectomies. There were no laser-related complications in his study. In 1989, McGee18 reported an update on 2500 tympanomastoid procedures, of which 510 were primary stapedectomies. By comparing 100 consecutive laser stapedectomies with a previous 139 small fenestra stapedectomies using instruments, McGee found that the laser technique permitted much shorter hospital stay, less vertigo, and excellent hearing results (93% air-bone gap closure ≤10 dB at 6 months). This large study indicated the safety of argon laser use for stapedectomy, and yielded comparable hearing results and less vertigo. This clinical evidence is inconsistent with experimental animal studies, in which temporary changes of cochlear microphonics and saccular perforations were noted.17,19,20 Clinical experience from several centers has illustrated the safe use of lasers in ear surgery; however, arguments and opinions persist regarding the best type of laser.21-25
Comparison of Lasers The visible spectrum lasers were used initially because they were the only ones that were precise and accurate enough for stapes surgery. Similar to any tool or instrument, each type of laser has advantages and disadvantages. The visible-wavelength lasers (argon and KTP) have the practical advantage of precision because the aiming beam and the working beam are the same. The blue-green (argon) or green (KTP) aiming beam has clear, crisp margins,
which allows extreme precision. Several years ago, some manufacturers of the KTP laser incorporated the use of a separate helium-neon laser aiming beam. This heliumneon aiming beam produces a red spot, and the margins are not as clear and crisp as the blue-green argon laser or green KTP laser. The alignment of the two beams (the helium-neon aiming beam and the visible laser beam) remains very precise because the wavelengths are similar (helium-neon is 632.8 nm wavelength). The CO2 laser beam is invisible; a separate laser aiming beam (helium-neon) is required. This aiming beam is coaxial and focuses in a different plane from the CO2 laser beam because of the differences in wavelength of the helium-neon and CO2 lasers. With these lasers, there is more potential for misalignment and a greater margin of error than with the KTP and argon lasers. It is crucial that the CO2 laser and its helium-neon aiming beam be calibrated precisely and checked frequently during the procedure to ensure maximum accuracy. The tissue absorption of the visible-laser energy is color dependent, and for the argon and KTP lasers, peak absorption is dark red. For lightly colored tissues, such as a neomembrane or bone, a significant amount of laser energy is reflected rather than absorbed. A practical solution involves placing a minute quantity of blood in the field, or applying several bursts (not in rapid succession) to get a dark char, increasing the laser energy absorption. With the CO2 laser, this absorption is not a problem because the laser beam is invisible, and absorption by water and tissue is not color dependent. The visible-wavelength laser beam can be carried by thin fiberoptic cables, which allow two modes of delivery: a micromanipulator attached to the microscope or a hand-held probe (Fig. 24-5). The hand-held probe provides greater angle of divergence of the laser beam (i.e., rapid deterioration of power density) and must be placed very close to the tissue. This instrument is directly in the operative field and can obstruct a portion of the visual operative field. With the micromanipulator, the angle of the laser beam is less divergent. It does not require an instrument in the field (except a suction for smoke plume removal), but does require the use of a “joystick” control mechanism to direct the beam. The CO2 laser energy cannot be carried by a standard fiberoptic cable; use of a micromanipulator system is the only choice. Older CO2 lasers have a system of articulated arms and mirrors that significantly reduces accuracy. Newer models allow the optical chamber to be attached to the side of the microscope, however, with fewer arms and mirrors, a significant technologic advance. The delivery of CO2 laser energy has potentially advanced with the introduction of an omnidirectional photonic bandgap reflector system developed by researchers at Massachusetts Institute of Technology in 1998. This technology allows for the CO2 laser beam to be transported within a hollow-core, flexible fiber, eliminating the need for articulating arms and mirrors.26
Chapter 24 • Laser Revision Stapedectomy
A
285
B
FIGURE 24-5. A, Micromanipulator delivery system. B, Hand-piece delivery system.
A commercial organization (OmniGuide, Inc.) was established in 2004, and the first otology device, OtoBeam, was used in June 2007. This optical fiber has an outer diameter of 0.9 mm, with a spot size of 0.25 mm. The angle of divergence when the CO2 beam leaves the end of the optical fiber is 7 degrees; the tip of the optical fiber must be relatively close to the tissue to avoid significant power loss. In recent years, questions have arisen regarding the safety of visible-spectrum lasers in stapedectomy and revision stapedectomy. The issue revolves around depth of penetration of laser energy into an open vestibule with potential injury to inner ear structures, such as the saccule or utricle. No laser surgeon advocates firing any type of laser beam into an open vestibule. These concerns have not been borne out, as experience with hundreds of patients and multiple authors has indicated.15,16,21-23,25,27 Serious theoretical safety issues related to visible spectrum lasers arose when Lesinski23 pointed out that the shorter wavelength of the visible wavelength (argon and KTP) lasers penetrates tissue more deeply than the longer wavelength (CO2) lasers. By constructing a model of the vestibule and placing a black painted thermocouple in it at a depth comparable to the saccule or utricle, he measured very high temperatures when the visible laser was allowed to strike the thermocouple directly. This effect was not noticeable with the CO2 laser. He concluded there was the potential for the visible lasers to traumatize the saccule or the utricle or both, risking sensorineural hearing loss. Years of clinical experience with visible lasers used for primary and revision stapes surgery had failed to show clinical evidence to support this concept. This discrepancy between theoretical and clinical experience is probably best explained by the fact that visible laser energy is maximally absorbed by darkly colored or pigmented structures. A thermocouple painted black would absorb most of the laser energy of the KTP (green) and the argon
(blue-green) laser. In the human inner ear, there are no darkly colored structures, however. The saccule and the utricle are a light pink; the laser energy is much more likely to be reflected rather than absorbed. A second possible reason that clinical experience fails to support the theoretical concern is the possibility that microperforations of the saccule or utricle or both do occur, but may have no clinical significance or measurable effect on hearing.
Laser Use: 1990s to 2007 The renewed interest in laser use for revision stapes surgery combined with the theoretical issues of “best” type of laser (visible versus invisible) sparked a series of selfappraisals by stapes surgeons. The two major concerns of revision stapes surgery—successful outcome (air-bone gap ≤10 dB) versus sensorineural hearing loss—continue to be important, regardless of use of the laser or not. In essence, revision stapes surgery has improved as surgeons have employed better selection criteria. A recurring theme throughout the literature is that patient selection for revision stapes affects the outcome. If a patient never had a good result to begin with, the chances are less that a good result would be obtained by revision. Likewise, patients with vertigo and reaccumulation of the air-bone gap do not fare as well. The results of revision stapes surgery without the laser are better than historical controls of the 1970s and 1980s (Table 24-2).28-34 The results of revision stapes surgery using the laser are also better than the historical controls, and show a lesser incidence of sensorineural hearing loss (Table 24-3).23,27,35-40
Analysis of Failed Stapes Surgery The analysis of the long-term success and failure rate of primary stapes surgery is difficult for several reasons. The original surgeon seldom has the opportunity to follow all of the patients over the long term; the surgeon would be
286
OTOLOGIC SURGERY
TABLE 24-2 Revision Stapes Procedure without Laser (1990s-2007) Study
No. Cases
No. (%) with ≤10 dB Air-Bone Gap
No. (%) with SNHL
No. (%) with Dead Ear
250 74 66 66 49 99 63 667
200 (80) 34 (46) 30 (46) 40 (61) 8 (16) 58 (59) 33 (52) 403 (60)
13 (5.2) 3 (4.1) 5 (7.6) 2 (3) 0 0 1 24 (3.6)
0 1 (1.4) 0 0 2 (4) 0 1 4 (0.6)
Hammerschlag et al28 Han et al29 Prasad and Kamerer30 Langman and Lindeman31 Cokesser et al32 Farrior and Sutherland33 Gros et al34 Total SNHL, sensorineural hearing loss.
TABLE 24-3 Revision Stapes Procedure with Laser (1990s-2007) Study Lesinski23 McGee et al27 Wiet et al35 Nissen36 Horn et al37 Haberkamp et al38 Silverstein et al39 Sarac et al40* Total
No. Cases
No. (%) with ≤10 dB Air-Bone Gap
No. (%) with SNHL
No. (%) with Dead Ear
59 77 23 21 32 25 38 36 311
39 (66) 62 (80.5) 12 (52) 9 (43) 24 (75) 16 (65) 19 (50) 16 (44) 203 (65)
0 1 (1.3) 0 1 (5) 0 2 (8) 0 1 5 (1.6)
0 0 0 0 0 0 1 (2.5) 0 1 (0.3)
*Laser
in most, but not all, cases. SNHL, sensorineural hearing loss.
unaware of some of the failures. Most of the revision stapes surgery performed is often done by another surgeon other than the original. In our mobile society, patients often move to other locations. If their original successful result deteriorates, they may simply seek a hearing aid. Younger and middle-aged patients may outlive the older experienced surgeon, or at least live longer than the surgeon’s active practicing years. Finally, it is common that the original operative reports and audiograms are unavailable years later when the revision stapes is being considered. Although these circumstances are not unique to otosclerosis and otology, they increase the difficulty of establishing the long-term outcome of stapes surgery and of comparing different techniques, prostheses, and surgeons’ outcomes. Nonetheless, several factors responsible for the reaccumulation of the air-bone gap have been repeatedly identified in almost all studies. The most common intraoperative finding associated with recurrence of conductive hearing loss is a displaced prosthesis.29,44,45 Findings at the time of revision stapes surgery generally can be classified as common or uncommon (Table 24-4). Although there are no excellent long-term follow-up data to support or implicate any one technique because of the previously mentioned social factors, certain conclusions can still be reached. Most of the primary stapes
surgery involved total or near-total removal of the footplate during the years of frequent stapes surgery. Regardless of the prosthesis used, the oval window tissue seal is much larger than the prosthesis. All of the soft tissue seals (lobule fat, areolar fascia, temporalis fascia, vein, tragal perichondrium) have to occlude the oval window to prevent perilymph leak. Subsequently, a snug fit of this tissue results in prolapse of the tissue into the vestibule to some degree, or a relative heaping up of the tissue in the oval window niche, or both. As this tissue heals, portions within the vestibule could easily band to the saccule or utricle41,42 because the vestibule is only a few millimeters deep. When this tissue in the oval window is removed mechanically during revision surgery, tears and avulsions of the saccule or utricle or both could easily occur, resulting in vertigo or significant sensorineural hearing loss, or both. This potential for inner ear damage can be estimated only because there is no good way to assess this possible condition preoperatively or intraoperatively. Accurate centering of the prosthesis in the oval window with these tissue grafts and total footplate removal is difficult because the margins of the oval window are obscured by the tissue grafts. As healing occurs, this tissue fibroses and matures, presumably resulting in scar contracture, which causes the distal end of the prosthesis to migrate to the margin of the oval window,
Chapter 24 • Laser Revision Stapedectomy TABLE 24-4 Findings at the Time of Revision
Stapes Surgery
Common
Uncommon
Displaced prosthesis Incus erosion Fibrosis of oval window New bone growth Prosthesis too short Residual footplate
Dislocated incus Prosthesis too long Fixation of malleus/incus Depressed footplate fragment Reparative granuloma Perilymph fistula
where it adheres to the bony margin of the oval window. Subsequently, the combination of this adhesion/fixation and the inefficient angle of vibration of the prosthesis results in a conductive hearing loss from fixation. With stapedotomy or small fenestra techniques, this is a much less frequently encountered cause of failure because the prosthesis has much less tendency to migrate within the oval window. With total or near-total footplate removal using mechanical techniques, footplate fragments can be left in the oval window or depressed into the vestibule. These fragments may be under-recognized because the purpose of revision stapes surgery is to re-establish an efficient conductive mechanism, not the total exploration of the oval window. These bone fragments, if contacting the prosthesis, impede the motion of the prosthesis, resulting in a conductive hearing loss. Bone fragments can be embedded in the soft tissue scar and be in contact with the saccule or the utricle or both. The removal of the scar tissue with mechanical techniques can concomitantly remove the embedded bone fragment and tear the membranous structures in the vestibule. There is no certain way to assess this possible condition preoperatively or intraoperatively. Incus erosion at the lenticular process at the attachment of the prosthesis is another common finding in revision cases.43-46 One theory of causation holds that a tightly crimped shepherd’s crook results in ischemia and subsequent necrosis of the lenticular process.7,47,48 Another theory holds that a poorly performed crimping initially or owing to the “spring-back” nature of stainless steel used in the shepherd’s crook results in a loose-fitting shepherd’s crook.49 This laxity results in differential vibration of the incus and wire, ultimately eroding a notch in the lenticular process. Several prostheses are available with platinum ribbons for the shepherd’s crook, supposedly for easier and better fitting crimping. Also, as noted earlier, migration of the distal end of the prosthesis, particularly with previous total stapedectomy, can result in fixation of the distal end of the prosthesis. Subsequent normal vibration of the incus within the prosthesis crook or bucket handle would likely result in erosion of the incus at its attachment. The appearance of the lenticular process is that it has been “sawn” through from repeated vibration.
287
Uncommon findings, such as a subluxated incus, previously unrecognized malleus/incus fixation, and improper prosthesis length, are avoidable causes of failure with proper training and clinical experience. Bony regrowth in the oval window owing to otosclerosis is a condition without a proven prevention strategy.
Case for Use of Lasers in Revision Stapes Surgery During the 1990s through 2007, several studies analyzed the success and complication rates of laser revision stapes surgery (see Table 24-3). The results are consistent regardless of the type of laser used or the method of delivery of the laser energy. The average success rate in closing the air-bone gap to within 10 dB is 65%; the incidence of significant sensorineural hearing loss is 1.6%, and the incidence of dead ears is 0.3%. There is no significant difference in outcomes using the visible spectrum or CO2 laser. Although these results do not equal the results of primary stapes surgery, they do represent an improvement of the historical controls. How could the laser account for some of this improvement, particularly when there has also been an improvement in the results of revision stapes surgery without the use of lasers? The answer lies in the management of the underlying cause of the original failure. Lasers give the surgeon an additional resource with which to deal with unforeseen or difficult circumstances. Proponents of use of the laser in revision stapes surgery maintain it is less traumatic and more precise than mechanical instruments. Laser energy in the soft tissue of the oval window results in less bleeding than with instrumentation, which improves visualization. Laser “hits” vaporize soft tissue, allowing for precise delineation of the margin of the oval window, identification of the prosthesis–soft tissue interface, recognition of regrowth of bone and existing bone fragments, and precise sizing of the fenestra. If a bony fragment is identified within the fibrous tissue of the oval window in the ideal site of planned prosthesis placement, it can be vaporized without having to manipulate it with instruments. Even with otosclerotic bony regrowth in the oval window, the laser can be used to create the fenestra and avoid the microdrill or pick. This ability to manage the oval window with less trauma partly accounts for the improved success rates and the decreased rate of sensorineural hearing loss.
Limitations of Lasers A primary limitation of lasers is the learning curve associated with the micromanipulator or the fiberoptic handpiece. Visible lasers have both options, whereas the CO2 laser uses the micromanipulator only, owing to the physics of the long wavelength of CO2 and fiberoptic cables. The potential for buildup of thermal energy in the oval window tissue is real; the rapid-fire sequence of laser
288
OTOLOGIC SURGERY
hits should be avoided. Experienced laser surgeons recommend a 2 second pause between hits to allow for adequate tissue cooling. Evacuation of the smoke plume with a small suction also dissipates heat from this area by providing regional airflow around the oval window. No laser surgeon advocates allowing laser energy delivery directly into an open vestibule, regardless of the laser type; the precision of contemporary lasers permits one to avoid this. The laser can also be useful in managing the incus because it can be used to vaporize the necrotic tip of the lenticular process. A prosthesis using nitinol for the shepherd’s crook has been introduced for primary stapes surgery.50 Nitinol is an alloy with shape memory made of nickel (55.3% by weight) and titanium (44.7% by weight).51 The shepherd’s crook “shrink wraps” around the incus or malleus with the introduction of heat at 45° C (113° F). This heat can easily be obtained with a single KTP laser shot at 0.5 W at 0.1 second duration or use of a battery-powered cautery unit (eye cautery). Current published literature is limited to primary stapedotomy with this prosthesis,50-54 but the author has used it on revision cases successfully with its attachment to the incus and the malleus. The use of a stapes prosthesis with a nitinol self-crimping shepherd’s crook eliminates the need for manual crimping, greatly simplifying this step. When this prosthesis is used for malleus attachment, care must be taken not to place the heat source too close to the tympanic membrane because a perforation could result. With the KTP laser, this possibility is greatly reduced because of the ability to place the laser spot accurately on the shepherd’s crook.
TECHNIQUE Primary or revision stapedectomies can be performed with the patient under local anesthesia with intravenous sedation or under a general anesthetic. For anesthesia with intravenous sedation, typically 50 to 100 μg of fentanyl citrate and 1 to 2 mg of midazolam are administered intravenously before the ear is prepared and draped. A short-acting barbiturate (50 to 100 mg of thiopental) is administered intravenously just before infiltration of the ear canal. This agent allows a brief somnolence, allowing painless infiltration of the local anesthetic. With a 27 gauge needle, 1% lidocaine (Xylocaine) with 1:15,000 dilution of epinephrine is administered to infiltrate the ear canal skin. Typically, only 0.3 to 0.5 mL of this solution is used. The high concentration of epinephrine is not necessary for adequacy of vasoconstriction, but is used for a more rapid onset of vasoconstriction. During the surgical procedure, if the patient is restless and seems inadequately sedated, additional medication can be given. Caution must be exercised because an impatient surgeon or an inexperienced anesthesiologist can overmedicate, which paradoxically increases restlessness and movement.
Surgeons performing stapes procedures in the past frequently avoided general anesthesia. Typically, the patient with general anesthesia would have an endotracheal tube, which is a very strong, noxious stimulus to the larynx and trachea. The anesthesiologist would have to give high doses of anesthetics to suppress coughing and bucking during surgery. During emergence from anesthesia and extubation, the patient would frequently buck, gag, and cough, however. There was concern by surgeons that transmitted pressure could dislodge the prosthesis or graft, resulting in a poor outcome. In addition, the anesthetic agents (especially narcotics) plus oval window manipulation would cause the patient to become nauseated, and retch or vomit in the recovery room and immediately postoperatively, also causing concern of a poor outcome. Since 1996, the author has almost exclusively used general anesthesia employing a laryngeal mask anesthesia system. With general anesthesia, the patient is completely motionless. Local infiltration of the canal skin is with the same mixture noted earlier. With the use of laryngeal mask airway, the patient has no laryngeal or tracheal irritation and is able to breathe spontaneously on his or her own. Particularly, extubation is much smoother without coughing, retching, and bucking. Severe, active gastroesophageal reflux disease is a contraindication to the laryngeal mask anesthesia technique. Patients with mild to moderate gastroesophageal reflux are given a histamine blocker (e.g., ranitidine) and metoclopramide intravenously in the immediate preoperative period. At the conclusion of the case, the laryngeal mask airway is removed with the patient still under deep anesthesia to avoid coughing and retching. As mentioned, the author has used this technique since 1996, and all patients undergoing primary and revision stapes surgery are discharged home within a few hours after surgery is completed. A transcanal tympanomeatal-stapedectomy flap is raised with the patient under local anesthesia with vasoconstriction and intravenous sedation. On entry into the middle ear, the cause of the conductive hearing loss is assessed (Fig. 24-6). The malleus and incus are inspected and gently palpated to rule out fixation. Any obstructing middle ear fibrous adhesions are lysed with the laser. The chorda tympani nerve, if present, is frequently adhered to the tympanomeatal flap and can be sharply dissected with the laser. At the appropriate setting, the obliterating tissue of the oval window surrounding the prosthesis is vaporized until the exact oval window margins and depth are identified. For the KTP laser, the spot size is 0.15 mm, the power setting is 1.2 to 1.4 W, and the pulse duration is 0.1 second. For the CO2 laser on the superpulse mode, the spot size is also 0.15 mm, the power setting is 0.8 to 1 W, and the pulse duration is 0.1 second. The attachment of the prosthesis at the incus is freed or loosened with a right angle hook. The prosthesis may be removed at this point or may require further lysis at its base (Fig. 24-7). A series of laser hits are applied to the
Chapter 24 • Laser Revision Stapedectomy
289
FIGURE 24-6. A, Elevation of tympanomeatal flap. B, Migrated prosthesis with oval window obliteration.
FIGURE 24-7. A, Dissection of distal stapes prosthesis. B, Loosening of prosthesis at the incus. C, Removal of prosthesis.
FIGURE 24-8. Opening of vestibule.
oval window neomembrane in a nonoverlapping rosette pattern in the center of the oval window (Fig. 24-8). A minimum of 2 second intervals between bursts is necessary to minimize heat buildup of the neomembrane and perilymph. A 0.6 mm stapedotomy is created and is incomplete until the vestibule and clear perilymph are identified. The stapedotomy size is confirmed by use of a
0.4 mm and 0.7 mm McGee rasp. If the incus long proc ess is satisfactory, a nitinol prosthesis with a fluoroplastic piston of 0.5 mm diameter is used. A length of 4.25 mm is used in 90% to 95% of cases. If the lenticular process is unsatisfactory, the same type of prosthesis is used from the malleus to the fenestra. The most commonly used length for this prosthesis is 4.50 to 4.75 mm. After laser
290
OTOLOGIC SURGERY
crimping, the ossicular chain is gently palpated to ensure freedom of movement and appropriateness of prosthesis length. The oval window is sealed with areolar fascia. The tympanomeatal flap is returned to its anatomic position and secured with saline-moistened absorbable gelatin sponges or antibiotic ointment.
SUMMARY The use of lasers for revision stapedectomy represents a significant advantage over standard instrument techniques. Not only has the clinical safety of properly used lasers for revision stapedectomy been shown, but also this technology has yielded superior audiologic results. Nitinol heatactivated self-crimping prostheses and delivery of CO2 laser energy through a hollow optical fiber hold potential to improve outcomes in revision stapedectomy. Any laser is simply a tool, and as such has benefits and limitations. The appropriate use of lasers for revision stapedectomy requires a thorough understanding of the principles of laser energy and technology, proper training, and hands-on experience in the laboratory and in clinical settings.
Acknowledgment The author thanks Roger Vail, CRNA, BS, for consultation and advice concerning the intravenous sedation anesthetic technique. The author thanks Bruce Leone, MD, for consultation and advice relating to laryngeal mask anesthesia principles.
REFERENCES 1. Shea JJ: Fenestration of the oval window. Ann Otol R hinol Laryngol 67:932-951, 1958. 2. Glasscock M E III, McKennan K X, Levine SC : Revision stapedectomy surgery. Otolaryngol Head Neck Surg 96:141-148, 1987. 3. Sheehy J L , Nelson R A, House H P: Revision stapedectomy: A review of 258 cases. Laryngoscope 91:43-51, 1981. 4. Feldman BA, Schuknecht HF: Experiences with revision stapedectomy procedures. Laryngoscope 80:1281-1291, 1970. 5. Derlacki E L : Revision stapes surgery: Problems with some solutions. Laryngoscope 95:1047-1053, 1985. 6. Crabtree J: An evaluation of revision stapes surgery. L aryngoscope 90:224-227, 1980. 7. Lippy WH : Stapedectomy revision. Am J Otol 2:15-21, 1980. 8. Lippy WH, Schuring AG: Stapedectomy revision of the wire-Gelfoam prosthesis. Otolaryngol Head Neck Surg 91:9-13, 1983. 9. Lippy WH, Schuring AG: Stapedectomy revision following sensorineural hearing loss. Otolaryngol Head Neck Surg 92:580-582, 1984. 10. Pearman K, Dawes J D K : Post-stapedectomy conductive deafness and results of revision surgery. J Laryngol Otol 96:405-410, 1982.
11. Sataloff J: Experimental use of the laser in otosclerotic stapes. Arch Otolaryngol Head Neck Surg 85:58-60, 1967. 12. Stahle J, Hoegberg L , Engstrom B : The laser as a tool in inner ear surgery. Acta Otolaryngol (Stockh) 73:27-37, 1972. 13. Wilpizeski C R : Otological applications of laser. In Wolbarsht M K (ed): Laser Applications in Medicine and Biology. New York, Plenum, 1977, vol. 3, pp 289-328. 14. Escudero L H, Castro AO, Drummond M, et al: Argon laser in human tympanoplasty. Arch Otolaryngol Head Neck Surg 105:252-253, 1979. 15. Perkins RC : Laser stapedotomy for otosclerosis. L aryngoscope 90:228-241, 1980. 16. DiBartolomeo J R , Ellis M : The argon laser in otology. Laryngoscope 90:1786-1796, 1980. 17. McGee TM : Argon laser in chronic ear and otosclerosis. Laryngoscope 93:1177-1182, 1983. 18. McGee TM : Lasers in otology. Otolaryngol Clin North Am 22:233-238, 1989. 19. Gantz B J, Kischimoto S, Jenkins H A, et al: Argon laser stapedotomy. Ann Otol Rhinol Laryngol 91:25-26, 1982. 20. Vollrath M, Schreiner C : Influence of argon laser stapedotomy on cochlear potentials. Acta Otolaryngol (Stockh) 385(Suppl):1-31, 1982. 21. Bartels L : KTP laser stapedotomy: Is it safe? Otolaryngol Head Neck Surg 103:685-692, 1990. 22. Horn K, Gherini S, Griffin G: Argon laser stapedotomy using an endo-otoprobe system. Otolaryngol Head Neck Surg 102:193-198, 1990. 23. Lesinski S : Lasers for otosclerosis. Laryngoscope 99 (Suppl):1-24, 1989. 24. McGee TM, Kartush J M : Laser-stapes surgery. L aryngoscope 100:106-108, 1990. 25. Vernick D M : CO2 laser safety. Laryngoscope 100: 108-109, 1990. 26. Jacobs S A, Temelkuran B, Weisberg O, et al: Hollowcore fibers. In Mendez A, Morse TE (eds): Specialty Optical Fibers. Amsterdam, Academic Press, 2007. 27. McGee TM, Diaz-Ordaz E A, Kartush J M : The role of KTP laser in revision stapedectomy. Otolaryngol Head Neck Surg 109:839-843, 1993. 28. Hammerschlag PE, Fishman A, Scheer A A : A review of 308 cases of revision stapedectomy. Laryngoscope 108:1794-1800, 1998. 29. Han WW, Incesulu A, McKenna M J, et al: Revision stapedectomy: Intraoperative findings, results, and review of the literature. Laryngoscope 107:1185-1192, 1997. 30. Prasad S, Kamerer D B : Results of revision stapedectomy for conductive hearing loss. Otolaryngol Head Neck Surg 109:742-747, 1993. 31. Langman AW, Lindeman RC : Revision stapedectomy. Laryngoscope 103:954-958, 1993. 32. Cokesser Y, Naguib M, Aristegui M : Revision stapes surgery: A critical evaluation. Otolaryngol Head Neck Surg 111:473-477, 1994. 33. Farrior J, Sutherland A : Revision stapes surgery. Laryngoscope 101:1155-1161, 1991. 34. Gros A, Vatovec J, Zargi M, Jenko K : Success rate in revision stapes surgery for otosclerosis. Otol Neurotol 26:1143-1148, 2005.
Chapter 24 • Laser Revision Stapedectomy 35. Wiet R J, Kubek DC, Lemberg P, Byskosh AT: A metaanalysis review of revision stapes surgery with argon laser: Effectiveness and safety. Am J Otol 18:166-171, 1997. 36. Nissen R L : Argon laser in difficult stapedotomy cases. Laryngoscope 108:1669-1673, 1998. 37. Horn K L , Gherini SG, Franz DC : Argon laser revision stapedectomy. Am J Otol 15:383-388, 1994. 38. Haberkamp TJ, Harvey S A, Khafagy Y: Revision stapedectomy with and without CO2 laser: Analysis of results. Am J Otol 17:225-229, 1996. 39. Silverstein H, Bendet E, Rosenberg S, Nichols M : Revision stapes surgery with and without laser: A comparison. Laryngoscope 104:1431-1434, 1994. 40. Sarac S, Mckenna MJ, Mikulec AA, et al: Results after revision stapedectomy with malleus grip prosthesis. Ann Otol Rhinol Laryngol 115:317-322, 2006. 41. Hohmann A : Inner ear reaction to stapes surgery (animal experiments). In Schuknecht H F (ed): Otosclerosis. Boston, Little Brown, 1982, pp 305-317. 42. Linthicum F: Histologic evidence of the cause of failure in stapes surgery. Ann Otol Rhinol Laryngol 80:67-68, 1971. 43. Krieger LW, Lippy WH, Schuring AG, Rizer FM : Revision stapedectomy for incus erosion: Long-term hearing. Otolaryngol Head Neck Surg 119:370-373, 1998. 44. Lippy WH, Battista R A, Berenholz L , et al: Twenty year review of revision stapedectomy. Otol Neurotol 24:560566, 2003. 45. Lesinski SG: Causes of conductive hearing loss after stapedectomy or stapedotomy: A prospective study of 279 consecutive surgical revisions. Otol Neurotol 23:281-288, 2002.
291
46. Lippy WH, Wingate J, Burkey J M, et al: Stapedectomy revision in elderly patients. Laryngoscope 112:1100-1103, 2002. 47. Morganstein K M, Manace E D: Incus necrosis following stapedectomy. Laryngoscope 78:600-619, 1968. 48. Alberti PW: The blood supply of the incus long process and the head and neck of the malleus. J Laryngol Otol 79:966-970, 1965. 49. McGee TM : The loose wire syndrome. Laryngoscope 91:1478-1483, 1981. 50. Rajan G P, Atlas M D, Subramaniam K, Eikelboom R H : Eliminating the limitations of manual crimping in stapes surgery: A preliminary trial with the shape memory Nitinol stapes piston. Laryngoscope 115:366-369, 2005. 51. Knox GW, Reitan H : Shape-memory stapes prosthesis for otosclerosis surgery. Laryngoscope 115:1340-1346, 2005. 52. Sorom A J, Driscoll C L , Beatty CW, Lundy L B : Retrospective analysis of outcomes after stapedectomy with implantation of a self-crimping nitinol prosthesis. Otolaryngol Head Neck Surg 137:65-69, 2007. 53. Rajan G P, Diaz J, Blackham R , et al: Eliminating the limitations of manual crimping in stapes surgery: Midterm results of 90 patients in the Nitinol stapes piston multicenter trial. Laryngoscope 117:1236-1239, 2007. 54. Brown K D, Gantz B J: Hearing results after stapedotomy with a nitinol piston prosthesis. Arch Otolaryngol Head Neck Surg 133:758-762, 2007.
25
Special Problems of Otosclerosis Surgery William H. Lippy and Leonard P. Berenholz Videos corresponding to this chapter are available online at www.expertconsult.com.
As the incidence of otosclerosis declines, fewer surgeons acquire adequate experience in stapes surgery, and even fewer surgeons treat the problems of either complicated or unsuccessful stapedectomy. This chapter presents a comprehensive approach and diagnostic criteria for selection of these unusual patients. Intraoperative problems and solutions are defined and illustrated. The solutions are presented in a logical, safe, stepwise manner to avoid irreversible results. Before embarking on the nuances of primary and revision stapedectomy, first let us summarize a more recent finding with regard to pregnancy and its effect on otosclerosis. Traditional teaching has assumed that pregnancy would exacerbate otosclerosis. Our evaluation of incidence and progression of otosclerosis in women with children versus childless women has conclusively shown no impact of the pregnancy on further hearing loss in otosclerosis.1 Before addressing technical aspects, a few practical and philosophical points should be presented. When surgery is scheduled, a significant family member or friend should accompany the patient so that another person fully understands the goals and risks of the proposed surgery. During surgery, the surgeon should terminate the procedure if he or she encounters a problem that might jeopardize the patient’s hearing further. The patient and the surgeon can accept termination more easily than a bad result. The surgeon should not lose focus just to be compulsively neat during stapedectomy. For example, one should not search for the superstructure if it falls into the hypotympanum, should not remove pieces of footplate floating in the perilymph, and should not force on a prosthesis or a wire keeper that is excessively tight. Second-stage procedures can be done. This chapter reviews the technique of stapedectomy and the principles that prevent misadventures and discusses solutions to unusual problems. In addition, revision techniques for failed stapedectomy are described in detail. Finally, experience is presented in specific areas, such as far advanced otosclerosis with little or no testable hearing, and stapedectomy in children, in elderly patients, in patients with small air-bone gaps, and in fighter pilots.
Stapedectomy in the presence of chronic otitis media, the need for promontory drilling, and findings in the other ear in patients with otosclerosis also are summarized.2-4
INTRAOPERATIVE AUDIOMETRY In the past, surgeons showed the success of stapedectomy when the patient, under local anesthesia, heard sound ranging from a soft whisper to a loud voice. More sophisticated methods are now applied with great success. By using a portable audiometer in the operating room, the surgeon can precisely measure a patient’s hearing before and after surgery. Such improved assessment benefits the surgeon and the patient. Any portable audiometer can be used. One of the earphones is removed from the headset and inserted into a sterile plastic sleeve, which is available as a disposable orthopedic drill sleeve. The surgeon holds the sterile earphone to the patient’s ear (Fig. 25-1). Testing begins with the presentation of a tone that is easily heard by the patient. Threshold is obtained by progressively decreasing the loudness of the presented tone until the patient cannot hear it. The frequency with the greatest air-bone gap is usually used for single-frequency testing. Circulating nurses can easily learn to operate the audiometer. Hearing is tested at the beginning and the end of the operation to measure changes in hearing resulting from surgery. Despite the disturbed eardrum and blood in the middle ear and in the perilymph, the hearing usually is within 15 dB and often 5 dB from the final hearing result. The result is qualitative, not quantitative, so one is testing for a hearing gain. Testing by an audiometer in the operating room offers several advantages. First, the surgeon and the patient have instant and accurate feedback on the success of the operation. Second, the improvement of hearing defines the end point of surgery. Third, in revision cases, the surgeon can explore the footplate area without opening the oval window by repositioning the prosthesis in various locations in the oval window. Finally, in difficult cases, 293
294
OTOLOGIC SURGERY
FIGURE 25-1. Intraoperative audiometer setup.
different techniques can be attempted to determine the best prosthesis and best placement for optimal hearing.5
ROUTINE STAPEDECTOMY The basic technique of our routine stapedectomy, which has remained largely unchanged for 43 years, illustrates the principle of a safe approach. Use of this technique and the application of the principles described earlier have closed the air-bone gap in 96% of 17,000 cases. More importantly, overclosure of the air-bone gap occurs in 75% of the cases. Worse hearing ears developed in only 0.5%. Before surgery, the surgical nurse carefully explains the procedure to each patient. This knowledge helps an otherwise anxious patient to be calm and cooperative. The anesthesiologist or nurse anesthetist begins an intravenous infusion and monitors the patient during the stapedectomy. The operation begins with the injection in four quadrants of the ear canal with a mixture of 0.5 mL of epinephrine 1:1000 solution and 4.5 mL of 2% lidocaine. This solution results in maximal control of bleeding and minimal cardiovascular changes or symptoms. If the patient remains anxious, intravenous medication is administered in a dose that keeps the patient comfortable, but awake enough to permit intraoperative audiometry. Each step of the operation should be completed carefully and exactly. Precision in one step makes the next step easier and results in perfection. The speculum holder, which is always used, is positioned so that each portion of the tympanomeatal flap incision is visible as it is made. The flap should be elevated carefully to prevent damage to the skin and
tympanic membrane. As the middle ear is entered, an absorbable gelatin sponge (Gelfoam) pledget soaked in the previously mixed anesthetic solution is placed into the middle ear to anesthetize the middle ear mucosa. As the drum is pushed back, the manubrium of the malleus and incus can be seen and palpated. Any fixation should not preclude completion of the operation, but should be noted so that the patient can be advised later if the hearing result is suboptimal.6 A sharp, strong curette simplifies the task of curetting the ear canal. Enough of the scutum should be removed to see the origin of the stapedius tendon, the facial nerve, and the entire footplate area. A control hole is placed in the footplate at the junction of the anterior one third and the posterior two thirds of the footplate. The hole may facilitate later removal of the footplate. It also permits early detection of a rare perilymph gusher, when it can be more easily controlled. In addition, if the footplate comes out with the superstructure, the hole reduces the sudden change of pressure in the vestibule. If the footplate is too thick for a needle, the argon laser is used to make the control hole. After the incudostapedial joint is severed and the tendon is cut, the superstructure is fractured toward the promontory and removed to expose the footplate. The control hole can now be extended across the footplate, and the footplate posterior to the hole is removed. A stapedotomy or partial stapedectomy is done. We have found no significant difference in outcome when comparing stapedotomy, partial stapedectomy, or total stapedectomy except for the lower rate of overclosure in stapedotomies.7 A vein from the forearm, previously harvested, pressed, and prepared, is immediately placed across the oval window with the adventitial side down to seal and protect the vestibule. Until more recently, the Robinson stainless steel piston prosthesis was used in all cases. This prosthesis comes with either a standard or large well, 0.4 or 0.6 mm stem width, and in various lengths. We have found that a prosthesis with a large well, narrow stem, and length of 4 mm is suitable in 99% of cases, eliminating the need to measure. Instead of the stainless steel prosthesis, we now use the titanium bucket handle prosthesis. With potential future advances in magnetic resonance imaging (MRI) technology in mind, we made the change from stainless steel prosthesis to the titanium prosthesis to optimize MRI compatibility. Results obtained using the titanium prosthesis are equal to the results obtained using the stainless steel prosthesis. In addition to increased MRI compatibility, an advantage offered by use of the titanium prosthesis is that there is no reflection of light from the titanium prosthesis.8 The absence of a reflection enables the surgeon to visualize the placement of the prosthesis better. The prosthesis is placed by use of a two-handed technique. One hand lifts the incus with an incus hook, while the other gently directs the prosthesis with a strut guide.
Chapter 25 • Special Problems of Otosclerosis Surgery
A controlled study evaluating hearing results with various prosthesis widths revealed similar hearing results in 0.4 and 0.6 mm prostheses. The narrow 0.4 mm stem prosthesis is used because the 0.6 mm prosthesis occasionally can be too wide for a narrow oval window niche.9 Because this prosthesis centers itself in the oval window opening, middle ear packing is not used. The patient’s hearing can be tested immediately after the tympanic membrane is replaced. If the wire keeper does not easily swing over the lenticular process, its use is unnecessary. Forcing it may displace the prosthesis from the center of the oval window (Fig. 25-2).
Intraoperative problems Tympanomeatal Flap Tear To avoid tearing the flap as it is lifted, the speculum is repositioned frequently for better vision, especially when dissection is near the annulus, where most tears occur. A torn flap need not stop the operation; it can be repaired by approximation or with tissue, such as vein, fascia, or perichondrium—whatever tissue is used to seal the oval window. In our surgical approach, placement of the vein tissue underneath the tear gives the best result. Enough tissue should have been previously harvested to cover the oval window and to repair any tears or perforations.
Tympanic Membrane Perforation Tears that involve the tympanic membrane are repaired in the same manner as are tympanomeatal flap tears. When a perforation develops centrally as a result of manipulation, a piece of tissue is placed under the perforation and packed against the tympanic membrane with Gelfoam. The edges of the perforation are not freshened.
Atrophic Tympanic Membrane An atrophic tympanic membrane may signal a poor blood supply to the incus. An atrophic membrane has been observed on exploration in revision stapedectomy with erosion of the lenticular process being a common finding.10 As in treatment of a perforation, the intact tympanic membrane is reinforced from the underside of the tympanic membrane with tissue. This may be vein, fascia, or, in more severe cases, perichondrium or cartilage. This procedure should thicken the tympanic membrane and protect the incus by providing a better blood supply.
Ossicular Dislocation During stapedectomy, the incus may be inadvertently loosened in several situations. Loosening may occur when the scutum is curetted away, when a wire is placed on the incus, or when an instrument strikes the incus. If curettage is initiated lateral to the scutum and progresses
295
medially, dislocation is much less likely. The practical solution is to attach a piston stapes prosthesis to the lenticular process to help hold the incus in place. Two thirds of these cases are successful; only a few unsuccessful cases need a revision with a different technique.
Fixed Malleus The malleus must always be routinely palpated with the same instrument under the surgeon’s direct vision from the underside of the tympanic membrane. The malleus may be slightly fixed, moderately fixed, or totally fixed. If fixation is slight or moderate, the final result of stapedectomy would be as if the malleus had not been fixed at all. The success rate would be the same (96% to 97%), but the overclosure rate would be substantially reduced.11 Partial malleus fixation should be ignored. When the malleus (and probably the incus) is totally fixed, a stapedectomy should be completed, if the patient also has a fixed footplate. Most of the footplate should be removed to create a large enough oval window opening for a second future procedure (malleus or tympanic membrane to oval window technique). Of totally fixed malleus cases, 68% are successful to within 10 dB, and the air-bone gap is closed to within 10 to 20 dB in an additional 15%. Cases with an air-bone gap of 25 dB or more should be considered for a second-stage procedure. Applying this simple solution over the past 20 years, we have had good hearing results, and no patient with otosclerosis and a fixed malleus has had a further sensorineural hearing loss (Table 25-1). A more complex or possibly traumatic procedure can be postponed until an oval window tissue seal is present. Procedures to free the head of a fixed malleus are usually nonrewarding on a permanent basis.
Fused Incudostapedial Joint If the joint cannot be separated with a joint knife, the laser can be used instead.
Partial Absence of the Incus When a partial absence of the long process of the incus is found in a patient with otosclerosis, a stapedectomy is still done. Incus erosion is the second most common finding in revision stapes surgery; a crimped wire prosthesis causes erosion twice as often as the Robinson prosthesis. In place of the standard prosthesis, the Lippy modified Robinson prosthesis is used. The fenestra should be larger than usual because the prosthesis does not selfcenter.11,12 The technique of prosthesis placement is important to success (see Video 25-1). The lower stem end of the prosthesis is placed on the vein graft, the upper end with the open well toward the eroded incus. The prosthesis is guided onto the remaining incus from the direction of the promontory. A significant foreshortening of the eroded incus or overhang
296
OTOLOGIC SURGERY
FIGURE 25-2. Routine stapedectomy. FIGURE 25-3. Lippy modified Robinson prosthesis. FIGURE 25-4. Modified 4.5 mm Robinson prosthesis in place on vein graft covering oval window. FIGURE 25-5. Robinson prosthesis in place on mobilized footplate. FIGURE 25-6. Revision stapedectomy wire prosthesis is pushed aside.
Chapter 25 • Special Problems of Otosclerosis Surgery
297
TABLE 25-1 Malleus-Incus Fixation and Otosclerosis Hearing Results (N = 102) AIR-BONE GAP (%) Degree of Fixation
Overclosed
Slightly (n = 40) Moderately (n = 28) Totally (n = 34)
70 29 24
Within 10 dB
Worsened Conduction
Sensorineural Loss
96 97 68
0 0 9
0 0 0
of the facial nerve necessitates use of an offset Lippy modified prosthesis, yielding more length to avoid the facial nerve.10 In long-term follow-up using the Lippy modified prosthesis in nonrevision cases, initial success (<10 dB air-bone gap) was 90% with long-term hearing (<10 dB air-bone gap) maintained in 86% of patients.10 If the lenticular process comes to a pointed rather than a blunted end, the laser is used to square the end. The laser can also be used to thin or sculpture an incus that is too thick to accept the Lippy modified prosthesis (Figs. 25-3 and 25-4).
Dehiscent Facial Nerve In otosclerosis surgery, the facial nerve rarely interferes with a stapedectomy except in a congenitally deformed middle ear, or when the facial nerve canal is completely dehiscent. In cases in which more than 50% of the footplate was covered by the facial nerve, the overall success of stapedectomy was similar to that of cases in which the footplate was not covered.13 If any part of the footplate is visible, a stapedectomy usually can be accomplished. A hole should be made first in the visible part of the footplate. Often, a large portion of the footplate can be removed from underneath the dehiscent nerve by retraction of the facial nerve with the shaft of the same instrument used to extract the footplate. If the footplate cannot be removed, the technique is to shatter the footplate with a pointed pick—even blindly, if necessary. After a vein graft is placed across the open oval window, the prosthesis is inserted by compressing the facial nerve with the prosthesis. In our experience, this technique of compressing the facial nerve has never caused permanent facial nerve paralysis. The success rate is slightly lower, however, probably because the nerve pushes the prosthesis out of optimal position.
Obliterative Otosclerosis An obliterative footplate is saucerized with a 0.5 mm diameter carbide burr. As large an area as possible is drilled to saucerize the footplate. A small opening of the footplate should be avoided until a wide area is saucerized because enlargement of a small footplate opening surrounded by thick, hard footplate may be impossible.
The aim is to develop a blue eggshell appearance over as large an area of the footplate as possible without penetrating the footplate. Occasionally, only the membrane under the footplate remains after drilling. In the 1960s, 30% of the footplates were drilled; in the 1970s, 9%; in the 1980s, 4%; and in the 1990s, 6%. Fifty-five percent of the cases overclosed, 80% were successful, and 0.2% were worse. Although the results of drill-out cases are acceptable, they are not as good as routine cases. Preoperatively, the surgeon should be more suspicious of a possible drill-out in an obliterated footplate if the patient presents with a more than 30 dB air-bone gap or has had otosclerosis for many years. Suspicion should also be high in patients whose hearing loss begins early in life.14
Narrow Oval Window Niche and Promontory Drilling It is helpful to drill off a small portion of the promontory adjacent to the footplate when the oval window niche is too narrow. When performing promontory drilling, the bit rests on the footplate, and several gentle outward strokes from the footplate are made. The footplate is opened with a needle and removed with picks. If mobilization of the footplate occurs, the drilling is terminated and a prosthesis is placed (see section on floating footplate). In a moderately thick footplate, a laser may be used. We evaluated a group of patients with otosclerosis who underwent promontory drilling to carry out stapedectomy.3 There was no deterioration in bone conduction hearing or in speech discrimination ability in this group. These patients did as well as a control group of patients with stapedectomies in whom the promontory was not drilled.
Floating Footplate The footplate can move when the surgeon is drilling a hole in the fixed footplate, fracturing the stapes superstructure, or extracting the fixed footplate. The most common cause of a floating footplate from the 1960s to the early 1980s was drilling on a solid footplate. From the early 1980s through the 1990s, the most common cause was fracturing the superstructure.15
298
OTOLOGIC SURGERY
When the footplate is mobilized, regardless of whether it is solid (white) or diffuse (blue), a vein graft is placed on top of the mobilized footplate, followed by placement of a Robinson prosthesis attached to the lenticular process (Fig. 25-5).16 This conservative method gives excellent long-term results. If the mobilized footplate is mostly blue with diffuse otosclerosis, the hearing success rate is 97%. Long-term hearing results at 3 years remained the same for the mobilized blue footplate with no refixation. If the footplate is thick, white, or biscuit shaped, the hearing success rate is 52%. If the thick, white footplate later refixes, it can be revised during a revised stapedectomy. Refixation occurred in 30% of mobilized thick, white, obliterative footplates. If we add the results of the unsuccessful cases that required revisions to the initially successful cases, the final success rate is 76%. If the footplate again moves during the revision procedure, no future surgery should be planned. Following this protocol, none of 147 cases of a floating footplate in a series of 8000 cases developed a further sensorineural loss (Table 25-2). In recent years, the laser has been used in the moderately thick footplate, preferably before it mobilizes.
REVISION STAPEDECTOMY Our experience is based on more than 1800 cases.19-21 Cases are divided into two major categories: sensorineural hearing loss and conductive hearing loss. These cases are divided further by the type of prosthesis and oval window covering used in the primary surgery. We do not revise our cases (Robinson-vein) during the first 6 weeks after surgery. Previously, when we attempted early revision, tissue reaction was found throughout the middle ear, hindering our effort to analyze the problem. No patient gained improved hearing, and many lost further hearing. Patients with a negative attitude might not be candidates for revision. Because the risks are slightly higher in revision than in primary surgery, the patient must be prepared to accept a potentially poor result.
Elderly patients may undergo revision stapedectomy. We compared a group of patients older than 65 years with a group younger than 65, and the success rate was very similar. There was no increase in side effects secondary to surgery in the patients older than 65.22
Sensorineural Hearing Loss In patients with delayed sensorineural hearing loss after stapedectomy with a piston and an oval window tissue seal, revision is indicated only for a history of trauma or dizziness. Before this directive, in 27 cases, 79% had negative surgical findings with no evidence of a surgical problem or oval window fistula. Only one case had a fistula.17 We now revise only an occasional case with unexplained persistent dizziness or rare cases in which a fistula is highly suspected. Of patients who are dizzy before revision, 20% gain at least some relief after revision. Parenthetically, we have never been able to improve a sensorineural hearing loss. In cases with a delayed or immediate sensorineural hearing loss and without an oval window tissue seal, the findings at revision surgery are more dramatic. Most cases had a wire prosthesis with either Gelfoam or a blood clot as an oval window seal; 50% of the cases had oval window fistulas. These cases were revised with an oval window tissue seal and a Robinson prosthesis. Hearing improved in a few cases, and dizziness improved in 75%. The next most common findings were cases with negative surgical findings and prostheses that were too long when placed. Table 25-3 illustrates findings in the cases with sensorineural hearing loss by comparing cases with and without a tissue graft. The surgeon should seriously consider a revision stapedectomy in patients with a sensorineural hearing loss without an oval window tissue seal. If a tissue graft was used to seal the oval window, the surgeon should be reluctant to do a revision.
Conductive Hearing Loss The deciding factor for revision of conductive hearing loss is the hearing history after the primary procedure.17,18 The most appropriate surgical candidates have hearing
TABLE 25-2 Prosthesis on Mobilized Footplate TABLE 25-3 Surgical Findings in Revision
FOOTPLATE Hearing Successful Conduction worse Sensorineural loss worse Successful after revision
Thick White (n = 56) (%)
Thin Blue-Mixed (n = 92) (%)
52 7 0
95 2 0
76
—
of Sensorineural Cases
Surgical Findings
Tissue Seal (n = 29) (%)
No Tissue (n = 42) (%)
Negative findings Fistula Long prosthesis Tissue reaction Lateral vein
79 4 0 10 7
24 50 21 5 0
Chapter 25 • Special Problems of Otosclerosis Surgery
299
improvement postoperatively and then development of another conductive hearing loss. Patients with the same or worse hearing postoperatively have a low success rate for revision surgery (Table 25-4). The operative experience of the previous surgeon is another factor. The greater the surgical experience, the less likely a reversible problem would be solved. Patients whose first surgeon was an experienced stapes surgeon and whose hearing did not improve are usually poor candidates for revision.19,20 The type of prosthesis used and whether an oval window tissue graft was used are also important. Conductive hearing loss with a Robinson prosthesis and vein is most frequently due to incus erosion, prosthesis malfunction, or negative findings.14,21 In cases of eroded incus, many tympanic membranes were atrophic. In these cases, the tympanic membrane is reinforced with fascia or vein. To decrease prosthesis malfunction, a large well with a narrow shaft is used in all stapedectomies, allowing the prosthesis more freedom of movement to center in the oval window.10 Incus erosion is common with a wire prosthesis and can be revised with good hearing results. Partial stapedectomy, in which the crus is mobilized into an opened covered oval window, is a procedure that is usually revised with a good result. Another successful revision occurs with or without conductive loss with distorted hearing or vibrations because of a prosthesis that is too short. The symptoms may be eliminated by the addition of a second vein graft and the placement of a new prosthesis 4 mm in length whether the original prosthesis was a piston or a wire.
vein graft may be placed on the oval window followed by a Robinson prosthesis. At this point, the tympanomeatal flap is replaced, and the hearing is tested. If the hearing improves, the surgical procedure is finished. If the hearing is not improved, the surgeon should consider other causes, such as previously inadequate footplate removal, otosclerosis regrowth, or a misdirected stapes prosthesis. The bottom of the prosthesis first should be moved slightly to search for an opening into the oval window. In the routine stapedectomy technique, the prosthesis is always self-centering. In a revision, because the tissue graft is healed and less compliant, adjustment may be necessary. If repositioning of the prosthesis does not improve the hearing measured by the audiometer, further exposure of the oval window covering with the laser may be necessary.
Surgical Technique
Prosthesis Malfunction
Intraoperative testing is an important tool in revision surgery, allowing surgeons to obtain the most information with the least amount of trauma to the labyrinth. The patient must be under local anesthesia and should be instructed to inform the surgeon of any dizziness. The surgical technique includes removing the previously placed prosthesis without disturbing the oval window seal, and placing a vein graft and a Robinson prosthesis. In cases with a wire prosthesis in which the patient experiences dizziness on manipulation, the end of the wire that is attached to the incus is detached and pushed aside, leaving the distal end of the wire in the oval window seal (Fig. 25-6). A vein graft is placed over the oval window area with a slit cut in the graft to accommodate the wire. A Robinson prosthesis is placed (see Video 25-1). The argon laser otoprobe is often used in revision stapedectomy. It is particularly useful in slowly removing tissue around the distal end of a wire prosthesis in the oval window. The prosthesis may be removed less traumatically using the laser at low wattage (1 W), and using brief bursts to avoid excessive vestibular stimulation. When perilymph is identified, and a large bluish area is seen, a
Prosthesis malfunction in cases with a Robinson prosthesis on an oval window tissue seal is uncommon because of the self-centering ability of the prosthesis. A lenticular process that is too large can misdirect the piston from self-centering. This unusual problem can be avoided in the original procedure by using a 4 mm polytef (Teflon) prosthesis with a large well. This well is 0.2 mm larger than that of the Robinson prosthesis with a large well. This prosthesis accommodates the occasional extra-large lenticular process without causing misdirection. Malfunction is likely to occur in cases in which the stapes prosthesis becomes fused to the incus lenticular process and directed out of its self-centering position. If the fused prosthesis cannot be easily removed from the lenticular process, it may be removed with the laser. A prosthesis that is too short and used with a tissue graft may give a good result at first. A conductive loss develops, however, as the tissue graft thins out. These problems can be corrected by revision. Wire prostheses, which lack rigidity, are more likely to migrate. The distal looped end of the wire commonly rests on the promontory, or is fixed to a margin of the oval window. In addition, a loose attachment of the wire
Surgical Findings Incus Erosion Incus erosion is the most common finding with wire prostheses. The pressure of the wire around the long process of the incus causes necrosis and erosion. If the only problem is a loose wire, the prosthesis can be crimped again. With prostheses that attach to the lenticular proc ess of the incus, such as the Robinson, incus erosion most often results from a previous infection, manifested by a healed perforation or an atrophic tympanic membrane and incus. For wires and pistons, the success rate of revision with the Lippy modified Robinson prosthesis is 80%.18
300
OTOLOGIC SURGERY
TABLE 25-4 Audiologic Patterns and
TABLE 25-5 Hearing Results in Cases with Negative
Hearing Results
Air-Bone Gap Delayed conduction No change in hearing Increased conduction
Findings
Successful Hearing (%) 70 35 25
rosthesis on the long process of the incus may occur p with incomplete crimping or gradual erosion of the incus. Lastly, short wire prostheses are frequently found, resulting from either improper measurement or inadvertent shortening when the prosthesis is crimped to the incus. Revision stapedectomy corrects these problems.
Negative Findings The so-called negative surgical finding category refers to the situation that occurs when no problem can be recognized. In cases with an oval window tissue seal and a piston prosthesis, a revision stapedectomy did not improve hearing. Hearing improved in 60% of cases, however, in which a wire prosthesis without a tissue seal was replaced by a Robinson prosthesis on a vein (Table 25-5). The reason for this disparity seems to be the efficiency of the Robinson prostheses. The heavier and more rigid piston prosthesis more closely resembles the stapes mass than does the wire prosthesis and is more efficient. With the use of laser, there are fewer cases that we call “negative findings” patients.
Malleus Fixation Malleus and incus fixations are discussed in the section on intraoperative problems. Malleus fixation may have been present but ignored at the initial surgery, or it may have progressed in the interim. Totally fixed cases can be revised with a footplate-to-drum prosthesis or a malleusto-footplate prosthesis when an oval window tissue seal is present. Total ossiculoplasty with footplate removal may be performed at the same setting if the middle ear is healthy. The tissue seal preferably is vein or areolar fascia.23
Other Findings Adhesions that are dense enough to impede the prosthesis are extremely rare. Adhesions are often found during revision, but the hearing rarely changes when they are removed. Intraoperative audiometry allows the surgeon to evaluate the effect of such adhesions when they are removed. Tissue graft lateralization, which is uncommon, can be revised by placing a new tissue graft on top of the old one, followed by a rigid piston prosthesis. Aggressive,
HEARING RESULTS (%) Prosthesis
Successful Hearing
Worse Hearing
60 0
0 9
Wire–no tissue Robinson-tissue
otosclerosis regrowth should not be removed because the chances of further regrowth or further sensorineural loss are high. In cases where the previous drilling was well described and limited, a new area may be drilled. Regrowth of otosclerosis is rare, however.
Fistula We have found only six fistulas in our cases with a Robinson prosthesis and a tissue seal. All six were women with smaller than usual vein grafts in whom fistulas occurred years after the original procedure. Each patient had a sudden hearing loss after descent either in a car or in an airplane. The fistulas were found at the margin where the facial nerve canal adjoins the oval window. We now take a larger vein in women and attempt to remove some of the mucosa from the facial nerve canal or the facial nerve to create a raw surface to which the tissue seal would adhere and not slide toward the promontory. Fistulas were common in conductive cases without tissue grafts.
Small Fenestra In revision stapedectomy of cases that have had a small fenestra technique, the most common surgical finding is displacement of the prosthesis from the small fenestra. The revision technique is to ignore the previous small fenestra to avoid manipulating adhesions that may have formed in the vestibule. Often, a larger fenestra can be placed in the remaining portion of the footplate with picks or the laser. Both fenestrae are covered by a tissue graft, and a Robinson prosthesis is placed in the new larger opening.
Recommendations In cases of revision stapedectomy, the following measures are recommended: 1. Local anesthesia should be used to monitor any dizziness, and to permit intraoperative audiometry. 2. Intraoperative audiometry should be performed. A hearing gain is evidence of a successful technical solution. The surgeon need not explore the oval window area further. If hearing is not improved, the prosthesis should be moved. If the hearing is still not improved, more of the oval window covering is exposed with the
Chapter 25 • Special Problems of Otosclerosis Surgery
laser, to search for an intact otosclerotic footplate or otosclerosis regrowth. 3. A tissue seal should be used over the oval window, whether or not it was opened, for three reasons. First, tissue provides a seal of the oval window and prevents fistulas. Second, fistulas of the oval window are not always evident because they can be minute or temporarily closed. Third, the seal produced by absorbable gelatin sponge or mucosa would not safely support a rigid piston prosthesis. 4. The oval window should not be routinely reopened. When the oval window is reopened, the incidence of hearing loss increases for several reasons. First, routine stapedectomy can cause vestibular adhesions from indentation of the oval window seal by the prosthesis. Reopening the oval window can cause vestibular trauma because manipulation of the adhesions damages the membranous labyrinth. Second, the most critical part of stapedectomy is removing the footplate. Reopening the oval window in the absence of the footplate as a landmark is more technically difficult, and carries the risk of greater surgical trauma. Third, delayed sensorineural hearing loss, which occurs rarely and inexplicably after stapedectomy, is probably related to a labyrinthine tissue reaction. The less trauma to the labyrinth, the less chance of a delayed sensorineural hearing loss. The oval window seal should not be opened routinely. 5. Aggressive otosclerosis regrowth should not be removed. When it is encountered, the surgical procedure should be terminated. The reasons for not reopening the oval window seal are supported further by the difficult task of drilling and removing the otosclerosis regrowth. Even when the otosclerosis regrowth is successfully removed, the hearing gain is temporary in most patients. The incidence of a greater hearing loss is more than 50% in our experience. 6. The distal loop of the wire prosthesis should be left in place in many cases. If the distal loop appears deep within the oval window seal, or if the patient experiences dizziness, the loop in the oval window seal is not removed. Removal may reopen the oval window, which increases the likelihood of a hearing loss or dizziness. If the patient’s main problem is significant incapacitating dizziness, however, the wire must be removed even if hearing is possibly sacrificed. 7. If the problem cannot be identified (negative findings), the case with a tissue seal should not be revised unless a laser is used. Cases without a tissue seal should be revised with good results.
FAR ADVANCED OTOSCLEROSIS Far advanced otosclerosis is defined as no measurable air or bone conduction, or, at best, air conduction no better than 95 dB and bone conduction at 55 to 60 dB
301
at one frequency only. The history may include a family member with otosclerosis, previous audiograms showing a conductive hearing loss, and progressive hearing loss. The patient may be wearing a hearing aid successfully or may have previously worn an aid. Findings would also include better than expected speech patterns and a softer voice than expected with such a severe sensorineural hearing loss. Most important is the ability to hear, not just feel, the 512 Hz tuning fork on the upper teeth. This practical test gives 10 dB more gain than when the tuning fork is placed on the mastoid. Edentulous patients are tested on their dentures or on their gums if they have no dentures.24 In some patients, this test is the only one to yield measurable hearing evidence of far advanced otosclerosis. Patients with some of these findings should be considered for a stapedectomy.
Surgical Technique In far advanced otosclerosis, the surgical techniques are the same as in routine stapedectomy, but the surgical findings are different. Fifty percent of the patients have obliterative otosclerosis; the surgeon must be prepared to drill the footplate extensively, as recommended previously.25
Results Success is measured by improved air conduction, improved speech discrimination, and more benefit from a hearing aid. In our surgical experience with 72 patients with far advanced otosclerosis, the average hearing gain was 20 dB, and 70% benefited more from a hearing aid.26 Discrimination was improved by more than 15% in 54% of cases. In addition, there is a high correlation of success between ears: a patient who gained hearing after surgery in one ear also did well in the other; a patient who did not achieve a good result in the first ear did not do well in the second ear. The results of stapedectomy for far advanced otosclerosis are often dramatic. They reinforce the surgeon’s resolve to double-check patients with no measurable hearing.
STAPEDECTOMY FOR SMALL AIR-BONE GAPS Although most otologic surgeons advocate at least a 20 dB pure tone average air-bone gap as an indication for stapedectomy, a review by the senior author (W.H.L.)27 reveals that even smaller air-bone gaps may be corrected. Among 136 cases, overclosure was achieved in more than 80% of the patients with mean overclosure of 8.1 dB.27 These patients had a 10 dB or slightly less preoperative air-bone gap with an average 16.7 dB improvement
302
OTOLOGIC SURGERY
ostoperatively. Five-year follow-up revealed that almost p all patients maintained their initial gains.
STAPEDECTOMY IN CHILDREN Approach and management of children with otosclerosis differ from those of adults with otosclerosis. In our review of 47 children 7 to 17 years old, footplate pathology was greater in children necessitating drill-outs in 27% of cases for obliterative otosclerosis.28 In cases where inner ear malformations were suspected, they were ruled out with computed tomography (CT), and no such cases were identified in this study. Some older children can be operated with a local anesthetic. There is a greater than 90% chance of closing the air-bone gap to within 10 dB. In 5-year follow-up, the mean pure tone average deteriorated an average of 8.1 dB. Overclosure is not as great as in adult patients.
STAPEDECTOMY IN ELDERLY PATIENTS Although otosclerosis usually manifests in patients younger than 40 years, some patients delay surgery until they are older than 70 years. In a review of 154 patients ranging in age from 70 to 92 years, 91% closed the airbone gap to within 10 dB.29 Results were stable at 5 years with 2.5 dB deterioration at 5 years. There was no increase in complications in the elderly compared with a comparison younger group, with transient dizziness occurring in less than 2% of the patients postoperatively.
STAPEDECTOMY IN PILOTS High-performance pilots with otosclerosis are a special group warranting attention. Six fighter pilots underwent stapedectomy, with three having bilateral stapedectomy.30 All the pilots returned to their active flight duties with no vestibular symptoms. Full flight status may be resumed after an altitude pressure test and a waiting period of 3 months.
STAPEDECTOMY IN PATIENTS WITH CHRONIC OTITIS MEDIA The need for stapedectomy in patients with chronic otitis media is rare.2 Before performing stapedectomy, the middle ear should be free of fluid, and the tympanic membrane should be intact. When these two conditions have been met, stapedectomy may be performed. The stapes fixation in these cases is due to tympanosclerosis secondary to chronic otitis media. In our review,2 these patients had significant improvement in their hearing, although the air-bone gap closure was less than that
achieved in patients with otosclerosis. This difference is probably due to some residual stiffness and thickening of the tympanic membrane.
FINDINGS IN BILATERAL STAPEDECTOMIES Because of the high percentage of patients developing bilateral otosclerosis, information on possible abnormal findings in the contralateral ear would be valuable. We reviewed 1800 patients and found that bilateral middle ear abnormalities occurred in 25% of the patients. The most common abnormality noted was obliterative otosclerosis (40% of the patients) requiring drill-out of the footplate. The second most common finding was overhanging facial nerve.4
REFERENCES 1. Lippy WH, Berenholz L P, Schuring AG, Burkey J M : Does pregnancy affect otosclerosis? Laryngoscope 115:1833-1836, 2005. 2. Berenholz L P, Lippy WH, Burkey J M, et al: Stapedectomy following tympanoplasty. Laryngol Otol 115:444446, 2001. 3. Lippy WH, Berenholz L P, Schuring AG, et al: Promontory drilling in stapedectomy. Otol Neurotol 23:439-441, 2002. 4. Daniels R L , Krieger LW, Lippy WH : The other ear: Findings and results in 1,800 bilateral stapedectomies. Otol Neurotol 22:603-607, 2001. 5. Lippy WH, Schuring AG, Rizer FM : Intraoperative audiometry. Laryngoscope 105:214-216, 1995. 6. Lippy WH, Schuring AG, Ziv M : Stapedectomy for otosclerosis with malleus fixation. Otolaryngol Head Neck Surg 104:338-389, 1978. 7. Rizer FM, Lippy WH : Evolution of techniques from the total stapedectomy to the small fenestra stapedectomy. Otolaryngol Clin North Am 26:443-451, 1993. 8. Lippy WH, Burkey J M, Schuring AG, Berenholz L P: Comparison of titanium and Robinson stainless steel stapes piston prostheses. Otol Neurotol 26:874-877, 2005. 9. Fucci M J, Lippy WH, Schuring AG, Rizer FM : Prosthesis size in stapedectomy. Otolaryngol Head Neck Surg 118:1-5, 1998. 10. Krieger LW, Lippy WH, Schuring AG, Rizer FM : Revision stapedectomy for incus erosion: Long-term hearing. Otolaryngol Head Neck Surg 119:370-373, 1998. 11. Lippy WH, Schuring AG: Solving ossicular problems in stapedectomy. Laryngoscope 93:1147-1150, 1983. 12. Lippy WH, Schuring AG: Prosthesis for the problem incus in stapedectomy. Otolaryngol Head Neck Surg 100:237-239, 1974. 13. Neff B A, Lippy WH, Schuring AG, Rizer FM : Stapedectomy in patients with a prolapsed facial nerve. Otolaryngol Head Neck Surg 130:597-603, 2004. 14. Lippy WH, Berenholz L P, Burkey J M : Otosclerosis in the 1960s, 1970s, 1980s, and 1990s. Laryngoscope 109:1307-1309, 1999.
Chapter 25 • Special Problems of Otosclerosis Surgery 15. Lippy WH, Fucci M J, Schuring AG, Rizer FM : Prosthesis on a mobilized stapes footplate. Am J Otol 17:713-716, 1996. 16. Lippy WH, Schuring AG: Treatment of the inadvertently mobilized footplate. Otolaryngol Head Neck Surg 98:8081, 1973. 17. Lippy WH, Schuring AG: Stapedectomy revision following sensorineural hearing loss. Otolaryngol Head Neck Surg 92:580-582, 1984. 18. Lippy WH, Schuring AG: Stapedectomy revision of the wire-Gelfoam prosthesis. Otolaryngol Head Neck Surg 91:9-13, 1983. 19. Lippy WH, Schuring AG, Ziv M : Stapedectomy revision. Am J Otol 2:15-21, 1980. 20. Lippy WH: Revision stapedectomy. In Jonas T. Johnson (ed): AAO-HNS Instructional Courses. St Louis, Mosby– Year Book, 1994. 21. Lippy WH, Battista R A, Berenholz L P, et al: Twentyyear review of revision stapedectomy. Otol Neurotol 24:560-566, 2003. 22. Lippy WH, Wingate J, Burkey J M, et al: Stapedectomy revision in elderly patients. Laryngoscope 112:1100-1103, 2002.
303
23. Berenholz L , Lippy WH : Total ossiculoplasty with footplate removal. Otolaryngol Head Neck Surg 130:120-124, 2004. 24. Lippy WH, Rotolo A L , Berger KW: Bone conduction measurement: Mastoid versus upper central incisor. Trans Am Acad Ophthalmol Otolaryngol 70:1084-1088, 1966. 25. Lippy WH, Battista R A, Schuring AG, Rizer FM : Faradvanced otosclerosis. Am J Otol 15(5 Pt 2):225-228, 1994. 26. Lippy WH, Burkey J M, Schuring AG, Rizer FM : Word recognition score changes following stapedectomy for far-advanced otosclerosis. Am J Otol 19:56-58, 1998. 27. Lippy WH, Burkey J M, Schuring AG, Rizer FM : Stapedectomy in patients with small air-bone gaps. Laryngoscope 107:919-922, 1997. 28. Lippy WH, Burkey J M, Schuring AG, Rizer FM : Stapedectomy in children: Short- and long-term results. Laryngoscope 108:569-572, 1998. 29. Lippy WH, Burkey J M, Fucci M J, et al: Stapedectomy in the elderly. Am J Otol 17:831-834, 1996. 30. Katzav J, Lippy WH, Shaniss A, Davidson B Z : Stapedectomy in combat pilots. Am J Otol 17:847-849, 1996.
26
Avoidance and Management of Complications of Otosclerosis Surgery Joseph B. Roberson, Jr. Videos corresponding to this chapter are available online at www.expertconsult.com.
Otosclerosis surgery is one of the most exciting and rewarding procedures an otologic surgeon performs. The physical demands of the operation are among the most refined of the surgical disciplines, and includes an extremely small tolerance for error even when the case goes along well with perfect equipment and perfect performance of support personnel. Small deviations from the surgical techniques illustrated in the remainder of this text regularly prove necessary to avoid complication. The knowledge base and technical skills needed to handle these deviations add difficulty to an already challenging undertaking seen with routine cases. With this difficulty comes a large reward, however, as patients experience improvement of their hearing with a successful outcome—frequently to the normal range. As with other difficult procedures, the learning curve for otosclerosis surgery is not steep. Complete preparedness for each potential surgical obstacle or complication is necessary to achieve results approaching those of experienced surgeons in centers of excellence located throughout the world. The atmosphere in the operating room is usually one of excitement and potentially slight tension with this procedure, especially for surgeons just beginning their careers. Thorough knowledge of and ability to deal with potential complications and surgical deviations help reduce anxiety and allow the operating surgeon to focus on and complete the job at hand. Reliable correction of otosclerotic conductive hearing impairment requires discipline, precision, knowledge, preparation of the operating facility, and judgment. Cognitive preparation should be complete before undertaking stapes surgery as the primary surgeon. Expert training and steady experience provide mastery of technical skill and development of operative judgment. Only then can the fullest potential as a stapes surgeon be reached. This chapter provides knowledge based on my own experience and that of other contributors to the field who have courteously shared their experience as colleagues, professors, mentors, and authors. The focus is on the preoperative, operative, postoperative, and reoperative situations a surgeon faces when he or she seeks to prevent or rectify complications.
PREOPERATIVE EVALUATION Medical Conditions A written or mental checklist is useful to avoid overlooking an important medical feature during preoperative evaluation. It is helpful (and recommended) to have any and all family members in the office present in the examination room during the interview. A useful practice is to write the names of the individuals present on the chart during the interview. In addition to a complete history and physical examination with attention to medical conditions germane to any surgical procedure, the prudent surgeon attempts to identify the following conditions. Fluctuating hearing loss, episodic vertigo, and lowfrequency sensorineural hearing loss (SNHL) may indicate endolymphatic hydrops. Care must be taken to avoid confusing the early conductive hearing loss of prior audiograms (which may falsely appear as SNHL) with endolymphatic hydrops. Patients with endolymphatic hydrops who undergo stapes surgery have a higher rate of SNHL (presumably from dilation of the saccule that contacts the stapes footplate where it is at risk during stapedectomy or stapedotomy) and chronic dizziness. This condition may be a contraindication to surgery.1 Surgeons should insist that acoustic reflex testing be performed and reviewed before entering the operating room because superior semicircular canal dehiscence manifests with conductive hearing impairment, which may be confused with impairment caused by otosclerosis on pure tone audiometry. This practice helps avoid the situation where a mobile and normal stapes is found intraoperatively after entering the middle ear. Clinical history also is very helpful if the patient has symptoms of vertigo with loud sounds. Diagnosis is established or refuted with a computed tomography (CT) scan in the plane of the superior semicircular canal. Patient symptoms such as vertigo with impulse noise may also alert the operating surgeon to the presence of this otosclerosis look-alike. A history of multiple fractures or blue sclera may allow the diagnosis of osteogenesis imperfecta to be made preoperatively.2 305
306
OTOLOGIC SURGERY
Lifelong hearing loss in one ear should alert the surgeon to the possibility of congenital footplate fixation. Congenital footplate fixation carries a higher than usual risk of gusher and SNHL.3 A CT scan should be performed preoperatively in patients with suspected congenital footplate fixation to look for abnormal cerebrospinal fluid (CSF)– perilymph connections that predispose to gusher. If a high risk of a gusher is found, amplification is recommended. A genetic pedigree focused on hearing loss is helpful in identifying patients with X-linked progressive mixed deafness. Patients with this disorder are rare and unique in that a conductive hearing loss is seen on the audiogram with intact stapedial reflexes.4,5 During surgery, a stapes gusher is encountered with the attendant risk of SNHL. Although males are typically affected, heterozygous females may exhibit milder audiologic abnormalities.6 Imbalance may occur after otosclerosis surgery despite a well-done surgical procedure and an excellent hearing result. Professional athletes, high-rise steelworkers, and painters might be best advised to avoid surgery until the completion of their careers. In similar fashion, patients who depend on their sense of taste for employment (e.g., workers in the wine industry, coffee tasters, and professional chefs) may choose to avoid surgery. Changes in chorda tympani function considered minor to most patients may be a source of disability in such professions. Commercial airline pilots usually are allowed to have stapes surgery without affecting their employability, but the operating surgeon or patient should secure written documentation of the patient’s employer’s policy before undertaking surgery. Military personnel who fly aircraft should consult with a flight surgeon. A history of or exhibited characteristics of severe anxiety, neuropsychiatric disease, claustrophobia, restless legs syndrome, and other conditions that would make the procedure difficult under sedation with an awake patient should be sought. When such patients have surgery, it should be performed under general anesthesia.
Physical Examination Various findings have an impact on upcoming surgery, including the following:
• Otitis externa must be completely treated for at least 1 month before undertaking surgery.
• Serous otitis media contains bacteria in approximately
50% of patients. Entrance into the vestibule in this situation puts the patient at risk for bacterial labyrinthitis, SNHL, and meningitis.7-9 • Tympanic membrane perforation or chronic otitis media must be repaired with a separate procedure before stapes surgery to reduce the bacteria present within the operative field as much as possible. Six to 12 months should pass after the first-stage repair for eustachian tube function to manifest, and for the hyperemia attendant with the primary procedure to fade.
• Exostosis, common in cold water swimmers, should be
corrected and allowed to heal before attempting stapes surgery. Occasionally, small exostoses or osteomas may be removed at the same time as stapes surgery. • A small or unusually angled ear canal should be noted in the clinic preoperatively. In extreme circumstances, enlargement may be necessary. More commonly, the case is simply made more difficult. • Limited neck rotation or obesity may require rotation of the surgical table to allow correct positioning of the ear for the procedure. If this rotation is more than about 15 degrees, patients having surgery under local anesthesia with sedation are concerned about sliding on the operating room table. Such situations are better handled with general anesthesia so that the patient can be positioned more securely on the table. • Congenital auricular or periauricular anomalies may indicate congenital malformation of the middle ear structures. • Abnormalities of the ossicles or of the chorda tympani nerve noted at otomicroscopy are markers for congenital abnormalities. • Severe myringosclerosis or an area of bimeric tympanic membrane (where the fibrous layer of the eardrum is missing, leaving only a mucosal lining and an epithelial surface) increases the chance of tympanic membrane perforation occurring as a result of tympanomeatal flap elevation. Note ������������������������������������� to the patient����������������������� should be made of the finding with an explanation of the procedure for correction intraoperatively or during a second procedure. Myringosclerosis may be an indicator of tympanosclerosis involving the stapes footplate (see later for tympanosclerosis management). T • uning forks should confirm a conductive hearing loss with Weber or Rinne testing consistent with the audiogram. Overestimation of the conductive hearing loss using some audiologic techniques occurs. Lack of a “flipped tuning fork” may not be a contraindication to surgery when bilateral disease is present.10 • Careful examination may reveal evidence of prior otologic surgery not mentioned by the patient.
Informed Consent Medicolegal situations over complications that are expected to occur in a small percentage of patients are avoided by obtaining written informed consent before the procedure. The reader is referred to other chapters for more information on the subject.
OPERATING ROOM Surgical Technique Prerequisites for Residents When surgeons are in training, much effort is put toward using proper methodology in the operating room. The technical difficulty associated with stapes surgery requires
Chapter 26 • Avoidance and Management of Complications of Otosclerosis Surgery
the surgeon to be facile with several key techniques. Residents and fellows should enter the operating room with previously demonstrated abilities in several areas. Preparation becomes more of an issue as the number of cases of surgically correctable otosclerosis decreases in most training programs.11 Limitation of hospital privileges may become more of an issue in the future if case availability precludes ascent to an acceptable level on an individual’s own learning curve. The following list is put forth for surgeons in training to use as preparation for successful performance of stapes surgery, while minimizing the risk of complications for the patient.
• Parfocal setup of the operating microscope: This allows the
surgeon to change magnification without losing focus in the operative field. Safety, accuracy, and speed all are improved. A • ppropriate microscope triangulation intraoperatively to allow depth of field perspective: To achieve depth of field information, the plane of view must be off-axis from the surgical instrument by several degrees. Movement of the microscope off-axis establishes a virtual triangle with its apex at the point of surgical interest (usually the tip of an instrument or an anatomic structure). The two points of the base of the triangle are formed by the microscope lens and the intersection of an imaginary line drawn perpendicular to the line of vision from the microscope where it intersects another imaginary line extending from the handle of the instrument (Fig. 26-1). Very small movements of the microscope can mean the difference between success and failure with delicate manipulations mandatory with stapedotomy. • Appropriate use of magnification balancing depth of field with surgical detail: Increasing magnification decreases depth of field and vice versa. The surgeon must balance the need for high magnification (e.g., needed with footplate work) with the need for depth of field (e.g., needed when securing the tympanomeatal flap with absorbable gelatin sponge [Gelfoam]). L • aser safety, use, beam sizing, and focus coincident with parfocal microscope: Delivery of laser energy is a function of laser wavelength, generated power from the laser base unit, and power density as determined by the size of the working spot size in the operative field. If the laser is scope mounted, it should remain focused precisely with the vision at the working magnifications in which it would be needed. The surgeon should also be familiar with beam defocusing. This technique can be useful with measures such as stopping bleeding and shrinking tissue. Hand-held lasers diverge from the fiberoptic tip. In this instance, laser power can be manipulated by moving the probe tip to or away from the intended target. T • wo-handed technique: It is crucial that the operative surgeon be able to use two hands to perform surgical maneuvers. This skill is learned and starts with the use of two hands for otologic procedures beginning with tympanostomy tubes. Likewise, vision with two eyes is
307
necessary to allow depth perception. Instruments must be grasped and used in a way that neither the right nor the left eye’s visual pathway from the microscope is obstructed. Observer side arms are monocular, and the surgeon in training may not realize when binocular vision is missing when using the binocular microscope unless instructed. Professors of otologic surgery can ensure binocular vision while looking through the monocular side arm by watching the video monitor and the side arm in succession. Typically, the side arm is what the operating surgeon sees through one eye, while the video shows what is seen through the other eye. • Lack of significant tremor: Although many fine surgeons have rhythmic variation in fine muscle control, delicate otologic surgery and significant tremor are mutually exclusive. Nervousness and anxiety make tremor worse. The surgeon and the professor of surgery should attempt to create and maintain an atmosphere of relaxed concentration and focus. Responsible surgeons would determine if caffeine avoidance on the morning of surgery is prudent. C • anal injection for hemostasis: Local anesthetic provides anesthesia for the patient under sedation and is crucial to procedure success. The mucosa of the middle ear receives innervation from deeper nerves and is not affected by the canal injection. A drop or two of local anesthesia instilled into the middle ear provides nearly instant relief. Anesthetic should be promptly removed from the middle ear to prevent absorption into the inner ear, avoiding the coincident severe vertigo, nausea, and vomiting. A high percentage of epinephrine in the canal injection (e.g., 1:20,000) provides a maximal amount of vasoconstriction with a minimum of volume of injection. Turning the needle such that the bevel is against the bone allows delicate instillation of solution under thin skin. Most innervation proceeds down the canal, but small nerves traverse the fissures of Santorini in the anteroinferior canal adjacent to the annulus. These nerves are more of an issue when dissection is carried out in this area, but they may also need to be surrounded with local anesthetic to provide a comfortable patient under sedation. Use of identical syringes for each procedure allows the surgeon to develop the feel necessary to perform excellent injections (we suggest a glass dental-type syringe holding 3 mL of solution). A perfect injection is an art that comes only with practice. Surgeons in training should perform and demonstrate proficiency with injections in cases where it is not as critical to the outcome of the procedure (e.g., tympanoplasty) before being allowed to inject for stapedotomy. Creation of large blebs cannot be rectified easily, and markedly increases case difficulty. • Tympanomeatal flap design and elevation: A recurring mistake of inexperienced surgeons occurs when the tympanomeatal flap is too short to reach anatomic position after curetting to expose the oval window niche (Fig. 26-2). One should avoid suctioning the elevated skin to prevent flap tears. Positioning the suction
308
OTOLOGIC SURGERY
Microscope Otologic instrument
FIGURE 26-1. Surgical technique of microscope triangulation.
Tympanomeatal flap
4
2
Bony scutum removed
Incision
FIGURE 26-2.
behind the round knife during flap elevation keeps the field dry and prevents inadvertent suctioning of the elevated skin. The vertical incisions of the flap should be kept 1 to 2 mm lateral to the annulus to reduce the likelihood of tearing the tympanic membrane.
• Chorda tympani nerve identification and preservation:
Surgeons should be able to identify and mobilize this nerve at the iter chordae posterius and from the posterior surface of the malleus to maximize the chance of its preservation.
Chapter 26 • Avoidance and Management of Complications of Otosclerosis Surgery
• Curetting: Proficiency must be shown in removing the
posterosuperior canal wall to expose the facial nerve and pyramidal process, while preserving the chorda tympani nerve. • Incudostapedial joint identification and division: Correct identification of the joint is necessary to prevent removal of the lenticular process of the incus. This becomes more of an issue when a bucket handle prosthesis is used. Slight anterior pressure on the incus while focusing on the joint under high power allows correct identification of this plane. • Prosthesis sizing: Understanding of the measurements used (Fig. 26-3) improves results. Measurement from the lateral surface of the footplate to the medial side of the incus is taken. To this figure is added 0.5 mm and an amount equal to the thickness of the stapes footplate. Footplate thickness may vary from 0.2 mm to several millimeters. The prosthesis should extend 0.5 mm beyond the medial surface of the footplate into the vestibule (Figs. 26-4 and 26-5). • Suction sizing to usage and risk: Proper two-handed technique turns the suction into an instrument. Suction size must be scaled to use. Damage to middle and inner ear structures may occur with use of too large a suction or with improper use of any suction. Intraoperative electro nystagmographic studies performed on stapedectomy patients in the prelaser era implicate suctioning over the vestibule, drilling near the oval and round windows, and manipulation of the stapes footplate as the great offenders for vestibular and presumably for cochlear damage.12 • Crus division and stapes superstructure removal: Fracture of the stapes superstructure with both crura intact and a fixed footplate is frequently possible. Division of the posterior crus before fracture decreases the chance of stapes footplate fracture, mobilization, and transmission of potentially damaging energy to the inner ear. Various methods are available for division of the posterior crus. Our preferred method is laser vaporization, which carries with it a much reduced risk for the inner ear. Although some techniques advocate laser vaporization of the anterior crus, there is no indication that fracture of a single crus with a fixed footplate carries any risk of footplate mobilization. P • rosthesis preparation and loading: For surgeons using a vein-clad technique, vein preparation, sizing, and veinclad prosthesis preparation are steps in which small variations can produce postoperative conductive hearing loss. The stapedotomy, prosthesis, and attached vein should be approximately 0.9 mm, 0.6 mm, and 0.2 mm. Variations in the thickness of the attached vein may cause the surgeon to vary the size of the stapedotomy. For this reason, the vein-clad prosthesis should be prepared before creation of the stapedotomy. In situations in which a vein is not attached to the prosthesis, the stapedotomy should exceed the size of the prosthesis by 0.2 to 0.3 mm. Bucket handle prostheses should be inserted with the bail inferior to the incus to
309
allow it to be swung superiorly and secured with small pieces of tissue. Shepherd’s crook pistons should be grasped with smooth forceps in a way that the open end of the wire hook can be placed directly over the incus. • Footplate removal: The surgeon must be able to create a round and concentrically located stapedotomy by melding multiple round individual laser bursts. The tiny movements used to move the laser are more reliably produced by moving the microscope with pressure from a hand or nose than with the joystick, although both methods work. Vaporization of the footplate with a laser leaves a small meshwork of char. This char does not need to be removed before prosthesis placement. Large fragments of footplate that enter the vestibule can give patients symptoms of positional vertigo. For this reason, all bony fragments within the stapedotomy that may mobilize with prosthesis placement should be vaporized. One should never attempt to “fish out” a fragment that has entered the vestibule. Use of a micro drill creates a round stapedotomy. Significant pressure on the footplate must be avoided to prevent drill overinsertion in the vestibule. • Prosthesis placement: Resident surgeons should practice crimping techniques in the laboratory before undertaking the maneuver on a patient. Likewise, placement of a bucket handle prosthesis should be performed in the temporal bone laboratory setting first. Excess vein from operative procedures can be taken to the laboratory to create a realistic practice model. • Prosthesis removal: Frequently with revision surgery, a prior prosthesis must be removed. It is also sometimes necessary, especially early in one’s career, to remove a prosthesis that is the wrong size during primary surgery. This is best accomplished for a prosthesis attached to the incus with a wire by inserting a right angle hook and rotating the hook to open the wire loop. In this way, traction forces on the incus are avoided. Laser ablation of the tissue surrounding the oval window segment of the prosthesis is a first step before removal to avoid traction damage of structures within the vestibule. Surgeons who have performed stapes surgery frequently understand that each individual step of the procedure itself must be performed to perfection, or the ill effects become additive. (For example, an imperfect injection makes every remaining portion of the procedure more difficult if blood continues to enter the operative field. Likewise, inadequate curetting limits exposure of the footplate.) A small amount of compromise in each step of the procedure quickly adds up to overall failure. This “law of additive inadequacy” has proven useful as a teaching concept. Viewing the procedure in this way emphasizes the need for perfection in each step before moving forward in the operation. Similar to a preflight checklist, every skill must be checked off before allowing performance of a stapedotomy. Surgeons who never master the listed techniques
310
OTOLOGIC SURGERY
4.25 mm
A = 3.5
B = 0.25
C = 0.5 (A = 3.5) + (B = 0.25) + (C = 0.5) = 4.25 mm
FIGURE 26-3. Stapedotomy prosthesis measurement lengths.
M I
Fixation
FIGURE 26-6. Malleus and incus fixation.
311
Chapter 26 • Avoidance and Management of Complications of Otosclerosis Surgery
probably should not perform stapes surgery as part of their surgical practice—just as not every otolaryngologist is able to perform procedures such as blepharoplasty, partial laryngectomy, and laryngotracheal reconstruction. Attending physicians are reticent to allow significant participation of residents and fellows if this means that several inadequacies have already summed to put the senior surgeon in a tough spot while completing the procedure. Residents and fellows gain the confidence of their attending surgeons as these skills are mastered allowing greater and greater participation. It is the duty of those of us who teach surgical technique to deliver to the next generation a cadre of well-trained individuals suited to stapes surgery without sacrificing success and safety for our current patients.
Surgical Equipment, Decisions, and Techniques
Comparative data from primary stapedotomy question this tenet, viewing both techniques as effective and safe.17 Use of a laser does improve results in revision cases.18 In all cases, proper use of the laser reduces bleeding associated with tissue ablation, which is an advantage. Both techniques are accepted within the standard of care.
Stapedectomy versus Stapedotomy For most otologists, the procedure of choice for otosclerosis has become stapedotomy. Compared with stapedectomy, the limited fenestra improves results in the high frequencies, and most authors report a reduction in SNHL as a result of the procedure.19-21 Stapedotomy carries a smaller rate of postoperative vestibular complaints. Stapedectomy remains a valuable alternative in the experience of some surgeons. Occasionally, a stapedotomy needs to be converted to a complete stapedectomy.
Prosthesis Type, Size, and Availability
Stapedotomy Site
Three general prosthesis types exist: piston/wire, bucket handle, and polytef (Teflon) varieties. Few comparative data exist to compare the different types, but bucket handle prostheses may have a smaller incidence of incus necrosis in long-term follow-up. Piston/wire and Teflon varieties are probably easier to place. Self-crimping prostheses offer a new option in design, which may reduce the need for manual crimping.13 Long-term results with attention to incus necrosis would be of interest because the metal nitinol used in these prostheses contains a small percentage of nickel—a known cause of hypersensitivity reactions in other uses in humans such as jewelry. Prostheses come in several different diameters, ranging from 0.3 to 0.8 mm most commonly. Experienced surgeons have indicated that 0.6 mm gives optimal results.14 Several studies have looked at alternative sizes, and there seems to be no degradation of results in the speech range down to 0.4 mm piston diameter.15 Prostheses measuring 0.3 mm show worse hearing results compared with prostheses measuring 0.4 mm.16 Lighter prostheses perform better in higher frequencies in situ in temporal bone studies. Prosthesis length varies from patient to patient. Availability of the correct prosthesis is crucial to successful outcome. We prefer to use nonferromagnetic materials, in particular titanium bucket handle prostheses. Table 26-1 outlines the prostheses stocked in our operating suite. Note the inclusion of the incus replacement prosthesis (see the later discussion on incus necrosis).
The stapedotomy should be placed posteroinferiorly in the central footplate region (see Fig. 26-2). This area does not overlay the saccule or utricle and gives the most margin for overpenetration on the medial side of the footplate. It is possible, and frequently necessary, to move the stapedotomy to other areas of the footplate because of anatomic concerns of the incus or structures surrounding the oval window niche.
Laser Stapedotomy versus Drill Stapedotomy It is generally agreed among most surgeons that use of the laser reduces the risk of mechanical transmission of vibratory energy to the inner ear, making it a safer technique.
Footplate/Vestibular Relationships Almost all patients have a minimum safe distance of 1 mm between the medial surface of the footplate and the utricle or saccule. Penetration of the vestibule by more than
TABLE 26-1 Prosthesis Type and Size Stocked
in Operating Suite
Length
A
3.5 3.75 4 4.25 4.5 4.75 5 5.75 6 6.25 6.5 6.75 7 7.25
X X X X X X X
B
C
X X X X
X X X X
D
X X X X X X X
A, Platinum-Teflon piston. B, Titanium bucket handle prosthesis 1 mm incus well × 0.6 mm piston. C, Titanium bucket handle prosthesis 0.9 mm incus well × 0.6 mm piston. D, Malleus-to-stapedotomy piston prosthesis (incus replacement prosthesis).
312
OTOLOGIC SURGERY
1 mm with instruments or the prosthesis may impinge the structures of the membranous labyrinth, producing vertigo with prosthesis movement. Perforation of the saccule or utricle may induce SNHL, vertigo, or both.22,23
Laser Type Controversy regarding laser type, efficacy, and safety has arisen since the first laser stapedotomy performed by Perkins.24 Visible wavelength lasers have the added advantage of absorption by blood, providing hemostasis. Appropriate adjustment of the CO2 laser also can provide hemostasis. Visible wavelength lasers have the technologic advantage of being able to use the laser light as the aiming beam; this avoids needing to produce a visible wavelength light that also serves as the aiming beam, which may not accurately reflect where the therapeutic beam is aimed (as can occur with the helium-neon aiming beam with CO2 lasers). Theoretical concerns regarding absorption of energy from the KTP or argon laser by the inner ear have been voiced by several authors (visible wavelength lasers are transmitted through clear fluid). Several studies have proven equal safety of visible wavelength and CO2 lasers for laser stapedotomy.25,26
are tolerated well by the patient and usually heard only as low-volume and low-frequency stimuli if performed appropriately, carrying minimal risk of injury to the inner ear. If a prosthesis is too long and touches the membranous labyrinth (which could produce vestibular symptoms with loud sounds postoperatively), gentle motion of the reconstruction induces vertigo in sedated patients allowing prosthesis replacement with a shorter variant during the original procedure. Prosthesis ballottement can be a useful measure for experienced surgeons to assess the function of the reconstruction on the operating table.
Prosthesis Displacement at Stapedotomy After prosthesis placement, a gentle posterior-to-anterior force applied to the shaft of the prosthesis at the footplate helps identify a prosthesis of inappropriate length. A prosthesis well situated and of appropriate length within a stapedotomy fenestra would not displace, whereas a too-short prosthesis would be moved from the stapedotomy. Because prosthesis displacement is the most common reason for failure of surgical success, proper sizing and positioning should receive special attention.
Local versus General Anesthesia
Oval Window Seal
Local anesthesia for laser stapedotomy is cost-effective, safe, and comfortable for patients. Under sedation, a patient can supply feedback to the surgeon if any event that stimulates the vestibular system occurs. Such feedback may help to prevent damage to inner ear structures. Additionally, the patient’s assessment of hearing ability after prosthesis placement and tympanomeatal flap replacement is helpful. Movement, which could be disastrous, is rarely experienced in the appropriately psychologically prepared and medicated patient. General anesthesia is used in patients unable to comply with the demands of local anesthesia with sedation, including claustrophobic patients, pediatric patients, or non– English-speaking patients. General anesthesia has appeal for teaching cases, where operative time may be long.
A tissue seal at the oval window (we prefer vein taken from the dorsum of the hand) nearly eliminates the risk of perilymphatic fistula.28
Stapedotomy in Children Stapedotomy can be performed safely in children with surgical results as least as good as the results expected in adults.27 Children should be beyond the age of otitis media, and surgery should be attempted only under general anesthesia. Congenital footplate fixation requires special considerations, as discussed later.
Prosthesis Ballottement After prosthesis placement, careful and gentle palpation of the incus reveals the degree of freedom of movement of the prosthesis. Gentle, nonsudden movements
MANAGEMENT OF COMPLICATIONS ENCOUNTERED AT PRIMARY SURGERY Tympanic Membrane Perforation and Flap Tear Skin or tympanic membrane tears can complicate elevation of the tympanomeatal flap. The case can be continued with an underlay graft of fascia or other collagencontaining tissue providing repair. With small tears, the defect heals if edges are approximated without an underlay graft. A similar repair can be used when the flap is made too short to cover the area of bone removal in the posterosuperior margin of the bony annulus. The flap incision should be at least 6 mm from the annular ligament to prevent inadequate flap length. The surgeon can estimate appropriate flap length to be twice the length of a large round knife.
Incus Dislocation Dislocation of the incus precludes routine stapedotomy. The case should be halted and time given for the incus and malleus to reattach. In 4 to 6 months, the patient is taken back to the operating room. If sufficient reattachment has occurred, stapedotomy is performed in the
Chapter 26 • Avoidance and Management of Complications of Otosclerosis Surgery
usual fashion. If not, the incus is removed, and a malleusto-footplate prosthesis is used.
Malleus and Incus Fixation Routinely, every case should include a check of the mobility of the malleus and incus. Patients with otosclerosis have a higher incidence of hyalinization of the anterior malleal ligament, which may be the sole cause or a contributing factor in the conductive hearing impairment.29 This is best performed after division of the incudostapedial joint with gentle upward pressure on the undersurface of the handle of the malleus, while watching for movement at the lenticular process of the incus. Documentation of the state of function of the first two ossicles is crucial when evaluating and planning revision for patients who do not experience adequate improvement of hearing with the primary procedure. In addition, in approximately 1% of cases, fixation of the malleus and incus is discovered, allowing rectification during the primary procedure. A small or moderate amount of limitation of motion of the malleus and incus usually produces little in terms of conductive hearing loss, which is usually low frequency in nature. In cases where doubt exists about the severity of malleus and incus fixation with clearly appreciated footplate fixation, performance of the stapedotomy with postoperative testing before work on the malleus or incus or both shows good judgment. Operative repair of the situation is best addressed with a mastoidotomy with extension into the root of the zygoma to allow exposure of the body of the incus and head of the malleus in the epitympanum (Fig. 26-6). Usually, fixation occurs at the superior malleolar ligament from the tegmen. The posterior process of the incus may also be involved, as can any suspensory ligament of the ossicular chain. Removal of offending bone and restoration of ossicular mobility can be performed with a laser.30 Larger wattages are necessary than those used on the stapes superstructure or footplate (we use the KTP laser with 5 to 8 W on continuous mode). A bloodsoaked Gelfoam is placed as a “backstop” to protect the facial nerve with higher laser settings. Although a drill can remove bone as well, laser use minimizes the risk of transmission of damaging vibratory energy to the vestibule. If the stapes is also fixed, it is prudent to remove the stapes superstructure before mobilization of the malleus and incus. If the conductive hearing loss resides solely in the fixation of the malleus and incus, disarticulation of the incudostapedial joint to produce discontinuity is mandatory in cases where the laser is not used to remove the fixation (sometimes interposition of Gelfoam is necessary to keep the lenticular process of the incus from touching the capitulum of the stapes). Removal of bone in the mastoid can be performed comfortably in most patients under sedation if necessary. If the suspensory ligaments cannot be reached safely, reconstruction can still be performed. Removal of the
313
incus allows access to the head of the malleus through the mastoid. Removal of the head of the malleus superior to the lateral ligament is performed with a malleus nipper or a tiny diamond drill. The chorda tympani nerve should be mobilized from the medial surface of the neck of the malleus and pushed inferiorly to prevent its injury with this maneuver. Reconstruction with a malleus-to-stapes footplate prosthesis completes the repair. Complete closure of the air-bone gap is frequently impossible with this technique, and the decision for surgery should take this into account.
Persistent Stapedial Artery and Vascular Anomalies Rarely, a persistent stapedial artery is encountered running from the facial nerve through the arch of the stapes to the carotid artery. Even less commonly, an aberrant carotid is seen. Although the persistent stapedial artery is not seen with otoscopy, an aberrant carotid is visible, and can be mistaken for a glomus tumor. Most vascular anomalies occur in women.31 If a persistent stapedial artery is small, fine bipolar cautery or laser coagulation may be used to remove the vessel from the field, allowing completion of the procedure. With larger arteries, it may be prudent to stop the procedure and prescribe amplification.
Tympanosclerosis One may discover ossicular fixation secondary to tympanosclerosis of the stapes footplate. Fixation of this type is not as vascular as otosclerosis and has a softer texture. Stapedotomy may be performed safely. Excellent early results may be achieved with some attrition over time.32-34
Osteogenesis Imperfecta Osteogenesis imperfecta is a congenital disorder of bone inherited via autosomal dominant or autosomal recessive patterns. The triad of blue sclera, multiple fractures, and conductive hearing loss is known as Van der Hoeve’s syndrome. The mean age of hearing impairment is in the early 20s.35 Footplate fixation accounts for the conductive hearing loss. Footplates are typically thick and frequently very soft with increased vascularity, whereas crura may be atrophic.36 Progressive SNHL can be seen in a few patients.37 Despite a small increase in risk of SNHL in some series, stapedotomy remains a viable treatment for conductive hearing loss associated with this disorder.38
Congenital Stapes Fixation Complete absence of the annular ligament of the stapes without evidence of otosclerotic bone growth may be congenital stapes fixation. This entity is discussed at the beginning of the chapter.
314
OTOLOGIC SURGERY
Round Window Obliteration
Biscuit Footplate
Severe otosclerotic overgrowth may completely obliterate the round window membrane. No remedy is available for the excess bone, and complete closure of the air-bone gap is unlikely postoperatively.
Manipulation of a biscuit footplate puts the patient at increased risk for footplate mobilization and SNHL (Fig. 26-8). Use of the laser allows removal of bone and performance of a footplate fenestra without significant energy transfer to the tenuous footplate. Care should be used to prevent heat transfer to the vestibule, which may put the inner ear at risk. Cases performed under local sedation allow the surgeon to sense when caloric stimulation begins for the patient as vertigo frequently begins. Cessation of laser use for several minutes allows cooling of the footplate and vestibule, allowing resumption of the bone removal process. Patients operated on under general anesthesia should have no more than 8 to 10 laser pulses in a row without allowing time for cooling. Excellent results are obtainable for the patient and surgeon.
Overhanging Facial Nerve The facial nerve can provide obstruction to completing successful stapes surgery and is encountered regularly (Fig. 26-7). Recognition of an aberrant or dehiscent facial nerve prevents injury. Twenty-five percent to 40% of temporal bone specimens include dehiscence of the bony fallopian canal. Inconveniently located dehiscence may be sites of trauma secondary to surgical instrumentation. Local anesthetic agents may also penetrate these areas more readily, causing postoperative facial paralysis. The reader is referred to Chapter 29 for management options. In some situations, coupling the technique used for narrow oval window enlargement becomes necessary for prosthesis placement. A dehiscent facial nerve rarely prevents successful surgical repair.
Narrow Oval Window Niche Anatomic abnormalities or impingement on the oval window area by the facial nerve can produce narrowing so that prosthesis placement is difficult or impossible. Enlargement of the inferior margin of the oval window niche at the midpoint of the footplate just anterior to the junction of the subiculum and promontory can facilitate surgical success.39 Intermittent laser pulses produce char that is then removed with a rasp. Significant extra space can be gained with this technique because the bone is quite thick in this area. Care must be taken to prevent caloric overstimulation of the vestibule. In addition, the stapedotomy can be shifted inferiorly to the margin of the oval window to facilitate prosthesis placement. Overlapping the margin of the oval window can be dangerous because Reissner’s membrane and the basilar membrane occupy this area in the cochlear hook region.40
Congenitally Ectopic Facial Nerve The facial nerve may be congenitally malpositioned entirely on the promontory side of the footplate, split with a portion on either side of the footplate, or pass through the arch of the stapes.41,42 The stapes crura are not attached to the footplate a high percentage of the time with abnormalities in the course of the facial nerve. If prosthesis placement can proceed without iatrogenic injury to the nerve, the case can be completed with the prosthesis placed above, between, or around the aberrant nerve. (Alternatively, few hearing devices produce facial paralysis!)
Obliterative Otosclerosis Obliteration of the oval window niche occurs from otosclerotic bone growth in some cases (Fig. 26-9). The area should not be drilled out using a diamond burr because reactivation and reformation of the otosclerotic growth may be stimulated, and a higher rate of SNHL may be realized. Bleeding is also frequently produced with such a technique.43-45 Footplate fenestration may be accomplished as with a biscuit footplate outlined previously.
Perilymphatic Gusher The normal anatomic arrangement of the human ear allows flow of CSF into the perilymphatic space. Normally, extremely small connections between these two spaces exist. With enlarged connections (usually through the fundus of the internal auditory canal, through the modiolus of the cochlea, and possibly through an enlarged cochlear aqueduct), a rapid outpouring of perilymph occurs when the stapes footplate is removed as a barrier in either a stapedectomy or a stapedotomy. Known as a “gusher,” such an event can be associated with SNHL.46 Conditions known to increase the chance of this condition include enlarged vestibular aqueduct syndrome, X-linked progressive mixed deafness, congenital footplate fixation, and Mondini’s dysplasia. Successful repair is most easily accomplished with a bucket-type piston over a vein graft seal of the stapedotomy. The bucket-type prosthesis is recommended because of the possible CSF pressure extrusion of a piston-shepherd’s crook–type prosthesis. If a watertight seal can be obtained, no further treatment is necessary except restricted activity for 1 week. Several authors recommend creation of a small control hole in the footplate before removal of the
Chapter 26 • Avoidance and Management of Complications of Otosclerosis Surgery
Facial nerve
315
316
OTOLOGIC SURGERY
f ootplate if the technique of stapedectomy is used. It is much easier to handle this complication through a small footplate hole than with the entire footplate out. In the event a gusher is encountered that cannot be controlled with a vein and prosthesis, more extensive management becomes necessary. In this situation, management is similar to that of a CSF leak. Tissue is used to provide a seal for the oval window fenestra, preferably with a prosthesis providing tamponade (which is not always possible). A lumbar drain is placed to decrease the cerebrospinal and perilymph pressure. Postoperatively, the head of the bed is elevated at least 30 degrees, and bed rest is employed. Fluids are restricted, and oral acetazolamide may be prescribed. Prophylactic antibiotics are given.47 The patient is discharged 24 hours after the lumbar drain is clamped with no further fluid leak as judged by CSF rhinorrhea. Bed rest at home is recommended for another week.
Fractured Footplate
offenders for vestibular and presumably for cochlear damage.49
COMPLICATIONS AFTER PRIMARY SURGERY Medical Management Acute Otitis Media Acute otitis media occurs in the postoperative period in patients who have had stapes surgery. The infection is usually successfully treated without sequelae. Entrance of pathogenic organisms into the perilymph can produce SNHL and vestibular damage. Progression into the CSF has been reported with meningitis.8,50-52 Prophylactic antibiotics have not been shown to reduce the incidence of this complication and carry some risk, and are not recommended.
Barotrauma
No attempt should be made to remove any footplate fragments in the vestibule. Although positional vertigo is rarely a sequela of these fragments, most cause no harm. Attempted removal with an instrument in the vestibule carries a high rate of SNHL.
Iatrogenic rearrangement of the normal anatomy may increase a patient’s chance of experiencing barotrauma of the inner ear after stapes surgery. We allow patients to fly in pressurized aircraft 2 days after surgery when a vein graft has been placed. No restrictions are placed for snorkeling or scuba diving after healing has occurred, provided that patients are able to self-insufflate through an open eustachian tube (as all patients should have for these activities even without previous ear surgery). A wide range of practices regarding postoperative restrictions exists among otologists. No significant differences have been shown in the prevalence of barotrauma based on individual physicians’ recommendations for these activities.53
Blood within the Vestibule
Dysgeusia
A bloodless field increases safety by allowing the surgeon to see all structures clearly. In addition, blood in the vestibule is associated with an increase in risk for sensorineural hearing impairment in animal studies and human experience.48 The ultimate surgical technique employs a completely bloodless oval window before entrance into the vestibule through the stapes footplate.
Approximately 20% of patients note taste changes in the postoperative period when asked. Chorda tympani nerve dysfunction is usually transient, and less than 5% of patients experience permanent deficits. “Second ears” and patients with a vocation involving taste warrant special consideration. The risk of economic hardship to a patient who works as a sommelier or coffee taster may not be worth the risk of surgery. Revision cases frequently involve a nerve adherent to the posterior surface of the tympanic membrane. Mobilization of an adherent nerve is best accomplished with a small sharp blade (e.g., a No. 5910 Beaver blade designed for corneal incisions).
Fragmentation occasionally occurs. Free-floating surface fragments should be removed, and the surgery should be converted to a partial or complete stapedectomy. A tissue seal is mandatory in this setting.
Footplate Fragments in the Vestibule
Sensorineural Hearing Loss SNHL is perhaps the most disappointing and devastating complication for the patient and the surgeon. Complete or partial hearing loss can result from the most meticulously and appropriately performed stapes procedure. Most cases probably are due to surgical trauma, however. Intraoperative electronystagmographic studies performed during stapedectomy in the prelaser era implicate suctioning over the vestibule, drilling near the footplate or oval window niche, and manipulation of the footplate as the great
Delayed Facial Nerve Paralysis A few cases of delayed facial nerve paralysis have occurred. Typically, the onset of paralysis occurs 7 to 10 days after surgery and is associated with pain. Treatment with a tapering dose of steroids (initial dosing of prednisone, 30 mg twice daily tapering over 2 weeks) has brought about resolution in all cases.54
Chapter 26 • Avoidance and Management of Complications of Otosclerosis Surgery
317
Hyperacusis
Vestibular Complaints
Almost all patients have some degree of phonophobia postoperatively. Reassurance is adequate treatment, with only a few patients experiencing a persistent problem.55
Vertigo with and after stapedotomy is less frequent than with stapedectomy. Approximately 1 or 2 in 10 patients experience some balance complaint immediately after laser stapedotomy. Standard sedative vestibular suppressants provide relief allowing the problem to resolve on its own. Patients rarely have symptoms for more than 2 days postoperatively.
Binaural Diplacusis The same tone is perceived as different pitches in each ear in one third of patients after stapes surgery. By 6 weeks postoperatively, the condition fades without treatment.56
Otosclerotic Inner Ear Syndrome Balance disturbance may result from ongoing growth of the otosclerotic focus postoperatively. Patients complain of brief episodes of motion not severe enough to be termed spinning or of diffuse, persistent unsteadiness. Such complaints are hard to distinguish from perilymphatic fistula and other causes of postoperative vestibular complaints. A high percentage of patients respond to the administration of fluoride when the complaint is due to ongoing otosclerosis.57 We currently use calcium fluoride (Monocal), 2 tablets twice daily. Sodium fluoride (Flurocal) has been associated with gastric intolerance in a higher percentage of patients.58
Sensorineural Hearing Loss Progression Otosclerosis induces a progressive SNHL in many patients.59-65 Otosclerosis can be a cause of SNHL without a conductive component.66 It can be extremely disappointing to see an excellent surgical result deteriorate because of progressive SNHL, leaving the patient with functionally significant hearing impairment. Convincing data exist that establish the role of fluoride in stabilizing SNHL associated with otosclerosis.67-73 We place patients with SNHL present at the time of surgery on calcium fluoride (2 tablets orally twice daily with meals) for 1 to 2 years after surgery. If hearing remains stable at that time, the treatment is stopped. Any further progression prompts further treatment. Patients who exhibit new-onset SNHL in the postoperative period are treated similarly. Patients whose otosclerosis produces severe to profound sensorineural hearing loss can be expected to do well with cochlear implants, although there is a higher incidence of facial nerve stimulation in such applications.74
Serous Labyrinthitis Patients who experience the onset of constant dysequilibrium several days after surgery are thought to have an inflammatory response in the inner ear. Frequently, a short course of steroids improves symptoms and alleviates dizziness.
Wound Infection Antibiotic-soaked Gelfoam is placed against the tympanomeatal flap incision after return of the eardrum and skin to anatomic position. It is unnecessary to fill the ear canal. The packing is removed 5 to 10 days after surgery. Antibiotic eardrops are prescribed once daily for 2 weeks. Water precautions are followed until 3 weeks postoperatively. With such a regimen, wound infections occur extremely rarely.
Upper Respiratory Infection Infection in the postoperative period with influenza virus has been associated with an increase in SNHL.75
Surgical Management Adhesions Postoperative adhesions form in virtually every surgical case. Significant conductive hearing loss is rarely a result of adhesions for stapedotomy patients, unless the healing process has displaced the prosthesis. When the middle ear is re-examined after primary surgery in these patients, a source of conductive hearing loss apart from adhesions must be sought. Stapedectomy patients have a higher tendency to form adhesions around the prosthesis foot in the oval window. Severe fibrosis in this area can displace or severely limit motion of the prosthesis producing conductive loss. One of the most useful features of the laser for use in revision surgery is removal of adhesions with little to no bleeding. For insignificant adhesions, laser vaporization allows identification of the real cause of recurrent conductive hearing impairment without troublesome bleeding. In the case of oval window fibrosis, therapeutic removal of the adhesions with maintenance of a thin membrane separating the vestibule allowing prosthesis placement is possible.
Acute Facial Nerve Paralysis As opposed to delayed facial nerve paralysis, which is a medically treated condition, acute paralysis may require surgical intervention. Adequate time (approximately 2 hours) should elapse after surgery to preclude a local anesthetic effect. Cases of persistent paralysis should be re-explored. Referral or including a colleague for
318
OTOLOGIC SURGERY
bjectivity and future support and collaboration may be o a prudent decision.
Conductive Hearing Loss Return of conductive hearing loss is the most commonly experienced postoperative complication of stapedotomy, accounting for 50% to 70% of revision surgical procedures.76-80 Although exact rates are difficult to determine, 10% to 20% of cases require revision sometime during the patient’s lifetime. With the decrease in primary surgery, many experienced stapes surgeons perform revision surgery a significant percentage of the time.
Prosthesis Displacement Hearing deterioration may be acute, as the prosthesis becomes dislodged, or chronic, as it is gradually displaced out of position (Fig. 26-10). This complication may be encountered many years after stapes surgery. The diagnosis is definitively made at reoperation, but may be suspected based on history and tuning fork and audiometric testing.
Incus Necrosis Division of the incudostapedial joint reduces the blood supply to the distal incus. Vessels from the stapes superstructure provide a little perfusion, whereas vessels coursing from the stapedial tendon bring most of the blood to the distal incus. Channels for blood flow exist within the interior of the long process of the incus. Necrosis of this incus leads to prosthesis displacement (Figs. 26-11 and 26-12). Prostheses that crimp onto the incus may reduce further the only remaining blood supply coming from the direction of the body of the incus, increasing the chance of this complication. For this reason, some surgeons prefer a bucket handle prosthesis. Incus necrosis still occurs, no matter which prosthesis is used. If an adequate amount of incus remains, a shepherd’s crook prosthesis can be used by crimping the prosthesis more proximally on the incus during revision surgery. Occasionally, it becomes necessary to bend the prosthesis around the facial nerve ridge to establish footplate contact. The amount of bend affects the length of prosthesis needed. In selected cases, application of an otologic cement may aid in prosthesis stabilization after repositioning or replacement.81 In the event the remnant incus cannot be used, the tympanic membrane is elevated from the lateral surface of the malleus, and a malleus-to-stapes footplate prosthesis is placed. Table 26-1 lists sizes of our preferred malleus-to-footplate prostheses.
extrusion becomes evident. A prosthesis that has partially extruded through the eardrum can be left in place as long as adequate function is maintained, infection is not present, and skin is not growing into the middle ear. In this event, the ear should be kept dry. Persistent tympanic membrane perforations should be repaired before replacing the stapes prosthesis to reduce the bacterial content of the middle ear before opening the vestibule. Placement of a tympanostomy tube or positioning of autologous tissue over the prosthesis helps discourage future extrusion.
Otosclerotic Regrowth Otosclerosis continues to grow in some patients, producing prosthesis fixation or displacement (Fig. 26-13). At revision surgery, the cause of the conductive hearing impairment is best handled with prosthesis removal, laser enlargement of the stapedotomy, and placement of a fresh prosthesis.
Wire Loop Prosthesis The dominant prosthesis used for many years for stapes surgery was the wire loop prosthesis. The wire loop prosthesis was used only in the technique of total stapedectomy. Return of conductive hearing impairment prompting revision frequently shows displacement of the oval window portion of the prosthesis. Because tissue was used to seal the stapedotomy in most instances, a large amount of scar usually exists in the oval window (Fig. 26-14). A properly adjusted laser allows removal of scar surrounding the oval window loop of the prosthesis. The wire loop prosthesis should not be grasped and removed because it frequently is attached to vital structures of the vestibule, producing a high chance of inner ear injury. It is necessary when revising these cases to leave a small membrane of tissue, if possible, over the vestibule, or to insert autologous tissue with the new prosthesis.
Long Prosthesis Overinsertion Overinsertion of the long prosthesis into the vestibule may cause a sensation of vertigo with loud sounds (Fig. 26-15). If this occurs, prosthesis removal and replacement with a shorter version usually fixes the difficulty. Palpation of the freshly placed prosthesis by putting gentle pressure on the incus in patients under sedation identifies a too-long prosthesis while still in the operating room. With a toolong prosthesis, the patient experiences vertigo with this maneuver. Care should be taken to avoid overstimulation of the vestibule, which could produce SNHL.
Prosthesis Extrusion
Loose Prosthesis Syndrome
Prosthesis extrusion may rarely occur. Usually, contact of the prosthesis is first established by retraction of the eardrum. Over time, a perforation occurs, and prosthesis
Loose coupling of the prosthesis to the incus produces characteristic sensations of sound distortion (see Fig. 26- 11).77 A large conductive hearing loss may not exist
Chapter 26 • Avoidance and Management of Complications of Otosclerosis Surgery
Wire loop prosthesis
Regrowth of otosclerosis
Saccule
Saccule
Prosthesis contacting membranous labyrinth
FIGURE 26-13. Otosclerosis regrowth. FIGURE 26-14. Adhesions from oval window to membranous labyrinth. FIGURE 26-15. Excess prosthesis length, which contacts the membranous labyrinth.
Scar attached to membranous labyrinth
319
320
OTOLOGIC SURGERY
on the audiogram; tuning forks may also be normal. Occasionally, severe symptoms of this type prompt revision surgery. Tightening the crimp or changing to a different prosthesis can completely alleviate the difficulty.
Perilymphatic Fistula Perilymphatic fistula is a cause of postoperative dysequilibrium and SNHL. The signs and symptoms of perilymphatic fistula may be indistinguishable from normal postoperative findings.78,79 In addition, a perilymphatic fistula may be discovered at revision surgery with an asymptomatic patient. Fistulas seem to be much less common after stapedotomy compared with stapedectomy. Progressive SNHL or unremitting vestibular complaints or both may prompt re-exploration. Repair of a symptomatic perilymphatic fistula relieves vestibular complaints in approximately half of patients.
Reparative Granuloma Reparative granuloma is defined as a histologically confirmed formation of granulation tissue involving the prosthesis and oval window in a symptomatic patient after stapes surgery. The lesion does not involve granulomatous inflammation.80 This unusual complication occurs in approximately 0.1% of all cases. Presenting symptoms usually surface 1 to 6 weeks after surgery, and most commonly involve vertigo, but may include SNHL, progressive mixed hearing loss, sudden hearing loss, and tinnitus.82 Management includes either immediate surgery with removal and replacement of prosthesis and grafting material or nonsurgical management comprising steroids and antibiotics. Surgical intervention seems to give a better outcome.
Introduced Ectopic Tissue Occasionally, material introduced into the operative field brings about a return of conductive hearing loss. Introduced squamous epithelium can produce a cholesteatoma,83 and perichondrium has been known to stimulate formation of cartilage.84
SURGICAL RISK REDUCTION IN REVISION SURGERY Although variation occurs based on the adequacy and technique of primary surgery, revision stapes surgery is needed in a significant percentage of patients. Series from nations with centralized health care provide the best data because of controlled follow-up. A revision rate of 13% in one series of 4000 cases amassed by several surgeons has been reported.85 Revision may be necessary immediately after primary surgery or many
years later, with an average time to revision of 8 to 12.5 years.86,87 Success rates for revision surgery have improved significantly over the past 20 years. Early reports found postoperative conductive hearing loss of 10 dB or less in less than half of patients.72,88 Modern results more closely approach those of primary surgery with closure to 10 dB or less in 90% of cases. Use of a laser improves surgical outcome. Metaanalysis of revision stapes surgery comparing 11 studies without laser use (N = 1147 patients) with 4 studies that employed laser technique (N = 170 patients) showed a statistically significant (P = .0002) advantage in terms of safety and efficacy. Postoperative air-bone gaps of 10 dB or less were accomplished in 69% of cases in which a visible wavelength laser was used, whereas only 51% of patients on whom standard techniques were used had the same results.89 The risk of SNHL is greater with revision surgery. Rates seem to be higher when revision of stapedectomy is undertaken compared with cases with stapedotomy as the primary technique. Oval window drill-out is associated with an unacceptably high rate of inner ear injury and hearing loss and should be avoided.72 Reported rates of SNHL with revision surgery vary from 0 to 7.6%.44,72,86,90 We currently quote patients a rate of SNHL twice that of primary surgery.
CONCLUSION Successful performance of surgery for otosclerosis includes thorough mental and physical preparation. A solid knowledge base of potential complications and their management is vital to patient counseling, treatment, and outcome. Although complications are unavoidable, and certain ones can be rectified, avoidance with proper surgical technique and planning is preferable for the patient and the surgeon.
REFERENCES 1. Smith M FW, Hopp M L : 1984 Santa Barbara State-ofthe-Art Conference on Otosclerosis: Results, conclusions, consensus. Ann Otol Rhinol Laryngol 95:1-4, 1986. 2. Garretsen TJTM, Cremers WR J: Ear surgery in osteogenesis imperfecta. Arch Otolaryngol Head Neck Surg 116:317-323, 1990. 3. Olson N R , Lehman R H : Cerebrospinal fluid otorrhea and the congenitally fixed stapes. Laryngoscope 78:352360, 1968. 4. Snik A F, Hombergen GC, Mylanus E A, et al: Air-bone gap in patients with X-linked stapes gusher syndrome. Am J Otol 16:241-246, 1995. 5. Cremers CW: Audiologic features of X-linked progressive mixed deafness syndrome with perilymphatic gusher during stapes surgery. Am J Otol 6:243-246, 1985.
Chapter 26 • Avoidance and Management of Complications of Otosclerosis Surgery 6. Cremers CW, Huygen PL : Clinical features of female heterozygotes in the X-linked mixed deafness syndrome (with perilymphatic gusher during stapes surgery). Int J Pediatr Otorhinolaryngol 6:179-185, 1983. 7. Brown J S : Meningitis following stapes surgery: The pathway of spread to the intracranial cavity. Laryngoscope 77:1295-1303, 1967. 8. Clairmont A A, Nicholson WL , Turner J S : Pseudomonas aeruginosa meningitis following stapedectomy. Laryngoscope 85:1076-1083, 1975. 9. Snyder B D: Delayed meningitis following stapes surgery. Arch Neurol 36:174-175, 1979. 10. Gordon M A, Silverstein H, Willcox TO, et al: A re-evaluation of the 512-Hz Rinne tuning fork test as a patient selection criterion for laser stapedotomy. Am J Otol 6:712-717, 1998. 11. Harris J P, Osborne E : A survey of otologic training in U.S. residency programs. Arch Otolaryngol Head Neck Surg 116:342-345, 1990. 12. Majoras M : Electronystagmography during stapedectomy. Int Surg 47:323-327, 1967. 13. Rajan G P, Atlas M D, Subramaniam K, Eikelboom R H : Eliminating the limitations of manual crimping in stapes surgery? A preliminary trial with the shape memory nitinol stapes piston. Laryngoscope 115:366-369, 2005. 14. Shea JJ: Thirty years of stapes surgery. J Laryngol Otol 102:14-19, 1988. 15. Fisch U: Stapedectomy versus stapedotomy. Am J Otol 4:112-117, 1982. 16. Grolman W, Tange R A, de Bruijn A J, et al: A retrospective study of the hearing results obtained after stapedotomy by the implantation of two Teflon pistons with a different diameter. Eur Arch Otorhinolaryngol 254: 422-424, 1997. 17. Sedwick J D, Louden C L , Shelton C : Stapedectomy versus stapedotomy: Do you really need a laser? Arch Otolaryngol Head Neck Surg 123:177-180, 1997. 18. Wiet R J, Kubek DC, Lemberg P, et al: A meta-analysis review of revision stapes surgery with argon laser: Effectiveness and safety. Am J Otol 18:166-171, 1997. 19. Persson P, Harder H, Magnuson B : Hearing results in otosclerosis surgery after partial stapedectomy, total stapedectomy, and stapedotomy. Acta Otolaryngol 117: 94-99, 1997. 20. Kursten R , Schneider B, Zrunek M : Long-term results after stapedectomy versus stapedotomy. Am J Otol 15:804-806, 1994. 21. Glasscock M E III, Storper I S, Haynes DS, et al: Twentyfive years of experience with stapedectomy. Laryngoscope 105:899-904, 1995. 22. Anson B J, Bast TH : Anatomical structure of the stapes and the relation of the stapedial footplate to vital parts of the labyrinth. Ann Otol Rhinol Laryngol 67:389-399, 1958. 23. Pauw B K H, Pollack A M, Fisch U: Utricle and saccule and cochlear duct in relation to stapedotomy: A histological human temporal bone study. Ann Otol Rhinol Laryngol 100:966, 1991. 24. Perkins RC : Laser stapedotomy for otosclerosis. Laryngoscope 91:228-241, 1980. 25. Vernick D M : A comparison of the results of KTP and CO2 laser stapedotomy. Am J Otol 17:221-224, 1996.
321
26. Antonelli PJ, Gianoli GJ, Lundy L B, et al: Early postlaser stapedotomy hearing thresholds. Am J Otol 19: 443-446, 1998. 27. Robinson M : Juvenile otosclerosis: A 20-year study. Ann Otol Rhinol Laryngol 92:561-565, 1983. 28. Lippy WH, Schuring AG: Stapedectomy revision following sensorineural hearing loss. Otolaryngol Head Neck Surg 92:580-582, 1994. 29. Nandapalan V, Pollak A, Langner A, Fisch U: The anterior and superior malleal ligaments in otosclerosis: A histopathologic observation. Otol Neurotol 23:854-861, 2002. 30. Seidman M A : A new approach for malleus/incus fixation: No prosthesis necessary. Otol Neurotol 25:669-673, 2004. 31. Pirodda A, Sorrenti G, Marliani A F, et al: Arterial anomalies of the middle ear associated with stapes ankylosis. J Laryngol Otol 108:237-239, 1994. 32. Tos M, Lau T: Tympanosclerosis of the middle ear: Late results of surgical treatment. J Laryngol Otol 104:685689, 1990. 33. Gormley PK : Stapedectomy in tympanosclerosis. Am J Otol 8:123-130, 1987. 34. Giddings N A, House JW: Tympanosclerosis of the stapes—hearing results for various surgical treatments. Otolaryngol Head Neck Surg 107:644-650, 1992. 35. Pedersen U, Elbrond O: Stapedectomy in osteogenesis imperfecta. ORL J Otorhinolaryngol Relat Spec 45: 330-337, 1983. 36. Garretsen TJTM, Cremers CWR J: Stapes surgery in osteogenesis imperfecta: Analysis of postoperative hearing loss. Ann Otol Rhinol Laryngol 100:120-130, 1991. 37. Garretsen TJTM, Cremers CWR J: Ear surgery in osteogenesis imperfecta. Arch Otolaryngol Head Neck Surg 116:317-323, 1990. 38. Vincent R , Gratacap B, Oates J, Sperling N: Stapedotomy in osteogenesis imperfecta: A prospective study of 23 consecutive cases. Otol Neurotol 26:859-865, 2005. 39. Inserra M M, Mason TP, Yoon PJ, Roberson J B Jr: Partial promontory technique in stapedotomy cases with narrow niche. Otol Neurotol 25:443-446, 2004. 40. Stidham K, Roberson J B Jr: Cochlear hook anatomy: Evaluation of the spatial relationship of the basal cochlear duct to middle ear landmarks. Acta Otolaryngol 119: 773-777, 2000. 41. Leek J H : An anomalous facial nerve: The otologist’s albatross. Laryngoscope 84:1535-1544, 1974. 42. Willis R : Conductive deafness due to malplacement of the seventh nerve. J Otolaryngol 6:1-4, 1977. 43. Gherini SG, Horn K L , Bowman C A, et al: Small fenestra stapedotomy using a fiberoptic hand-held argon laser in obliterative otosclerosis. Laryngoscope 100: 1276-1282, 1990. 44. Derlacki E L : Revision stapes surgery: Problems with some solutions. Laryngoscope 95:1047-1053, 1985. 45. Farrior D: Abstruse complications of stapes surgery: Diagnosis and treatment. In Henry Ford Hospital International Symposium on Otosclerosis. Chicago, Little, Brown, 1962, pp 509–521. 46. Glasscock M E : The stapes gusher. Arch Otolaryngol Head Neck Surg 98:82-91, 1973. 47. Brodie H A : Prophylactic antibiotics for post-traumatic cerebrospinal fluid fistulae: A meta-analysis. Arch Otolaryngol Head Neck Surg 123:749-752, 1997.
322
OTOLOGIC SURGERY
48. Radeloff A, Unkelbach M H, Tillein J, et al: Impact of intrascalar blood on hearing. Laryngoscope 117:58-62, 2007. 49. Majoras M : Electronystagmography during stapedectomy. Int Surg 47:323-327, 1967. 50. Gristwood R E : Acute otitis media following the stapedectomy operation. J Laryngol Otol 80:55-60, 1966. 51. Brown J S : Meningitis following stapes surgery: The pathway of spread to the intracranial cavity. Laryngoscope 77:1295-1303, 1967. 52. Snyder B D: Delayed meningitis following stapes surgery. Arch Neurol 36:174-175, 1979. 53. Harrill WC, Jenkins H A, Coker NJ: Barotrauma after stapes surgery: A survey of recommended restrictions and clinical experiences. Am J Otol 17:835-846, 1996. 54. Althaus S R , House H P: Delayed post-stapedectomy facial paralysis: A report of five cases. Laryngoscope 83:1234-1240, 1973. 55. Matthisen J: Phonophobia after stapedectomy. Acta Otolaryngol (Stockh) 68:73-77, 1969. 56. Bracewell A : Diplacusis binauralis—a complication of stapedectomy. J Laryngol Otol 80:55-60, 1966. 57. Cody T, Baker H : Otosclerosis: Vestibular symptoms and sensorineural hearing loss. Ann Otol Rhinol Laryngol 87:778-796, 1978. 58. Das TK, Susheela A K, Gupto I P, et al: Toxic effects of chronic fluoride ingestion on the upper gastrointestinal tract. J Clin Gastroenterol 18:194-199, 1994. 59. Cole J M, Bartels L J, Beresny G M : Long-term effect of otosclerosis on bone conduction. Laryngoscope 89: 1053-1060, 1979. 60. Vartiainen E, Virtaniemi J, Kemppainen M, et al: Hearing levels of patients with otosclerosis ten years after stapedectomy. Otolaryngol Head Neck Surg 108:251-255, 1993. 61. Linthicum FH : Correlations of sensorineural hearing impairment and otosclerosis. Ann Otol Rhinol Laryngol 75:512-524, 1966. 62. Balle V, Linthicum FH Jr: Proven cochlear otosclerosis: Sensorineural without conductive hearing loss. Ann Otol Rhinol Laryngol 93:105-111, 1984. 63. Causse J R , Uriel J, Berges J, et al: The enzymatic mechanism of the otospongiotic disease and NaF action on the enzymatic balance. Am J Otol 3:297-314, 1982. 64. Causse J R , Causse J B, Uriel J, et al: Sodium fluoride therapy. Am J Otol 14:482-490, 1993. 65. Forquer B D, Linthicum FH, Bennett C : Sodium fluoride: Effectiveness of treatment for cochlear otosclerosis. Am J Otol 7:121-125, 1986. 66. Bretlau P, Causse J, Causse J B, et al: Otospongiosis and sodium fluoride: A blind experimental and clinical evaluation of the effect of sodium fluoride treatment in patients with otospongiosis. Ann Otol Rhinol Laryngol 94:103107, 1985. 67. Bretlau P, Salomon G, Johnsen NJ, et al: Otospongiosis and sodium fluoride: A clinical double-blind, placebocontrolled study of sodium fluoride in otospongiosis. Am J Otol 10:2-20, 1989. 68. Shambaugh G E Jr, Scott A : Sodium fluoride for arrest of otosclerosis. Arch Otolaryngol 80:263-270, 1964. 69. Linthicum FH, House HP, Althaus SR: The effect of sodium fluoride on otosclerotic activity as determined by strontium 85. Ann Otol Rhinol Laryngol 82:609-613, 1973.
70. Pedersen C B, Felding JU: Stapes surgery: Complications and airway infection. Ann Otol Rhinol Laryngol 100:607-611, 1991. 71. Sheehy J L , Nelson R A, House H P: Revision stapedectomy: A review of 258 cases. Laryngoscope 91:43-51, 1981. 72. Feldman BA, Schuknecht HF: Experiences with revision stapedectomy procedures. Laryngoscope 80:1281-1291, 1970. 73. Glasscock M E, McKennan K X, Levine SC : Revision stapedectomy surgery. Otolaryngol Head Neck Surg 96:141-148, 1987. 74. Marshall A H, Fanning N, Symons S, et al: Cochlear implantation in cochlear otosclerosis. Laryngoscope 115:1728-1733, 2005. 75. Pearman K, Dawes J D K : Post-stapedectomy conductive deafness and results of revision surgery. J Laryngol Otol 96:405-410, 1982. 76. Farrior J, Sutherland A : Revision stapes surgery. Laryngoscope 101:1155-1161, 1991. 77. McGee TM : The loose-wire syndrome. Laryngoscope 91:1478-1483, 1981. 78. Moon C N: Perilymphatic fistulas complicating the stapedectomy operation: A review of 49 cases. Laryngoscope 80:515-535, 1970. 79. Lippy WH, Schuring AG: Stapedectomy revision following sensorineural hearing loss. Otolaryngol Head Neck Surg 92:580-582, 1984. 80. Fenton J E, Turner J, Shirazi A, et al: Post-stapedectomy reparative granuloma: A misnomer. J Laryngol Otol 110:185-188, 1996. 81. Goebel J A, Jacob A : Use of Mimix hydroxyapatite bone cement in ossicular reconstruction. Otolaryngol Head Neck Surg 132:727-734, 2005. 82. Seicshnaydre M A, Sismanis A, Hughes G B : Update of reparative granuloma: Survey of the American Otological Society and the American Neurotology Society. Am J Otol 15:155-160, 1994. 83. Von Haacke N P: Cholesteatoma following stapedectomy. J Laryngol Otol 101:708-710, 1987. 84. Benecke J E, Gadre A K, Linthicum FH: Chondrogenic potential of tragal perichondrium: A cause of hearing loss following stapedectomy. Laryngoscope 100:1292-1293, 1990. 85. Pedersen C B : Revision surgery in otosclerosis: Operative findings in 186 patients. Clin Otolaryngol 19:446-450, 1994. 86. Pedersen C B : Revision surgery in otosclerosis—an investigation of the factors which influence the hearing result. Clin Otolaryngol 21:385-388, 1996. 87. Prasad S, Kamerer D B : Results of revision stapedectomy for conductive hearing loss. Otolaryngol Head Neck Surg 109:742-747, 1993. 88. Crabtree J A, Britton B H, Powers WH : An evaluation of revision stapes surgery. Laryngoscope 90:224-229, 1980. 89. Wiet R J, Kubek DC, Lemberg P, et al: A meta-analysis review of revision stapes surgery with argon laser: Effectiveness and safety. Am J Otol 18:166-171, 1997. 90. Hammerschlag PE, Fishman A, Scheer A A : A review of 308 cases of revision stapedectomy. Laryngoscope 108:1794-1800, 1998.
27
Perilymphatic Fistula George T. Singleton and Teresa M. Lo
Perilymphatic fistula (PLF) is a condition resulting from an abnormal connection between inner ear perilymph and the middle ear space. Its etiology has been described as being congenital, acquired, or idiopathic. Congenital middle or inner ear defects are believed to play a role in the development of PLF. PLFs exist in association with stapedectomies and other invasive procedures of the cochlea. Likewise, severe head injury, abdominal blows, and rapid shifts in environmental pressure are accepted causes of PLF. Acute and chronic mastoiditis with erosion into the labyrinth, and chronic granulomatous diseases, such as syphilis and tuberculosis, have also led to the development of PLF. In the absence of a clear antecedent traumatic event, idiopathic PLF remains a diagnostic and treatment challenge. Iatrogenic poststapedectomy PLFs were first described in the mid-1960s. Steffen and colleagues1 reported findings of gross perilymph flow at the oval window in poststapedectomy patients with hearing loss, tinnitus, and vertigo. Fee2 reported three patients who presented with vertigo, fluctuating hearing loss, and tinnitus, who also had known or suspected recent head trauma. Intraoperative findings showed perilymph leak at the oval window. Repair of the leak resulted in significant improvement in symptoms. In 1971, Goodhill3 coined the terms implosive (Valsalva-induced) and explosive (increased intracranial pressure) to describe pressure changes that can result in PLF. In reference to implosive, Goodhill stated that increased negative pressure from the tubotympanic region can be directed via the ossicles to the perilymphatic space. These external forces can result in a tear in the oval or round window membranes, driving air into the inner ear and displacing perilymph into the middle ear. The explosive route results from increased intracranial pressure possibly secondary to a severe blow to the head or abdomen. Increased intra-abdominal pressure transmits via the vertebral veins to increase the cerebrospinal pressure with resultant increase in intracranial pressure. Increased intracranial pressure can be transmitted to the inner ear via the cochlear aqueduct or internal auditory
canal, potentially causing subsequent rupture, from inside out, of the oval or round window, or both.3 Homeostasis of the pressure differentials between endolymph and perilymph is maintained by patency of the endolymphatic duct and sac. If this pressure/volume balance is disrupted, as might occur in the case of a PLF, endolymphatic hydrops may result in damage to the hearing apparatus. Animal models with experimental PLF have shown hydrops, usually resolving within 3 weeks.4,5 The cochlear aqueduct provides an important communication between the subarachnoid and perilymphatic space. The only outlet from increased intracochlear pressure is via a tear of the oval or round window membranes. Tears of the oval window annular ligament occur most frequently anteroinferiorly, rarely superiorly around the stapes footplate. The round window membrane tears less frequently.6 Tears of the round window membrane have occurred when its position was 45 degrees to the promontory, and there was little or no overhanging promontory, allowing direct visualization of the round window membrane transtympanically (Fig. 27-1B).7,8 The cribriform areas at the depths of the internal auditory canal are another area of potential transmission of increased cerebrospinal fluid pressure to the perilymphatic space. Rarely, these sites of communication are widely open, allowing direct connection of cerebrospinal fluid to perilymph, such as in Mondini’s malformation. Cerebrospinal fluid gushers can result in these patients after any invasive operative procedure of the cochlea. Microfissures of the otic capsule in the area of the oval window, round window niche, and posterior canal ampulla have been previously theorized to lead to PLF. This theory was based on temporal bone studies by Kohut and colleagues,9 who noted “loose fibrous composition in the fissula ante fenestra” in patients with PLF. A followup study by El Shazly and Linthicum10 showed that these microfissures are commonly present in temporal bones. No association between the presence of microfissures and sudden sensorineural hearing loss was found, and no evidence of PLF was noted. 323
324
OTOLOGIC SURGERY
FIGURE 27-1. A, Right middle ear: surgeon’s view with tympanomeatal flap and drum folded forward. Sites of oval window fistula are numbered in order of occurrence (1 and 2 most common). B, Abnormal round window (RW) membrane at 45 degree angle to promontory and no overhanging lip. C, Normal RW membrane hidden from view in depths of niche, with lip of bone overhang. Mucosal folds frequently appear to seal niche partially, with a hole in the center. EAC, external auditory canal; PLF, perilymphatic fistula; TM, tympanic membrane.
Chapter 27 • Perilymphatic Fistula
Animal studies have tried to examine the pathophysiologic mechanism contributing to the hearing loss noted in PLF patients. Weisskopf and colleagues11 generated perforations in the round window membrane of guinea pigs. The resultant hearing loss recovered with time, and the authors concluded that perforations alone did not explain the degree of hearing loss present in PLF. Some research has shown pneumolabyrinth to cause sizable auditory threshold shifts, with removal of the air bubble from the cochlea resulting in resolution of hearing loss.12 Early stapedectomy procedures were more likely to result in PLF than methods practiced today. The entire stapes and footplate were often removed and replaced with a pointed polyethylene strut or wire loop prosthesis placed atop a thin piece of absorbable gelatin sponge (Gelfoam) or tissue seal. The wire could migrate, and the thin tissue membrane could rupture, both increasing chances of PLF. The current small fenestra, small piston, tissue seal techniques have dramatically reduced the incidence of PLF.13-17 Lesinski18 prospectively reviewed 19 cases of revision stapedectomy performed for symptoms of possible oval window fistula. The 14 of 19 patients who were noted to have an active PLF had undergone prior complete stapedectomy with migration of the prosthesis eccentrically, close to the edge of the oval window. Invasive procedures of the stapes footplate, including the Fick or Cody tack procedure for Meniere’s disease (sacculotomy) were ultimately abandoned because of the high incidence of sensorineural hearing loss. Later exploration of many of these ears revealed a persistent PLF.19 Likewise, the 20% sensorineural hearing loss associated with cochleostomy or cochleosacculotomy, as it was originally described, was probably related to a persistent round window PLF.19 As awareness of PLFs increased, the number of diagnoses and surgical explorations for PLF did as well. Lower rates of positive exploration and of successful repair were being reported, however. In many cases of re-exploration for possible persistent PLF, prior operative records had described the presence of a hole in the round window membrane. Re-exploration findings showed a normalappearing round window membrane deep in the niche (Fig. 27-1C). It is likely the prior surgeon had sealed the mucosal folds surrounding the round window niche. Similarities in symptoms may have led to the misdiagnosis of Meniere’s disease or superior semicircular canal dehiscence as PLF. Attempts at identifying predictive tests, such as glycerin challenge testing or electrocochleography, to differentiate Meniere’s disease from PLF have been unsuccessful.20-23 The classic symptoms of Meniere’s disease are episodic vertigo, fluctuant hearing loss, tinnitus, and, in 80%, a sense of fullness in the involved ear. Patching of windows in these cases with no visible leak would give the same result as allowing time to pass. In Minor’s24 review of 65 patients with superior semicircular canal dehiscence, 54 (83%) had vestibular
325
symptoms elicited by loud sounds, and 44 (67%) had pressure-induced (sneezing, coughing, and straining) symptoms. In addition, a 10 dB or greater conductive hearing loss was present in 70%. (See Chapter 42 for further discussion.) Previous studies of children with sensorineural hearing loss of unknown etiology showed a rate of 6% to 16% positive PLFs on exploration.25-27 In the largest prospective series by Reilly25 of 244 children with sensorineural hearing loss, 17% underwent exploration, all of whom had prior abnormal computed tomography (CT) scan findings. Of the children who underwent exploration, 26% had active congenital PLF. A more recently published retrospective series reported 64% of ears with suspected PLF to have a leak confirmed visually.28 Middle ear malformations have been noted in 81% of PLF patients at the time of surgery, most commonly an anomalous stapes.29
PATIENT SELECTION Patient presentation is variable with PLF. Symptoms have been reported to be present from days to decades. The typical patient with a PLF presents with a sudden onset of hearing loss or mild vertigo or dysequilibrium, or both hearing loss and vertigo, associated with a traumatic event. Trauma includes invasive middle or inner ear surgical procedures, abdominal or head blows, blast injuries, or severe changes in environmental pressure, particularly in the presence of an upper respiratory infection or an acute allergic attack. In some series, one third to one half of patients recall no potential triggering event.30,31 In cases of surgically proven PLF, patients present with some degree of vestibular and auditory disturbances. Patients reported vestibular complaints including episodic or positional vertigo, dysequilibrium, lightheadedness, and motion intolerance 46% to 91% of the time.31-35 A chief complaint of hearing loss was present in 28% to 93% of patients.31-35 Although only 28% reported noting hearing loss in the Iowa series, 54% had abnormal audiograms.32 Hearing losses are most commonly reported to be sudden or rapidly progressive; however, in some series they were also noted to be fluctuant, raising debate as to whether the disease was actually Meniere’s disease. Tinnitus has been reported as a chief complaint in 25% to 76% of patients with PLFs, always in combination with vestibular or auditory symptoms or both.31,32,35,36 In some series, no patients reported tinnitus as a presenting symptom.34
PREOPERATIVE EVALUATION Vestibular signs are commonly present in PLF. Positional nystagmus may have a very short or no latency and relatively long duration. Repeated testing shows nystagmus
326
OTOLOGIC SURGERY
with minimal or no fatigue, no direction reversal when changing from inducing position to sitting position, and less violent than seen with benign paroxysmal positional vertigo. The nystagmus is rarely rotatory. From most to least frequent, the nystagmus may be horizontal, diagonal, or vertical.27,37 The direction of the nystagmus is of no diagnostic value in determining which ear is involved. Hearing loss is typically sensorineural, without a specific pattern. Conductive hearing loss is rarely observed. This hearing loss has been attributed to a slipped stapes prosthesis or possible presence of air within the labyrinth. The speech reception threshold is usually worse than anticipated from the pure tone average, and the discrimination score is usually lower than expected.38 Most series report usage of a fistula test by applying positive or negative pressure to an intact tympanic membrane with a pneumatic otoscope. Resultant nystagmus indicates a positive test and is present only about 25% of the time.45 A sensation of dysequilibrium suggests a positive test. Modifications, such as administration of the test with the patient standing and eyes closed30,39 or on a moving posture platform (platform fistula test), have been used to try to improve test sensitivity. Electronystagmography has also been performed concurrently. Several studies noted no significant difference with electronystagmography-enhanced fistula testing.34,41 High positive exploration rates have been reported in cases using the moving platform fistula test; however, these findings have not been supported by others.42,43 The Quix test involves having the patient stand erect with feet together, eyes closed, and arms outstretched. The examiner looks for deviation of the arms to the side of the lesion. The result is positive in only about 20% of cases.44 The eyes-closed-turning test has shown high sensitivity and specificity for PLF. It is performed by having the patient walk in a straight line with eyes closed. The examiner taps the subject’s shoulder, indicating to the patient to turn 180 degrees either right or left and stop in a position of attention with the eyes still closed (Fig. 27-2). A positive test is readily recognized by the patient’s swaying or having a tendency to lose balance when he or she has turned to the side of the lesion.45,46 Electrocochleography has been used in attempts at improving diagnostic ability; however, results have shown limited clinical utility. Because of pathophysiologic overlap with other conditions such as Meniere’s disease, electrocochleography lacks sensitivity in identifying PLF.21,47 Radiographic assessment has not proven to be helpful in diagnosis of PLF because adequate evaluation of the oval and round window regions is difficult. In patients with confirmed PLF, congenital anomalies of the middle or inner ear or both were able to be visualized on highresolution CT only 50% of the time.48 Rigid and flexible endoscopy via myringotomy have been proposed to help in middle ear visualization before
definitive exploration. Some authors report adequate visualization, whereas others have reported limited view of the round and oval window niches, often with mucosal adhesions obscuring the site.49-51 Negative findings were not followed up with formal exploration. Visualization of clear fluid in the middle ear does not confirm presence of PLF. Injected local anesthetic pooling in the middle ear can confuse the picture. The volume of perilymph has been reported to be approximately 75 μL; visualization of such small volume can be difficult. Large fluid volume pooling in the middle ear raises the question of cerebrospinal fluid leak. Provocative testing, such as Trendelenburg position, Valsalva maneuver, and jugular vein compression, and application of mineral oil in the middle ear have been attempted; the usefulness of these techniques has not been proven.52,53 Numerous objective assays have been attempted to improve identification of perilymph in the middle ear. Technetium-labeled albumin was tried early on, but the albumin did not migrate into the perilymph. Intrathecal and intravenous fluorescein administration has not been shown to be useful given the potential neurologic complications and uncertainty of accurate perilymph labeling.54,55 Glucose measurement of approximately 100 mg/dL of middle ear fluid, in the absence of blood product, has been used to help identify perilymph. Silverstein56 reported on the use of indicator paper to identify protein concentration in middle ear fluid. The color change and correlation with known differing protein concentrations in perilymph, cerebrospinal fluid, and serum were used to determine presence of perilymph. Possible errors in identification could be due to local anesthetic dilution. Protein markers such as β2-transferrin57,58 and beta-trace protein,59 which are present in cerebrospinal fluid and perilymph, have also been investigated as possible diagnostic markers for PLF. Measurements of these markers in perilymph have been unreliable, owing to difficulty with collection methods and small sample volumes. In addition, processing time precludes use of these assays for intraoperative decision making.
PATIENT COUNSELING Most clinicians consider PLF to be a surgically correctable problem, particularly in cases in which antecedent surgery or trauma is present. Conservative or medical therapy has been advocated in patients seen soon after presentation with stable hearing. Recommendations are for bed rest for 5 to 7 days, elevation of the head, stool softeners, avoidance of Valsalva maneuvers, and serial audiometric evaluation. If hearing deteriorates, or the vestibular symptoms fail to abate, surgical intervention should be considered. If hearing loss is present, waiting more than 2 weeks from the onset of PLF significantly reduces the likelihood of improving hearing with operative intervention.
Chapter 27 • Perilymphatic Fistula
327
FIGURE 27-2. Eyes-closed-turning test. The patient walks with eyes closed, turns quickly right or left, then stops in position of attention with eyes still closed. A positive test result consists of staggering or swaying on turning to the involved side.
The goal of surgical intervention is to cover the leak with autologous tissue, such as fat, fascia, perichondrium, or areolar tissue. As previously discussed, however, identification of the fistula site can be a challenge. Patients are recommended to undergo the procedure under local anesthesia with monitored anesthetic care in an ambulatory surgical setting. If there are concerns of claustrophobia, anxiety, or difficulty with positioning, a general anesthetic should be considered. No preoperative laboratory work or testing is required for healthy young adults; older patients and patients with systemic diseases should undergo testing as relevant to their respective medical comorbidities. Patients are advised to discontinue aspirin and nonsteroidal analgesic therapy 10 to 14 days before surgery. For anticoagulants such as warfarin or clopidogrel, clearance should be obtained from the prescribing physician to discontinue in an appropriate time frame. Patients are instructed to wash their hair the night before surgery. Patients are counseled regarding possibility of a small postauricular or tragal incision if graft material needs to be harvested. If a leak is found intraoperatively, patients require bed rest for 5 days with bathroom privileges, and restrictions of no heavy lifting, straining, and sexual activity until at least 2 weeks postoperatively.
SURGICAL PROCEDURE Preoperative Preparation Patients have an intravenous line started in the preoperative holding area. They are asked to go to the bathroom immediately before coming back to the operative suite. The operative ear is confirmed with the patient and marked accordingly. If unusually nervous, patients may receive a small amount of midazolam before coming back
to the operative suite. Dexamethasone (Decadron) can be administered at the beginning of the case to help alleviate potential postoperative nausea. No routine systemic antibiotic therapy is administered. Patients are placed backward on a standard operating table, allowing room for the surgeon’s legs. The anesthesiologist places the electrocardiographic monitors and pulse oximetry on the patient. An automatic blood pressure cuff is placed on the arm opposite the operative ear to prevent accidental movement of the surgeon’s arm during insufflation. Any loose hair from around the ear is taped back with wide paper adhesive tape. No shaving is usually necessary. The patient’s head is positioned with the nonoperative ear turned toward the shoulder and gently secured with paper adhesive tape, followed by silk adhesive tape for reinforcement. A sedative-hypnotic agent is administered by the anesthesiologist (e.g., methohexital, propofol). Supplemental oxygen is administered via nasal cannula or mask as needed. Local anesthetic (1% lidocaine with 1:100,000 epinephrine) is administered via a ring block completely around the operative ear. The operating microscope is brought into position, the ear canal is cleaned, and fourquadrant injections are performed. Care is taken to produce no excessive injection or blistering of the ear canal skin. The inferior injection is placed posteroinferiorly at the bony cartilaginous junction, with care taken to place the bevel of the needle against bone and under the periosteum. Administering the injection is a slow, deliberate process, with the surgeon watching carefully for blanching, and ensuring that the fluid diffuses all the way to the annulus inferiorly. All desquamated epithelium from the ear canal is removed with a small suction. The operative microscope and all instrumentation are removed from the patient to allow for sterile preparation of the patient and operative site. An “ether screen”
328
OTOLOGIC SURGERY
Chapter 27 • Perilymphatic Fistula
329
FIGURE 27-3. Details of oval window fistula sites in order of occurrence. Sites are in the annular ligament site. A, Pants graft of areolar tissue 3 × 1.5 mm with legs 2 mm long. B, Pants graft in place over denuded mucosa around perilymphatic fistula (PLF).
FIGURE 27-4. Congenital defects in posterior footplate.
is secured to the table, opposite the operative site, to keep the drapes from resting on the patient’s face. The operative site is painted with povidone-iodine (Betadine) solution and blotted from around the ear. A 3M adhesive drape with a 2 inch hole is placed over the ear. The drape is precut so that it does not fall over the patient’s mouth and nose. A disposable, lint-free, paper ear-draping pack is used to cover the patient and the remainder of the operative field. A small hole is cut near the patient’s upper face so that he or she can see the anesthesiologist’s or nurse’s face.
Surgical Instruments A standard tympanostomy setup is used. Disposable straight and angled Beaver ear blades are used for flap incisions. A skeeter microdrill with 1 to 1.4 mm diamond burrs is used for removing the scutum. A sharp 0.5 mm right angle pick is altered slightly by bending its shaft 20 degrees, 2 inches from the end to allow better visualization of the tip. A straight pick with an angled handle is used for work around the stapes footplate and to place grafts.
Surgical Technique The sterile-draped operative microscope is brought into the field. Ear canal injections are repeated as necessary. The ear canal is copiously irrigated with normal saline to remove povidone-iodine. The inferior incision is made first by use of the No. 1 straight Beaver blade. A cut is made from the 6 o’clock position and is angled to about 1 cm lateral to the annular ring on the posterior ear canal. The No. 2 angled Beaver blade is used to make an incision superiorly from 2 mm lateral to the short process of the malleus to join the tip of the other incision in the posterior ear canal. A duckbill elevator is used to elevate the skin of the posterior canal down to the annulus. A House No. 2 knife can be used to elevate fibers adherent in the suture line. Hemostasis must be achieved before entering the middle ear and can be completed by using a 20 gauge suction placed on the bleeder and low-setting cautery. The middle ear is opened at the notch of Rivinus by use of a Rosen needle. The chorda tympani nerve is identified, and the beginning of the annular ring is raised with the Rosen needle or drum elevator. The chorda tympani is gingerly dissected free of the tympanic membrane, and the posterior half of the tympanic membrane and ear canal flap are folded forward so that the middle ear may be inspected (see Fig. 27-1A).
Frequently, clear fluid is seen in the oval window recess and in the round window niche. The fluid is usually local anesthetic that has seeped into the middle ear space, and a 24 gauge suction should be used to remove it. The patient is asked to perform a Valsalva maneuver to observe if fluid reaccumulates. If the fluid reaccumulates, and it is unclear whether a fistula is present, the fluid may be quickly checked to try to determine if it is perilymph. A small piece of Gelfoam can be placed on an angled straight pick to soak up the solution, taking care to avoid any blood product. The Gelfoam is placed on a Clinistrip for glucose testing. If the reading is 100 mg/dL, this is most consistent with perilymph. Another method is to measure protein concentration in the perilymph using the Xomed Treace protein-measuring kit. A micropipette is used to obtain fluid, which is placed on indicator paper. The color change is compared with a standard, which can help to identify the presence of perilymph. No bony work or mucosal manipulation (other than initial suctioning of the middle ear) has been performed to this point. The scutum has been left intact, and the round window niche and lip of the promontory have not been disturbed. If the round window membrane is immediately visible when the middle ear is opened—that is, at about a 45 degree angle to the plane of the promontory—the surgeon should become suspicious of a probable PLF in the round window membrane. After checking for the recurrence of fluid in these recesses, the surgeon removes the scutum and gets complete exposure of the oval window area. If no fluid was noted in the oval window region earlier with Valsalva maneuver, a straight pick can be placed on the lenticular process of the incus and pressed gently, looking for accumulation of fluid around the annular ligament or in the round window niche. If no fluid accumulates in either place with Valsalva maneuver and pressure on the stapes, no repair is performed. We believe there is a risk for creating some degree of conductive hearing loss and the possibility of injuring the inner ear by patching the oval or round window with no PLF. Other authors advocate patching the oval and round windows without obvious leak present. Most PLFs at the oval window are located directly anterior to the anterior crus or immediately below it; a few are superior to the anterior crus (Fig. 27-3). Generally, there is good visualization of this region. Leaks in this area are best repaired by teasing away the surface mucosa either with a straight pick or with a small right angle pick. A graft of adventitia is obtained from over the temporalis fascia. This graft is compressed and cut into the shape of a small set of trousers, approximately
330
OTOLOGIC SURGERY
3 mm long and 1.5 mm wide, and a 2 mm slit is created up the middle to form the pants’ legs (see Fig. 27-3A). The graft is draped around the anterior crus and packed in place with Gelfoam (see Fig. 27-3B) soaked in Ringer solution. Some authors have advocated adding fibrin glue to the graft for increased stability. Gelfoam is packed to the level of the tympanic membrane followed by a sheet of absorbable gelatin film (Gelfilm) to prevent adhesions to the tympanic membrane. If there is a congenital defect of the stapes footplate, it may be present in the middle of the footplate, or may incorporate the entire posterior half of the footplate (Fig. 27-4). In these cases, the mucosa must be denuded completely around the footplate. With larger perforations, perichondrium from the tragus is used because it is thicker and easier to handle to effect a seal. The area between the crura is packed full with Gelfoam to hold the graft in place. If there is a round window PLF, the membrane is in clear sight with no overhang of the round window niche. Generally, the tear is readily visible around the annular ring of the membrane. Occasionally, it is in the center, particularly if the injury was created by a myringotomy knife or foreign body. If the fistula is less than 2 weeks old, a fibrin clot or granulation tissue may be seen around the leak. The area surrounding the perforation is roughened with a small right angle pick. A thicker graft of tragal perichondrium is used. The graft is held in position by packing with Gelfoam and placement of Gelfilm to prevent formation of adhesive bands. The tympanic membrane is replaced to its native position, and the skin flap in the ear canal is laid flat. Gelfoam moistened in antibiotic solution is placed over the tympanic membrane and skin flap to fill the ear canal, and a cotton ball is placed. The postauricular and tragal graft harvest sites are closed with absorbable sutures.
Dressing and Postoperative Care Patients are returned to the recovery room in a semisitting position. If a PLF was identified, patients are counseled regarding the need for bed rest for 5 days with bathroom privileges only. They are to sleep either in a recliner chair or in a bed with the head elevated 4 to 6 inches. Patients are advised that when they rise from bed, they are to roll on to their side and push up with their arms to avoid tensing abdominal musculature. They are to use ototopical antibiotics, 5 drops twice daily, until their postoperative appointment in 2 weeks. The cotton ball should be changed as necessary. Patients are placed on postoperative stool softeners for 2 weeks and sedatives as necessary. After 5 days of bed rest, patients should avoid heavy lifting, straining, and sexual activity until 2 weeks postoperatively. At 2 weeks, they may return to sedentary work. Avoiding heavy manual labor is recommended, however. The patient is advised to sneeze with the mouth open and to avoid blowing the nose for 2 weeks.
Pitfalls There are two potential pitfalls of PLF surgery: (1) inability to obtain adequate hemostasis and anesthesia, particularly in the lower portion of the flap, and (2) inability to obtain adequate visualization of the oval window area without removing the scutum. Some patients also experience discomfort when the surgeon denudes the area around the stapes or over the round window membrane in preparation for graft placement. We avoid placing 4% lidocaine in the middle ear, as is often done for stapedectomy or singular neurectomy. If lidocaine enters the inner ear, the patient becomes violently dizzy and may completely lose hearing in the ear for a short time. Adhesive band formation between the tympanic membrane and round window graft should be avoided if possible. We advise placement of Gelfilm just medial to the tympanic membrane to help prevent this. If adhesions do form, the patient experiences an apparent sensorineural hearing loss combined with a conductive loss, both of which clear when the adhesion is lysed.
RESULTS Most series report subjective improvements in vestibular and hearing symptoms, noting better response of vestibular disturbances to PLF surgery. After repair of PLF, 72% to 100% of patients report significant improvement of vertigo or dysequilibrium.31,32,34,40,60 Black and colleagues60 noted objective improvement in dynamic posturography after PLF surgery, with 12 of 32 patients having normal tests after PLF repair. Hearing outcomes after PLF surgery have not shown as much improvement. Hearing improvement has been reported in 13% to 49% of PLF repairs,32,34,40 and stabilization of hearing has been reported in 40% to 67%.33,35,60 Deterioration of hearing has been noted in 11% to 17% of PLF repairs.31,32,40 Negative exploration rates of 40% to 50% for suspected PLFs have been reported. Management of these suspected PLFs differ; some surgeons place grafts in the oval and round windows,34,35 whereas others avoid patching if no leak is visualized. Fitzgerald and associates35 reported a negative exploration in 78% in their series of 197 patients. They patched the oval and round windows in all cases, and noted positive outcomes in 86% of their patients, leading them to define a proven PLF based on a positive surgical outcome, as opposed to positive intraoperative findings. Hughes and colleagues61 surveyed 167 otologists in the United States, and noted 78% would place grafts on the oval and round windows at the time of exploration. Advocates of patching both windows in cases of negative exploration cite the possibility of minute62 or intermittent33 leaks. Overall improvement has been reported to
Chapter 27 • Perilymphatic Fistula
range from 29% to 44% of negative explorations with prophylactic patching.33,34 The possibility of placebo effect of PLF surgery has also been considered. Tissue allografts to patch the oval and round windows have included fat, fascia, perichondrium, and areolar tissue. Multiple authors recommend avoiding use of fat because it has been noted to have increased failure rate.32 Graft sites have also been covered with fibrin glue in hopes of improving postoperative outcome.31,35,63
RECURRENT PERILYMPHATIC FISTULA If instability and mild vertigo persist after the repair has healed, and if the turning or fistula test results remain positive, re-exploration should be considered. Six weeks should be allowed for the wound to heal completely. Failure occurs most frequently in cases with deep recesses of the oval window, where good access to the anterior crus cannot be obtained to denude the bed adequately and pack the graft tightly into the fistula tract. Surgeons should attempt re-repair of an oval window PLF at least three times before considering doing a stapedectomy to close the fistula. PLF ears seem more sensitive than others to stapedectomy. High-frequency sensorineural hearing loss has occurred commonly when stapedectomy has been done for a PLF. Some patients have postoperative improvement with the balance disturbance and motion intolerance symptoms typical of PLF, but develop episodic whirling vertigo typical of Meniere’s disease. Potter and Conner64 raised the issue of possible endolymphatic hydrops secondary to membranous labyrinthine damage caused by PLF. Damage to the membranous labyrinth resulting in hydrops has been theorized to be a result of the initial trauma causing the PLF, the PLF leak itself causing stresses on the membranous labyrinth, or from trauma during surgical repair of the PLF.32 It is far more likely this group of patients had Meniere’s disease to begin with, not PLF. Lollis and coworkers65 reported a retrospective review of a small series of patients with PLF refractory to otologic surgical management. For patients who showed improvement of their vertiginous symptoms, a ventriculoperitoneal shunt placement was performed. Subjective improvement in symptoms was reported in all patients so treated. No preoperative and postoperative objective measures were available for comparison in this study, however.
SUMMARY PLF remains a diagnostic and therapeutic challenge. Diagnosis relies on a combination of history-taking, physical examination, and objective audio and vestibular testing. Ultimately, surgical exploration is the gold standard in confirming the presence of a fistula. Patients must
331
be counseled regarding the controversies surrounding the need for and extent of surgical intervention. Alternative diagnoses must also be considered.
REFERENCES 1. Steffen TN, Sheehy J L , House H P: The slipped strut problem. Ann Otol Rhinol Laryngol 72:191-205, 1963. 2. Fee G A : Traumatic perilymphatic fistulas. Arch Otolaryngol 88:477-480, 1968. 3. Goodhill V: Sudden deafness and round window rupture. Laryngoscope 81:1462-1474, 1971. 4. Nomura Y, Hara M, Funai H, Okuno T: Endolymphatic hydrops in perilymphatic fistula. Acta Otolaryngol 103:469-476, 1987. 5. Nomura Y, Hara M, Young YH, Okuno T: Inner ear morphology of experimental perilymphatic fistula. Am J Otol 13:32-37, 1992. 6. Singleton GT: Perilymph fistula. In Sharpe J A, Barber HO (eds): The Vestibulo-Ocular Reflex and Vertigo. New York, Raven Press, 1993. 7. Pullen FW: Round window membrane rupture: A cause of sudden deafness. Trans Am Acad Ophthalmol Otolaryngol 76:1444-1450, 1972. 8. Rybak L P: “How I do it”—otology and neurotology. L aryngoscope 90:2049-2050, 1980. 9. Kohut R I, Hinojosa R , Ryu J: Sudden-onset hearing loss in eleven consecutive cases: A temporal bone histopathologic study with identification of perilymphatic fistulae. Trans Am Otol Soc 77:81-88, 1989. 10. El Shazly M A, Linthicum FH : Microfissures of the temporal bone: Do they have any clinical significance? Am J Otol 12:169-171, 1991. 11. Weisskopf A, Murphy JT, Merzenich M M : Genesis of the round window rupture syndrome: Some experimental observations. Laryngoscope 88:389-397, 1978. 12. Nishioka I, Yanagihara N: Role of air bubbles in the perilymph as a cause of sudden deafness. Am J Otol 7:430438, 1986. 13. Douek E : Perilymph fistula. J Laryngol Otol 89:123-130, 1975. 14. Goodhill V: Stapedectomy revision commandments: Posterior arch stapedioplasty. Trans Pac Coast Otoophthalmol Soc 55:35-59, 1974. 15. Harrison WH, Shambaugh G E, Derlacki E L , et al: Perilymph fistula in stapes surgery. Laryngoscope 77:736849, 1967. 16. Hemenway WG, Hildyard VH, Black FO: Post stapedectomy perilymph fistulas in the Rocky Mountain area: The importance of nystagmography and audiometry in diagnosis and early tympanotomy in prognosis. Laryngoscope 78:1687-1715, 1968. 17. House H P: The fistula problem in otosclerotic surgery. Laryngoscope 77:1410-1426, 1967. 18. Lesinski SG: Causes of conductive hearing loss after stapedectomy or stapedotomy: A prospective study of 279 consecutive surgical revisions. Otol Neurotol 23:281-288, 2002. 19. Singleton GT: Perilymph fistulas. Adv Otolaryngol Head Neck Surg 2:25-38, 1988.
332
OTOLOGIC SURGERY
20. Lehrer J F, Poole DC, Sigal B : Use of the glycerin test in the diagnosis of post-traumatic perilymphatic fistulas. Am J Otolaryngol 1:207-210, 1980. 21. Arenberg I K, Ackley R S, Ferraro J, Muchnik C : ECoG results in perilymphatic fistula: Clinical and experimental studies. Otolaryngol Head Neck Surg 99:435-443, 1988. 22. Meyerhoff WL , Yellin MW: Summating potential/action potential ratio in perilymph fistula. Otolaryngol Head Neck Surg 102:678-682, 1990. 23. Campbell KC M, Karker L A, Abbas PJ: Interpretation of electrocochleography in Ménière’s disease and normal subjects. Ann Otol Rhinol Laryngol 101:496-500, 1992. 24. Minor L B : Clinical manifestations of superior semicircular canal dehiscence. Laryngoscope 115:1717-1727, 2005. 25. Reilly J S : Congenital perilymphatic fistula: A prospective study in infants and children. Laryngoscope 99:393-397, 1989. 26. Parnes L S, McCabe B F: Perilymph fistula: An important cause of deafness and dizziness in children. Pediatrics 89:524-528, 1987. 27. Pappas DG, Simpson LC, Godwin G H : Perilymphatic fistula in children with pre-existing sensorineural hearing loss. Laryngoscope 98:507-510, 1988. 28. Weber PC, Bluestone C D, Perez B : Outcome of hearing and vertigo after surgery for congenital perilymphatic fistula in children. Am J Otol 24:138-142, 2003. 29. Weber PC, Perez B A, Bluestone C D: Congenital perilymphatic fistula and associated middle ear abnormalities. Laryngoscope 103:160-164, 1993. 30. Cole GG: Validity of spontaneous perilymphatic fistula. Am J Otol 16:815-819, 1995. 31. Goto F, Ogawa K, Kunihiro T, et al: Perilymph fistula 45 case analysis. Auris Nasus Larynx 28:29-33, 2001. 32. Seltzer S, McCabe B F: Perilymph fistula: The Iowa experience. Laryngoscope 96:37-49, 1986. 33. Shelton C, Simmons FB : Perilymph fistula: The Stanford experience. Ann Otol Rhinol Laryngol 97:105-108, 1988. 34. Rizer FM, House JW: Perilymph fistulas: The House Ear Clinic experience. Otolaryngol Head Neck Surg 104:239243, 1991. 35. Fizgerald DC, Getson P, Brasseux CO: Perilymphatic fistula: A Washington, DC experience. Ann Otol Rhinol Laryngol 106:830-837, 1997. 36. Glasscock M E III, Hart M J, Rosdeutscher J D, Bhansali S A : Traumatic perilymphatic fistula: How long can symptoms persist? A follow-up report. Am J Otol 13:333338, 1992. 37. Bluestone C D: Otitis media and congenital perilymphatic fistula as a cause of sensorineural hearing loss in children. Pediatr Infect Dis J 7:S141-S145, 1988. 38. Singleton GT: Correlation of clinical symptoms of vertigo and/or hearing loss with anatomic site of surgically confirmed PLF. In Arenberg I K (ed): Inner Ear Surgery. Proceedings of the Third International Symposium and Workshops on Surgery of the Inner Ear. Snowmass-Aspen, CO, 1990. New York, Kugler, 1991, pp 395-397. 39. Kohut R I : Perilymph fistulas: Clinical criteria. Arch Otolaryngol Head Neck Surg 118:687-692, 1992.
40. Healy G B, Strong M S, Sampgna D: Ataxia, vertigo, and hearing loss: A result of rupture of inner ear window. Arch Otolaryngol 100:130-135, 1974. 41. Podoshin L , Fradis M, Ben-David J, et al: Perilymphatic fistula—the value of diagnostic tests. J Laryngol Otol 108:560-563, 1994. 42. Black FO, Lilly DJ, Nashner L M, et al: Quantitative diagnostic test for perilymph fistulas. Otolaryngol Head Neck Surg 96:125-134, 1987. 43. Meyerhoff WL : Spontaneous perilymphatic fistula: Myth or fact. Am J Otol 14:478-481, 1993. 44. Hart CW: The evaluation of vestibular function in healing and disease. In: Hoeber (ed.) Otolaryngology. Hagerstown, MD, Harper & Row, 1972, pp 1-63. 45. Singleton GT, Karlan M S, Post K N, et al: Perilymph fistulas: Diagnostic criteria and therapy. Ann Otol Rhinol Laryngol 87:797-803, 1978. 46. Singleton GT: Perilymph fistula. In Sharpe J A, Barber HO (eds): The Vestibulo-Ocular Reflex and Vertigo. New York, Raven Press, 1993. 47. Campbell KC, Parnes L : Electrocochleographic recordings in chronic and healed perilymphatic fistula. J Otolaryngol 21:213-217, 1992. 48. Weissman J L , Weber PC, Bluestone C D: Congenital perilymphatic fistula: Computed tomography appearance of middle ear and inner ear anomalies. Otolaryngol Head Neck Surg 111:243-249, 1994. 49. Rosenberg S I, Silverstein H, Willcox TO, Gordon M A : Endoscopy in otology and neurotology. Am J Otol 15:168172, 1994. 50. Poe DS, Bottrill I D: Comparison of endoscopic and surgical explorations for perilymphatic fistulas. Am J Otol 15:735-738, 1994. 51. Pyykko I, Selmani Z, Ramsay H : Middle ear imaging in neurotological work-up. Acta Otolaryngol Suppl 520:273276, 1995. 52. Althaus S R : Long-term results of perilymph fistula repair. Laryngoscope 23:1502-1507, 1973. 53. Todd NW, Jackson RT: Oil-on-water: Proposed method for intraoperative identification of perilymphatic fistula. Am J Otol 16:539-542, 1995. 54. Poe DS, Gadre A K, Rebeiz E E, Pankratov M M : Intravenous fluorescein for detection of perilymphatic fistulas. Am J Otol 14:51-55, 1993. 55. Bojrab D I, Bhansali S A : Fluorescein use in the detection of perilymphatic fistula: A study in cats. Otolaryngol Head Neck Surg 108:348-355, 1993. 56. Silverstein H : Rapid protein test for perilymph fistula. Otolaryngol Head Neck Surg 105:422-426, 1991. 57. Levenson M J, Desloge R B, Parisier SC : Beta-2 transferrin: Limitations of use as a clinical marker for perilymph. Laryngoscope 106:159-161, 1996. 58. Buchman C A, Luxford WM, Hirsch B E, et al: Beta-2 transferrin assay in the identification of perilymph. Abstr Am Otol Soc 131:40, 1998. 59. Michel O, Petereit H, Klemm E, et al: First clinical experience with beta-trace protein (prostaglandin D synthase) as a marker for perilymphatic fistula. J Laryngol Otol 119:765-769, 2005. 60. Black FO, Pesznecker S, Norton T, et al: Surgical management of perilymphatic fistulas: A Portland experience. Am J Otol 13:254-262, 1992.
Chapter 27 • Perilymphatic Fistula 61. Hughes G B, Sismanis A, House JW: Is there consensus in perilymph fistula management? Otolaryngol Head Neck Surg 102:111-117, 1990. 62. Kohut R I, Waldorf R A, Haenel J L , Thompson J N: Minute perilymph fistulas: Vertigo and Hennebert’s sign without hearing loss. Ann Otol Rhinol Laryngol 88:153159, 1979. 63. Black FO, Pesznecker S, Norton T, et al: Surgical management of perilymph fistulas: A new technique. Arch Otolaryngol 117:641-648, 1991.
333
64. Potter C R , Conner G H : Hydrops following perilymphatic fistula repair. Laryngoscope 93:810-812, 1983. 65. Lollis S S, Dudley JW, Phillips J M, Roberts DW: Ventriculoperitoneal shunt insertion for the treatment of refractory perilymphatic fistula. J Neurosurg 105:1-5, 2006.
28
Management of Bell’s Palsy and Ramsay Hunt Syndrome Bruce J. Gantz, Miriam I. Redleaf, Brian P. Perry, and Samuel P. Gubbels
Of the multitude of causes of facial paralysis, Bell’s palsy and Ramsay Hunt syndrome are two of the most common. Bell’s palsy alone accounts for almost three quarters of all acute facial palsies. Although the diagnosis of Ramsay Hunt syndrome is generally obvious, a thorough evaluation and close follow-up are required to establish a firm diagnosis of Bell’s palsy. This chapter discusses the pathology, pathophysiology, epidemiology, evaluation, and management of these common disorders.
BELL’S PALSY Bell (1774-1842)1 first described a patient with a facial paralysis in 1818; subsequently, all patients with facial palsy of unknown etiology have come to bear his name. The etiology of this “idiopathic” disorder has become much clearer in recent years. Although first proposed in 1972 by McCormick,2 herpes simplex virus (HSV) has been identified as the disease vector only recently,3-5 and an animal model has been designed.6 Murakami and associates5 identified HSV type 1 (HSV-1) DNA fragments in perineural fluid in 11 of 14 patients undergoing facial nerve decompression. In this study, no control subjects had HSV-1 DNA in perineural fluid. Using polymerase chain reaction to analyze the saliva of patients with Bell’s palsy, Furuta and colleagues3 identified HSV-1 DNA in 50% of patients, which was significantly more often than controls, a finding confirmed by other groups.7 Polymerase chain reaction has also been used to isolate HSV-1 genomic DNA from the geniculate ganglion of a temporal bone in a patient dying during the acute phase of Bell’s palsy.4 Sugita and coworkers6 proposed an animal model of Bell’s palsy. Six days after inoculation of HSV-1 into either the auricle or the tongue of mice, a temporary ipsilateral facial paralysis was identified that recovered spontaneously within 3 to 7 days. Histopathologically, neural edema, vacuolar degeneration, and inflammatory cell infiltrate with associated demyelination or axonal degeneration were shown in the affected facial nerve and nucleus.8 HSV-1 antigens were identified within the
facial nerve, geniculate ganglion, and facial nucleus 6 to 20 days after inoculation.6 Similar pathologic findings have been shown in rabbits after HSV-1 inoculation, but without the associated facial paralysis.9 The histopathologic changes seen in autopsy specimens of patients who died with acute idiopathic facial paralysis have provided some insight into the underlying cellular mechanisms in Bell’s palsy. Reddy and coworkers10 found degeneration of the myelin sheath and axons, perivascular inflammation, and a phagocytic cell infiltrate in 10% to 30% of nerve fibers in a patient 17 days after the onset of an acute idiopathic facial paralysis. Fowler11 found diffuse vascular engorgement throughout the intratemporal facial nerve and evidence of acute hemorrhage within the intracanalicular, labyrinthine, and geniculate portions of the facial nerve in a patient who died shortly after the onset of Bell’s palsy. In examining an autopsy specimen of a patient who died 13 days after the onset of Bell’s palsy, McKeever and colleagues12 found a diffuse lymphocytic infiltration of the intratemporal facial nerve with ongoing myelin phagocytosis. They later re-examined the same specimen and noted the most pronounced lymphocytic infiltration of the nerve at the labyrinthine segment, findings that these authors believed were most consistent with an ongoing compression-type injury to the facial nerve. Multiple reports have found that there seems to be evidence of constriction of the nerve at the meatal foramen in cases of Bell’s palsy. In specimens examined during the acute phase of Bell’s palsy, lymphocytic infiltration, perineural edema, and myelin degeneration have been noted.13,14 In a nerve specimen examined 1 year after the diagnosis of Bell’s Palsy ��������������������� lymphocytic���������� ��������� infiltration, perineural edema, ��������������������������������� and fibrotic changes were noted.15 Intraoperative biopsy specimens of the greater superficial petrosal nerve from patients undergoing nerve decompression procedures for Bell’s palsy have shown axonal demyelination and degeneration with lymphocytic infiltration.16 Together, the histopathologic findings in Bell’s palsy show demyelination and axonal loss with lymphocytic or phagocytic infiltration, findings that seem to be more pronounced at the labyrinthine segment of the facial nerve. 335
336
OTOLOGIC SURGERY
In considering this evidence, the pathogenesis of Bell’s palsy becomes more apparent: a virally induced, inflammatory response that produces edema within the nerve. Fisch and Felix16 first proposed that the facial nerve was entrapped at the meatal foramen as a result of neural edema. Intraoperative conduction studies have shown an electrophysiologic blockage at this site.17 The constriction imposed produces a conduction block at first; however, with prolonged or increased constriction, ischemia results. Subsequently, wallerian degeneration occurs, producing axonotmesis or neurotmesis or both. A spectrum of injury within the nerve from neurapraxia to neurotmesis may occur in Bell’s palsy.11,18 The proportion of each of these determines the amount of facial function that returns when the acute phase of the disease subsides. Bell’s palsy accounts for nearly three quarters of all acute facial palsies.19 The incidence of Bell’s palsy is 20 to 30 cases per 100,000 per year.20 The median age is 40 years, but it can occur at any age.21 The incidence is highest in patients older than 70 years and lowest in children younger than 10 years. The left and right sides are affected equally. Men and women are equally affected, but the incidence of Bell’s palsy is higher in pregnant women (45 cases per 100,000). The clinical presentation of Bell’s palsy is well known; however, the clinician must exclude other causes of facial paralysis based on history and physical examination findings. Patients describe an abrupt onset of unilateral paresis that occurs over 24 to 48 hours. The paresis can progress over 1 to 7 days to complete paralysis. Bilateral involvement, either simultaneously or consecutively, has been described.23 A history of progression of weakness over weeks to months, repeated episodes of paralysis, and twitching of the facial muscles should not be considered symptoms of Bell’s palsy. Other associated symptoms of hearing loss, vestibular symptoms, or other cranial nerve neuropathies also rule out the diagnosis of Bell’s palsy. On physical examination, the patient displays unilateral weakness or flaccid paralysis of all branches of the facial nerve. If forehead movement is normal, and there is strong eye closure and symmetric blinking, a central origin of paralysis should be suspected. The tympanic membrane should be of normal color and mobility. Careful bimanual palpation of the parotid gland may reveal a deep lobe parotid neoplasm. Oral cavity examination may show loss of papillae on the ipsilateral tongue. Cranial nerve testing is normal with the exception of the involved CN VII. Serial examinations are essential; if some evidence of recovery is not noted within 3 to 6 months, an aggressive search for an underlying neoplasm should be undertaken. Audiometric evaluation should reveal symmetric function except for an absent ipsilateral acoustic reflex. If unilateral hearing loss or acoustic reflex decay is present, further evaluation for retrocochlear pathology is necessary. If vestibular complaints are present,
an electronystagmogram is performed, and diagnosis of Bell’s palsy should be questioned. If the history and clinical presentation are highly suggestive of Bell’s palsy, magnetic resonance imaging (MRI) and computed tomography (CT) are not performed. High-resolution CT scans and MRI are obtained if patients have associated symptoms of otorrhea, vestibular complaints, and hearing loss. Planned surgical decompression and persistent dense paralysis after 6 months are also indications for imaging. Serologic tests for Lyme disease (IgG, IgM) are an important part of the work-up for unexplained facial paralysis in endemic areas.24 Serologic testing for antibodies to HSV or varicella zoster have provided some correlative evidence for a viral etiology in Bell’s palsy, although laboratory investigations in general have not been found to provide clinically relevant data. Electrodiagnostic testing is an important element of the diagnostic evaluation of facial paralysis. Testing can determine the extent of facial nerve injury and provide useful prognostic information for the development of management strategies. The technique of electroneuronography (ENoG), developed by Esslen,25 can distinguish the nerve fibers that have undergone wallerian degeneration from fibers that are temporarily blocked (neurapraxia). ENoG testing is not performed until 3 to 4 days after the development of complete unilateral paralysis because wallerian degeneration does not become apparent until 48 to 72 hours after an acute injury to the nerve. Electrodiagnostic testing is not performed when the patient exhibits paresis only. If the paresis progresses to total paralysis, electrodiagnostic testing is performed 3 days after the onset of total paralysis. Presence of voluntary movement 4 to 5 days after the onset of paresis indicates only minor injury, and complete recovery should be anticipated. ENoG is most accurate when it is performed within 3 weeks of the acute injury. The test is performed using standard electromyography (EMG) equipment, but requires the use of special surface stimulating and recording electrodes.26 The recording electrodes are in a handheld carrier and are manipulated in the nasolabial fold with the Esslen technique. The recording electrodes are not taped to the skin, as has been described by others.27 An evoked electric stimulus generates synchronous facial muscle movement that can be recorded from the skin surface (termed the compound muscle action potential [CMAP]). The amplitude of the biphasic CMAP has been found to correlate with the number of blocked or neurapraxic nerve fibers. As the percentage of degenerated fibers within the nerve increases, the amplitude of the CMAP decreases compared with the normal side of the face. Fisch and Esslen28 determined that if 90% or more of the fibers within the facial nerve degenerate within the first 14 days of an acute paralysis, a severe injury has occurred, and the chances of complete recovery are less than 50%. Patients who do not reach the 90% degeneration level by 3 weeks have a very good prognosis
Chapter 28 • Management of Bell’s Palsy and Ramsay Hunt Syndrome Acute paresis
337
Acute paralysis
Days 0–14
Days >14
Days 0–3
Prednisone + valacyclovir
Observation follow-up 6 months
Prednisone + Valacyclovir
Days 3–14
Days >14
Follow-up 6 months ENoG
Follow-up 3 days
Follow-up 5 days
Paresis
Paralysis
<90% Degeneration
>90% Degeneration
Follow-up 1 month
ENoG Follow paralysis protocol
Prednisone + valacyclovir
Recommend MCF decompression
Prednisone = 1 mg/kg x 7 days, Valacyclovir = 500 mg tid x 10 days
Follow-up depends on ENoG up to 14 days (See Fig. 28–2)
FIGURE 28-1. Algorithm for management of Bell’s palsy. ENoG, electroneuronography; MCF, middle cranial fossa.
and are likely to regain normal facial motion without synkinesis. The time course of degeneration is also important; the more rapid the degeneration, the more severe the injury.29 A patient showing greater than 90% degeneration at 5 days would have a worse prognosis than a patient with 90% degeneration at 14 days. Patients exhibiting 90% to 100% degeneration or no response to electric stimulation, in addition to ENoG, must undergo EMG testing. An EMG needle is placed in the orbicularis oris and oculi muscles, and the patient is asked to make a forceful contraction. Any voluntary motor unit activity indicates deblocking of the conduction block and a favorable prognosis. When deblocking occurs, fibers may not discharge at the same rate because of the previous injury, and may fail to generate a surface CMAP, resulting in a false-positive test result on ENoG testing. As movement returns to the face, the surface CMAP may also be absent for the same reason. Voluntary evoked EMG testing is mandatory if a surgical decompression is planned. If a facial paralysis has been slowly progressive over weeks to months, degeneration and regeneration of nerve fibers occur within the nerve, resulting in similar dyssynchronous discharge of evoked impulses and an inaccurate CMAP recording. Topognostic testing is widely reported to be useful in determining the site of injury in acute facial paralysis; however, intraoperative studies have shown that the Schirmer test is not accurate in diagnosing Bell’s palsy.26 The Schirmer test may be used to determine the extent of lacrimation and the need for eye protection.
A review of the natural history of Bell’s palsy shows that approximately 85% of patients begin to display some return of facial movement within 3 weeks of the onset of paresis.30 The remaining 15% begin to improve 3 to 6 months after the onset of the disease. Most patients show a complete return of facial function, but 10% to 15% have residual unilateral weakness and develop secondary deformities of synkinesis, tearing, or contracture. Some motion returns in almost all individuals with Bell’s palsy by 6 months. If no movement returns, a vigorous search for another etiology should begin anew. The management of patients with Bell’s palsy varies, depending on the type of specialist initially seeing the patient and on the training of the individual specialist. Figure 28-1 presents an overview of our management strategy. Patients presenting within the first week of facial weakness with paresis are placed on steroid therapy (prednisone, 60 to 80 mg/day for 7 days) and an antiviral agent (valacyclovir, 500 mg three times per day for 10 days). If patients are seen 7 to 10 days after onset and motor function is stable or improving, medical treatment is unnecessary. Patients are instructed to return in 1 week for re-evaluation to determine if neural degeneration has occurred. Electrodiagnostic testing is unnecessary as long as voluntary facial movement is present. If total paralysis occurs in the interim, the total paralysis protocol of Figure 28-1 is followed. Stable or improving patients are seen in 1 month. The use of steroids in Bell’s palsy has been the subject of much debate. Numerous studies of varying designs in adults have shown better outcomes in patients
338
OTOLOGIC SURGERY
treated with steroids,21,31,32 especially when initiated early in the course of the disease.33-35 Other randomized studies36 and meta-analyses,37 including many studies in children,38-40 have concluded that steroids did not affect the ultimate facial function outcomes in Bell’s palsy. Grogan and Gronseth,41 in a comprehensive, evidence-based review, concluded that there seemed to be a beneficial effect in the use of steroids in Bell’s palsy. Ramsey and coworkers42 performed a meta-analysis of 47 trials of steroid therapy for Bell’s palsy, and concluded that there seemed to be an improved odds of recovery in patients treated with steroids (49% to 97%) compared with untreated controls (23% to 64%). We use prednisone in a dose of 1 mg/kg daily for 7 days in all cases in which it is not medically contraindicated, in anticipation of speeding recovery, reducing the number of degenerating axons, and reducing the number of patients needing decompression. The combination of steroids and antivirals may be superior to either one alone. A double-blind, randomized, controlled trial of acyclovir and prednisone versus prednisone alone in the treatment of Bell’s palsy showed better results with the combination therapy.43 This study documented poor facial function recovery in 23% of the prednisone-only group compared with 7% in the acyclovir-plus-prednisone group. Interpreting the results of this study, Grogan and Gronseth41 reported that patients who receive antiviral therapy in addition to steroids were 1.22 times as likely to have a good facial nerve outcome. A multicenter, randomized, placebo-controlled study comparing treatment of patients with Bell’s palsy with steroids and antivirals with treatment with steroids alone concluded that the addition of valacyclovir improved the recovery rate from 75% to 90% in cases of complete palsy, and from 89% to 96% in all cases of facial palsy.44 Many other reviews of the topic have made similar conclusions.45-48 Other studies have found no significant difference between this combination of drugs and the natural history of the disease.24,49 One study found a negative impact of treatment of patients with Bell’s palsy with antivirals alone versus steroids alone.50 Because of the paucity of potential side effects and good patient tolerance of antiviral medications, the addition of antiviral medications to steroids in the treatment of patients with Bell’s palsy seems prudent. Patients presenting within 1 week of the onset of total unilateral paralysis undergo electrodiagnostic testing (if at least 3 days have passed since the onset of paralysis) and are started on medical therapy. If the patient is seen in the first 3 days after the onset of paralysis, steroid and antiviral therapy are initiated, and follow-up electrodiagnostic studies are arranged. The frequency of follow-up electrodiagnostic examinations is determined by the result of testing and the time interval after paralysis, as suggested by Fisch51 (Fig. 28-2). Patients exhibiting nearly 90% neural degeneration on ENoG examination,
FIGURE 28-2. Recommended electroneuronography (ENoG) testing schedule in acute facial paralysis.
or who are degenerating quickly, undergo frequent electrodiagnostic testing (every 1 or 2 days). If greater than 90% degeneration is reached, and there are no motor unit potentials on voluntary EMG testing, the patient is considered a candidate for middle cranial fossa decompression. When 90% degeneration is not reached within 2 weeks (14 days) after the onset of paralysis, no further electrodiagnostic studies need to be performed. Patients seen for the first time more than 2 weeks after the onset of paralysis undergo EMG evaluation to determine if regeneration has begun. They are scheduled for a 6 month follow-up to ensure that some motor function has returned. If no movement is evident at 6 months, it must be assumed that Bell’s palsy was an incorrect diagnosis, and a search for another disease process is begun.
RAMSAY HUNT SYNDROME Herpes zoster oticus refers to a syndrome of acute otalgia accompanied by a herpetic, vesicular rash. When accompanied by facial paralysis, the syndrome is known as Ramsay Hunt syndrome.52 Ramsay Hunt syndrome is the second most common cause of facial paralysis (after Bell’s palsy), and is induced by the reactivation of the varicella-zoster virus that remains latent in the geniculate ganglion after primary infection with chickenpox.53 Classically, patients present with severe otalgia and unilateral facial paralysis. Vesicular eruptions may or may not be present initially, but usually appear within 3 to 5 days of the paralysis. The vesicular lesions can appear on the concha, ear canal, postauricular skin, and tympanic membrane. Occasionally, the oral cavity, neck, and shoulder are also involved. The disease can affect other cranial nerves, including auditory, vestibular, trigeminal, glossopharyngeal, and vagus, prompting the name herpes zoster cephalicus.54 Also in contrast to Bell’s palsy, the symptoms are more severe, and the prognosis is worse in Ramsay Hunt syndrome. The frequency of complete neural degeneration of the facial nerve is substantially higher than with
Chapter 28 • Management of Bell’s Palsy and Ramsay Hunt Syndrome
Bell’s palsy, and complete recovery of facial motor function has ranged from 10% to 31% in several studies.55-57 Patients with auditory and vestibular dysfunction in addition to facial paralysis generally have a worse prognosis. In addition, patients with diabetes, hypertension, and advanced age have been reported to have a poorer prognosis in Ramsay Hunt syndrome.58 The diagnosis of Ramsay Hunt syndrome is based on the history of otalgia, vesicular lesions or eschars, and facial paralysis. MRI shows enhancement of a large portion of the facial nerve, often the vestibular and cochlear nerves, the labyrinth, and the dura lining the internal auditory canal.59 Imaging as part of routine evaluation is unnecessary. Electrodiagnostic studies have been unreliable in herpes zoster oticus. The management of Ramsay Hunt syndrome has changed with the development of antiviral agents. The natural history of the disease was evaluated by Devriese and Moesker,56 who identified only a 10% rate of complete facial nerve recovery. Many studies have identified a superior recovery rate of 75% using a combination of steroids and antiviral agents.45,60-62 The benefit of early initiation of steroid and antiviral treatment was clearly shown in a study by Murakami and colleagues,60 in which 75% of patients treated within 3 days of the onset had complete recovery, whereas only 30% of patients treated after 7 days experienced a complete recovery. We have experienced similar successful results using intravenous acyclovir (10 mg/kg three times daily), oral acyclovir (800 mg five times daily), or oral valacyclovir (500 mg three times daily) for 10 days, in combination with a 3 week tapering course of prednisone (60 to 80 mg/kg daily). Patients report rapid reduction in pain, and occasionally experience return of facial movement during the medical therapy. Because of the presence of “skip” regions and diffuse neuritis of the facial nerve,63,64 surgical decompression of RHS is not recommended.
FACIAL NERVE DECOMPRESSION FOR BELL’S PALSY Decompression of the facial nerve for Bell’s palsy has been reported since the 1930s,65 and is recommended in patients who exhibit greater than 90% degeneration on ENoG and no voluntary EMG activity within 14 days of the paralysis. Preliminary results suggested that early decompression of the labyrinthine, geniculate, and tympanic segments of the facial nerve through a middle cranial fossa approach improves recovery of severely degenerated cases,25 whereas decompression of the mastoid segment alone did not alter the natural history of the disease.66 Gantz and associates67 showed a statistically significant difference in facial nerve outcome when decompression is performed within 2 weeks of onset of the paralysis. The patients undergoing decompression exhibited HouseBrackmann grade I (n = 14) or II (n = 17) in 91% of
339
cases; two patients had a grade III, and one patient had a grade IV outcome. There were no grade V or VI results in the surgical group. Patients who met surgical criteria but elected not to undergo surgery had a 58% chance of a poor outcome. Within the nonsurgical group, 19 patients had a House-Brackmann grade of either III or IV at 7 month follow-up. Intraoperative evoked EMG is used to localize the region of the conduction block.17 Intraoperative direct nerve stimulation is useful when the nerve is not completely degenerated, which includes most instances when preoperative ENoG reveals 100% degeneration. This finding suggests that a few nerve fibers remain capable of stimulation even when ENoG shows total nerve degeneration. In most cases, the nerve conduction block is localized between the geniculate ganglion and the internal auditory canal segments of the facial nerve. Failure to generate a motor unit potential when the tympanic segment of the nerve is stimulated indicates that the nerve is 100% degenerated, or that the conduction block is more distal. If the nerve appears normal in the tympanic portion, while there is edema and erythema of the internal auditory canal segment, total degeneration should be suspected. If erythema and edema are apparent in the tympanic segment, a mastoid decompression is added.
Preoperative Preparation The risks of middle fossa decompression of the facial nerve are discussed with the patient, and include cerebrospinal fluid leak (4% to 6%), infection (1%), hearing loss (1%), dizziness (1%), intracranial hemorrhage (<1%), and aphasia (<1%). When the decision is made to proceed with surgical decompression, the procedure should be performed as soon as possible. Appropriate preoperative laboratory and imaging studies are obtained, along with a Stenvers projection plain radiograph of the temporal bone to identify the floor of the middle cranial fossa and superior semicircular canal. Cefazolin and dexamethasone are given before the skin incision and are continued for six doses. General anesthesia and transoral endotracheal intubation are accomplished; thereafter, the bed is rotated 180 degrees for access (Fig. 28-3). Paralytic agents must be reversed before the skin incision. A urinary catheter is placed to monitor fluids and diuresis. Hair is shaved approximately 10 cm above and 5 cm behind the ear. The entire side of the head and face are prepared, and EMG needles are placed in the orbicularis oculi and oris muscles (Fig. 28-4A) for facial nerve monitoring. A clear drape is placed over the prepared area to allow visualization of the entire side of the face in the case of monitoring equipment failure (Fig. 28-4B). Standard auditory brainstem–evoked recording electrodes are placed in the right and left mastoid tips and the vertex and forehead (ground), and insert headphones are placed in both external auditory canals. The auditory brainstem response is monitored throughout the procedure.
340
OTOLOGIC SURGERY
FIGURE 28-3. Operating room setup for middle cranial fossa approach.
Surgical Technique The skin incision is marked as shown in Figure 28-5 and carried down to the level of the temporalis fascia. Meanwhile, mannitol (0.5 g/kg body weight) and hyperventilation (partial pressure of carbon dioxide of 25 mm Hg) are initiated by the anesthesiologist to relax the brain. After the posteriorly based skin flap is elevated, a 4 × 6 cm piece of temporalis fascia is harvested and set aside in a moist gauze for use at the time of closure. An anteriorly based, inferiorly staggered muscle flap is elevated down to the level of the linea temporalis and reflected forward with an Adson cerebellar retractor. Staggering the incisions prevents dural exposure if wound dehiscence occurs. The zygomatic root identifies the floor of the middle cranial fossa and is the central landmark of the craniotomy. The skin and muscle flaps should be wrapped with moist sponges and secured with temporary retraction stitches. The craniotomy should be approximately 4 cm in anteroposterior dimension and 5 cm cephalocaudal (Fig. 28-6A). The anteroposterior margins must be kept parallel for stability of the middle cranial fossa retractor. The craniotomy can be created with either cutting burrs or a craniotomy saw. The bone flap should be elevated with care by use of a blunt dural elevator. Occasionally, the middle meningeal artery is embedded within the bone, requiring bipolar coagulation to free it. The bone flap is wrapped in a moist gauze and set aside for use at closure. Dural elevation from the floor of the middle cranial fossa is accomplished with a Freer or joker elevator, always in a posterior-to-anterior direction, which prevents injury to the greater superficial petrosal nerve and geniculate ganglion. The petrous ridge is identified at the posterior margin of the craniotomy, and the dura is slowly elevated over the arcuate eminence and meatal plane. Dural reflections are cauterized and sharply transected to allow elevation to the anterior petrous ridge. The hiatus of the facial canal with the greater superficial petrosal nerve and artery is the anterior margin of elevation. Further anterior elevation exposes the foramen spinosum and the pterygoid plexus of veins, which can cause troublesome oozing throughout the procedure.
After dural elevation, cottonoid sponges can be placed at the anterior and posterior margins of the elevation to help retract the dura during placement of the selfretaining middle cranial fossa retractor (Fig. 28-6B). Before the bony exposure of the facial nerve, the Stenvers projection radiograph is re-examined to determine the depth of the superior semicircular canal in the temporal bone. The superior semicircular canal is the first structure to be located. When its blue line is identified, the remaining intratemporal structures have consistent anatomic locations. Landmarks of the middle cranial fossa floor can be quite subtle. The arcuate eminence may not be apparent and, in many instances, is not parallel with the superior semicircular canal. One consistent anatomic feature to remember is that the plane of the superior semicircular canal is almost always perpendicular to the petrous ridge (Fig. 28-7). If the arcuate eminence is not initially apparent, drilling is begun posterior to the semicircular canal, slowly removing the tegmen mastoideum with a moderate-sized diamond burr. The whitish color of the membranous temporal bone can be distinguished from the yellow, dense otic capsule bone of the superior canal. When the superior canal is identified, drilling in a parallel direction with the canal gradually exposes the blue line. When the location of the superior canal is confirmed, an anterior line 60 degrees to the blue line locates the position of the internal auditory canal. The depth of the internal auditory canal varies, but drilling medially near the petrous ridge provides a safe route to the canal. Drilling laterally near the anterior ampulla of the superior semicircular canal places the geniculate ganglion, labyrinthine segment of the facial nerve, and cochlea at risk. When the blue line of the internal auditory canal is exposed, bone is removed in a lateral direction until the vertical crest is found (Fig. 28-8). Bone can now be removed over the geniculate ganglion and tegmen tympani. The labyrinthine segment of the facial nerve is the narrowest portion of the fallopian canal and lies in an anterosuperior plane from the vertical crest to the geniculate ganglion. A 1 mm diamond burr is used to remove bone over the labyrinthine segment, while the area
Chapter 28 • Management of Bell’s Palsy and Ramsay Hunt Syndrome
341
FIGURE 28-4. A, Electrode placement for intraoperative electromyography. B, Draping with exposure of half the face for visual monitoring of facial movement. FIGURE 28-5. Skin incision for middle cranial fossa approach. Mastoid exposure can be obtained by extending incision postauricularly. Incision of anteriorly based temporalis muscle flap is offset to keep suture lines from being directly in line.
342
OTOLOGIC SURGERY
FIGURE 28-6. A, Craniotomy for middle cranial fossa exposure. Inferior expansion of craniotomy allows more exposure during dural elevation. Vertical margins must be parallel for stability of retractor. Craniotomy should be centered on the temporal root of the zygoma (dashed line). B, Placement of House-Urban middle cranial fossa retractor.
FIGURE 28-7. Exposure of superior semicircular canal blue line and position of internal auditory canal (IAC) 60 degrees anterior to a line through the blue line. Superior canal blue line almost always is perpendicular to petrous ridge.
directly anterior to the segment is closely observed for the blue line of the basal turn of the cochlea. The final layer of bone is removed from the labyrinthine, geniculate, and proximal tympanic segments of the facial nerve with thin, angled hooks and blunt microelevators. At this point, marked swelling of the nerve in the internal auditory canal, labyrinthine segment, geniculate ganglion,
and proximal tympanic segments is usually observed. Further swelling is apparent after neurolysis. A disposable microscalpel (Beaver No. 59-10) is used to slit the periosteum and epineural sheath. Intraoperative EMG is used to localize the region of the nerve conduction block. Direct stimulation of the most distal exposed tympanic segment of the nerve is performed
Chapter 28 • Management of Bell’s Palsy and Ramsay Hunt Syndrome
343
FIGURE 28-8. A-D, Middle cranial fossa exposure of CN VII and surrounding anatomy. Bone removed from tegmen tympani to expose tympanic segment of CN VII. Note edema of internal auditory canal segment of CN VII, frequently found in Bell’s palsy. SSC, superior semicircular canal.
with monopolar or bipolar microforceps (Fig. 28-9). If the conduction block is medial to the point of stimulation, and the nerve is not completely degenerated, a motor unit potential is observed. Stimulating more medially toward the internal auditory canal fails to elicit a motor unit potential if the conduction block is at the meatal foramen. On completion of the decompression and confirmation of the site of the conduction block, the opened internal auditory canal is covered with a piece of temporalis muscle. The previously harvested temporalis fascia is placed over the opened floor of the middle fossa after bone waxing the opened mastoid air cells. The dural retractor is removed, and the anesthesiologist is asked to re-establish a normal partial pressure of carbon dioxide. A corner of the craniotomy bone flap is harvested and placed superior to the fascia to prevent herniation of the
temporal lobe dura into the attic and internal auditory canal. The temporal lobe is allowed to re-expand into the middle cranial fossa floor. The remainder of the craniotomy flap is placed on the dura, and the temporalis is closed, creating a watertight seal with interrupted absorbable sutures. Skin is closed with a deep layer of interrupted absorbable sutures and an outer layer of nylon absorbable sutures. A mastoid-type dressing is applied. No drains are used.
Postoperative Care Postoperative care includes observation in the intensive care unit overnight, limitation of fluids (1500 to 1800 mL/ day), dexamethasone (6 mg every 6 hours for 36 hours), cefazolin (1 g every 8 hours for 36 hours), routine neural
344
OTOLOGIC SURGERY
FIGURE 28-9. Intraoperative evoked electromyography to identify site of nerve conduction block. If conduction block is medial to geniculate ganglions, stimulation of tympanic segment results in motor unit potential (1); stimulation of internal canal segment (3) results in no response. Conduction block usually is at meatal foramen (MF). LSC, lateral semicircular canal; SSC, superior semicircular canal; SVN, superior vestibular nerve.
checks, and limitation of analgesia to codeine. There is little postoperative pain, and narcotics stronger than codeine may mask intracranial complications. The patient is transferred to a routine postoperative floor the next morning, encouraged to begin ambulation, and started on a diet as tolerated. On postoperative day 3, and daily thereafter, observations for cerebrospinal fluid rhinorrhea are made by asking the patient to lean forward with the head between the knees. If cerebrospinal fluid rhinorrhea occurs, a spinal drain must be placed for 4 to 5 days. Following this regimen, only very rarely has a patient required surgical closure of the leak. Patients are usually discharged from the hospital on postoperative days 5 to 7. No intracranial complications, including intracranial hemorrhage, aphasia, or seizures, occurred in the Iowa series.67
Limitations and Special Considerations The anatomy of the middle cranial fossa floor is varied and presents some difficulty in identification of landmarks. The surgeon must have a precise knowledge of the threedimensional anatomy of the temporal bone. Many hours in the temporal bone dissection laboratory are required to attain the delicate microsurgical skills necessary for this type of surgery. Dural elevation can be difficult, especially in patients older than 65 years. If a dural tear occurs, a temporalis fascia repair must be performed. Hearing loss can occur from contact of the rotating burr with an intact ossicular chain or by entrance into the cochlea or labyrinth. Vestibular dysfunction can occur in a similar fashion. If the membranous labyrinth is violated, immediate placement of a small amount of bone wax may preserve auditory and vestibular function. Correct positioning of the craniotomy and maintaining parallel vertical margins are essential. Meticulous
hemostasis must be maintained with bipolar cautery, oxidized cellulose, and pressure. A dry surgical field is crucial for microscopic dissection of subtle landmarks and prevention of complications. If large apical air cells are opened, they must be plugged with temporalis muscle to prevent cerebrospinal fluid leakage.
SUMMARY Bell’s palsy is a clinical entity encountered with relative frequency by practitioners of many specialties. Before making the diagnosis of Bell’s palsy, other causes of facial paralysis must be excluded through thorough history taking and physical examination. Bell’s palsy and Ramsay Hunt syndrome seem to be due to viral etiologies, and seem to benefit from medical therapy with steroids and antiherpetic medications. Electrodiagnostic testing in cases of Bell’s palsy with complete paralysis is indicated and provides crucial information to guide surgical decision making and patient counseling. In cases of Bell’s palsy with greater than 90% degeneration on ENoG and absence of CMAP on EMG, middle fossa craniotomy with decompression of the labyrinthine, geniculate, and proximal tympanic segments of the nerve should be offered to patients without medical contraindications to an intracranial procedure. Familiarity with the anatomy of the middle cranial fossa is essential for effective decompression of the facial nerve and avoidance of complications.
REFERENCES 1. Bell C : On the nerves, giving an account of some experiments on their structure and function, which led to a new arrangement of the system. Philos Trans 111:398-424, 1821.
Chapter 28 • Management of Bell’s Palsy and Ramsay Hunt Syndrome 2. McCormick D P: Herpes-simplex virus as a cause of Bell’s palsy. Lancet 1:937-939, 1972. 3. Furuta Y, Fukuda S, Chida E, et al: Reactivation of herpes simplex virus type 1 in patients with Bell’s palsy. J Med Virol 54:162-166, 1998. 4. Burgess RC, Michaels L, Bale JF Jr, et al: Polymerase chain reaction amplification of herpes simplex viral DNA from the geniculate ganglion of a patient with Bell’s palsy. Ann Otol Rhinol Laryngol 103:775-779, 1994. 5. Murakami S, Mizobuchi M, Nakashiro Y, et al: Bell palsy and herpes simplex virus: Identification of viral DNA in endoneurial fluid and muscle. Ann Intern Med 124(1 Pt 1):27-30, 1996. 6. Sugita T, Murakami S, Yanagihara N, et al: Facial nerve paralysis induced by herpes simplex virus in mice: An animal model of acute and transient facial paralysis. Ann Otol Rhinol Laryngol 104:574-581, 1995. 7. Lazarini PR, Viamna MF, Alcantara MPA, et al: [Herpes simplex virus in the saliva of peripheral Bell’s palsy patients]. Rev Bras Otorrinolaringol (Engl Ed) 72:7-11, 2006. 8. Liston S L , Kleid M S : Histopathology of Bell’s palsy. Laryngoscope 99:23-26, 1989. 9. Carreno M, Llorente JL, Hidalgo F, et al: [Application of the polymerase chain reaction to an experimental model of infection by herpes simplex virus type 1]. Acta Otorrinolaringol Esp 49:15-18, 1998. 10. Reddy J B, Liu J, Balshi S, et al: Histopathology of Bell’s palsy. Eye Ear Nose Throat Mon 45:62-66, 1966. 11. Fowler E P Jr: The pathologic findings in a case of facial paralysis. Trans Am Acad Ophthalmol Otolaryngol 67:187-197, 1963. 12. McKeever P, Proctor B, Proud G: Cranial nerve lesions in Bell’s palsy. Otolaryngol Head Neck Surg 97:326-327, 1987. 13. Podvinec M : Facial nerve disorders: Anatomical, histological and clinical aspects. Adv Otorhinolaryngol 32:124-193, 1984. 14. Donoghue O: Histopathologic aspects of Bell’s palsy. Presented at American Academy of Otolaryngology– Head and Neck Surgery. Anaheim, CA, Association for Research in Otolaryngology, 1983. 15. Jackson CG, Johnson GD, Hyams VJ, et al: Pathologic findings in the labyrinthine segment of the facial nerve in a case of facial paralysis. Ann Otol Rhinol Laryngol 99(5 Pt 1):327-329, 1990. 16. Fisch U, Felix H : On the pathogenesis of Bell’s palsy. Acta Otolaryngol 95(5-6):532-538, 1983. 17. Gantz B J, Gmur A, Fisch U: Intraoperative evoked electromyography in Bell’s palsy. Am J Otolaryngol 3: 273-278, 1982. 18. Proctor B, Corgill D A, Proud G: The pathology of Bell’s palsy. Trans Sect Otolaryngol Am Acad Ophthalmol Otolaryngol 82:ORL70-ORL80, 1976. 19. Peitersen E, Bell’s palsy: The spontaneous course of 2,500 peripheral facial nerve palsies of different etiologies. Acta Otolaryngol Suppl 549:4-30, 2002. 20. Hauser WA, Karnes WE, Annis J, et al: Incidence and prognosis of Bell’s palsy in the population of Rochester, Minnesota. Mayo Clin Proc 46:258-264, 1971. 21. Katusic S K, Beard CM, Wiederholt WC, et al: Incidence, clinical features, and prognosis in Bell’s palsy, Rochester, Minnesota, 1968-1982. Ann Neurol 20:622-627, 1986.
345
22. Adour K K : Current concepts in neurology: Diagnosis and management of facial paralysis. N Engl J Med 307:348-351, 1982. 23. Cwach H, Landis J, Freeman JW: Bilateral seventh nerve palsy: A report of two cases and a review. S D J Med 50:99-101, 1997. 24. Holland NJ, Weiner G M : Recent developments in Bell’s palsy. BMJ 329:553-557, 2004. 25. Esslen E : Electromyography and electroneurography. In Fisch U (ed): Facial Nerve Surgery, Birmingham, AL, Aesculapius. 1977, pp 93-100. 26. Gantz B J, Gmuer AA, Holiday M, et al: Electroneurographic evaluation of the facial nerve: Method and technical problems. Ann Otol Rhinol Laryngol 93(4 Pt 1):394-398, 1984. 27. Blumenthal F, May M : Electrodiagnosis. In May M (ed): The Facial Nerve, New York, Thieme, 1986. 28. Fisch U, Esslen E : Total intratemporal exposure of the facial nerve: Pathologic findings in Bell’s palsy. Arch Otolaryngol 95:335-341, 1972. 29. Fisch U: Prognostic value of electrical tests in acute facial paralysis. Am J Otol 5:494-498, 1984. 30. Peitersen E : Natural history of Bell’s palsy. Acta Otolaryngol Suppl 492:122-124, 1992. 31. Adour K K, Wingerd J, Bell DN, et al: Prednisone treatment for idiopathic facial paralysis (Bell’s palsy). N Engl J Med 287:1268-1272, 1972. 32. Wolf S M, Wagner JH, Davidson S, et al: Treatment of Bell palsy with prednisone: A prospective, randomized study. Neurology 28:158-161, 1978. 33. Williamson IG, Whelan TR : The clinical problem of Bell’s palsy: Is treatment with steroids effective? Br J Gen Pract 46:743-747, 1996. 34. Shafshak TS, Essa AY, Bakey FA : The possible contributing factors for the success of steroid therapy in Bell’s palsy: A clinical and electrophysiological study. J Laryngol Otol 108:940-943, 1994. 35. Brown J S : Bell’s palsy: A 5 year review of 174 consecutive cases: An attempted double blind study. Laryngoscope 92:1369-1373, 1982. 36. May M, Wette R, Hardin WB Jr, et al: The use of steroids in Bell’s palsy: A prospective controlled study. Laryngoscope 86:1111-1122, 1976. 37. Salinas R A, et al: Corticosteroids for Bell’s palsy (idiopathic facial paralysis). Cochrane Database Syst Rev (1): CD001942, 2002. 38. Salman M S, MacGregor D L : Should children with Bell’s palsy be treated with corticosteroids? A systematic review. J Child Neurol 16:565-568, 2001. 39. Prescott C A : Idiopathic facial nerve palsy in children and the effect of treatment with steroids. Int J Pediatr Otorhinolaryngol 13:257-264, 1987. 40. Unuvar E, Oguz F, Sidal M, et al: Corticosteroid treatment of childhood Bell’s palsy. Pediatr Neurol 21:814-816, 1999. 41. Grogan PM, Gronseth G S : Practice parameter: Steroids, acyclovir, and surgery for Bell’s palsy (an evidence-based review): Report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology 56:830-836, 2001. 42. Ramsey M J, DerSimonian R, Holter MR, et al: Corticosteroid treatment for idiopathic facial nerve paralysis: A meta-analysis. Laryngoscope 110(3 Pt 1):335-341, 2000.
346
OTOLOGIC SURGERY
43. Adour K K, Ruboyianes JM, Von Doersteu PG, et al: Bell’s palsy treatment with acyclovir and prednisone compared with prednisone alone: A double-blind, randomized, controlled trial. Ann Otol Rhinol Laryngol 105:371-378, 1996. 44. Hato N, Yamada H, Kohno H, et al: Valacyclovir and prednisolone treatment for Bell’s palsy: a multicenter, randomized, placebo-controlled study. Otol Neurotol 2007 28(3): 408-13. 45. Dickins J R , Smith JT, Graham S S : Herpes zoster oticus: Treatment with intravenous acyclovir. Laryngoscope 98:776-779, 1988. 46. Sweeney C J, Gilden D H : Ramsay Hunt syndrome. J Neurol Neurosurg Psychiatry 71:149-154, 2001. 47. Morrow M J: Bell’s palsy and herpes zoster oticus. Curr Treat Options Neurol 2:407-416, 2000. 48. Axelsson S, Lindberg S, Stjernquist-Desatnik A : Outcome of treatment with valacyclovir and prednisone in patients with Bell’s palsy. Ann Otol Rhinol Laryngol 112:197-201, 2003. 49. Ramos Macias A, de Miguel Martine Z, I Martin Sanchez AM: Incorporacion del aciclovir en el tratamiento de la paralisis periferica: Un estudio en 45 casos. Acta Otolaryngol Espanola 43:117-120, 1992. 50. De Diego J I, Prim MP, De Sarriá MJ, et al: Idiopathic facial paralysis: A randomized, prospective, and controlled study using single-dose prednisone versus acyclovir three times daily. Laryngoscope 108(4 Pt 1):573-575, 1998. 51. Fisch U: Surgery for Bell’s palsy. Arch Otolaryngol 107: 1-11, 1981. 52. Sachs E Jr, House R K : The Ramsay Hunt syndrome, geniculate herpes. Neurology 6:262-268, 1956. 53. Ikeda M, Hiroshige K, Abiko Y, et al: Impaired specific cellular immunity to the varicella-zoster virus in patients with herpes zoster oticus. J Laryngol Otol 110:918-921, 1996. 54. Adour K K : Otological complications of herpes zoster. Ann Neurol 35(Suppl):S62-S64, 1994. 55. Devriese P: Herpes zoster causing facial paralysis. In Fisch U (ed): Facial Nerve Surgery. Birmingham, AL, Aesculapius, 1977, pp 419-420.
56. Devriese PP, Moesker WH : The natural history of facial paralysis in herpes zoster. Clin Otolaryngol Allied Sci 13:289-298, 1988. 57. Peitersen E : Spontaneous course of Bell’s palsy. In Fisch U (ed): Facial Nerve Surgery. Birmingham, AL, Aesculapius, 1977, pp 337-343. 58. Yeo SW, Lee DH, Jun BC, et al: Analysis of prognostic factors in Bell’s palsy and Ramsay Hunt syndrome. Auris Nasus Larynx 34:159-164, 2007. 59. Brandle P, Satoretti-Schefer S, Bohmer A, et al: Correlation of MRI, clinical, and electroneuronographic findings in acute facial nerve palsy. Am J Otol 17:154-161, 1996. 60. Murakami S, Hato N, Horiuchi J, et al: Treatment of Ramsay Hunt syndrome with acyclovir-prednisone: Significance of early diagnosis and treatment. Ann Neurol 41:353-357, 1997. 61. Stafford FW, Welch A R : The use of acyclovir in Ramsay Hunt syndrome. J Laryngol Otol 100:337-340, 1986. 62. Uri N, Meyer W, Greenberg E, et al: Herpes zoster oticus: Treatment with acyclovir. Ann Otol Rhinol Laryngol 101(2 Pt 1):161-162, 1992. 63. Denny-Brown D: Pathologic features of herpes zoster: A note on geniculate herpes. Arch Neurol Psychiatry 20:149-159, 1944. 64. Honda N, Yanagihara N, Hato N, et al: Swelling of the intratemporal facial nerve in Ramsay Hunt syndrome. Acta Otolaryngol 122:348-352, 2002. 65. Balance C : The operative treatment of facial palsy: By the introduction of nerve grafts into the fallopian canal and by other intratemporal methods. Acta Otolaryngol 15: 1-79, 1932. 66. May M, Klein S R , Taylor FH : Idiopathic (Bell’s) facial palsy: Natural history defies steroid or surgical treatment. Laryngoscope 95:406-499, 1985. 67. Gantz B J, Rubinstein JT, Gidley P, et al: Surgical management of Bell’s palsy. Laryngoscope 109:1177-1188, 1999.
29
Traumatic Facial Paralysis Herman A. Jenkins, Gregory A. Ator, and Henry H. Chen
The facial nerve may be injured by many blunt and pene trating mechanisms. Common causes include motor vehicle accidents, stab or gunshot wounds to the face (frequently seen in urban areas), and iatrogenic inju ries during head and neck surgical procedures. Primary mechanisms of injury include stretching, compression, and transection of the nerve. The course of the nerve from the brainstem to the facial musculature can be divided into three segments: intracranial, intratemporal, and extratemporal or periph eral (Fig. 29-1). The pathophysiology of facial nerve disorders varies according to the segment of the nerve involved. Each segment is discussed individually.
INTRACRANIAL INJURY TO THE FACIAL NERVE The intracranial facial nerve, extending from the brain stem to the fundus of the internal auditory canal, is rarely damaged by penetrating trauma because of the excellent protection afforded by the petrous bone and the cranial vault, but with severe trauma, stretch and shock wave– type injuries may occur. Penetrating trauma to this region is likely to be accompanied by extensive central nervous system pathology, which must be evaluated and treated first. Evaluation of the injured nerve begins with a careful examination of motor function as soon as possible after the injury is sustained. If the nerve is functional at presentation and becomes progressively paretic, a complete transection injury is unlikely. If the nerve manifests any motion, regu lar clinical observation can be used to follow the status of the nerve. After the onset of complete paralysis, surgery is considered if the nerve shows electric signs of near-total degeneration (see section on timing of surgery).
INTRATEMPORAL INJURY TO THE FACIAL NERVE The intratemporal facial nerve, extending from the inter nal auditory canal fundus to the stylomastoid foramen, is frequently damaged from blunt trauma to the skull that
leads to temporal bone fracture. Fractures produced by blunt trauma have been traditionally grouped into longi tudinal and transverse varieties (Table 29-1), although almost any type of fracture can be encountered. Frac tures with the main component parallel to the long axis of the petrous pyramid are classified as longitudinal (Fig. 29-2), whereas fractures perpendicular to the long axis (Fig. 29-5) are classified as transverse. Longitudinal fractures are produced by trauma to the lateral aspects of the skull in the temporoparietal region, and compose 80% of fractures in most series.1 Transverse fractures are produced by trauma to the occipital or frontal regions of the skull, and compose about 20% of fractures. Many fractures are oblique or combine elements of longitu dinal and transverse fractures.2 Severely comminuted and complex fractures of the temporal bone are com monly produced by penetrating gunshot wounds of the temporal bone.3 A longitudinal fracture (Figs. 29-2 and 29-3) is sus pected when a step-off is present in the external audi tory canal and is frequently accompanied by blood in the external auditory canal. A perforation or tear of the tympanic membrane may be present, and cerebrospinal fluid (CSF) otorrhea is occasionally seen. Sterile instru ments should be used during the examination of the external auditory canal to avoid introducing contami nation into the area and producing retrograde menin gitis. A conductive hearing loss usually is present and can have numerous causes. Perforation of the eardrum, hematoma in the middle ear cleft, disruption of liga ments supporting the ossicles in the attic region, and disruption of the ossicular joints all can lead to vary ing degrees of conductive hearing loss (see section on ossicular damage). A longitudinal fracture often extends to the foramen ovale. Facial paralysis is seen in only 20% of longitudinal fractures, but is the most common cause of facial paralysis in blunt trauma of the temporal bone because of the rela tive infrequency of transverse fractures. The facial nerve is involved in the perigeniculate region in 90% of cases4,5 and less commonly in the mastoid segment by fractures of the posterior external auditory canal (Fig. 29-4). The pathology of the facial nerve injury in blunt temporal 347
348
OTOLOGIC SURGERY
FIGURE 29-1. A, Facial nerve anatomy. B, Axial view. C, Lateral view. CT, chorda tympani nerve; EAC, external auditory canal; OW, oval window; RW, round window.
bone trauma, in decreasing order of occurrence, consists of intraneural hemorrhage, bony fragment impingement, and nerve transection.6 A transverse temporal bone fracture is suspected when a patient presents with sensorineural hearing loss and vertigo accompanied by facial paralysis. The external canal is frequently intact, and no evidence of canal wall discontinuity and hemotympanum may be present (Figs. 29-5 and 29-6). Transverse fractures can extend into
the foramen spinosum or lacerum. These patients have a 50% incidence of facial paralysis, which occurs from damage to the geniculate ganglion region (Fig. 29-7).7 The causes of injury are the same as for longitudinal frac tures, and intraneural hemorrhage is the most common. The traditional classification scheme, based on ana tomic cadaveric studies performed more than 50 years ago,8 has been criticized because of its inability to predict complications of temporal bone fractures. More recently,
Chapter 29 • Traumatic Facial Paralysis
349
TABLE 29-1 Types of Temporal Bone Fractures
Mechanism Incidence Incidence of facial paralysis Type of hearing loss External auditory canal Tympanic membrane Ossicular damage Vertigo Skull base foramen
Longitudinal
Transverse
Temporal or parietal blow 80% 20% Conductive Torn, bloody Perforated Common Uncommon Ovale
Frontal or occipital blow 20% 50% Sensorineural Intact Intact, hemotympanum Uncommon Common Spinosum, lacerum
several authors have proposed new classification schemes with predictive ability.9-12 These schemes are similar and are based on whether the otic capsule or petrous apex or both are violated. Fractures involving the otic capsule were significantly more likely to result in the complica tions of facial nerve injury, sensorineural hearing loss, and CSF leak than fractures that did not. The new clas sification systems seem to have utility for predicting these serious sequelae of temporal bone fractures. Gunshot wounds to the temporal bone region typi cally produce extensive damage, the degree of which is determined by the velocity of the projectile. Low-velocity civilian projectiles have low energy and produce mainly locally destructive manifestations. In contrast, highvelocity, large-caliber weapons, which are increasingly being seen on city streets, are capable of widespread destruction, with extensive local and regional manifesta tions produced by the concomitant shock wave. Severe life-threatening injuries including vascular and intracra nial damage are seen in one third to one half of patients and must be stabilized first. Angiography and computed tomography (CT) scan of the head are needed as part of the initial evaluation.13,14 In gunshot wounds to the temporal bone, the inci dence of facial nerve injury is about 50%, with the verti cal segment being the most frequently damaged.3,13 Less frequent sites of injury include the tympanic segment, the main trunk just distal to the stylomastoid foramen, and the labyrinthine segment. At the time of surgery, two thirds to three fourths of patients were found to have com plete transection of the facial nerve13; interposition grafts and transmastoid decompression have been the primary modalities of treatment. Because residual bullet fragments can remain lodged in the temporal bone and can become a nidus for infection, a canal wall down or radical mastoid ectomy has been advocated as the approach of choice.13,15 Gunshot wounds of the temporal bone frequently result in loss of a segment of the nerve, usually in the vertical por tion, requiring interposition grafting for repair. There is also a high incidence of concomitant vascular and central nervous system injuries.
EXTRATEMPORAL INJURY TO THE FACIAL NERVE Facial paralysis after laceration or iatrogenic injury to the parotid region is best repaired primarily and as soon as the patient’s condition permits. If no loss of nerve substance has occurred, the nerve should be repaired by direct anastomosis. When nerve substance loss has occurred, an interpositional graft should be used.
PATIENT EVALUATION The presentation of facial nerve injuries greatly affects their management. The presence of a tightly enclosing fallopian canal around the intratemporal facial nerve makes the nerve much more susceptible to all types of trauma. Lack of any space to accommodate edema, which inevitably accompanies soft tissue trauma, leads to further neural injury. An injured nerve may not mani fest significant clinical dysfunction initially, but later, after sufficient edema has occurred to prevent axoplas mic flow, the injury manifests. Fisch6 and others16-18 have shown that the area of the fallopian canal with the least expansion room for neural swelling is in the region of the meatal foramen. Because most injuries to the facial nerve occur in the perigeniculate area just dis tal to the meatal foramen, the edema produced in facial nerve injury is quite critical in the pathophysiology of this disorder. The findings elicited from a careful history and physical examination on the patient’s presentation to the emergency department provide prognostic data and determine appropriate management. Eyewitness accounts of facial nerve function immediately after the injury and of any progression during transport to the emergency department are often unreliable and likely to be fraught with inaccuracy, but still can provide important infor mation, especially if an initial examination is impossible because of other life-threatening injuries. An accurate analysis of facial nerve function might be impossible if
350
OTOLOGIC SURGERY
A
B FIGURE 29-2. Longitudinal fracture of temporal bone. A, Fracture line is parallel to the long axis. B, CT scan of fracture.
the patient has been intubated and sedated as part of the primary survey, but every effort should be made to elicit some sort of facial movement, even a grimace, in a coma tose patient. Patients with any facial movement after the injury and before the onset of paralysis rarely need surgi cal intervention.19 A nerve with diffuse weakness in all branches can be observed clinically, and if some function persists, expectant management can be employed. If this situation deteriorates to total paralysis, electric testing
should be used to follow the nerve to ensure that total degeneration does not occur. Audiometric evaluation should also be performed as soon as the patient’s condition permits. The type of hearing loss can corroborate CT scan findings. If surgical exploration is warranted, the severity of the hearing loss in the affected ear guides the surgeon in determining the best approach,19-21 and serves as a baseline with which to compare postoperative results.22
Chapter 29 • Traumatic Facial Paralysis
351
FIGURE 29-3. Longitudinal fracture of temporal bone. EAC, external auditory canal. FIGURE 29-4. Location of lesion to facial nerve in 15 cases of longitudinal fractures. (From Coker NJ, Kendall KA, Jenkins HA, Alford BR:Traumatic intratemporal facial nerve injury: Management rationale for preservation of function. Otolaryngol Head Neck Surg 97:262-269, 1987.)
Prognosis Based on Electric Studies Fisch23 and Esslen16 have postulated that surgery can facilitate return of facial nerve function if performed before complete degeneration. A level of 90% degen eration or less, as determined by electroneuronography (ENoG), has been correlated with a uniformly good prognosis for return of function. If the nerve is nonfunc tional at the initial examination, the chance of a complete transection is high and will likely require surgery. Patients with complete facial paralysis at the initial examination are screened daily with nerve excitability testing. This test uses direct transcutaneous stimulation of the nerve on each side of the face and determines a stimulation threshold that produces perceptible move ment. The normal side is used as a control. If the thresh old difference between the normal and dysfunctional sides exceeds 2.5 mA, ENoG is performed regularly thereafter. ENoG uses transcutaneous supramaximal stimulation of the facial nerve while recording the evoked potential from anterograde stimulation in the periphery of the face.24 The maximal evoked response of the nerve is measured on each side by use of a nonfixed recording electrode
technique. A side-to-side comparison is made, with the normal side serving as the control. The percentage of degeneration is calculated as the difference between the two sides. More recent data have shown that a correla tion exists between ENoG and nerve excitability testing: a 90% degeneration score on ENoG correlates to about a 3.5 mA difference on nerve excitability testing.25 Chang and Cass,26 in a comprehensive and criti cal assessment of the available literature on facial nerve injury secondary to temporal bone trauma, proposed that serial ENoG be performed in any patient with acute-onset complete facial nerve paralysis or acute-onset incomplete paralysis that subsequently progresses to complete paraly sis. Only patients progressing to greater than 95% degen eration within 14 days are at risk for poor outcomes and should be offered facial nerve exploration. Patients with incomplete paralysis at presentation who do not progress to complete paralysis and patients who have a normal initial examination with subsequent delayed-onset facial nerve paralysis, whether complete or not, have an excel lent prognosis for recovery and can be observed. In a nonacute injury, ENoG can be relied on for up to 3 weeks, but after this period, a desynchronization
352
OTOLOGIC SURGERY
A
B FIGURE 29-5. Transverse fracture of temporal bone. A, Fracture line is perpendicular to the long axis. B, Radiography of transverse fracture.
(deblocking) of electrically evoked facial nerve dis charge can occur, preventing a single unified discharge of all neurons in the trunk. This effect occurs because of the differing time courses over which recovering neu rons re-establish electric conductivity and the capabil ity to conduct an action potential. At this stage, it is no longer possible to compare the diseased, asynchronously discharging side with the unaffected, synchronously
ischarging side, making accurate determination of the d severity of degeneration by this technique alone impos sible. If by 3 weeks after injury the patient has not pro gressed to 90% degeneration, however, it is unlikely surgery would be needed.26 After 3 weeks, electromyography (EMG) may be used to establish whether recovering axons are pres ent. This information is useful if delayed intervention is
Chapter 29 • Traumatic Facial Paralysis
353
FIGURE 29-6. Transverse fracture of temporal bone. EAC, external auditory canal. FIGURE 29-7. Location of lesion in three cases of transverse temporal bone fracture. (From Coker NJ, Kendall KA, Jenkins HA, Alford BR:Traumatic intratemporal facial nerve injury: Management rationale for preservation of function. Otolaryngol Head Neck Surg 97:262-269, 1987.)
being considered. Voluntary motor units and polypha sic potentials indicate that regeneration is in progress. Lack of the foregoing and fibrillation potentials indicate a fully degenerated nerve without evidence of ongoing recovery. EMG can detect signs of wallerian degenera tion, such as fibrillation potentials, only after the 10th day following nerve injury.6,7,21 A limitation of EMG is that it cannot be used initially following injury; how ever, it does not seem to have the reliability and repro ducibility issues that several authors have found with ENoG.27,28
Radiologic Evaluation Thin-cut CT scanning of the temporal bone is routinely required for evaluation of trauma to the facial nerve. Evaluation of bone detail often establishes the anatomy of the fractures and allows prediction of neural seg ment damage. The geniculate ganglion region is most frequently involved in blunt trauma, and nondisplaced fractures across the tegmen may be difficult to recognize on CT scan. If facial nerve injury is suspected, special temporal bone views are necessary because the resolution in standard brain CT scans is insufficient to delineate the intricate bony features of the fallopian canal.
Carotid arteriography is indicated if major vascular injury is suspected, particularly in gunshot injuries to the temporal bone. Traumatic pseudoaneurysms and arteriovenous fistulas are occasional sequelae and are readily identified by arteriography.
Timing of Surgery Timing and even the necessity of surgery in some cases of facial nerve injury remain controversial. McCabe29 suggested that exploration and repair be done at 21 days after injury based on studies of motor neuron proteosyn thetic activity levels and maximal repair activity at a neu ral anastomosis. More recent evidence does not support this theory, but does show a trend toward lower regenera tion rates with increasing time after onset of injury.30 Facial nerve paresis arising from blunt trauma to the peripheral portion of the facial nerve should be man aged expectantly. If no recovery is evident at the end of 6 months, reconstitution of the dysfunctional portion of the nerve may be required. Facial paralysis ensuing after laceration or iatrogenic injury to the parotid region is likely a transection, and is best repaired primarily and as soon as the patient’s condition permits. If a divided nerve cannot be repaired as soon as possible after the
354
OTOLOGIC SURGERY
injury, at least limited exploration of the wound should be performed to identify the severed ends of the nerve for subsequent repair. The use of an electric nerve stimulator can be helpful for 48 hours after injury for stimulation of the distal ends of the severed nerve. The regional twitch ing of facial musculature can be used as an aid to nerve identification.4 The role of surgery in delayed facial nerve injuries after blunt trauma remains controversial. Significant sequelae from the injury exist in these patients, however. Currently, the timing of surgery is largely determined by electrophysiologic testing. Surgery is performed when a significant level of degeneration of the facial nerve is evi dent (i.e., 90%) on ENoG. The reasoning behind this strategy is to prevent conversion of the neural injury from a Sunderland class II to a class III. The latter has pro nounced synkinesis in a large percentage of cases. The goal of surgery is to relieve the pressure on the nerve, per mitting the nerve to expand, and to decrease damage to the endoneural tubules from the external constriction. Patients with immediate onset of complete facial nerve paralysis after a penetrating injury should also be offered surgery as soon as their condition permits because of the high likelihood of facial nerve transection and the need for nerve repair.20,25 This approach allows facial nerve recovery to proceed without delay.
ASSOCIATED TRAUMA Other structures in the vicinity of the facial nerve may be injured from the trauma. These injuries may require either immediate or delayed management, depending on the circumstances.
Ossicular Damage Trauma to the ossicular chain frequently occurs in con cert with damage to the intratemporal facial nerve. The ossicles are most often damaged in longitudinal tempo ral bone fractures as the fracture line passes through the vicinity of the attic and posterosuperior external audi tory canal wall. Ossicles may be damaged by disloca tion brought about by relative movement of supporting structures or by inertial factors associated with sudden movements of the supporting structures.31 Many differ ent types of injury to the ossicles can occur, but the most frequent are dislocation of the incudostapedial joint, frac tures of the stapes crura, and subluxation of the stapes footplate.32 The malleus is rarely injured, but occasional fractures of the long process of the malleus are seen. The treatment of ossicular trauma depends on the nature of the injury. Fractures of the distal long process of the malleus near the umbo are treated by excision of the fractured segment and reconstruction of the drum with temporalis fascia. More proximal fractures near the head of the malleus must be treated by removal of the malleus
and incus and placement of a partial or total ossicular replacement prosthesis if the stapes is abnormal. Dislocation of the incus can be from incudostape dial disarticulation, incudomalleolar disarticulation, or both. Two approaches have been advanced for treatment of this condition. The traditional approach is an incus interposition, in which the incus is shaped into a strut with the former incudomalleolar joint area transformed into a notch for the malleus handle, engaging it near the region of the insertion of the tensor tympani. The tip of the former short process of the incus is fashioned into a cup to fit over the stapes capitulum. The overall length of the incus strut is determined by trial and error, and care is taken to ensure a slight amount of tension exists after placement between the malleus handle and stapes capitulum to enhance retention of the prosthesis, and to promote good sound conduction. Bone dust from the shaping operation should be allowed to remain on the incus remnant to encourage fixation of the prosthesis in good position. An alternative approach is to reduce the dislocation at the malleus and stapes with careful packing of the incus in reduction from the mastoid and middle ear aspects. Good results are obtained by some authors using this approach, particularly when the joint is not completely separated. We favor incus interposition in most of these cases, however. Fractures of the crura of the stapes render the ossi cle ineffectual as a conductor of sound to the inner ear, and a stapes replacement procedure of some type must be undertaken to restore acoustic function. Stapedec tomy or stapedotomy, at the surgeon’s preference, can be performed in cases of a normal footplate, but if the footplate is fractured or subluxed, stapedectomy may be required. Fenestration of the footplate, or most ossicular reconstruction, should not be performed in the presence of an infected middle ear cleft, and consideration should be given to using a staged reconstruction approach. The choice between autograft versus prosthetic ossicular reconstruction methods is usually resolved in favor of autograft because of extrusion problems and the greater incidence of long-term tolerance problems with prosthetic materials. When the incus is unavailable, how ever, an incus/stapes replacement prosthesis can be used effectively to reconstruct the ossicular chain. In cases in which the entire long process of the malleus is missing, a total ossicular replacement prosthesis or partial ossicu lar replacement prosthesis made of hydroxyapatite with a platform is used. The incidence of extrusion is greatly reduced by cartilage reinforcement of the platform sur face at the interface with the tympanic membrane.
Traumatic Otorrhea The presence of CSF drainage from the ear or nose of a patient with head trauma is not unusual and represents a defect in the dural covering of the brain.33 The inci dence of meningitis in patients with CSF leaks lasting
Chapter 29 • Traumatic Facial Paralysis
longer than 7 days is 23% to 88%,19,34-36 and mortality may be 10% in traumatic cases.35 Only otologic sources are considered in this chapter, but an anterior or middle cranial fossa defect in the sinus region should always be considered in the differential diagnosis of CSF rhinor rhea, especially in traumatic injuries. Fluid originating from a posterior or middle fossa fracture defect may enter the mastoid and middle ear and drain into the nose or the oropharynx. CSF otorrhea is frequently associated with longitudinal temporal bone fractures because the fracture defect may result in a dural tear, whereas a step-off in the external ear canal provides a direct channel for egress of the fluid from the middle ear. Clear fluid in the ear should alert the examiner to the presence of CSF, and the diagnosis should be straight forward. The diagnosis of CSF rhinorrhea is typically confounded, however, by the appearance of clear nasal secretions frequently accompanying nasal trauma. An informal test to make this differentiation is the halo test, whereby the fluid is placed on filter paper and allowed to diffuse. Blood in the sample is left behind, and a ring of clear fluid surrounds the red ring of blood products. The glucose levels in nasal secretions have also been used to identify CSF; high levels (>50 mg/100 mL) are con sidered to be indicative of CSF. The most sensitive and specific method of differentiation seems to be protein electrophoresis of the sample: the β2 fraction of transfer rin is specific to CSF.37 Potential help in localizing the source of the leak can be obtained from contrast-enhanced CT, radioiso tope studies, and intrathecal dye instillation. Intrathecal dye methods are used infrequently because of potential adverse reactions, whereas radioisotope studies are more useful in studies of the anterior skull base. A metrizamideenhanced CT scan provides the best localization for otorrhea because of the good detail of the bony defect usually accompanying the dural defect. Pneumocephalus is a dreaded, potentially treach erous complication of a defect in the dural protective barrier of the brain.38 It occurs when air is introduced into the cranial cavity via a fistula in the dura. Fre quently, pneumocephalus is accompanied by CSF leak age, but it may occur in the absence of clinical evidence of fluid leakage. The major difficulty in this entity is the potential formation of a tension pneumocephalus from a ball-valve defect in the dura. Continued accu mulation of air may induce intracranial hypertension, with resultant brain herniation. Aggressive management is required in CSF fistulas accompanied by persistent pneumocephalus. The treatment of CSF otorrhea must take into account the natural history of the entity and, in particular, the incidence of meningitis. Conservative management is possible because of the high probability that the condi tion will heal spontaneously, and because of the efficacy of present generation antibiotics in treating meningitis. A more recent meta-analysis has shown that the use of
355
rophylactic antibiotics has led to a significant reduction p in the incidence of meningitis.39 In a series of anterior cranial fossa CSF fistulas, 35% resolved by 24 hours, and 85% healed by 1 week.34 The most common organisms isolated from post-traumatic meningitis cases are Pneumococcus spp., and antibiotics are effective against them. Several measures may be implemented to aid spon taneous closure of CSF fistula. The most important treatment is bed rest and avoidance of activities that increase intracranial pressure, such as straining, lifting, or bearing down. Placement of an indwelling lumbar sub arachnoid drainage catheter may be useful in resistant cases. A closed drainage system allows removal of fluid daily and permits regular monitoring of CSF cell counts to allow early identification of meningitis. Avoidance of the need for repeated lumbar punctures is also a major benefit with these systems. The indications for operative management of CSF fistula include persistent leakage despite adequate con servative measures, recurrent meningitis, and persistent pneumocephalus.40 These cases may be approached via the middle ear, mastoid, or middle fossa, depending on the extent and localization of the defect. Generally, the defect should be identified and repaired with some form of soft tissue reinforced by bone, where possible. Discrete defects in the dura of less than 1 cm should be closed, and the area should be reinforced with fascia. If a mastoid tegmen defect is identified, the dural defect is repaired, reinforced with ���������������������������������������������� soft tissue, and finally ��������������������� a�������������������� bone graft is used to stabilize the repair in the face of continuous CSF pul sation pressure and gravity.
SURGICAL TREATMENT Preoperative Preparation Patients with traumatic facial paralysis are often quite ill because of the multiple sequelae of severe head trauma. Trauma sufficient to produce fracture of the base of the skull and resulting facial nerve injury often damages intracranial contents and other structures throughout the body. Assessment of the entire patient must be made, with particular emphasis on the cervical spine, airway, and circulation. A thorough neurologic examination to rule out intracranial pathology, such as hematoma or parenchymal injury to the brain, must be performed, and treatment must be rendered in a timely manner. Increased intracranial pressure is frequently seen as a result of head injury. High-dose steroids, hyperventila tion, and head elevation are frequently required for pres sure control, and intracranial pressure may be monitored by placement of an intraventricular catheter. Neurosur gical consultation should be obtained early in the treat ment of these disorders. After stabilization of the overall neurologic status and treatment of any acute medical problems, the patient can be prepared for surgery to treat the facial nerve.
356
OTOLOGIC SURGERY
Patient Positioning and Preparation for Surgery The patient is placed on the operating room table in the supine position with anesthesia located away from the head and neck region. In addition to the standard post auricular access, the head of the table must be available in case a middle cranial fossa approach is required. The head is placed on a head holder with a recess that allows positioning of the opposite ear without compression. The entire table, and not the head, is moved during the course of the procedure to prevent flexion vascular compromise to the opposite auricle, which can occur if the ear is folded when the patient’s head alone is moved. Approximately half of the scalp is prepared and the hair shaved after the patient is asleep. The entire hemi facial area to the midline is included in the preparatory area, and the lower neck and face are included if vas cular control will be required. Prophylactic antibiotics are given at this time, and 1:100,000 epinephrine solu tion is injected subcutaneously into the line of incision. If intraoperative EMG is to be used, the skin lateral to the oral commisure and lateral canthus is prepared with povidone-iodine (Betadine) swabs. Bipolar electrodes are placed in the orbicularis oris and oculi muscles, and the electrodes are gently tapped while recording to ensure that a discharge is elicited. A characteristic audio signal (“pop”) is elicited by this maneuver, which is used as a check for proper electrode and recording system func tion. The electrodes are sutured in place using sterile technique, and the rest of the preparation and draping are performed. Care is taken to drape the entire half of the face out into the sterile field so that direct obser vation of the face can be used to confirm electrophysi ologic events. No special instruments are employed in these procedures.
Technique Intratemporal Nerve Segment The surgical technique for mastoid facial nerve explora tion is discussed in Chapter 16. Details specific to posttraumatic treatment for paralysis are presented here. Surgery for intratemporal traumatic facial paralysis cen ters on exposure of the damaged segment of the fallo pian canal, facilitating surgical repair and providing the nerve room to expand. In treating damage to the nerve in the intratemporal segment, several factors must be taken into account, including status of individual nerve fibers, percentage of nerve loss, and length of nerve trunk loss. The techniques used for the repair of common lesions are shown in Figure 29-8. In injuries to the fallopian canal, the bone fragments may be in good reduction, betray ing the true extent of injury to the nerve. In these cases, an intraneural hematoma can be produced by free blood within the intact nerve epineurium. Blood staining is fre quently observed, but in some cases, the nerve appears
diffusely enlarged without areas of visible sheath staining. The nerve sheath should be incised using sharp, atrau matic technique and taking special care to preserve all underlying nerve fascicles (see Fig. 29-8A). The Ziegler ophthalmic knife (Storz) is useful for incising the nerve sheath. After the nerve sheath has been opened, the degree of nerve loss is assessed. If a significant portion of the nerve fibers seems to be divided, clean division of the remaining trunk followed by an interposition graft should be considered. Bony fragments impinging on the nerve sheath should be atraumatically removed, and the nerve should be assessed for intraneural hematoma and treated as discussed earlier (Fig. 29-8B). Complete transection situations can be repaired using nerve rerouting, when the anatomy permits, to avoid the need for an interposi tion graft (Fig. 29-8C and 29-9B). Gunshot wounds to the temporal bone with a tympanic or labyrinthine seg ment injury and a dead ear are the typical situation (Fig. 29-9).6 The interposition graft should be performed with a donor nerve graft of the appropriate diameter. In many situations, creation of a bony channel can enhance nerve and graft alignment without the need for multiple stabili zation sutures and attendant postoperative inflammation (Fig. 29-9A).5 In all cases, the nerve should be widely decompressed until nonedematous nerve is encountered. In situations in which the loss of mastoid segment nerve tissue is minimal, the nerve can be dissected from the sty lomastoid foramen region, and the parotid facial nerve can be mobilized proximally to obviate an interposition graft. Electric monitoring of the facial nerve is helpful in determining when surgical trauma is occurring to the nerve. If the nerve is completely transected, this tech nique is of no value, but if the injury has occurred within 3 days, use of intraoperative facial nerve EMG can be helpful because the nerve remains electrically active dis tally even though complete transection has occurred. Generally, if the results of ENoG do not reveal 100% degeneration, facial nerve monitoring may have some value because a few neurons would be able to produce discharge in the facial musculature, although this activity is not detectable clinically. Facial nerve EMG monitoring is performed by plac ing a pair of transcutaneous needle electrodes in the orbicularis oculi and oris muscles. The electrodes are connected directly by a short length of wire to a pream plifier and sent to the main recording unit. Processing and display options vary among commercially available units, but all process the signal to provide audio output of the signal and perhaps a visual record of the actual action potential or at least a visual indication of the amplitude of the response. The output signal is usually filtered, and artifact rejection mechanisms may eliminate signals pro duced by electrocautery and other electric noise sources in the operating room. We employ a Nicolet Compact Four (Nicolet Biomedical Instruments, Madison, WI) evoked potential unit, which requires a technician for operation
Chapter 29 • Traumatic Facial Paralysis
357
FIGURE 29-8. Surgical management of common facial nerve injuries in perigeniculate region. A, Intraneural hemorrhage. Inset, After opening nerve sheath. B, Bony fragment impingement. Inset, After fragment removal and opening of nerve sheath. C, Perigeniculate transection. Inset, Direct anastomosis after section of superior petrosal nerve.
and setup and for monitoring and recording the wave forms during the operative case. Audio output is avail able to the surgeon throughout the procedure, and care is taken to ensure that the same filter settings are used at all times so that the sound of facial nerve discharge can be identified consistently. Occasional discharges of
the nerve are not considered as significant as prolonged trains of discharge; the latter usually indicate a more severe, potentially nonreversible insult to the nerve. The face is always exposed via a clear plastic drape so that visual confirmation and correlation of electrophysiologic events can be obtained.
358
OTOLOGIC SURGERY
FIGURE 29-9. A, Loss of nerve length in vertical segment in a nonhearing ear. A trough is created based on the fallopian canal, which helps main tain the graft and nerve in position and minimizes the need for suture fixation. B, Extensive loss of tympanic and proximal vertical segment of facial nerve with rerouting from the internal auditory canal (IAC) to the vertical segment in a nonhearing ear. The superficial petrosal nerve is sectioned.
Chapter 29 • Traumatic Facial Paralysis
The most common area of damage to the intratem poral segment is in the perigeniculate and labyrinthine regions, but location of the precise area of damage should be ascertained on preoperative CT scans. The middle fossa approach to decompression is used when labyrin thine function must be preserved. If clinical suspicion is confirmed by preoperative CT scans, and intraopera tive findings warrant, decompression of the mastoid seg ment of the facial nerve can be performed via a mastoid approach (see Chapter 16). During middle fossa decom pression of the proximal facial nerve, the condition of the distal nerve can be determined by examination of the tympanic portion as it is exposed. If the nerve appears to be in good condition with no hemorrhage or edema evident, and no other indications exist for exploration of the mastoid portion of the nerve, exploration need not be done.4 In many cases of transverse fractures and gunshot wounds of the temporal bone, the patient has minimal or no residual hearing on audiometric testing. In these cases, a translabyrinthine approach to expose the entire facial nerve is possible without resorting to middle cranial fossa surgery. The labyrinth is removed, and the inter nal auditory canal is exposed. The nerve is identified in the tympanic segment adjacent to the lateral semicircular canal and in the labyrinthine segment distal to the inter nal auditory canal, and in this fashion, the entire intra temporal facial nerve is exposed. During exposure of the nerve in the vertical and tym panic segments, surgical trauma to an already diseased nerve should be minimized. This is accomplished by leaving a thin shell of bone over the nerve throughout the decompression. At the conclusion of the gross exposure, a dissector is used to lift off the thin shell of bone as a large piece, avoiding contact of the diamond burr with the nerve sheath. Small fragments of the shell sometimes may be left in place without harm, and the goal of nerve decompression is still attained. Repair should be performed at all anastomoses by use of neuroscopic technique, atraumatic handling of tissues, exact end-to-end anastomosis, tension-free clo sure, and the use of monofilament 9-0 or 10-0 suture.5 Interposition grafts are employed when nerve tissue loss would produce tension in a direct anastomosis, or when diseased nerve segments must be excised. The greater auricular nerve is adequate for defects less than 7 cm, and the sural nerve can bridge defects 30 cm long. The supe riority of perineurial over epineurial repair has not been shown, but meticulous techniques are useful in reconsti tution of the facial nerve.41
Extratemporal Nerve Segment Establishing a functionally intact nerve in the frontal and marginal mandibular distribution of the face is important because of the limited anastomotic interconnection from adjacent branches of the facial nerve in these regions.
359
Conversely, small branches in the midface region do not require extensive procedures to re-establish continu ity because of the rich anastomotic network that already exists. This quality leads to a much higher probability of a favorable outcome. Details of neural anastomosis tech niques can be found in Chapter 30. The technique for exposure of the peripheral nerve is based on techniques commonly used in parotid sur gery. Penetrating trauma to the nerve at the stylomastoid foramen region may prevent identification of the nerve proximally; the nerve would need to be found in the periphery and traced back proximally. Finding a branch of the marginal mandibular or frontal distribution is fre quently useful.42 If the exploration is performed within 2 days of the injury, the distal branches of the nerve can be stimulated by electric current, and observation of facial musculature twitching can help find the distal stump.
POSTOPERATIVE CARE Postoperatively, a modified mastoid dressing is applied, extending superiorly to apply pressure over the midfossa incision if needed. The dressing is changed daily, and the wound is examined for signs of infection or CSF collec tion in the wound. A dressing is left in place for 2 to 3 days in the case of midfossa surgery, and for 1 day for mastoid surgery in which the dura is not violated. Intravenous antibiotics are continued for 24 hours postoperatively. The dressing is left in place for 24 hours and then removed, and the wound is left exposed to allow easy observation. Any evidence of CSF collection in the wound is treated with pressure dressing and daily or twice-daily observation to ensure that the fluid has not recurred. Needle aspiration is performed if indicated. Patients are asked to walk on the evening of surgery or the first day after surgery. Total hospital stay is 4 to 5 days.
PITFALLS OF SURGERY The primary area of pathology in intratemporal injuries of the facial nerve is the labyrinthine segment. Frequently, the nerve must be exposed in this area if surgical inter vention is required. If preservation of hearing is a goal, a middle fossa surgical approach is required. Middle fossa surgery is a difficult technique because the anatomic approach is unfamiliar, and the indications for its use are infrequent. These factors contribute to the difficulty in mastering this technique. The lack of good alterna tive procedures to address the anatomic region of great est interest—the labyrinthine and meatal segment in a patient with intact hearing—makes maintenance of skills in middle fossa surgery a necessity for every active neu rotologist.
360
OTOLOGIC SURGERY
RESULTS
REFERENCES
The postoperative result depends largely on the severity of injury to the nerve and to a lesser extent the timing of intervention. Recovery is excellent for Sunderland’s neurapraxia (class I) or axonotmesis (class II) lesions, which involve intact endoneurial tubules. Primary anas tomosis or interposition is required for neurotmesis (class III-V) injuries because complete separation of the nerve has occurred.43 The best surgical result in this condition is a House-Brackmann grade III or grade IV (American Academy of Otolaryngology–Head and Neck Surgery), with the latter being more typical. Crush and compres sion injuries are Sunderland class III or IV lesions, with varying degrees of endoneurial and perineurial disrup tion and no gross disruption of the nerve. These inju ries have varying degrees of recovery, but grades III to V are usually attainable. The decision to resect and per form an anastomosis or simply to decompress is left to the surgeon; few physiologic data on which to base such a decision exist.43 The best prognosis is obtained with early intervention in the acute phase of degeneration and before the onset of continuing endoneurial degeneration and fibrosis, which impede the eventual re-establishment of competent axons.
1. Cannon C R , Jahrsdoerfer R A : Temporal bone fractures: Review of 90 cases. Arch Otolaryngol 109:285-288, 1983. 2. Ghorayeb BY, Yeakley JW: Temporal bone fractures: Longitudinal or oblique? The case for oblique temporal bone fractures. Laryngoscope 102:129-134, 1992. 3. Duncan NO 3rd, Coker NJ, Jenkins H A, Canalis R F: Gunshot injuries of the temporal bone. Otolaryngol Head Neck Surg 94:47-55, 1986. 4. Coker NJ, Kendall K A, Jenkins H A, Alford B R : Trau matic intratemporal facial nerve injury: Management rationale for preservation of function. Otolaryngol Head Neck Surg 97:262-269, 1987. 5. Coker NJ: Management of traumatic injuries to the facial nerve. Otolaryngol Clin North Am 24:215-227, 1991. 6. Fisch U: Facial paralysis in fractures of the petrous bone. Laryngoscope 84:2141-2154, 1974. 7. Lambert PR , Brackmann D E : Facial paralysis in lon gitudinal temporal bone fractures: A review of 26 cases. L aryngoscope 94:1022-1026, 1984. 8. Gurdjian E S, Lissner H R : Deformation of the skull in head injury studied by “stresscoat” technique: Quanti tative determinations. Surg Gynecol Obstet 83:219-233, 1946. 9. Yanagihara N, Murakami S, Nishihara S : Temporal bone fractures inducing facial nerve paralysis: A new classifica tion and its clinical significance. Ear Nose Throat J 76: 79-80, 83-86, 1997. 10. Dahiya R , Keller J D, Litofsky N S, et al: Temporal bone fractures: Otic capsule sparing versus otic capsule violat ing clinical and radiographic considerations. J Trauma 47:1079-1083, 1999. 11. Ishman S L , Friedland D R : Temporal bone fractures: Traditional classification and clinical relevance. Laryn goscope 114:1734-1741, 2004. 12. Little SC, Kesser BW: Radiographic classification of tem poral bone fractures: Clinical predictability using a new system. Arch Otolaryngol Head Neck Surg 132:13001304, 2006. 13. Shindo M L , Fetterman B L , Shih L , et al: Gunshot wounds of the temporal bone: A rational approach to evaluation and management. Otolaryngol Head Neck Surg 112:533-539, 1995. 14. Moore PL , Selby G, Irving R M : Gunshot injuries to the temporal bone. J Laryngol Otol 117:71-74, 2003. 15. Bento R F, de Brito RV: Gunshot wounds to the facial nerve. Otol Neurotol 25:1009-1013, 2004. 16. Esslen E : The acute facial palsies: Investigations on the localization and pathogenesis of meato-labyrinthine facial palsies. Schriftenr Neurol 18:1-164, 1977. 17. Lang J: Anatomy of the brainstem and the lower cranial nerves, vessels, and surrounding structures. Am J Otol 6(Suppl):1-19, 1985. 18. Ge X X, Spector GJ: Labyrinthine segment and genicu late ganglion of facial nerve in fetal and adult human tem poral bones. Ann Otol Rhinol Laryngol 90(Suppl):1-12, 1981. 19. Brodie H A, Thompson TC : Management of compli cations from 820 temporal bone fractures. Am J Otol 18:188-197, 1997.
COMPLICATIONS CSF leakage is the most significant complication of sur gical repair of traumatic facial nerve injuries. Leakage of CSF after middle fossa surgery is usually caused by access of the fluid to the air cell system of the mastoid. These cells can be exposed at the tegmen simply by lifting the dura over a natural dehiscence in the bone. Frequently, cells are opened when the dissection of the labyrinthine segment of the facial nerve is performed. Prevention is the best treatment, and careful attention must be given to blockage of any exposed air cell by the use of bone wax or fascia where needed. Infection of a middle fossa wound has never occurred in the senior author’s (H.A.J.) series.44 Because of the low incidence of postoperative infection, it has not been necessary to modify our surgical approach to produce a staggered incision through the skin and temporalis mus cles,44 but rather we use an incision straight through the skin down to bone. A conductive hearing loss can occur after middle cra nial fossa surgery if dura is allowed to contact the heads of the ossicles, producing a reduction in motion. This reduction is generally minimal, but may occasionally require revision surgery with interposition of a bone plate to prevent contact. Hearing loss and vestibular dysfunc tion are always possible with middle cranial fossa and mastoid surgery, and can be prevented by extensive ana tomic dissection in the laboratory and care in the operat ing room.
Chapter 29 • Traumatic Facial Paralysis 20. McKennan K X, Chole R A : Facial paralysis in temporal bone trauma. Am J Otol 13:167-172, 1992. 21. Darrouzet V, Duclos JY, Liguoro D, et al: Management of facial paralysis resulting from temporal bone fractures: Our experience in 115 cases. Otolaryngol Head Neck Surg 125:77-84, 2001. 22. Nosan D K, Benecke J E Jr, Murr A H : Current perspec tive on temporal bone trauma. Otolaryngol Head Neck Surg 117:67-71, 1997. 23. Fisch U: Prognostic value of electrical tests in acute facial paralysis. Am J Otol 5:494-498, 1984. 24. Gantz B J, Gmuer A A, Holliday M, Fisch U: Electro neurographic evaluation of the facial nerve: Method and technical problems. Ann Otol Rhinol Laryngol 93:394398, 1984. 25. Coker NJ, Fordice JO, Moore S: Correlation of the nerve excitability test and electroneurography in acute facial paralysis. Am J Otol 13:127-133, 1992. 26. Chang CY, Cass S P: Management of facial nerve injury due to temporal bone trauma. Am J Otol 20:96-114, 1999. 27. Adour K K, Sheldon M I, Kahn Z M : Maximal nerve excitability testing versus neuromyography: Prognostic value in patients with facial paralysis. Laryngoscope 90:1540-1547, 1980. 28. Sittel C, Guntinas-Lichius O, Streppel M, Stennert E : Variability of repeated facial nerve electroneurography in healthy subjects. Laryngoscope 108:1177-1180, 1998. 29. McCabe B F: Injuries to the facial nerve. Laryngoscope 82:1891-1896, 1972. 30. Barrs D M : Facial nerve trauma: Optimal timing for repair. Laryngoscope 101:835-848, 1991. 31. Hough JV, Stuart WD: Middle ear injuries in skull t rauma. Laryngoscope 78:899-937, 1968. 32. Bellucci R J: Traumatic injuries of the middle ear. Otolar yngol Clin North Am 16:633-650, 1983.
361
33. Canniff J P: Otorrhoea in head injuries. Br J Oral Surg 8:203-210, 1971. 34. Mincy J E : Posttraumatic cerebrospinal fluid fistula of the frontal fossa. J Trauma 6:618-622, 1966. 35. Leech PJ, Paterson A : Conservative and operative man agement for cerebrospinal-fluid leakage after closed head injury. Lancet 1:1013-1016, 1973. 36. Westmore G A, Whittam D E : Cerebrospinal fluid rhi norrhoea and its management. Br J Surg 69:489-492, 1982. 37. Oberascher G: Cerebrospinal fluid otorrhea—new trends in diagnosis. Am J Otol 9:102-108, 1988. 38. Andrews JC, Canalis R F: Otogenic pneumocephalus. Laryngoscope 96:521-528, 1986. 39. Brodie H A : Prophylactic antibiotics for posttraumatic cerebrospinal fluid fistulae: a meta-analysis. Arch Otolar yngol Head Neck Surg 123:749-752, 1997. 40. Marentette L J, Valentino J: Traumatic anterior fossa cerebrospinal fluid fistulae and craniofacial consider ations. Otolaryngol Clin North Am 24:151-163, 1991. 41. Orgel MG: Epineurial versus perineurial repair of peripheral nerves. Clin Plast Surg 11:101-104, 1984. 42. Lore J M Jr: The parotid salivary gland. In Lore J M Jr (ed): An Atlas of Head and Neck Surgery. Philadelphia, Saunders, 1988, pp 708-713. 43. Coker NJ, Jenkins H A, Psifidis A : Electrophysiological prognostication of acute facial nerve trauma. In Fisch U, Valavanis A, Yasargil MG (eds): Neurological Surgery of the Ear and the Skull Base. Amsterdam, Kugler & Ghedini, 1989, pp 355-362. 44. Holsinger FC, Coker NJ, Jenkins H A : Hearing preserva tion in conservation surgery for vestibular schwannoma. Am J Otol 21:695-700, 2000.
30
Facial Nerve Tumors Clough Shelton and Frank M. Warren Videos corresponding to this chapter are available online at www.expertconsult.com.
Tumors of the facial nerve are rare causes of facial paralysis.1-5 The two most common tumors are facial nerve neuromas, which are intrinsic to the nerve, and facial nerve hemangiomas, which are extraneural in origin. This chapter focuses on these two types of facial nerve tumors. The techniques discussed can also be applied to other facial nerve neoplasms. Because of their subtle presentation, facial nerve tumors may be difficult to diagnose and require a high degree of clinical suspicion.6,7 The presenting symptoms vary with the tumor location, size, and histology. Facial nerve neuromas that arise in the internal auditory canal (IAC) and cerebellopontine angle may manifest with a progressive sensorineural hearing loss similar to that caused by an acoustic tumor,8 and the true diagnosis may be established only at surgery.9,10 In a patient with a suspected acoustic tumor, the presence of coexisting facial nerve symptoms (rare with acoustic tumors) should warn the surgeon that a facial nerve tumor may be present.11,12 Facial nerve neuromas usually do not cause symptoms until they are fairly large and may manifest initially with hearing symptoms.12-14 Tumors arising in the middle ear may contact the ossicles and cause conductive hearing loss.15 In such cases, when a mass behind the tympanic membrane is invisible, the patient may be thought to have otosclerosis, and the correct diagnosis is made during tympanotomy for stapedectomy.6 When a middle ear mass is encountered in such a situation, a biopsy must not be done because a facial nerve paralysis is likely to develop postoperatively (see section on pitfalls of surgery).1,6,13,16,17 If the tumor erodes into the labyrinth (typically, the lateral semicircular canal at the external genu), the patient may present with dizziness.3 On examination, the fistula test result may be positive. Facial nerve hemangiomas characteristically cause severe symptoms when very small.18,19 Similar to facial nerve neuromas, hemangiomas can occur in the IAC and manifest as acoustic tumors.20 They usually cause a sensorineural hearing loss that is severe for the size of the tumor. Extremely small hemangiomas of the geniculate ganglion can cause profound facial nerve symptoms.21-23
Hemangiomas are extraneural and cause paralysis by compression,24 but the small size of the tumor can make diagnosis by imaging studies very difficult. A high degree of clinical awareness is required.25 Patients with either tumor type may present with facial nerve symptoms. Typically, these patients have recurrent Bell’s palsy, although the facial nerve recovery is less complete with each episode.14 Facial nerve twitching can be present in some patients, and others may have facial paralysis with no recovery or a slowly progressive facial paralysis.26,27 When evaluating a patient with atypical facial nerve symptoms, one must always suspect facial nerve tumor. Electric testing can help establish the diagnosis of a facial nerve tumor. Electroneuronographic results may be abnormal in patients with facial nerve tumors, even in the face of clinically normal function. Facial electromy ography may show a pattern of simultaneous denervation and reinnervation in patients with tumors.18 This pattern is seen in slowly progressive, pathologic processes, and would not be expected with a rapid insult to the facial nerve, such as seen in Bell’s palsy. High-resolution magnetic resonance imaging (MRI) with gadolinium may be sufficient to detect most facial nerve tumors (Fig. 30-1). Imaging of a facial nerve neuroma generally reveals a mass lesion and enlargement of the fallopian canal. Small facial hemangiomas at the geniculate ganglion may require high-resolution computed tomography (CT) (Fig. 30-2). These tumors exhibit characteristic bony changes, termed honeycomb bone.28 The medial extent of tumors arising at the geniculate ganglion can be assessed best with gadoliniumenhanced MRI.29 This medial extension has important ramifications regarding selection of surgical approach.
PATIENT SELECTION The timing of surgery is perhaps the most difficult aspect of planning. Many patients with facial nerve tumors have normal or nearly normal facial function. The best 363
364
OTOLOGIC SURGERY
FIGURE 30-1. MR image of facial neuroma (arrowhead).
FIGURE 30-2. CT scan showing hemangioma at geniculate ganglion with honeycomb bone (arrowhead).
a nticipated facial function after a facial nerve graft is a House-Brackmann grade III to IV.16 Waiting too long to remove the tumor can adversely affect the ultimate facial nerve results, however. Patients with a long-standing facial nerve paralysis have worse results after facial nerve grafting than patients who have grafting when they have normal facial function.9,13 Extraneural tumors, such as facial nerve hemangiomas, can be removed with preservation of facial nerve continuity; early surgery in such cases may offer the best hope for excellent facial nerve function. Labyrinthine fistula can develop from bony erosion by the tumor and can lead to deafness and dizziness if the tumor is neglected too long.13,30 For an older patient in poor health who has a small tumor and good facial function, observation may be the best strategy. Younger patients with good facial function may also be followed, but they must be aware of the possible risk to the ultimate facial nerve outcome and to their
hearing if surgery is delayed.12 In some cases, we follow patients until they show greater than 50% denervation by electroneuronography. When they reach this degree of denervation, we are concerned that facial nerve grafting after further loss of neural “firepower” would ultimately result in poor facial function. Another option is planned surgical decompression to reduce pressure on nerve fascicles.31 This is an option for selected patients with small tumors for whom a facial palsy is an unacceptable outcome because it can give them several more years of normal facial function.32 When the tumor becomes neurosurgically significant, it is then resected. Decompression can be accomplished through the middle fossa or translabyrinthine approach, depending on the segments of the facial nerve that are involved. Facial nerve neuromas in the IAC may be recognized during surgery for what was presumed to be an acoustic
Chapter 30 • Facial Nerve Tumors
365
tumor.33 Tumor in the labyrinthine fallopian canal and the course of the facial nerve over the posterior aspect of an “acoustic tumor” are clues to the existence of a facial nerve neuroma. Preoperatively, we counsel patients with acoustic tumors about this unlikely possibility. When a facial neuroma is diagnosed intraoperatively, the options are to resect the tumor and place a nerve graft, and decompress the tumor. In some cases, the diagnosis of a facial nerve tumor is not made before the continuity of the facial nerve has been lost.
Radiation Therapy and Radiosurgery Stereotactic radiotherapy and radiosurgery has been performed on several facial neuromas.34 Reports are anecdotal, and the outcome of this therapy for facial nerve tumors is completely unknown. Because this technique is investigational, treatment of facial tumors with radiosurgery should be performed only as part of an institutional review board–approved protocol. A case of malignant transformation after radiosurgery has been reported.35 This patient had partial excision of a facial nerve neuroma with postoperative radiation. Ten years later, the patient developed the cancer and had a poor outcome.
Patient Counseling The most important aspect of preoperative patient counseling before tumor resection is the expected postoperative facial function. Patients are told that a facial nerve graft will be needed, although patients with small he mangiomas are informed that it may be possible to preserve the continuity of the facial nerve. Patients should expect a postoperative facial paralysis lasting 6 to 12 months. Ultimate facial function after a facial nerve graft will not be “normal,” and the consequences of synkinesis are discussed. To describe a “good” result after facial nerve grafting, I tell my patients that they will look normal at rest and have active voluntary movement, but it will not be entirely symmetric motion. This asymmetry will be to the extent that family members will notice a difference, but a stranger on the street would not likely turn and stare (Fig. 30-3). Patients with a preoperative facial nerve palsy are told to expect worse postoperative facial function after nerve grafting. The longer the duration or the greater the severity of the preoperative palsy, the worse the ultimate result. The other potential risks and complications of surgery are also discussed according to the surgical approach needed. For middle fossa and translabyrinthine cases, the risks are similar to risks for patients with acoustic tumors treated through these approaches (see Chapters 48 and 49). Patients with tumor involvement of the ossicles may need ossicular reconstruction. In some cases, this procedure is best carried out at a second stage, and the patients are also informed of this possibility.
FIGURE 30-3. Patient postoperatively exhibiting House-Brackmann facial function grade IV 1 year after facial nerve graft.
SURGICAL TECHNIQUES Preoperative preparation, patient positioning, and instrumentation are as described in Chapter 1. For patients needing a facial nerve graft, the upper neck is also prepared for harvesting of the greater auricular nerve. The surgical approach is selected on the basis of tumor location, tumor size, and level of residual hearing. For small tumors around the geniculate ganglion in patients with good hearing, the middle fossa approach can be used. Besides providing access into the IAC, it can expose the horizontal facial nerve to approximately the midtympanic portion. Because of the limited access, however, sewing a graft into the IAC can be difficult, and posterior fossa access is not provided. For patients with larger tumors or with poor hearing and involvement of the IAC, posterior fossa, or geniculate ganglion, the translabyrinthine approach can be used. This approach provides wide access for tumor removal, and allows placement of a facial nerve graft. Involvement of the horizontal and vertical facial nerve may also require a transmastoid facial recess approach. Erosion of the external auditory canal may necessitate a canal wall down procedure. The facial nerve can be followed into the parotid if tumor extension necessitates the exposure. A staged procedure may be required in a chronically infected ear that requires an intracranial surgical approach for tumor removal.
Middle Fossa Approach The initial middle fossa approach is carried out as described for acoustic tumors in Chapter 48. After the craniotomy window is made, the temporal lobe is supported by a retractor, and the greater superficial petrosal nerve and the skeletonized superior semicircular canal are identified. The greater superficial petrosal nerve is followed posteriorly to the geniculate ganglion (Fig. 30-4). For patients with tumor involvement in this area, extreme
366
OTOLOGIC SURGERY
Chapter 30 • Facial Nerve Tumors
367
FIGURE 30-4. Exposure of geniculate ganglion through middle fossa approach. A, Greater superficial petrosal nerve and superior semicircular canal are skeletonized. B, Tumor is encountered at geniculate ganglion. FIGURE 30-5. Internal auditory canal and labyrinthine facial nerve are exposed medial to geniculate ganglion. FIGURE 30-6. After removal of tegmen bone, the facial nerve can be exposed to approximately the midtympanic portion. FIGURE 30-7. Facial nerve neuroma is seen through transmastoid approach. Facial recess is opened, and thin eggshell of bone over distal facial nerve is removed.
care must be taken during the dissection because a tumor can distort the anatomy. The IAC is dissected, and the labyrinthine facial nerve is identified. The amount of IAC exposure varies with the degree of tumor extension in that area, and the need for access to place a graft (Fig. 30-5). The tegmen bone is also removed to expose the tympanic facial nerve (Fig. 30-6). Care is taken not to touch the ossicular heads with a burr during this removal because a sensorineural hearing loss would result.
Transmastoid Approach If additional distal exposure is needed, a transmastoid approach to the facial nerve can also be performed, as described in Chapter 16. The facial recess is opened, and the nerve is exposed to the stylomastoid foramen. Around the facial nerve, it is best to thin the bone over it with a diamond burr, to use copious irrigation, and to remove the overlying “eggshell” of bone with an instrument such as a sickle or a whirlybird knife (Fig. 30-7). Because of tumor involvement or the need for additional room, the malleus head and incus can be removed and ossicular reconstruction performed at the end of the procedure.
Translabyrinthine Approach The translabyrinthine approach is carried out as described in Chapter 49, and the facial nerve is skeletonized. The amount of medial bone removal varies with the tumor extent and the need for access for facial nerve grafting. The bone over the geniculate ganglion can be removed anteriorly to expose the greater superficial petrosal nerve (Fig. 30-8).
Tumor Removal Tumor removal is accomplished with sharp and blunt dissection. Posterior fossa tumor removal is handled in a way similar to the technique used for acoustic tumors. After the tumor has been removed, frozen sections are taken from the remaining nerve ends to ensure total tumor removal. In some extraneural tumors (hemangiomas), a plane can be developed between the tumor and the nerve, allowing facial nerve preservation. In selected neuroma cases, some authors have reported tumor removal while maintaining partial continuity of the facial nerve.36
Graft Material The greater auricular nerve serves as an excellent donor nerve for grafting. The diameter match is good, and the donor deficit is minimal. Adequate length can be obtained to graft from the IAC to the stylomastoid foramen. The greater auricular nerve can be found between the angle of the mandible and the tip of the mastoid process on the lateral surface of the sternocleidomastoid muscle,37 posterosuperior to the external jugular vein (Fig. 30-9). The nerve can be harvested through an oblique skin incision placed in a skin crease. By dissecting the nerve from the posterior aspect of the sternocleidomastoid muscle, ample length is obtained. Another option for a donor graft is the sural nerve (Fig. 30-10), which has a larger diameter and length than the greater auricular nerve. The ends of the graft are trimmed sharply, and excess fibrous tissue and epineurium are removed from the nerve stumps. Some surgeons advocate reversing the nerve direction to prevent axons from growing out branches of the graft.
Nerve Grafting Within the temporal bone, if enough of the fallopian canal remains as a trough, placement of the graft in approximation to the nerve end is usually sufficient (Fig. 30-11). The anastomosis can be packed into place with microfibrillar collagen (Avitene), which forms a clot over the anastomosis.38 When sutures are needed, two or three epineurial sutures of 9-0 polypropylene (Prolene) on a cardiovascular needle work well (Fig. 30-12). The suture length is trimmed to 6 inches to make it easier to work under the microscope, and background material is cut, wetted, and placed beneath the anastomosis to give a smooth working surface. We prefer to cut the graft slightly longer than needed to ensure that no tension exists on the anastomosis. If possible, the graft is placed along the normal course of the facial nerve so that if further surgery is required, it can be easily identified. For patients requiring an ossicular reconstruction at a second stage, the graft of the horizontal facial nerve is placed in the epitympanum, superior to the oval window, so that it would not need to be manipulated during later ossicular reconstruction. The placement of an anastomosis in the IAC or posterior fossa is much more difficult than an anastomosis performed within the temporal bone because the intracranial
368
OTOLOGIC SURGERY
Chapter 30 • Facial Nerve Tumors
369
FIGURE 30-8. Translabyrinthine approach with exposure of geniculate ganglion and greater superficial petrosal nerve. Tumor extends from internal auditory canal (IAC) to tympanic facial nerve. FIGURE 30-9. The greater auricular nerve is located on the lateral surface of the sternocleidomastoid muscle, posterior to the external jugular vein. It lies between the tip of the mastoid and the angle of the jaw.
FIGURE 30-10. The sural nerve can be found on the lateral surface of the ankle posterior to the lateral malleolus. FIGURE 30-11. Facial nerve graft is placed in apposition to proximal facial nerve stump at labyrinthine segment. Remaining fallopian canal holds graft in place, and sutures are not needed. FIGURE 30-12. Nerve graft is held in approximation to distal mastoid facial nerve stump with two epineurial sutures.
facial nerve has no epineurium (Fig. 30-13). Usually, a single through-and-through suture of 9-0 Prolene is sufficient.39 The suture is placed on the graft side first. When the suture is placed through the intracranial facial nerve stump, a fenestrated suction is used to support the facial nerve, and the needle is passed into a side hole of the suction (Fig. 30-14).40
Rerouting Depending on the location and length of the facial nerve defect, it may be possible to remove the facial nerve from its canal and reroute it to gain extra length. This maneuver subjects the nerve to a great deal of manipulation and may interfere with its blood supply. For a defect in the cerebellopontine angle, rerouting (Fig. 30-15) gains additional nerve length, however, and allows primary anastomosis. The wounds are closed as described in Chapter 29, and abdominal fat packing is used when indicated. The standard mastoid dressing is used. The postoperative care for tumor removal from the translabyrinthine and middle fossa approaches is similar to care given to patients with acoustic tumors removed through these approaches (see Chapters 48 and 49). An important aspect of postoperative care is attention to the paralyzed eye. This topic is discussed in detail in Chapter 61. Depending on the anticipated time and quality of facial function recovery, some patients require the placement of an upper eyelid spring or gold weight.
PITFALLS OF SURGERY Facial nerve tumors, particularly neuromas, tend to involve the nerve over long distances of its course. Underestimation of tumor extent is an important problem and can be overcome by high-resolution imaging studies.28 High-resolution CT scanning of the temporal bone can show enlargement and erosion of the fallopian canal and indicate tumor involvement. When needed, gadoliniumenhanced MRI shows extension of the tumor into the IAC and posterior fossa.41 Some cases in our series were diagnosed elsewhere when a middle ear mass was encountered during tympanotomy to correct a conductive hearing loss. A biopsy
must not be done on a middle ear mass in this situation. Of nine patients with tumors for which biopsies were done by the referring surgeons in our series, 88% experienced a postbiopsy facial paralysis.42 For a patient going to surgery for a stapedectomy, the development of a postoperative facial paralysis from the biopsy can be extremely disturbing. When a middle ear mass is encountered, the ear should be closed, and a radiologic, not histologic, evaluation should be done. A high-resolution CT scan showing enlargement of the fallopian canal yields a diagnosis of facial nerve tumor, and subsequent surgery can be planned based on the extent of the tumor. The labyrinth is at risk for fistulization by bone erosion caused by tumor growth. Particularly vulnerable are the inferior surface of the lateral semicircular canal at the external genu and the cochlea near the geniculate ganglion. A patient with a facial nerve tumor (particularly one that does not involve the IAC) who exhibits dizziness or sensorineural hearing loss may have a labyrinthine fistula. Such fistulas frequently can be detected on high-resolution CT. The surgeon should always be alert to the possibility of encountering a fistula during surgery, however, and must be prepared to repair it. In patients with a known fistula preoperatively, temporalis fascia can be harvested in anticipation of covering the fistula. For a nerve graft to function, the anastomosis must be made to viable neural tissue at the nerve stump and not to residual tumor. Frozen section examination of the nerve stump is necessary not only to validate complete tumor removal, but also to ensure that no tumor remains at the anastomotic site to impede axonal regrowth.
RESULTS Facial Nerve Neuromas A review of a series of 64 facial nerve neuromas revealed that the facial nerve required repair (graft or primary anastomosis) in 72% of these patients.42 Of cases with 1 year or more of follow-up, 83% of the patients who had undergone repair had facial function graded as HouseBrackmann grade IV or better (Table 30-1). For patients with preservation of facial nerve continuity, 70% had postoperative facial function of House-Brackmann grade
370
OTOLOGIC SURGERY
FIGURE 30-13. Facial nerve graft is placed into internal auditory canal through translabyrinthine approach. Single through-and-through suture holds nerve ends together.
FIGURE 30-14. Fenestrated suction supports intracranial facial nerve. The suture is passed through the nerve and into a side hole of the suction. FIGURE 30-15. Facial nerve is rerouted in translabyrinthine approach to gain approximately 1.5 cm of length.
III or better at 1 year (see Table 30-1). Only one patient in long-term follow-up experienced no recovery of postoperative facial function. Hearing was preserved near the preoperative level (within 10 dB pure tone average and 16% speech
iscrimination score) in 53% of patients, excluding d patients undergoing translabyrinthine tumor removal. Fistulas of the inner ear occurred in 16 patients; 11 involved a lateral semicircular canal, and 4 involved the cochlea. Hearing was lost in four patients, all with fistulas.
Chapter 30 • Facial Nerve Tumors TABLE 30-1 Postoperative Facial Nerve Function
for 34 Facial Nerve Neuroma Patients with >1 Year Follow-up
Facial Nerve Function*
FACIAL NERVE STATUS Repaired (n = 24)
Intact (n = 10)
Total (N = 34)
— — 9 11 4 0
3 1 3 1 1 1
3 1 12 12 5 1
I II III IV V VI *House-Brackmann
classification.
TABLE 30-2 Postoperative Facial Nerve Results for
23 Facial Nerve Hemangioma Patients with >1 Year Follow-up
Facial Nerve Grade* I II III IV V VI
FACIAL NERVE STATUS Repaired
Intact
— — 2 8 1 2
7 2 — 1 — —
*House-Brackmann
classification. From Shelton C, Brackmann DE, Lo WW, Carberry JN: Intratemporal facial nerve hemangiomas. Otolaryngol Head Neck Surg 104:116-121, 1991.
Facial Nerve Hemangiomas Facial nerve hemangiomas tend to occur at the geniculate ganglion or in the IAC, with facial nerve repair most often required for tumors at the geniculate ganglion. In a report of 23 facial nerve hemangiomas, 47% required facial nerve repair.18 Most of the repaired nerves achieved a House-Brackmann grade III or IV by 1 year or more after surgery (Table 30-2). Hearing was maintained near the preoperative level in 64% of patients. One patient had postoperative anacusis from a cochlear fistula caused by the tumor.
COMPLICATIONS AND MANAGEMENT Postoperative cerebrospinal fluid leaks occurred in 6% of the neuroma series, and postoperative meningitis developed in 3%. Meningitis did not occur in patients who also experienced a cerebrospinal fluid leak. The management of these complications is similar to management after acoustic tumor removal by the translabyrinthine and middle fossa approaches (see Chapters 48 and 49).
371
REFERENCES 1. Pulec J L : Facial nerve neuroma. Laryngoscope 82:11601176, 1972. 2. Rosenblum B, Davis R, Camins M : Middle fossa facial schwannoma removed via the intracranial extradural approach: Case report and review of the literature. Neurosurgery 21:739-741, 1987. 3. Pearman K, Welch AR: Schwannoma of the intratemporal facial nerve: Case report. J Laryngol Otol 94:779-784, 1980. 4. Liliequist B : Neurinomas of the labyrinthine portion of the facial nerve canal: A report of two cases. Adv Otorhinolaryngol 24:58-67, 1978. 5. Pulec J L : Facial nerve tumors. Ann Otol Rhinol Laryngol 78:962-983, 1969. 6. Jackson CG, Glasscock M E III, Hughes G, Sismanis A : Facial paralysis of neoplastic origin: Diagnosis and management. Laryngoscope 90:1581-1595, 1980. 7. Wiet R J, Pyle G M, Schramm D R : Middle fossa and intratemporal facial nerve neuromas. Otolaryngol Head Neck Surg 104:141-142, 1991. 8. Lee K S, Britton B H, Kelly D L Jr: Schwannoma of the facial nerve in the cerebellopontine angle presenting with hearing loss. Surg Neurol 32:231-234, 1989. 9. King TT, Morrison AW: Primary facial nerve tumors within the skull. J Neurosurg 72:1-8, 1990. 10. Dort JC, Fisch U: Facial nerve schwannomas. Skull Base Surg 1:51-55, 1991. 11. Nelson R A, House WF: Facial nerve neuroma in the posterior fossa: Surgical considerations. In Graham M D, House WF (eds): Disorders of the Facial Nerve. New York, Raven Press, 1982, pp 403-406. 12. Bailey C M, Graham M D: Intratemporal facial nerve neuroma: A discussion of five cases. J Laryngol Otol 97:65-72, 1983. 13. O’Donoghue G M, Brackmann DE, House JW, Jackler R K : Neuromas of the facial nerve. Am J Otol 10:49-54, 1989. 14. Pillsbury HC, Price HC, Gardiner L J: Primary tumors of the facial nerve: Diagnosis and management. Laryngoscope 93:1045-1048, 1983. 15. Neely JG, Alford B R : Facial nerve neuromas. Arch Otolaryngol Head Neck Surg 100:298-301, 1974. 16. Lipkin A F, Coker NJ, Jenkins H A, Alford B R : Intracranial and intratemporal facial neuroma. Otolaryngol Head Neck Surg 96:71-79, 1987. 17. Wiet R S, Lohan A N, Brackmann D E : Neurilemmoma of the chorda tympani nerve. Otolaryngol Head Neck Surg 93:119-121, 1985. 18. Shelton C, Brackmann D E, Lo WW, Carberry J N: I ntratemporal facial nerve hemangiomas. Otolaryngol Head Neck Surg 104:116-121, 1991. 19. Mangham C A, Carberry J N, Brackmann D E : Management of intratemporal vascular tumors. Laryngoscope 91:867-876, 1981. 20. Pappas DG, Schneiderman TS, Brackmann D E, et al: Cavernous hemangiomas of the internal auditory canal. Otolaryngol Head Neck Surg 101:27-32, 1989. 21. Fisch U, Ruttner J: Pathology of intratemporal tumors involving the facial nerve. In Fisch U (ed): Birmingham, AL, Aesculapius, Facial Nerve Surgery, 1977, pp 448456.
372
OTOLOGIC SURGERY
22. Balkany T, Fradis M, Jafek BW, Rucker NC : Hemangioma of the facial nerve: Role of the geniculate capillary plexus. Skull Base Surg 1:59-63, 1991. 23. Isaacson B, Telian S A, McKeever PE, Arts H A : Hemangiomas of the geniculate ganglion. Otol Neurotol 26:796802, 2005. 24. Ylikoski J, Brackmann D E, Savolainen S : Pressure neuropathy of the facial nerve: A case report with light and electron microscopic findings. J Laryngol Otol 98:909914, 1984. 25. Glasscock M E, Smith PG, Schwaber M K, Nissen A J: Clinical aspects of osseous hemangiomas of the skull base. Laryngoscope 94:869-873, 1984. 26. Tew J M Jr, Yeh H S, Miller GW, Shahbabian S : Intratemporal schwannoma of the facial nerve. Neurosurgery 13:186-188, 1983. 27. Valvassori G E : Neuromas of the facial nerve. Adv Otorhinolaryngol 24:68-70, 1978. 28. Lo WW, Brackmann D E, Shelton C : Facial nerve hemangioma. Ann Otol Rhinol Laryngol 98:160-161, 1989. 29. Kertesz TR , Shelton C, Wiggins R H, et al: Intratemporal facial nerve neuroma: Anatomical location and radiological features. Laryngoscope 111:1250-1256, 2001. 30. Sanna M, Zini C, Gamoletti R , Pasanisi E : Primary intratemporal tumours of the facial nerve: Diagnosis and treatment. J Laryngol Otol 104:765-771, 1990. 31. Angeli S, Brackmann D: Is surgical excision of facial nerve schwannomas always indicated? Otolaryngol Head Neck Surg 117:s144-s147, 1997. 32. Liu R , Fagan P: Facial nerve schwannoma: Surgical excision vs conservative management. Ann Otol Rhinol Laryngol 110:1025-1029, 2001.
33. Murata T, Hakuba A, Okumura T, Mori K : Intrapetrous neurinomas of the facial nerve: Report of three cases. Surg Neurol 23:507-512, 1985. 34. Litre C F, Gourg G P, Tamura M, et al: Gamma knife surgery for facial nerve schwannomas. Neurosurgery 60:853-859, 2007. 35. Shirazi M A, Leonetti J P, Marzo S J, Anderson D E : Surgical management of facial neuromas: Lessons learned. Otol Neurotol 28:958-963, 2007. 36. Perez R , Chen J M, Nedzelski J M : Intratemporal facial nerve schwannoma: A management dilemma. Otol Neurotol 26:121-126, 2005. 37. Pulec J L : Facial nerve grafting. Laryngoscope 79:15621583, 1969. 38. House JW: Facial nerve tumors and grafting. In Brackmann D E (ed): Neurological Surgery of the Ear and Skull Base. New York, Raven Press, 1982, pp 77-80. 39. Barrs D M, Brackmann D E, Hitselberger WE : Facial nerve anastomosis in the cerebellopontine angle: A review of 24 cases. Am J Otol 5:269-272, 1984. 40. Arriaga M A, Brackmann D E : Facial nerve repair techniques in cerebellopontine angle tumor surgery. Am J Otol 13:356-359, 1992. 41. Wiggins R H, Harnsberger H R , Salzman K L , et al: The many faces of facial nerve schwannoma. AJNR Am J Neuroradiol 27:694-699, 2006. 42. Alavi S, Shelton C : Facial nerve neuromas: Comparison of results of facial nerve preservation versus facial nerve repair. Trans Pacific Coast Otoophthalmol Soc 74�������������������������� , 1993.
31
Surgery for Cochlear Implantation William M. Luxford and Robert D. Cullen Videos corresponding to this chapter are available online at www.expertconsult.com.
There have been remarkable advances in cochlear implan tation. Although the first attempt to stimulate the auditory system electrically occurred nearly 2 centuries ago, the development of a cochlear prosthesis to restore hearing to patients with sensorineural hearing loss has happened only over the past 5 decades. The early pioneering work of Simmons, Michaelson, and House provided the stimulus to encourage others, including Bonfai, Chouard, Clark, Eddington, and the Hochmairs.1 The initial acceptance of cochlear implants was slow; safety and efficacy were the concerns of the early investigators, and the greatest champions of the implant were the patients themselves. Time and technology have increased the benefits gained by most patients from their cochlear implants. As a result, cochlear implants have become the most successful pros thesis ever used to attempt to restore a sensory deficit. As of the writing of this chapter, more than 120,000 pa tients have received cochlear implants worldwide. More than 50,000 patients have received cochlear implants in the United States. In the United States, the Food and Drug Admin istration (FDA) regulates the manufacturers of cochlear implants. The FDA has approved devices from Advanced Bionics Corporation (Sylmar, CA), Cochlear Corporation (New South Wales, Australia), and MED-EL Corporation (Innsbruck, Austria) for use in adults and children. Incremental improvements in technology have resulted in the approval of multiple devices over the years. As perfor mance with cochlear implants has improved over the years, the criteria for cochlear implantation have expanded (Table 31-1). In more challenging situations, bilateral cochlear implantation has been shown to provide additional benefit beyond a single cochlear implant, and is now considered an accepted medical practice outside of research protocols.2 Prospective studies continue to examine the benefits of bilateral cochlear implantation in adults and children.
PATIENT SELECTION Candidate selection for cochlear implantation has evolved as the devices and patient performance have improved. Generally, adults and children with bilateral severe to
profound sensorineural hearing loss, who receive little to no benefit from conventional hearing aids, are in good physical and mental health, and possess the aptitude and motivation to participate meaningfully in the audi tory rehabilitation program are potential candidates for cochlear implantation.
MEDICAL EVALUATION A thorough medical evaluation is necessary to detect problems that may contraindicate surgery or interfere with the patient’s ability to complete postimplantation rehabilitation. Rarely, the cause of hearing loss may be a contraindication to surgery. Aplasia of the cochlea and aplasia of the cochlear nerve are contraindications to surgery because there are no auditory nerve elements to receive stimulus from the implant. A prior history of meningitis with cochlear ossification or fibrosis does not exclude the patient from implantation, but may neces sitate modification of the surgical technique. Two crucial factors influencing auditory perfor mance after cochlear implantation include age of onset of deafness and duration of profound hearing loss. The ideal adult candidate has profound acquired sensorineu ral hearing loss. A period of auditory experience adequate for development of normal speech, speech perception, and language offers a significant advantage in learning to use the implant. These postlingually deafened patients represent most adults undergoing cochlear implantation. In these patients, there is a significant correlation between duration of profound hearing loss and performance.3 Patients with prolonged auditory deprivation receive similar auditory information as do other implant patients, but are unable to use the information as effectively; this is thought to be due to the loss of central auditory process ing. A few adult implant recipients are congenitally or prelingually deafened, with prolonged auditory depriva tion and little to no experience with sound. These patients typically have greater difficulty assimilating the new audi tory information, and generally have performed less well than patients with some degree of auditory memory. 373
374
OTOLOGIC SURGERY
TABLE 31-1 Criteria for Cochlear Implantation 1985
1990
1998
Current
Age Onset of hearing loss
Adults Postlingual
Degree of hearing loss
Profound
Adults; children (>2 yr) Postlingual adults; pre- and postlingual children Profound
Adults; children (>18 mo) Adults and children pre- and postlingual Adults—severe-profound; children—profound
Adult open-set sentences
0%
0%
<40%
Pediatric speech scores
NA
0% open-set
<20% (MLNT/LNT); lack of auditory progress
Adults; children (>12 mo) Adults and children pre- and postlingual >2 yr old—moderate to profound; <2 yr old— profound <50% in implanted ear; <60% contralateral ear <30% (MLNT/LNT); lack of auditory progress
NA, not applicable; MLNT = Multisyllabic Lexical Neighborhood Test; LNT = Lexical Neighborhood Test.
The presence of auditory memory associated with delayed onset of deafness also correlates with improved speech and language skills in children after cochlear implantation. Congenitally deafened children receive the maximal auditory benefit from cochlear implantation by undergoing cochlear implantation at the youngest age possible. Current FDA guidelines require a child to be at least 1 year old, have profound sensorineural hearing loss, show no benefit from conventional hearing aids, and be free from medical contraindications. In addition to these criteria, the commitment of the family and the child’s educational setting to postimplantation rehabilita tion is a crucial determinant of successful implant use.
Audiologic Assessment The basic testing battery included aided and unaided pure tone and speech detection thresholds, environmen tal sound recognition, and speech perception tests. Gen erally, a patient is considered an audiologic candidate for cochlear implantation if the following criteria are met: bilateral severe to profound sensorineural hearing loss, with an unaided three-frequency pure tone average of 70 dB hearing loss or poorer in the better ear, a speech discrimination score of less than 50% in the ear to be implanted, and a speech discrimination score of less than 60% in the best-aided binaural condition. Audiologic screening in children requires auditory brainstem response testing and otoacoustic emissions test ing, in addition to conventional behavioral audiometry. An aggressive trial of appropriately fitted hearing aids and intensive auditory and speech training are integral com ponents of candidacy assessment in children. The global evaluation of cochlear implant candidacy in children is considerably more challenging than in adults, and is best approached by a dedicated team comprising speech and hearing professionals, social workers, psychologists, and educators. The ultimate candidacy of a child is deter mined not only by a demonstrated physiologic need, but also by the strength of the child’s social and educational background.
Physical Examination It is important to identify preoperatively any external or middle ear diseases, including perforations of the tym panic membrane, that must be treated before cochlear implantation. For young children, the size of the implant in relation to the size of the child’s skull must be evalu ated, and issues involved with skull maturation must be considered. The distance between the cochlear prom ontory and the mastoid cortex, the approximate sites of the electrode array, and the receiver-stimulator increases about 1.7 cm from birth to adulthood, with one half of the increase occurring during the first 2 years of life.4 The electrodes must be long enough to tolerate the increase in height and width of the skull that occurs with the child’s growth. The accommodation occurs through the gradual straightening of the excess electrode length within the aircontaining mastoid cavity.5
Imaging Preoperative imaging completes the candidacy evaluation process and assists in surgical planning. Choice of imag ing modality varies at individual implant centers. Highresolution computed tomography (CT) scanning of the temporal bone and magnetic resonance imaging (MRI) are currently used. CT scanning provides excellent detail of inner ear morphology, the course of the intratemporal facial nerve, and aeration of the mastoid. Limitations of CT include the inability to evaluate retrocochlear pathol ogy and cochlear patency in patients with fibrosis of the cochlea as a precursor to cochlear ossification. The abil ity to detect an absent cochlear nerve also is limited. MRI addresses these limitations,6 but this imaging modality incurs additional expense and may require general anes thesia in young patients. Complete agenesis of the cochlea and an abnormal acoustic nerve resulting from congenital malformation, trauma, or surgery are contraindications for cochlear implant placement.7 Cochlear hypoplasia (present in Mondini’s deformity) is not a contraindication for
Chapter 31 • Surgery for Cochlear Implantation
cochlear implantation. Adults and children with incom plete congenital cochlear malformations have received implants successfully.8 Ossification or fibrous occlusion of the cochlea or the round window does not exclude a patient from implantation, but it may influence outcome. Occlusion of the cochlea may lead to partial insertion of the electrode carrier. MRI has become more useful than CT in the evaluation of the membranous inner ear in detecting cochlear fibrosis.
375
2.
3.
1.
B.
Promontory Evaluation A few implant teams perform an electric stimulation test at either the promontory or the round window mem brane.9 A positive response is a perception of sound on stimulation. Some investigators do not believe that such testing is crucial in the selection of candidates because patients with a negative response, particularly at the promontory, may respond to intracochlear stimulation with an implant. Promontory stimulation may be helpful in cases of suspected cochlear nerve agenesis.
SELECTION OF EAR Selection of the side for implantation is governed by sev eral factors. The most patent cochlea is typically chosen for implantation. It is generally believed that the ear with the shortest duration of deafness may serve as the best ear for implantation. If the patient uses a hearing aid in only one ear (the side that is perceived as the better hearing ear), implanting the contralateral “worse” ear does not have a negative impact on performance. When no spe cific factors lead to the choice of one ear over the other, the ear on the side of the dominant hand is chosen to facilitate device manipulation. If there is no difference acoustically, we place the implant in the better surgical ear, based on CT evalua tion. The side with the least ossification or fibrosis within the scala tympani is chosen. Results from vestibular tests should be given the least weight in the selection of the side of cochlear implantation. The ear with the least caloric response should receive the implant. For patients undergoing bilateral cochlear implantation, the need to select one ear over the other is obviated.
SURGICAL TECHNIQUE The cochlear implant should be implanted only by quali fied surgeons specifically trained to perform the proce dure. Although specific details of implantation vary based on the device implanted, the general approach to implan tation is the same. The implant is inserted via a transmastoid facial recess approach to the round window and scala tympani. Placement of cochlear implants in children is essentially
A.
1. Postauricular crease 2. Skin incision 3. Periosteal incision
A. 1 cm between skin-periosteal incisions B. 1 cm between periosteal/implant incision
FIGURE 31-1. Incision. The shape of the incision may vary from surgeon to surgeon, but it is important to maintain 1 to 2cm from the edge of the implant to the incision. A small postauricular incision (2) is planned close to the postauricular crease (1). A periosteal inci sion (3) is made 1 cm posterior to the skin incision. By separating these incision lines, device extrusion is less likely to occur. The periosteum is undermined to allow placement of the implant in a tight sub-periosteal pocket. This serves to immobilize the implant tightly against the skull.
the same as in adults. By age 1 year, the mastoid antrum and facial recess are adequately developed. In patients with mastoid cavities or absent posterior ear canal, oblit eration of the mastoid cavity with blind sac closure of the external auditory canal is preferably done at the time of disease removal. Cochlear implant placement is per formed at a second stage approximately 4 to 6 months later. Surgery is performed with the patient under general anesthesia with the use of continuous intraoperative facial nerve monitoring. Perioperative antibiotics are routinely administered before making the initial incision. Many different incisions have been designed to allow placement of the receiver-stimulator (Fig. 31-1). Generally, the skin flap must be large enough to cover the receiver-stimulator completely. The length of the incision and size of the skin flap have been reduced over the years. By reducing inci sion and flap size, there is less interruption of the vascular supply to the operative field; this seems to correlate with fewer wound and flap complications. Skin and periosteal incisions should overlap by at least 1 cm; the skin incision should be at least 1 cm anterior to the front edge of the receiver-stimulator (Fig. 31-5). The internal receiver-stimulator is placed into a well that is drilled into the calvaria of the skull. This well helps to prevent migration of the implant. The antenna of the
376
OTOLOGIC SURGERY
Short head of incus
Lateral semicircular canal
Chorda tympani Buttress Facial recess opened Round window
FIGURE 31-2. Cortical mastoidectomy. Superior and posterior mar gins are not saucerized. Middle fossa plate, sigmoid sinus, antrum, lateral semicircular canal, and short process of the incus are identified.
implant is placed in the subcutaneous tissue of adults, but remains in the subperiosteal layer in children. This differ ence in technique allows for better coupling of the exter nal magnet to the internal device in adults, while reducing skin and scalp complications in children. Device manufac turers and many surgeons recommend securing the device to the calvaria with sutures; however, in our experience, this is unnecessary. Great care must be taken to place the internal receiver-stimulator far enough away from the auricle that a behind-the-ear processer may be worn with out compressing the skin against the internal device. All implant systems use the transmastoid, facial recess approach to the round window and scala tympani. The mastoidectomy is done using conventional burrs and suc tion-irrigation techniques (Fig. 31-2). In contrast to sur gery for chronic otitis media, the superior and posterior mastoid cortical margins are not saucerized. The margins can be undercut to create a bony overhang that stabi lizes the coiled electrode within the mastoid cavity. The bone removal extends back to the sigmoid, but retraction of the sigmoid is not required unless it is far forward. Enough of the bone in the attic is removed so that the top of the incus can be clearly seen. The incus should not be dislocated or removed because this method does not increase surgical exposure. The short process of the incus and its buttress are important landmarks in the develop ment of the facial recess. The posterior bony ear canal wall is thinned without exposing the overlying vascular strip tissue. Thinning of the bony ear canal is necessary because in viewing the round window area, the direction of vision is parallel to the external auditory canal. The facial recess is opened (Fig. 31-3). The facial nerve is carefully skeletonized at the mastoid genu to avoid exposure of the nerve sheath. When the facial recess is opened, the lip of the round window niche is usually visible just inferior to the stapedius tendon and
Sinodural angle
FIGURE 31-3. Developing facial recess. The facial nerve in its verti cal portion is identified. Care should be taken not to expose the nerve sheath.
oval window. To get a good look at the round window, one must open the facial recess more inferiorly and poste riorly. Usually, removing the chorda tympani is unneces sary for adequate visualization of the round window niche area. If the facial recess is very restricted, the chorda can be removed, but because the chorda enters the middle ear at the level of the annulus, care must be taken not to damage the tympanic membrane. With a small diamond stone and continuous suctionirrigation, the lip of the niche is removed, and the round window membrane comes into clear view. To avoid pos sible damage to the facial nerve, the diamond stone is not rotated when it is passed through the facial recess to the round window area. In cases in which the round win dow niche is almost hidden under the pyramidal process, one must drill forward and thin the promontory until the scala tympani is entered. The cochleostomy is made at the anteroinferior aspect of the round window. The size of the cochleos tomy is determined by the device being implanted and any insertion tools required to implant the device. Gener ally, a small cochleostomy is preferred; however, it must be large enough to allow full insertion of the electrode array without causing trauma to the electrode contacts. After insertion of the electrode array, the cochleostomy is sealed with muscle or fascia. In some cases, the round window niche and mem brane are replaced with new bone growth.10 This con dition is more common in patients whose deafness is attributable to meningitis, rather than to other diseases.11 In these cases, the surgeon must drill forward along the basal coil for 4 to 5 mm. Usually, the new bone is white and can be demarcated from the surrounding otic capsule. Following this white plug of bone with the drill usually leads to the patent scala, allowing placement of the elec trode array.12 If new bone growth completely obliterates
Chapter 31 • Surgery for Cochlear Implantation
377
Remove “hook” of round window
FIGURE 31-4. Opening into scala tympani. Anteroinfe VII
the scala tympani, the surgeon can drill superiorly and possibly enter a patent scala vestibuli.13,14 In cases with complete ossification of the cochlea, the surgeon may choose to perform a canal wall down mastoidectomy and close the ear canal. A trough around the modiolus is cre ated, and the electrode is placed in it.15,16 When drilling the round window niche or attempting to create an opening into the scala tympani through new bone growth, the surgeon must direct the burr anteriorly toward the nose (Fig. 31-4). Drilling superiorly may lead to damage to the basilar membrane and osseous spiral lamina, which may result in the loss of ganglion cells. If the surgeon directs the burr inferiorly, a hypotympanic air cell may be accidentally entered, and the active electrode would be placed improperly into this area. Postopera tively, these cases may fail to stimulate. Temporal bone imaging shows that the active electrode is extracochlear. Revision surgery with placement of the electrode array into the scala tympani remedies this situation. If the sur geon is uncertain of the placement of the electrode, an intraoperative anteroposterior transorbital plain film can be taken to check the electrode position.
rior area of the true round window membrane is removed to allow entry into scala tympani beyond hook region of the cochlea.
The techniques required to insert the cochlear implant electrode array vary based on electrode design. Despite this, some general principles correspond to all inser tion techniques. Every attempt must be made to avoid insertion trauma to the cochlea. Excessive force during insertion may lead to damage to inner ear structures and distortion of the shape of the electrode. Both of these problems can adversely affect outcomes. A well is drilled into the calvarium to accommo date the internal receiver package [Figure 5]. The well is drilled deep enough to prevent migration of the device; however, decompression of the dura is generally avoided in order to prevent intracranial complications. The postauricular flap is closed in layers without drainage. For purposes of hemostasis, only bipolar elec trocautery should be used after insertion of the cochlear implant device. A standard mastoid dressing is applied and removed the day after surgery. Surgery is usually performed on an outpatient basis in both adults and children. Patients return for the first postoperative visit in one week. Approximately 2 to 4 weeks after surgery, allowing for resolution of the edema
378
OTOLOGIC SURGERY
A
Top view
B Bottom view
C
C.I.
FIGURE 31-5. Electrode insertion and placement of internal receiver package. Electrodes are gently inserted into the cochlea. It is important that force not be used to insert the electrode into the cochlear in order to prevent intracochlear trauma. After elec trode placement the internal receiver package is placed into the well drilled into the calvarium (A). The implant is placed within the well to help prevent migration of the receiver package (B). While it is important to drill this well deep enough to prevent migration of the device, decompression of the dura is not necessary (C).
in the postauricular flap, fitting the patient with the signal processor begins.
COMPLICATIONS The risks of the implant procedure are the same as the risks for chronic otitis media surgery and include infec tion, facial paralysis, cerebrospinal fluid drainage, men ingitis, and the usual risks of anesthesia. All of these risks are remote in chronic otitis media surgery and have proved to be so in implant surgery as well.14,17 Failure of the incision to heal and associated minor infections are the most common problems associated with implant surgery.18 In a few patients in whom the internal receiver has been placed too close to the wound’s edge, or in patients in whom the flap over the internal receiver is too thin, the internal receiver has extruded. As noted earlier, at least 1 cm must be maintained between the incision and the edge of the internal receiver. The ideal thick ness for the flap is 6 to 7 mm. Although too thin a flap may necrose, too thick a flap may diminish device per formance by decreasing the transcutaneous transmission of information. Problems with the facial nerve can occur as the result of surgery and stimulation.19 Good surgical landmarks must be maintained when the facial recess is created. Although the facial nerve is identified, it usually does not have to be uncovered with the facial recess approach. Adequate irrigation at the facial recess must be main tained to help dissipate the heat generated by the turning shaft of the diamond burr used to create the exposure of the round window and entrance into the scala tympani, especially in drill-out cases. To help alleviate the problem
1.
2. 3.
Side view
1. Well drilled in skull 2. Thin bone over dura 3. Dura
of the drill shaft turning against the facial nerve, a drill such as the Skeeter® drill manufactured by Medtronic, Inc., Jacksonville, FL could be used. The width of the drill bit shaft is smaller, and a sleeve around most of the length of the shaft protects the surrounding tissues as well. In cases in which the heat from the rotating burr shaft has led to facial nerve problems, the paralysis has been temporary, and has resolved over several weeks to several months. Facial nerve paralysis has occurred in patients with congenital malformation of the cochlea and in patients who have undergone radical mastoidectomy many years before the implant procedure. In these cases, the facial nerve, either because it is congenitally displaced or because it is exposed by previous surgery, is at greater risk.20 The use of facial nerve monitoring may decrease the chance of facial nerve trauma. Cerebrospinal fluid drainage has occurred at the internal receiver site and the cochlea. In some patients, the temporal squama can be quite thin. In these cases, creating an adequate seat for the internal receiver package requires bony dissection down to the dura. If small dural tears occur, they should be covered with temporalis fas cia, and the fascia should be supported with the internal receiver. After insertion of the intracochlear electrode, the cochleostomy is closed with strips of temporalis fascia to prevent perilymphatic fistula development. A gush of cerebrospinal fluid is more likely to occur in patients with congenitally malformed inner ears. In these patients, small pieces of fascia can be placed through the facial recess opening into the eustachian tube orifice to occlude the eustachian tube temporarily. This proce dure should be done before opening the scala tympani. Removal of the incus allows improved visualization of the eustachian tube orifice if packing of the eustachian
Chapter 31 • Surgery for Cochlear Implantation TABLE 31-2 Centers for Disease Control
and Prevention Guidelines for Immunization in Patients with Cochlear Implants
Age (yr)
Vaccine*
Streptococcus pneumoniae <2 2-5 >5
Prevnar Prevnar and Pneumovax Pneumovax
Haemophilus influenzae <5
Hib
*Prevnar,
7-valent pneumococcal conjugate vaccine (PCV-7); Pneumovax, 23-valent pneumococcal polysaccharide vaccine (PPV-23); Hib, Haemophilus influenzae type B conjugate.
tube is required. Also, closure of the wound should be done in layers, without a drain. Lumbar drainage is rarely necessary. Meningitis is serious complication that is more fre quent in patients with cochlear implants. Children with cochlear implants are at an elevated risk for meningitis compared with the general population.21,22 In addition, children implanted with a positioner distributed with Advanced Bionics Corporation devices before July 2002 are at greater risk than children without positioners. Streptococcus pneumoniae is the primary bacterial isolate in this patient group. In light of this increased risk, it is strongly advised that all patients undergoing cochlear implanta tion be vaccinated against S. pneumoniae in accordance with current Centers for Disease Control and Prevention guidelines (Table 31-2). Cochlear implant recipients, along with their families, educators, daycare providers, and health care providers, need to be aware of the signs of meningitis. Otitis media should be promptly diagnosed and treated with the appropriate antibiotics.
REVISION COCHLEAR IMPLANTATION With any surgically implanted medical device, there is an inherent risk that the device will necessitate revision sur gery for the patient. This risk exists for cochlear implants as well. The pattern of events that lead to revision surgery can be broken down into the following categories: device failure, suspected device malfunction, medical or surgical problems, and the desire for technologic upgrades. Revision surgery accounts for approximately 5% to 10% of adult and pediatric cochlear implant surgeries in many centers. The reasons for revision fall into the above-listed categories. Performance has been shown to be good after revision surgery, including cases that require replacement of the cochlear implant.23 Special precautions must be taken during revision cochlear implant surgery. The primary goal in revision surgery for device failure is to remove the implanted
379
device safely and replace it with a functional implant. A secondary goal should be to remove the device as intact as possible for evaluation by the device manufac turer to assess the cause of failure. Monopolar cautery should not be used during the procedure because this may damage the device further and conduct high electric current into the cochlea. Every effort should be made to remove the internal receiver and the proximal elec trode intact. The distal electrode should not be removed intact with the rest of the device, however. Often a thick fibrous sheath surrounds the electrode and extends from the facial recess to the cochleostomy. If the electrode is removed prematurely, this fibrous pathway may be lost, making reinsertion very difficult or impossible. For this reason, we recommend severing the electrode array at the facial recess. The original electrode should be removed only after the new device is seated in its seat, and the electrode is ready to be inserted. In this way, the new electrode may be passed immediately into the original cochleostomy. Sealing of the cochleostomy and wound closure is performed in the same manner as the initial procedure.
BILATERAL COCHLEAR IMPLANTATION Most recipients of cochlear implants have undergone unilateral cochlear implantation. Because there are defi nite advantages to binaural acoustic hearing, one would expect to find advantage with bilateral cochlear implan tation. This has been found to be the case in numer ous studies.24-27 Patients undergoing bilateral cochlear implantation have shown improved sound localization and speech detection in noise. Binaural summation, or the ability of two ears to hear the sound that a single ear cannot, has also been shown in most binaural cochlear implant recipients. There are valid arguments against bilateral cochlear implantation. Providing two cochlear implants to patients may produce a financial strain on individuals or health care systems or both. There is also the hope that patients might benefit from future scientific advances, such as hair cell regeneration. Placement of a single cochlear implant would likely eliminate that opportunity for that ear; placement of bilateral cochlear implants would completely eliminate that opportunity for the patient. Although this potential should be recognized, it must also be recognized that there seems to be a window of opportunity to introduce useful auditory information to the auditory system. Bilateral cochlear implantation may be safely per formed during a single operative setting (simultaneous) or in a delayed approach (sequential). The operative technique is the same as for unilateral cochlear implan tation, with certain special considerations. It is crucial that monopolar cautery not be used once a cochlear implant has been performed. Bipolar electrocautery,
380
OTOLOGIC SURGERY
the Hemostatix Thermal Scalpel System (Shaw Scal pel; Hemostatix Medical Technologies, Bartlett, TN), and the Harmonic Scalpel (Ethicon Endo-Surgery, Inc., Cincinnati, OH) are suitable alternatives to monopolar cautery. Extra care must be taken to avoid injury to the chorda tympani nerve, particularly if it was sacrificed on the initial side. Bilateral chorda tympani denervation can produce excessive loss of taste. Symmetric placement of the implant receiver-stimulators should also be per formed.
COMBINED ELECTROACOUSTIC STIMULATION Cochlear implantation has been very successful in pro viding meaningful auditory information to patients who meet the current cochlear implant criteria. Criteria for cochlear implantation continue to expand as the per formance of patients with cochlear implants improves. As patients with more residual hearing have been implanted, it has been found that hearing may be pre served in a many of these patients. This has led to the development of a new model of cochlear stimulation. Shorter cochlear implant electrodes are placed atraumat ically into the basal turn of the cochlea to provide elec tric stimulation for high-frequency sound representation. When low-frequency hearing is preserved, a hearing aid is used to provide acoustic information to these frequen cies. This strategy has been termed hybrid stimulation or combined electroacoustic stimulation (EAS).28 EAS has been shown to provide improvement of recognition of speech in noise and improved appreciation of music.29 Both of these factors were linked to the ability to distinguish fine pitch differences because of preserved residual low-fre quency acoustic hearing. EAS is currently under FDA trial and is being per formed only by cochlear implant centers as a part of the FDA study. Patient selection criteria are different from the criteria for standard cochlear implantation. For stan dard cochlear implantation, monosyllabic word scores are less than 40%. For EAS candidates, the monosyl labic word score range is 10% to 60% in the ear to be implanted, and 80% in the contralateral ear. Pure tone thresholds incorporate near-normal low-frequency hear ing with a high-frequency loss greater than 65 to 70 dB at 2 kHz and higher frequencies. A trial of well-fit hear ing aids is essential to show that acoustic amplification provides little benefit. Appropriate surgical technique must be used to maximize the potential for hearing preservation. Implan tation for EAS must be treated as carefully as a stapes procedure where preservation of hearing is essential. The scala tympani must be accessed through an appropriately placed cochleostomy or round window insertion to avoid damage to the basilar membrane and stria vascularis. A slow rotation diamond burr should be used for the
c ochleostomy. Contact between the burr and the endos teum of the cochlea should be limited because the energy transmitted to the cochlea is similar to drill contact with the incus.30 The electrode must be oriented appropri ately in relation to the modiolus. Aspiration of perilymph must be avoided, and a gentle insertion technique must be used to avoid rapid changes to cochlear fluids during insertion. Electrode design is also essential for the EAS strat egy. The ideal electrode would provide extensive cover age of the cochlea and produce little to no shift in hearing thresholds. Shorter electrodes with fewer contacts pro vide less auditory benefit than longer electrodes.29,31 Generally, the longer the electrode, the more potential there is for trauma during insertion with resultant hear ing loss. Optimally, the electrode should be long enough to provide similar benefit as a standard cochlear implant if a total hearing loss occurs as a result of implantation. This ideal electrode design has not yet been established. A single one-size-fits-all electrode may not be appropri ate for all patterns and degrees of hearing loss.
REFERENCES 1. Luxford W, Brackmann DE : The history of cochlear implants. In Gray RF (ed)����������������������������������������������� : Cochlear Implants. London, Croom Helm, 1985, pp 1-26. 2. Balkany T, Hodges A, Telischi F, et al: William House Cochlear Implant Study Group: Position statement on bi lateral cochlear implantation. Otol Neurotol 29:107-108, 2008. 3. Dowell RC, Mecklenburg DJ, Clark G M : Speech rec ognition for 40 patients receiving multichannel cochlear implants. Arch Otolaryngol Head Neck Surg 112:10541059, 1986. 4. O’Donoghue G M, Jackler R K, Jenkins WM, et al: Cochlear implantation in children: The problem of head growth. Otolaryngol Head Neck Surg 94:78-81, 1986. 5. Marks D R , Jackler R K, Bates GJ, et al: Pediatric cochlear implantation: Strategies to accommodate for head growth. Otolaryngol Head Neck Surg 101:38-46, 1989. 6. Adunka O F, Jewells V, Buchman C A : Value of computed tomography in the evaluation of children with cochlear nerve deficiency. Otol Neurotol 28:597-604, 2007. 7. Shelton C, Luxford WM, Tonokawa L L , et al: The nar row internal auditory canal in children: A contraindica tion to cochlear implants. Otolaryngol Head Neck Surg 100:227-231, 1989. 8. Slattery WH 3rd, Luxford WM : Cochlear implanta tion in the congenital malformed cochlea. Laryngoscope 105:1184-1187, 1995. 9. Waltzman S B, Cohen N L , Shapiro WH, et al: The prog nostic value of round window electrical stimulation in cochlear implant patients. Otolaryngol Head Neck Surg 103:102-106, 1990. 10. Green J D Jr, Marion M S, Hinojosa R : Labyrinthitis ossificans: Histopathologic consideration for cochlear implantation. Otolaryngol Head Neck Surg 104:320-326, 1991.
Chapter 31 • Surgery for Cochlear Implantation 11. Novak M A, Fifer RC, Barkmeier JC, et al: Labyrinthine ossification after meningitis: Its implications for cochlear implantation. Otolaryngol Head Neck Surg 103:351-356, 1990. 12. Balkany T, Gantz B, Nadol J B Jr: Multichannel cochlear implants in partially ossified cochleas. Ann Otol Rhinol Laryngol Suppl 135:3-7, 1988. 13. Steenerson R L , Gary L B : Multichannel cochlear implantation in children with cochlear ossification. Am J Otol 20:442-444, 1999. 14. Webb R L , Lehnhardt E, Clark G M, et al: Surgical com plications with the cochlear multiple-channel intraco chlear implant: Experience at Hannover and Melbourne. Ann Otol Rhinol Laryngol 100:131-136, 1991. 15. Gantz B J, McCabe B F, Tyler R S : Use of multichannel cochlear implants in obstructed and obliterated cochleas. Otolaryngol Head Neck Surg 98:72-81, 1988. 16. Lambert PR , Ruth R A, Hodges AV: Multichannel cochlear implant and electrically evoked auditory brain stem responses in a child with labyrinthitis ossificans. L aryngoscope 101:14-19, 1991. 17. Cohen N L , Hoffman R A : Complications of cochlear implant surgery in adults and children. Ann Otol Rhinol Laryngol 100:708-711, 1991. 18. Haberkamp TJ, Schwaber M K : Management of flap necrosis in cochlear implantation. Ann Otol Rhinol Lar yngol 101:38-41, 1992. 19. Niparko J K, Oviatt D L , Coker NJ, et al: Facial nerve stimulation with cochlear implantation. VA Cooperative Study Group on Cochlear Implantation. Otolaryngol Head Neck Surg 104:826-830, 1991. 20. House J R 3rd, Luxford WM : Facial nerve injury in cochlear implantation. Otolaryngol Head Neck Surg 109:1078-1082, 1993. 21. Reefhuis J, Honein M A, Whitney CG, et al: Risk of bacterial meningitis in children with cochlear implants. N Engl J Med 349:435-445, 2003.
381
22. Biernath K R , Reefhuis J, Whitney CG, et al: Bacterial meningitis among children with cochlear implants beyond 24 months after implantation. Pediatrics 117:284-289, 2006. 23. Cullen R D, Fayad J N, Luxford WM, et al: Revision cochlear implant surgery in children. Otol Neurotol 29:214-220, 2008. 24. Litovsky R , Parkinson A, Arcaroli J, et al: Simultaneous bilateral cochlear implantation in adults: A multicenter clinical study. Ear Hear 27:714-731, 2006. 25. Litovsky RY, Johnstone PM, Godar S P: Benefits of bilateral cochlear implants and/or hearing aids in chil dren. Int J Audiol 45(Suppl 1):S78-S91, 2006. 26. Litovsky RY, Johnstone PM, Godar S, et al: Bilateral cochlear implants in children: Localization acuity mea sured with minimum audible angle. Ear Hear 27:43-59, 2006. 27. Peters B R , Litovsky R , Parkinson A, et al: Importance of age and postimplantation experience on speech per ception measures in children with sequential bilateral cochlear implants. Otol Neurotol 28:649-657, 2007. 28. Soriano A F, Helfrich B, Chan DC, et al: Synergistic effects of new chemopreventive agents and conventional cytotoxic agents against human lung cancer cell lines. Cancer Res 59:6178-6184, 1999. 29. Gantz B J, Turner C, Gfeller K E : Acoustic plus electric speech processing: Preliminary results of a multicenter clinical trial of the Iowa/Nucleus Hybrid implant. Audiol Neurotol 11(Suppl 1):63-68, 2006. 30. Pau HW, Just T, Bornitz M, et al: Noise exposure of the inner ear during drilling a cochleostomy for cochlear implantation. Laryngoscope 117:535-540, 2007. 31. Gantz B J, Turner C : Combining acoustic and electrical speech processing: Iowa/Nucleus Hybrid implant. Acta Otolaryngol 124:344-347, 2004.
32
Implantable Hearing Devices William H. Slattery III
It is estimated that 32 million Americans have a hearing loss severe enough to cause problems with communication. The severity of this loss ranges from mild, in which the individual may have difficulty only when significant background noise is present, to profound, in which the patient is unable to understand and communicate even in the quietest situation. Most hearing loss is sensorineural in nature. Less than 10% of hearing-impaired individuals have losses correctable by medical or surgical means. Although hearing loss may affect all frequencies, it is most common for individuals to have some component of high-frequency sensorineural hearing loss. Patients with a mild loss may receive no treatment other than instructions on modifying their acoustic environment to diminish background noise when selecting seating arrangements to allow improved listening conditions. Individuals with conductive losses usually have the opportunity to undergo surgical therapy to have the loss corrected. In some cases, owing to congenital abnormalities or infection, surgical correction is not an option. Individuals with profound sensorineural loss may receive cochlear implants, which provide electric stimulation directly to the cochlear nerve. Most individuals with sensorineural hearing loss must rely on amplification to provide a better means of improving communication. This amplification is most commonly accomplished with conventional air conduction hearing aids. In recent years, hearing aids have decreased in size, and improved microchips have been developed, resulting in improved signal processing capabilities. Hearing aids have become easier to program with digital processors; they are able to provide much better individualization of amplification according to each patient’s needs. Despite the improvement in conventional hearing aid technology, only approximately 20% of hearing-impaired individuals who could receive benefit from amplification actually use these devices. There are many reasons why patients do not wear a hearing aid. One of the most common reasons is that conventional air conduction hearing aids do not provide enough amplification of the sounds the individual actually
wishes to hear. In addition, they produce troubling amplification of unwanted sounds, especially when background noise is present. Another major problem with conventional hearing aids is the limited high-frequency response. High-frequency output is limited by the highfrequency feedback that occurs when the microphone and receiver of conventional air conduction hearing aids are in close proximity. Although new circuitry (feedback cancellation) has diminished some of these complaints, some users of conventional air conduction hearing aids still complain of feedback problems. Poor fit can result when the ear canal enlarges because of long-term use of the device, causing feedback problems. As microchip technology has allowed hearing aids to become smaller to improve their cosmetic acceptance, it can result in greater problems with feedback as the microphone and speaker are placed closer together. Limited frequency output may produce problems with distortion, and, in particular, limited high-frequency amplification restricts sound localization abilities. Cosmesis is still a major complaint of conventional air conduction hearing aids. Many patients believe that wearing such a device indicates a disability or carries a stigma of old age. Ear molds may cause an occlusive effect in blocking residual hearing, and produce discomfort, skin irritation, and the potential for increased infections of the ear canal (external auditory canal [EAC]). Discomfort is especially noted with smaller devices that fit more medially in the EAC. The occlusion effect, in addition to being uncomfortable, can result in loss of low-frequency information. There is a significant breakdown rate in conventional hearing aids because of wax in the receiver. Cerumen not only can occlude the receiver of the hearing aid, but also may cause wax impaction by the medial displacement of wax in the EAC. Implantable hearing devices have become a viable alternative to conventional hearing aids. There are two main types of implantable devices. The more commonly used device is the bone-anchored cochlea stimulator (Baha). Used in Europe since 1977, the Baha received FDA clearance in 1996 as a treatment for conductive 383
384
OTOLOGIC SURGERY
and mixed hearing losses. In 2002, the Baha was also approved for treatment of unilateral sensorineural hearing loss or single-sided deafness. The other class of implantable devices is the middle ear implant, in which stimulation of the ossicular chain or direct stimulation to the cochlea is performed. This chapter reviews both of these devices.
REQUIREMENTS OF IMPLANTABLE DEVICES The goal of the middle ear implant is to improve the efficiency of the device by improving gain, sound quality, hearing, and noise, and eliminate acoustic feedback. The implant should improve the quality of life. The ideal implantable hearing device should be easy to implant. It should cause no trauma or damage to the normal auditory system. It is crucial that the auditory system remain intact in case of device failure. Experience with cochlear implants has shown the reliability of devices implanted in the postauricular and mastoid area, and revealed potential complications. Potential risks of implantation include further sensorineural hearing loss, damage to the dura, cerebrospinal fluid leak, and the possibility of cholesteatoma or skin implantation into the middle ear or mastoid area. Skin complications can occur with any type of implant, and infection of the device is always a risk. Additionally, difficulties arising from coupling of the device to the ossicular chain or inner ear may directly damage these structures. The facial nerve and chorda tympani may also be at risk with surgical approaches for implantation, resulting in facial paralysis or weakness, or taste disturbance. Increased stiffness of the ossicular chain resulting from device fixation to the ossicles may impede low-frequency response, whereas implants that increase the mass effect on the ossicular chain reduce the highfrequency response. The ideal device must prove to be safe over the longterm. Batteries are required for these fully implantable devices; changes should be able to be easily performed in the office or on an outpatient basis. It is to be expected that these devices, worn for years, will require future upgrades. The devices should allow technology that is easily upgraded allowing better speech processing strategies to be programmed when made available. Theoretically, a device that worked all the time—24 hours a day, 7 days a week—would offer a significant advantage over conventional air conduction hearing aids. The ability to use a device in all normal daily activities such as water exposure during bathing or swimming is also a significant benefit. Patients who wear conventional air conduction hearing aids are unable to use them while sleeping or during water exposure. The ideal implant offers significant cosmetic benefit by being as invisible as possible. The lack of maintenance or need to clean the EAC is a significant advantage to hearing impaired patients. A long battery life is essential
to reduce the need for additional surgical procedures to change the battery.
CONVENTIONAL VERSUS IMPLANTABLE HEARING AIDS Gain is the amount of acoustic energy a device is able to deliver above the incoming signal. An implantable hearing aid must provide better gain than conventional air conduction devices, or some other real benefit to justify its use in an individual patient. Amplification of high frequencies is expected, and gain must be significant enough to provide real benefit to the patient. Most patients with significant hearing loss require significant gain levels in the high frequencies to receive any benefit from a device. If the device does not provide enough amplification in the high frequencies, the amount of benefit the patient receives would be reduced. The maximum level of output in decibels SPL (Sound Pressure Level) required to accommodate various levels of hearing loss is approximately 50 dB above the hearing threshold. To provide benefit to patients with moderate hearing loss to moderate to severe hearing loss, a hearing aid must provide a maximum output level equivalent to 90 to 115 dB SPL, whereas a flat frequency response of 8 kHz is desirable for maximum speech comprehension. Conventional air conduction hearing aids amplify sound before it reaches the middle ear. A microphone converts the incoming acoustic signal into an electric signal, and the amplifier and signal processor modify the electric signal to increase its strength. The receiver converts the amplified electric signal into an acoustic signal for presentation to the tympanic membrane for transmission via the middle ear to the inner ear in the normal physiologic manner. In contrast, implantable devices provide acoustic energy to the middle ear or inner ear, bypassing the external ear canal, and in some cases the middle ear space. The microphone converts the incoming acoustic signal into an electric signal. The amplifier and signal processor modify the electric signal to increase its strength. The receiver converts the amplified electric signal into a vibratory signal for presentation to the ossicular chain or to the cochlea directly, bypassing the tympanic membrane. Middle ear implants take advantage of the direct vibration of the middle ear ossicles to drive the ossicular chain. Middle ear implants may be totally or partially implantable. The partially implantable device consists of a microphone and a speech processor connected to a transmitter fitted with an external coil that transmits electric energy transcutaneously to the internal device. An internal receiving coil connected to a receiver provides electric energy to a transducer connected to the ossicular chain. The external device also has a battery to power the system. A fully implantable device contains essentially all of the elements of a partially implantable device with the
Chapter 32 • Implantable Hearing Devices
exception of a transducer coil and receiver. A microphone is placed under the skin, or is attached to the middle ear space and is connected to the internal speech processor. The entire system is powered by a rechargeable battery. All middle ear implantable devices stimulate the ossicular chain or cochlear fluids; however, they differ primarily by the type of transducer used to connect to the ossicular chain or cochlea. These devices provide a broad frequency response with low linear and nonlinear distortion. They are able to amplify high frequencies without the problem of acoustic feedback seen in conventional hearing aids, allowing the potential for better hearing in background noise with a more natural sound quality. A fully implantable device can also eliminate the perceived social stigma of visible conventional hearing aids.
TYPES OF MIDDLE EAR IMPLANTS An implantable hearing device is any surgically implanted device that converts acoustic energy to mechanical energy, which delivers vibratory stimulation to the inner ear. A transducer is a device that converts one form of energy to another. Essentially two types of transducers currently are used in middle ear implants: piezoelectric and electromagnetic. Each type has advantages and disadvantages related to power, efficiency, frequency response, and reliability. Piezoelectric devices make use of ceramic crystals that change shape when voltage is applied. This change in shape can provide mechanical energy to stimulate the ossicular chain or inner ear. The change in the shape of the ceramic is temporary; the ceramic reverts to its original shape when the electric current is no longer applied. There are two types of piezoelectric ceramic crystals: monomorph and bimorph. The monomorph piezoelectric crystal consists of a single layer, which expands and contracts to create the vibrations directly. The bimorph consists of two bonded ceramic layers, which are arranged in opposing electric polarities. When the current is passed through the bonded layers, the entire structure bends, creating the vibration. Anatomic size restrictions limit use of piezoelectric ceramic materials because the amount of bending is proportional to the length of the crystal, reducing the amount of transductive power available. The electromagnetic transduction of sound involves the creation of mechanical vibration by passing a current through a coil proximal to a magnet. As the electricity passes through the coil, an electromagnetic field is created, vibrating the magnet, which by direct or indirect contact causes movement of the middle ear structures or cochlear fluids or both. The magnet may be attached directly to the middle ear vibratory pathway—the tympanic membrane, incus, or stapes. A fluctuating magnetic field is generated when the coil is energized by electric signals that correspond to the acoustic input. The magnetic field causes the magnet to vibrate, inducing vibration of the
385
ossicular chain or the cochlear fluids directly. The force generated by this system is directly proportional to the proximity of the magnet with the induction coil. One method to produce electromagnetic stimulation is to separate the magnet from the induction coil. The coil is housed in a separate device, usually within the external ear canal while the magnet is attached to the ossicular chain. It can sometimes be challenging, however, to control the spatial relationship of the magnet and the coil when the magnet is attached to one part of the ear, and the coil is located within the ear canal, which may result in a wide variation in device performance. This can manifest as varying frequency responses, or fluctuation of output levels, if the distance between the alignment of the coil and magnet changes. Another method of electromechanical stimulation is to house the coil and magnet together, often in a single assembly. If the magnet and coil are housed together, a probe extending from this assembly must be in contact with the ossicular chain. As current is passed through the assembly, vibrations from the probe are sent directly to the ossicular chain. This form of electromechanical stimulation optimizes the spatial and geometric relationships to avoid the problem of changing alignment that may occur between the coil and magnet. The major limitation of housing the magnet and coil together is the attachment of the stimulating device, or coupler, to the ossicle or inner ear. If the device shifts relative to the position of the ossicles or inner ear, there may be a reduction in the optimal transmission of auditory stimulus.
HISTORY OF MIDDLE EAR IMPLANTS The use of a magnetic field to stimulate the ossicles is not new. This concept can be traced back to 1935, when Wilska placed iron particles directly on the tympanic membrane. A magnetic field was generated by an electromagnetic coil inside an earphone, which caused the iron fillings to vibrate in synchrony with the magnetic field, producing vibration of the tympanic membrane, simulating hearing. Later, Rutschmann glued 10 mg magnets onto the umbo, causing it to vibrate via the application of a modified magnetic field with an electromagnetic coil. The resulting vibration of the ossicles produced hearing sensation. Attachment of devices into the middle ear space did not occur until the 1970s with the RION device, which is discussed later in this chapter. Frederickson and colleagues developed the first mechanical device at Washington University in St. Louis, in 1973. This device, which used a multichannel digital signal processor that transmitted power to the implanted coil via a transcutaneous link, was implanted in 12 rhesus monkeys. After 2 years of implantation, there was found to be no damage to the cochlea or peripheral auditory system. Frederickson and colleagues found that the results from mechanical stimulation were similar to results produced
386
OTOLOGIC SURGERY
by acoustic stimulation, and that high-intensity signals could be delivered to the middle ear effectively. Mamiglia at Case Western Reserve University in Cleveland, Ohio, demonstrated the use of an electromagnetic device in the cat. This group used the malleus as the microphone for a totally implantable device, which was implanted in cats for approximately 9 months. As with Frederickson’s experiments, the results showed thresholds from mechanical stimulation that were comparable to acoustic stimulation with no adverse effects on the middle or inner ear. Dumond, in Bordeaux, France, worked on piezoelectric devices placed in contact with the round window membrane. Twelve guinea pigs were stimulated over a 7 month period. This group attached the piezoelectric devices to the round window without removal of any other component of the ossicular chain.
SPECIFIC DEVICES A variety of implantable devices have undergone investigation. Devices that are currently approved and that have been implanted into humans are reviewed here (Table 32-1). Patient selection criteria depend on the type of transducer and the method by which the sound is stimulated.
RION The RION (Fig. 32-1), developed at Ehime University and Teikyo University in Japan in collaboration with the Rion Company by Yanagihara and colleagues, was first implanted in 1984. This is a partially implantable middle ear device that uses a piezoelectric transducer approach. The device consists of a microphone, speech processor, and battery that are contained in an external behindthe-ear unit. The internal component consists of an ossicular vibrator and internal coil, which are coupled. The essential component is the vibratory element consisting of a bimorph, or two piezoelectric ceramic elements pasted together with opposite polarity, which have been coated with layers of biocompatible material. The free end of the bimorph is attached to the stapes, and is attached to a housing unit screwed into the mastoid cortex providing
fixation. The bimorph vibrates in response to applied electric current. The indications for the RION include mixed hearing loss and significant mastoid disease. Its primary use has been in patients with conductive hearing loss from chronic otitis media. Patient selection criteria are listed in Table 32-2. Frequency responses are attenuated after approximately 5000 kHz, so patients with significant sensorineural hearing loss above this frequency may not receive as much benefit as patients with pure conductive hearing loss. The RION has been implanted in patients in Japan, but is not currently available in the United States. The device may be placed in the postauricular area during a canal wall down mastoidectomy for treatment of chronic ear disease or through a transmastoid facial recess approach. The housing unit fits in the mastoid with a tip that extends to the stapes. The open mastoid cavity is the preferred approach because it allows adequate exposure to the ossicles, but this requires the ear canal to be closed off. The device is fixed to the mastoid cortex, and the bimorph container is placed over the stapes. A seat is surgically created for the internal coil and electric unit. The RION device has been worn by some patients for more than 10 years. Implanted patients report natural sound quality without feedback or discomfort, which is very close to perceived normal hearing. Long-term sensorineural hearing loss has not occurred with the use of the RION. There have been no complications during implantation in more than 39 patients in Japan.
Totally Implantable Cochlear Amplifier The Totally Implantable Cochlear Amplifier (TICA) device (Fig. 32-2) was developed at the University of Tubingen, Germany, in collaboration with Implex Corporation in Munich in the mid-1990s. The company went bankrupt, and the device is no longer available for use. The TICA is reviewed for historical purposes because it represents the first fully implantable device that was used in humans, and allowed the development of some important concepts that have been used elsewhere. The TICA was a fully implantable device that used a piezoelectric transducer to stimulate the
TABLE 32-1 Specific Implantable Devices Device Name
Transducer
Implantable
Approval
RION Implex/TICA Envoy SoundTec Otologics/MET Soundbridge
Piezoelectric Piezoelectric Piezoelectric Magnet implant Electromagnetic Electromagnetic
Semi-implantable Fully implantable Fully implantable Semi-implantable Fully implantable Semi-implantable
Only in Japan Bankrupt Esteem I approved in Europe; Esteem II phase 2 in U.S. U.S. FDA approved September 2001 Approved in Europe; fully implantable phase 2 in U.S. Approved in Europe; U.S. FDA approved July 2000
FDA, Food and Drug Administration.
Chapter 32 • Implantable Hearing Devices
387
Inner unit Secondary coil
Ossicular vibrator
Outer unit Primary induction coil Microphone Amplifier Battery
A External link coil
Internal link coil
Microphone
Biomorph stapes
Ceramic biomorph Lead wire
Mastoidectomy
VII
1
Facial recess
EAC
3 VII
B
Ossicular Amplifier Battery
2
FIGURE 32-1. RION system. A, Outer unit contains the microphone, amplifier, and battery. Internal component contains the ossicular vibrator. B, Canal wall down mastoidectomy (1). Canal wall up mastoidectomy (2). Cross section, canal wall up mastoidectomy (3). EAC, external auditory canal.
TABLE 32-2 RION Device Patient Selection Criteria Average bone conduction speech frequency hearing level (500, 1000, 2000 Hz); ≤50 dB Moderate to severe deafness in contralateral ear Intraoperative vibratory hearing test shows effectiveness of unit
ossicular chain. In addition, it used a microphone implanted in the ear canal that picked up sound. This sensor provided an electric input into the fully implanted unit or can, which was implanted subcutaneously in the mastoid area. The can was a hermetically sealed titanium container that included the speech processor, battery, and a receiving coil. The actuator attached to
A Microphone Totally implantable cochlear amplifier Transducer and titanium rod
Receiver
Receiver Transducer Titanium rod
M I 1 Mastoidectomy
B
S Microphone
2
FIGURE 32-2. TICA system. A, TICA device with the fully implantable microphone in the ear canal. The speech processor battery unit is housed in a subcutaneous pocket in the postauricular area. The stimulator attaches to the incus to stimulate vibration. B, Surgical view (1). Coronal view showing position of transducer (2). I, incus; M, malleus; S, stapes.
Chapter 32 • Implantable Hearing Devices
the body of the incus, causing vibration of the ossicular chain. The induction coil within the titanium can was used to receive electric impulses to permit recharging of the battery. The patient would recharge the battery by wearing a small headband that could also be used for programming the device. A wireless remote control could be used by the patient to select four programs, adjust the volume, and switch the device on and off. The battery life was 50 hours, and recharging took approximately 2 hours. The speech processor consisted of a digitally programmable three channel audioprocessor. The induction coil could be used to receive input similar to cochlear implants for processing. Also similar to cochlear implants, fitting would start about 8 weeks postoperatively, and several programming sessions were usually required. There were several advantages to the TICA, such as having no external components and being MRIcompatible. It had a wide frequency range from 100 Hz to 10,000 kHz. The battery life was estimated to be 3 to 5 years before a replacement was needed. Approximately 20 patients were implanted with the TICA device. All patients had bilateral moderate to severe sensorineural hearing loss, and the patients had not benefited from hearing aids. One problem that arose after implantation of the TICA was feedback. The incus was stimulated by the actuator. The feedback occurred as sound was generated by the stimulation of the ossicular chain and picked up by the microphone implanted in the ear canal. The stimulation of the ossicular chair caused sound vibration to be transported to the cochlea, but this also resulted in the eardrum acting as a speaker, generating sound into the ear canal. This generation of sound necessitated disarticulation of the ossicular chain between the malleus and incus. To avoid feedback or sound coming out the ear canal from vibration of the malleus and eardrum, the neck of the malleus was removed. Patients with the TICA implant described their hearing as being distortion-free and transparent. They reported excellent speech intelligibility, and an improved ability to listen to music, especially in the presence of background noise. Patients were able to use the device during sporting events, including swimming, and in the shower. The experience of the TICA implant showed that a fully implantable device was possible, and allowed the development of many of the components used in subsequent devices.
Vibrant Soundbridge The Vibrant Soundbridge (Fig. 32-3) is the first FDA-approved implantable middle ear hearing device to treat sensorineural hearing loss, and has been implanted in thousands of patients worldwide. It is a partially implantable middle ear hearing device initially
389
developed by Symphonix Devices, Inc. (San Jose, CA). Subsequently, Med-El Corporation of Innsbruck, Austria, took over the production and distribution of the device. The Vibrant Soundbridge is a semi-implantable hearing aid consisting of two parts: the speech processor worn externally, and the implantable vibrating ossicular prosthesis. The vibrating ossicular prosthesis is surgically placed subcutaneously in the postauricular area. The floating mass transducer (FMT) is connected to the internal receiver, and is attached to the stapes. The FMT is a unique electromagnetic transducer that contains a magnet of inertial mass within two electromagnetic coils. When activated, the magnet mass vibrates within the FMT between the two coils causing the entire unit to vibrate. Titanium strips are attached around the long process of the incus to hold the device in place. The FMT is oriented in the direction of the stapes so that the device vibrates directly into the inner ear, parallel to the plane of the stapes. The external auditory processor is held in place over the internal receiver by a magnet. The auditory processor contains a microphone that picks up sound from the environment and converts it into an electric signal. The auditory processor is contained within the external unit, which also contains an induction coil to transmit the electric signal to the internal vibrating ossicular prosthesis. A receiving coil picks up the signal and transmits it to the FMT, causing it to vibrate, which stimulates the cochlea. Placement of the internal device requires an outpatient mastoidectomy similar to cochlear implantation. The facial recess is widely opened to visualize the incudostapedial joint, and to allow the FMT to pass through easily. The FMT is crimped onto the incus after the vibrating ossicular prosthesis is embedded in the cortical bone in a seat behind the mastoid posterior to the sigmoid sinus. The external processor is attached 6 weeks after surgery, at which time the device is programmed. Table 32-3 lists current indications for the Vibrant Soundbridge. Clinical trials for the Vibrant Soundbridge began in 1996, and the device was approved by the FDA in 2000 The FDA trials showed greater than 94% of patients reported improvement in their signal quality satisfaction rating with the Vibrant Soundbridge compared with their previous conventional air conduction hearing aids. Of patients who had complained of feedback problems with their presurgery air conduction hearing aids, 97% reported no feedback with the Vibrant Soundbridge. Eighty-eight percent of patients reported improved sound quality satisfaction rating of their own voice. Satisfaction with the overall fit and comfort of the Vibrant Soundbridge was reported by 98% of patients. The Vibrant Soundbridge has been implanted more recently over the round window membrane as a means to stimulate the cochlear fluid directly. Colletti implanted
390
OTOLOGIC SURGERY
A Transducer placed through tympanotomy
Sound bridge
M
Temporalis muscle Audio processor Incus Conductor link
I S
Internal receiver
VII
Conductor link
Demodulator
1
B
Internal receiver Centering magnet
Mastoidectomy Transducer 2
FIGURE 32-3. Symphonix Vibrant Soundbridge system. A, Vibrant Soundbridge with external microphone, with coil for transcutaneous stimulation of internal device. Internal device connects to the floating mass transducer, which attaches to the incus. B, Location of internal receiver (1). Position of transducer (2). I, incus; M, malleus; S, stapes.
Chapter 32 • Implantable Hearing Devices TABLE 32-3 Vibrant Soundbridge Patient Selection
Criteria
Adult (≥18 yr) Word recognition score ≥50% Normal middle ear function Realistic expectations Pure tone air conduction thresholds within frequencies shown in Figure 32-4
Frequencies Frequencies (Hz)
500
Upper limits (dB)
65
A
1000
2000
3000
4000
75
80
85
85
Frequency [kHz] 0
0.25
0.5
1
1.5
2
3
4
Hearing loss [dB HL]
20
40
60
80
B
100
FIGURE 32-4. Frequencies for Vibrant Soundbridge.
seven patients with atresia or chronic otitis media in which the stapes was unavailable to be stimulated. The classic transmastoid approach was used; however, the bony lip of the round window was drilled out to allow the FMT to fit over this area. Thresholds in the normal range were accomplished for all patients implanted with the FMT placed over the round window. Long-term effects of this type of stimulation are unknown; however, additional studies are currently investigating this mode of stimulation.
Soundtec The Soundtec Direct Drive Hearing System (Fig. 32-5) system was developed by Hough of Oklahoma City, Oklahoma. Although new devices are no longer available, it is reviewed for historical purposes because many patients in the United States have this device. The Soundtec used an electromagnetic approach that separated
391
the magnet and the induction coil. A tiny magnet, approximately the size of a grain of rice, was attached to the incudostapedial joint. The magnetic coil, which would drive the magnet, was located in the ear canal in a conventional in-the-ear mold. The sound processor, microphone, and battery were located in a behindthe-ear external unit and were linked to the coil in the external ear canal. The magnet was placed during an outpatient procedure via the transcanal approach, with the tympanic membrane elevated to attach the magnet to the incudostapedial joint A deep-seated ear canal fitting was required for the in-the-ear unit after complete healing occurred. A total of 103 patients were implanted at 10 sites for the FDA trial before approval was given in 2001 for treatment of sensorineural hearing loss. The Soundtec direct system (see Fig. 32-4B) gave an average of 7.9 dB increase in functional gain over optimally fitted air conduction hearing aids. Patients’ AFAB and speech discrimination scores were much higher for the Soundtec direct system compared with the optimally fitted air conduction hearing aids. This device is no longer available for implantation.
Envoy Esteem The Envoy Esteem (Fig. 32-6) has been developed by St. Croix Medical, Inc. (Minneapolis, MN). This is a fully implantable system that uses a piezoelectric transducer for reception and transduction. The transducer consists of two internal plates separated by a thin conduction material. The first transducer (sensor) detects movement of the malleus in response to sound stimulation. This sensor acts as a microphone sending signals to the processor. The second piezoelectric transducer (the driver) is placed on the stapes. The piezoelectric units use the bimorph design for the sensor and driver. One of the sensors is fixed to the malleus to detect vibration and stabilized by fixation to the cortical skull bone. A small amplifier in the base increases the gain. The fully implantable device also contains a speech processor powered by a lithium iodine battery. The Esteem device is implanted through a postauricular mastoidectomy, with the attic area opened widely to allow adequate room for placement of the sensor and driver. A bed is created in the cortical bone to attach the sound processor and battery unit. The facial recess is opened up to allow adequate visualization of the chorda tympani, facial nerve, and stapes, and to allow adequate space for the piezoelectric driver to be positioned in contact with the stapes. The bone of the posterior ear canal must also be thinned carefully to allow adequate visualization. The incudostapedial joint is separated, and a 2 mm section of the long process of the incus is resected. Disarticulation is required to separate the vibrating malleus from the stapes, which is to be stimulated. In some cases,
392
OTOLOGIC SURGERY
A
Hough-Dormer
Goode’s “ear lens”
2
Contactless electromagnetic devices M
1
M
I
I
Tm Tm Magnet 3
Cochlea
B
Amplifier
Coil
Magnetic torp
FIGURE 32-5. A, SoundTec device. The magnet is attached to the incus between the incus and stapes joint. B, Direct Drive Hearing System. Electromagnetic driver in ear canal (1). Magnet at incudostapedial joint or on tympanic membrane (2). Magnetic middle ear problems (3). I, incus; M, malleus; Tm, tympanic membrane.
Chapter 32 • Implantable Hearing Devices
A Driver Sound processor
Sensor
M
Envoy Sound processor
Stapes
Tm 3
M
1
Piezo sensor
EAC
Piezo driver
Tm
2
B
Mastoidectomy
Stapes
393
394
OTOLOGIC SURGERY
FIGURE 32-6. Envoy esteem fully implantable system. A, Eardrum is used as a microphone to pick up sound via a transducer attached to the malleus head. Internal component consists of a battery and speech processor. There is an internal coil for programming the device. The stimulation of the ossicles occurs via attachment to the stapes. B, Implant assembly (1). Coronal view showing position of transducer (2). Sensor attached to malleus, driver in contact with stapes (3). EAC, external auditory canal; M, malleus; Tm, tympanic membrane.
full removal of the incus may be necessary, with the sensor attached to the malleus head. The Envoy has gone through two generations of devices. The first device, the Esteem 1, had a battery life of 2.5 to 5 years, and no recharging of the battery was required. The Esteem 2, or second-generation device, has been used more recently in clinical trials. This device has a longer battery life—5 to 8 years without the need to recharge. There is also a broader fitting range resulting in improved gain of 10 to 20 dB over the Esteem 1. The first devices were implanted in Europe in March 2000. Phase I trial (safety study) for the Esteem 1 began in March 2002. The Esteem 1 is currently approved with its CE mark for European implantation (the CE mark is equivalent to the U.S. FDA). The Esteem 2 was approved for CE mark in 2008. Phase II clinical trials for the Esteem 2 began in 2008. At this time, neither the Esteem 1 nor the Esteem 2 is approved by the FDA for use in the United States.
MET The MET (Fig. 32-7) is a fully implantable ossicular stimulator produced by Otologics (Boulder, CO). The initial device tested was a partially implantable device that consisted of an external digital speech processor and an implanted unit. The external components consisted of a microphone, speech processor, battery, and transmitter that were housed in a disc that fit in the internal implanted device. The external unit is held in place with magnets that align it to the internal unit, similar to a cochlear implant. The implanted components consist of a subcutaneous electric package containing a transcutaneous receiver and a transducer motor in a hermetically sealed case. The electromagnetic motor drives a biocompatible probe tip, which is placed in a hole in the body of the incus. Activation of the device causes mechanical motion of the probe tip, which vibrates the ossicular chain. The transducer was found to have a linear input/output curve to beyond 1000 dynes at 1 kHz with low distortion. The frequency response is flat, varying only about 10 dB up from 1 kHz to 10,000 kHz. The device is placed during an outpatient surgical procedure through a postauricular transmastoid approach. A mounting device is placed into the mastoid cortex that secures the electromechanical motor, and a seat is created to place the electronic housing container. A laser hole is created in the incus body for the probe tip to rest. The device can be activated and programmed 6 to 8 weeks postoperatively.
The partially implantable device is currently available in Europe. FDA trials for the partially implantable device were not completed in the United States because the fully implantable device has subsequently become available. The fully implantable device is currently undergoing European and U.S. FDA trials at the time of this writing for the treatment of conductive and mixed hearing loss and congenital atresia. The fully implantable device consists of the same transducer stimulator as the partially implantable device. In addition, the internal case contains the battery, speech processor, receiving coil for programming and battery recharging, and a separate microphone that is attached to the receiving coil. The microphone is placed in the postauricular area subcutaneously. It is anticipated that the battery will last approximately 12 years. The battery requires daily recharging that takes less than 1 hour, during which time the patient can still use the device. The induction coil in the internal receiver receives a transcutaneous charge from a battery charger, which is also the method by which the device is programmed. The MET has a remote control that is used for volume control and turning the device on and off. The receiver coil on the internal unit is used to access information. The MET has several different attachments that may be used on the transducer for stimulation. The classic or original device was designed to attach to the incus. Subsequently, additional adapters have been created, allowing the device to attach to the stapes, oval window directly, or round window for direct stimulation. Clinical trials of these new attachments have not yet begun.
CONCLUSION Much work has been done over the past several decades to develop implantable hearing devices. Although there have been many successes, this field must be considered still in its infancy, with virtually all of these devices undergoing modification. Although development is rapid, and the outcomes from current devices are encouraging, patients must be informed that the longterm results are presently unknown. At the time of this writing, the field seems to be moving toward the use of these devices for conductive or mixed hearing losses, as opposed to pure sensorineural hearing loss. Further clinical trials need to be conducted to learn more about this emerging field. Experience has taught that clinical trials for implantable devices always take longer and cost more than originally estimated. Additionally, many manufacturers have
395
Chapter 32 • Implantable Hearing Devices
A
Receiving coil and electronics
Anchoring and positioning mechanism
Temporalis muscle Linea temporalis
Receiving coil Transmitting coil
Probe tip 3 I S M
1
EAC M M Mounting ring
B
I
Transducer
2
FIGURE 32-7. Otologic MET. A, Microphone is placed in subcutaneous pocket behind the ear. Internal device consists of a speech processor, battery, and coil. Electromechanical stimulator is attached to the body of the incus. Ossicular chain is kept intact. B, Surgical view (1). Coronal view showing position of transducer (2). Probe tip seated in body of incus (3). EAC, external auditory canal; I, incus; M, malleus; S, stapes.
396
OTOLOGIC SURGERY
had difficulty in the mass production of reliable products. In addition to manufacturing problems, cost has been a major issue because many government agencies and insurance carriers have not routinely paid for these devices. As outcome studies show measurable improvement by the use of implantable devices, however, the trend is beginning to turn for coverage of the cost of surgery and the device. Subjects describe the results from these devices as being better than their conventional air conduction hearing aids. The patient’s subjective assessment of improvement far outweighs the objective measures that have been used in clinical trials. It seems that the benefit described by patients is not measured by functional gain alone. Although the exact rationale for this perceived improvement is unknown, better amplification of high frequencies than conventional air conduction hearing aids is theorized to be responsible for much of this benefit. Development of new measurement tools may be necessary to measure adequately the device’s benefit as this area of otology continues to grow.
REFERENCES 1. Fredrickson J M, Coticchia J M, Khosla S : Current status in the development of implantable middle ear hearing aids. Advances in Otolaryngology, vol. 10 St Louis, Mosby, 1996, pp. 189-204. 2. Miller D A, Fredrickson J M : Implantable hearing aids. In Valente M (ed): Audiology Treatment. New York, Thieme, 2000, pp. 489-510. 3. Wilska A : Einmethode zur bestimmung der horsch wellanamplituden des trommelfells bei verscheiden frequenzen. Skand Arch Physiol 72:161-165, 1935. 4. Rutschmann J: Magnetic audition: Auditory stimulation by means of alternating magnetic fields acting on a permanent magnet fixed to the eardrum. IRE Transactions Med Electron 6:22-23, 1959.
5. Fredrickson J M, Tomlinson D R , Davis E R , Odkuist L M : Evaluation of an electromagnetic implantable hearing aid. Can J Otolaryngol 2:53-62, 1973. 6. Fredrickson J M, Coticchia J M, Khosla S : Ongoing investigations into an implantable electromagnetic hearing aid for moderate to severe sensorineural hearing loss. Otolaryngol Clin North Am 28:107-120, 1995. 7. Maniglia A J, Ko WH, Garverick S L , et al: Semiimplantable middle ear electromagnetic hearing device for sensorineural hearing loss. Ear Nose Throat J 76:333338, 340-341, 1997. 8. Dumon T, Zennaro O, Aran J M, Bebear J P: Piezoelectric middle ear implant preserving the ossicular chain. Otolaryngol Clin North Am 28:173-187, 1995. 9. Yanagihara N, Aritomo H, Yamanaka E, Gyo K : Implantable hearing aid: Report of the first human applications. Arch Otolaryngol Head Neck Surg 113:869-872, 1987. 10. Yanagihara N, Sato H, Hinohira Y, et al: Long-term results using a piezoelectric semi-implantable middle ear hearing device: The Rion device E-type. Otolaryngol Clin North Am 34:389-400, 2001. 11. Zenner H P, Leysieffer H : Total implantation of the Implex TICA hearing amplifier implant for high frequency sensorineural hearing loss: The Tubingen University experience. Otolaryngol Clin North Am 34:417-446, 2001. 12. Luetje C M, Brackman D, Balkany TJ, et al: Phase III clinical trial results with the Vibrant Soundbridge implantable middle ear hearing device: A prospective controlled multicenter study. Otolaryngol Head Neck Surg 126:97-107, 2002. 13. Colletti V, Soli S D, Carner M, Colletti L : Treatment of mixed hearing losses via implantation of a vibratory transducer on the round window. Int J Audiol 45:600608, 2006. 14. Hough J, Dyer R , Matthews P, Wood M : Early clinical results: SOUNDTEC implantable hearing device phase II study. Laryngoscope 111:1-8, 2001. 15. First chronic implant performed. Envoy-Voices. St. Croix Medical, Minneapolis, MN, March 2000.
33
The Bone-Anchored Cochlea Stimulator (Baha) Anders M.R. Tjellström Videos corresponding to this chapter are available online at www.expertconsult.com.
The bone-anchored cochlea stimulator (Baha) system is based on the concept of direct bone conduction stimulation of the cochlea. Baha combines an osseointegrated implant and a percutaneous abutment placed behind the external ear, and a specially designed impedance matched electromagnetic temporal bone stimulator (transducer). Using this system, acoustic energy is transferred directly to the fluids of the inner ear, bypassing the external ear and the middle ear. The damping effect from soft tissues of the mastoid process, which is in the range of 7 to 15 dB depending on the frequency, is eliminated.1 Elimination of this effect is important because a significant amount of sound energy is lost in the soft tissues, especially in the high-frequency range because this is where most of the consonant sounds are located. These sounds are important for speech understanding, especially in noisy surroundings. A direct coupling between the transducer and the skull without any soft tissue between is of great acoustic advantage.2-6 The Baha eliminates many of the drawbacks associated with the old-fashioned bone transducer, which is placed over the mastoid and requires pressure from a steel spring over the head or frames of eyeglasses. With Baha, patients can expect better sound quality and a stable position of the transducer, and do not experience any of the discomfort from the pressure required for transcutaneous stimulation.
HISTORICAL ASPECTS Several factors combined to promote the development of Baha. From the mid-1970s, the use of osseointegrated implants in the treatment of edentulousness was in clinical practice. In conjunction with this, Brånemark sought an acoustic means to evaluate the degree of osseointegration. In one experiment, a patient with implants in the upper jaw had a special adapter secured to a dental implant and an Oticon bone transducer attached. Although it was impossible to measure the stability of the implant at that time, it was clear that the patient experienced sound
very clearly, even when stimulation was of a low level; this had important implications regarding the ability to transmit sound via bone. A Ph.D. thesis by Kylén, who measured the sound reaching the cochlea during ordinary mastoid drilling, also contributed to the Baha concept.7 Kylén and Arlinger7 found that high noise levels produced a temporary threshold shift. A third factor was the problems patients with transcutaneous bone vibrators experienced. In 1977, in close cooperation with Håkansson at Chalmers University of Technology in Göteborg, the first prototype of the Baha was manufactured and tested on three patients with chronic ear disease. The initial results were encouraging, and during the following 5 years another 14 patients underwent surgery. Because the patients were very satisfied with the hearing results, the program expanded. In June 2009, the total number of Baha patients worldwide was estimated to be greater than 70,000. By following the original guidelines set by Brånemark for establishing osseointegration, the success rates for stable implants were as high for the Baha as for the oral cavity. It was also clear that the handling of soft tissues at the implant site was key to establishing a lasting and reactionfree skin penetration. Subcutaneous soft tissue reduction proved to be the solution; without it, there was a clear tendency toward inflammation and irritation that could occur at any time, ranging from a few months to a few years after treatment. For patients who have not undergone soft tissue reduction, subcutaneous soft tissue proliferation was also common and resulted in skin irritation.
HEARING THROUGH BONE CONDUCTION In his Ph.D. thesis, Stenfelt8 discussed the different bone conduction theories of von Békésy, Barany, Tonndorf, and others. Most of the studies cited refer to patient groups with a normal ear, which is seldom the case for Baha patients. One exception is after acoustic tumor surgery, in which the eighth cranial nerve has been 397
398
OTOLOGIC SURGERY
s acrificed, but the ear canal, middle ear, and cochlea are normal. Research has proven hearing through bone conduction to be highly complex. First, for frequencies less than 1.5 kHz, the relative movement of the ossicular chain dominates the bone conduction response; this is the inertia of the middle ear ossicles. Even with damage to the middle ear and ossicles, however, there is a bone conduction response in the low frequencies because the skull moves like a solid body. Second, for frequencies greater than 1.5 kHz, the response is attributed to the compression of the labyrinth. A third means of bone conduction is due to the relative movement of the mandible. Air in the external ear is set in motion and transmits sound through the intact drum and ossicles to the cochlea. Several other factors probably also play a role, such as the oval and round window release, inner ear fluid inertia, and cochlea aqueduct effect. These factors are often interrelated, and it is difficult to isolate one from another. Almost regardless of the location of bone stimulation, the waves of the basilar membrane travel from the base of the cochlea, where the membrane is stiffer, toward the helicotrema, as is the case with air conducted sound. This means that the cochlea has difficulties in differentiating between bone conducted and air conducted sound. Cancellation experiments have verified this difficulty. Stenfelt8 also showed that bone conducted sound from 0.1 to 10 kHz for levels up to 77 dB HL is linear; this is important because the distortion level of the bone conducted sound would be low.
When a commercially pure titanium implant is machined, the surface is covered with an oxide layer within milliseconds. The implant can be regarded as a ceramic. The oxide layer has unique biocompatible properties; it is also dense and adheres to the bulk metal. Proteins from the host would not denature when in contact with the oxide. Titanium oxide has a high electrochemical value, which means that the surface attracts foreign material that could contaminate the implant.11 This is why the implant should be handled with the utmost care, avoiding contact even with sterile particles from gloves and draping during surgery. The surgical procedure, which is described in detail subsequently, is focused on reducing surgical trauma as much as possible. Excessive heat interferes with healing, and an osteocyte tolerates only 42° C for 1 minute before apoptosis. If too many osteocytes are damaged, there is a risk of fibrous tissue healing. In the treatment of edentulousness with implants, early loading often caused loss of osseointegration. This was the reason why the two-stage procedure was initially recommended for the Baha.10 The load produced by the Baha is very low, however, compared with the forces produced during chewing. Based on a study comparing a onestage with a two-stage procedure in adults, the one-stage procedure is now recommended because no difference in osseointegration could be found.15 In irradiated bone, a two-stage procedure with a 6-month interval is suggested, often in combination with hyperbaric oxygen treatment.16
OSSEOINTEGRATION
Two groups of patients are potential candidates for the Baha: (1) patients with conductive or mixed hearing loss, and (2) patients with single-sided deafness.
The term osseointegration was coined by Brånemark and refers to direct contact between bone and an implant that can withstand a functional load. Numerous papers on osseointegration were published in the late 1970s.912 Osseointegration is a fundamental prerequisite for direct bone conduction; without it, the Baha would not function. Among the factors identified by Brånemark as key to establishing osseointegration was choice of implant material. Commercially pure titanium has been the material of choice since the start. Commercially pure titanium has a purity of 99.75%. The level of ferromagnetic contamination is 0.02% to 0.05% in the bulk metal, but no traces are found in the oxide layer, which means that the implant would not jeopardize magnetic resonance imaging (MRI).13,14 Because of the difference in density between bone and titanium artifacts, less than 1 mm would appear around the implants. The screw-shaped design provides initial stability. Micromovements during the healing phase, if too large, could result in soft tissue encapsulation and prevent osseointegration.
PATIENT SELECTION
Conductive or Mixed Hearing Loss Patients who have mixed hearing loss with a conductive element greater than 30 dB may not be helped by conventional air conduction aids, but could benefit from the Baha.
Chronic Ear Disease The Baha was originally designed for patients with bilateral or unilateral conductive or mixed hearing loss who needed amplification, but who could not be helped using reconstructive surgery, and who could not use an air conduction hearing aid. Suitable candidates included patients with ears that drain despite trials to make the ear dry, and patients whose ears start to drain when the external ear canal is occluded with a hearing aid mold. Patients with dry radical cavities who experience acoustic feedback when an air conduction aid is tried may also benefit from the Baha.
Chapter 33 • The Bone-Anchored Cochlea Stimulator (Baha)
Congenital Malformations Patients with bilateral atresia often have normal or nearnormal cochlea function and are ideal patients for the Baha. Because reconstructive atresia surgery is complex and high risk, a lasting hearing improvement is not always possible to achieve. In patients with a Jahrsdoerfer rating of 7 or worse, we often suggest a bone-anchored hearing implant instead of reconstructive trials.17 One advantage of the Baha for these patients is that the treatment does not interfere with atresia surgery if this becomes necessary later on. Although regulations regarding the minimum age for implantation vary from country to country, Baha has been successfully used in children 2 years old. Children can benefit from hearing with the Baha until they are old enough for an evaluation of the final anatomic situation to be made. The final decision can be made whether or not atresia surgery should be recommended. Unilateral external ear canal atresia has been said to have little or no importance for child development. In a review published in 2004, Cho Lieu18 found, however, that speech development and the learning capacity of these children are at risk, even when hearing is normal in the contralateral ear.
Conductive Loss in Only Hearing Ear A conductive loss in a patient with a deaf contralateral ear could be a problem. If the conductive component is large, an air conduction hearing aid may have difficulty overcoming this gap. A Baha bypassing the middle ear could be an alternative. As in all middle ear surgery, there is always a risk of damage to the cochlea; the author uses Baha in these cases. A patient with a deaf ear on one side, from stapes surgery, and a maximum conduction loss on the other side is an excellent candidate for a Baha.
Chronic External Otitis Preventing Use of an Air Conduction Hearing Aid Some patients in need of amplification experience problems with air conduction ear canal molds. Several different materials have often been tried, but for some patients the mold causes irritation, infection, and inflammation. As a result, these patients can use their hearing aid only for very limited periods or not at all. For them, the Baha is a good alternative. This group represents about 10% of the Baha patients at our Implant Unit.
Down Syndrome Conductive hearing impairment is common in Down syndrome. Narrow external ear canals and middle ear malformations are prevalent in patients with Down syndrome. Serous otitis media is also frequent in these children and can be hard to treat sometimes. It is clinically
399
proven that the Baha improves the quality of life for these patients.19
Single-Sided Deafness Cochlear deafness in one ear is not unusual. One cause is acoustic tumor surgery. Patients with single-sided deafness often experience verbal communication problems even when the opposite ear functions normally. Problems with speech understanding are most commonly experienced in noisy surroundings. One reason is the head shadow effect, but several other factors are involved. Proc essing is highly complex and is not yet fully understood. One finding is that patient satisfaction is often greater than indicated by test results.20-25
PATIENT COUNSELING Patients considering a Baha should be thoroughly counseled in all aspects. The patient must have realistic expectations and be aware that this sound processor also has its limitations. The surgical procedure should be described, and possible intraoperative and postoperative complications should be discussed. A dummy implant may be helpful for showing to the patient the small size of the implant when discussing the implant procedure. Many patients experience some postsurgical numbness around the implant site. This numbness is more common in patients in whom a large amount of soft tissue has been removed. The patient should also be able to come for regular follow-up visits. One of the main causes of adverse skin reactions is inadequate hygiene. The necessity of a high level of personal hygiene must be stressed preoperatively. The sound processor requires proper maintenance. Detailed instruction of its function and physical care should be given. Age is not a contraindication for the Baha per se. The oldest patient who has undergone surgery at the Sahlgrenska unit was more than 90 years old, and the youngest was 18 months old. Because surgery becomes easier and less risky with age, it is suggested that the Baha Softband be used until the child is 3 years old (Fig. 33-1) (see subsequent section �Baha in Children �.��) When clinical and audiometric criteria have been evaluated, and the final decision to proceed with implantation has been made, the patient should be informed that the Baha procedure is completely reversible. If the patient is dissatisfied with the Baha, it is easy to remove the abutment and, if so desired, the implant, returning the patient to his or her original state. Very few procedures in otology offer this possibility. Psychiatric disease can be considered a relative contraindication. When in doubt about a patient’s condition, psychiatric consultation could be very helpful for avoiding pitfalls.
400
OTOLOGIC SURGERY
FIGURE 33-1. A 3-year-old girl with a Soft-Band wearing the transducer in her own way.
FIGURE 33-2. A patient with the Baha Divino.
AUDIOMETRIC CRITERIA Generally, the better the cochlea function, the better the result. The air-bone gap is not important. As of September 2007, there was only one manufacturer of direct bone conduction cochlea stimulators, Cochlear Bone Anchored Solutions (a division within Cochlear Ltd, Göteborg, Sweden). Two ear-level devices and one body-worn device are available. The major difference between the ear-level devices is that they target different patient groups. The Baha Divino (Fig. 33-2) is for patients whose average bone conduction thresholds are at 0.5 to 3 kHz should be better than 30 to 35 (45) dB.20 The Divino also has a directional microphone setting. The other ear-level device, the Baha Intenso (Fig. 33-3), is recommended when the cochlea reserve is worse. Patients with a bone conduction threshold of 40 to 45 (55) dB can benefit from this hearing instrument. The body-worn device, the Baha Cordelle II (Fig. 33-4), has the transducer in a separate housing that is worn at ear level, connected to the Baha coupling. The microphone, electronics, and batteries are in the bodyworn unit. Patients with cochlea hearing loss of 50 to 55 (60) dB could benefit from this device. The levels of hearing impairment discussed here are only guidelines. Some patients have not been satisfied even when well within the audiometric criteria, whereas others are satisfied even though their hearing capacity is lower than technically required.
FIGURE 33-3. A patient with the Baha Intenso.
Chapter 33 • The Bone-Anchored Cochlea Stimulator (Baha)
FIGURE 33-4. Body-worn aid Baha Cordelle II. FIGURE 33-5. Test band with Baha.
In the preoperative evaluation, a testband with a Baha can be very helpful. The Baha is attached to a steel spring that the patient wears over his or her head (Fig. 33-5). With this test device, the patient still has skin between the transducer and the skull bone. If the patient is satisfied with this testband, the chances of implant surgery being successful are very high. The test rod can also be used as an alternative means to give the patient an idea of the outcome (Fig. 33-6). The Baha is attached to the coupling of a plastic rod and pressed onto the skin over the mastoid. To reduce the damping effect of the skin, the patient can hold the rod firmly between his or her teeth, which would simulate direct bone conduction.
CONTRAINDICATIONS There are no absolute contraindications to providing a patient with a Baha. Psychiatric disease or psychological problems rendering the patient unable to follow instructions and participate in follow-up could be discussed, however. The same could be said for patients with extremely poor general hygiene. Patients with diabetes, psoriasis, scleroderma, or other skin problems could experience a slightly increased frequency of adverse skin reaction, which should be mentioned during patient counseling. Bone diseases, such as osteogenesis imperfecta and Paget’s disease, might constitute a higher risk of implant losses.
FIGURE 33-6. Test rod with Baha.
401
402
OTOLOGIC SURGERY
SURGICAL TREATMENT The surgical technique is presented in the accompanying DVD. For the surgical technique in children, see the section on Baha in children.
Preparation of the Patient The surgical procedure is simple, generally performed under local anesthesia and as a day procedure. Premedication with morphine and scopolamine is often, but not always, given. No antibiotics or steroids are used, unless the patient has been irradiated or has artificial implants elsewhere, such as a hip prosthesis. The author prefers magnifying lenses to the otomicroscope because this makes it easier to find the exact angle of direction for the different steps of the procedure. An indicator for the Baha is used to make a mark for the implant site (Fig. 33-7A). It is important that the sound processor not touch the pinna because this could cause acoustic feedback. The implant should be placed as close as possible to the external ear canal opening, however. This is often about 50 to 55 mm from the center of the external ear canal opening. A suitable position in the cranial-caudal direction is at the level of the superior auricular fold, or just below the linea temporalis. A mark for the implant site is made with a pen, and the area is shaved. Before cleaning and draping the patient, a small scratch with a needle allows the surgeon to identify the implant site even if the ink mark has disappeared. Cleaning and draping is done in the same manner as for standard mastoid surgery. If there is secretion from the ear, the pinna could be folded anteriorly and kept in place with some adhesive draping to seal off the ear canal.
Surgical Instruments The instruments used consist of both general surgical instruments and instruments specially designed for this type of implant surgery. These include a drill unit with high-speed and low-speed functions and a feature to reverse the direction of rotation. The torque for the low speed can be adjusted according to the quality of the bone. The special instruments, implants, and sound processors are produced by Cochlear Bone Anchored Solutions (Göteborg, Sweden).
2 Self-retaining retractors, small 1 Periosteal elevator Bipolar cautery
Special Instruments Raspatory Dissector Baha dermatome Driver pin Connection to handpiece to pick up Baha coupling Titanium organizer Surgical wrench set Counter torque wrench Screwdriver Unigrip, 95 mm Screwdriver for internal hexagon, 20 mm Drill indicator Indicator for Baha Fixture mount Unigrip standard Machine screwdriver Unigrip, 25 mm Abutment inserter
Disposable Instruments, Sterile Packed Dermatome blade Guide drill with removable plastic spacer Countersink, 3 or 4 mm Healing cap Biopsy punch, 4 mm
Implant Sterile, Packed Titanium implant with pre-mounted Baha coupling, 3 or 4 mm
Drilling Equipment The drill is placed in a position where the surgeon can easily see the display and with the foot-pedal placed comfortably for adjusting speed, torque, rotation direction, and irrigation. Control unit with stand, foot control, and motor for high and low speeds Shank and head (handpiece)
Standard Instruments The standard instruments listed are placed by the scrub nurse on a sterile draped Mayo stand: 5 Mosquitoes 1 Mayo scissor 1 Curved scissor, small 1 Adson forceps 1 Needle holder 2 Bard-Parker blades 2 Skin hooks, small
SURGICAL TECHNIQUE The surgical procedure is illustrated in Figure 33-7B through L.
Preparing Thin Hairless Flap The implant site is clearly marked including the skin for the Baha dermatome flap. The author prefers to have the flap pedicle anteriorly. This is a matter of routine,
Chapter 33 • The Bone-Anchored Cochlea Stimulator (Baha)
A
B
C
D
E
F
FIGURE 33-7. A-F, Steps in the surgical procedure. See text for details.
403
404
OTOLOGIC SURGERY
G
H
I
J
K
L
FIGURE 33-7—cont’d G-L, Steps in the surgical procedure. See text for details.
Chapter 33 • The Bone-Anchored Cochlea Stimulator (Baha)
owever, and of no importance. The flap should be h called a pedicle graft because there is no clear arterial blood supply to the flap. The area indicating the amount of soft tissue to be removed is also marked. The size of this area depends on the amount of subcutaneous tissue. Less soft tissue has to be removed in a lean patient than in an obese patient. The aim is for the edges around the flap to slope gently down to the implant site. A local anesthetic of 10 mL of lidocaine with epinephrine is injected, making sure that some of the anesthetic is administered in and under the periosteum, and that it also reaches the periphery of the intended surgical field. The disposable sterile blade for the Baha dermatome is taken out of its pack. There are two grooves on one side of the base of the blade and only one on the other side. The side with the single groove should face upward when placed in the dermatome. The driver pin is placed in the handpiece, and the dermatome is firmly secured to the handpiece. The thickness and width of the dermatome are set, and no adjustment is needed; the width is 25 mm, and the thickness is 0.6 mm. This thickness produces a flap where the hair follicles are left for removal in conjunction with soft tissue reduction (see Fig. 33-7B). The drill is set to 2000 rpm. A little liquid paraffin is applied to the flap area to make the dermatome slide gently. A small cut with a knife along the starting point also facilitates creating the flap. A slight pressure is applied as the dermatome is pushed forward. It is important to maintain the correct angle of the dermatome—neither too shallow nor too steep. When the intended base of the flap is reached, the engine is stopped, and the dermatome is withdrawn. The hair remaining on the flap must be removed by scraping with a blade. The next step is to expose the implant site. A cut is made along the three sides of the subcutaneous flap leaving the base still attached (see Fig. 33-7C). The cut should be down to, but not through, the periosteum. Two small self-retaining retractors keep the flap out of the implant area. The periosteum is trimmed especially close to the actual implant site. The outer layer of the periosteum should be removed, leaving only the innermost layer in place. A hole about 6 mm in diameter is made in the thinned periosteum for the implant.
Guide Hole Drilling The sterile packed guide drill is attached to the handpiece (see Fig. 33-7D). The white plastic spacer is initially left in place. This spacer prevents drilling deeper than 3 mm. The drill speed is 2000 rpm. During drilling, generous cooling is used to reduce heat trauma to the bone. The drill should be moved up and down in short sequences about 3 seconds long. In addition, it is important to widen the hole to be able to inspect and check the bottom for any underlying soft tissues, such as the dura mater or the sigmoid sinus. The width should be almost 3.75 mm, which is the diameter of the implant. This widening also
405
allows the irrigation to reach the area where the cutting is taking place. The drill flutes should be kept free from bone because this would otherwise induce heat. If there is still bone at the base of the implant site when the spacer is all the way down, the spacer is removed to allow an additional 1 mm of drilling so that the 4 mm implant can be used.
Countersink Drilling The next step is to widen the hole to the final diameter of 3.75 mm; this is done with a spiral drill, which also has countersink edges (see Fig. 33-7E). Drill speed is still 2000 rpm. The surgeon should work up and down in two to four steps with generous cooling. The grooves on the side of the drill should be cleaned of bone, which collects here. If the grooves are not kept clean, the drilling temperature may become too high. The countersink edges are designed to provide an even surface for the flange of the implant to improve initial stability. The depth of the countersink should be kept at a minimum, however, because the cortical shell has high density and provides good implant stability; this is especially important if the bone is thin and soft as in children (see section �Baha in Children�).
Subcutaneous Soft Tissue Reduction The implant site has now been secured, and the soft tissue reduction can start (see Fig. 33-7F). This is one of the most important steps for establishing a lasting and reaction-free skin penetration for the years to come. The soft tissue reduction starts when the implant site has been secured, but before placing the implant. The selfretaining retractors are removed, and two small skin hooks are used. The assistant stretches out the skin flap with the hooks, and by folding the flap back and forth the surgeon can remove the soft tissue under the base of the flap (see Fig. 33-7G). The other three sides are undermined in the same way with the aim of creating tension-free sloping edges down to the periosteum (see Fig. 33-7H).
Placing the Baha Coupling The drill speed is set on low, which is 15 to 30 rpm, and the torque is adjusted to suit the bone quality. In the sclerotic mastoid of a patient with chronic ear disease, the torque often needs to be 40 to 45 NCM. In soft bone and with a thin cortical shell, a much lower setting should be used, 20 to 25 N/cm2, to avoid trauma to the implant site. The Baha abutment inserter is placed in the handpiece. The self-tapping implant with premounted abutment is picked up from its plastic container and should not be touched by anything (see Fig. 33-7I). The implant is kept over the implant site, and the engine is started. With a slight pressure, the self-tapping implant finds its way. After the first few turns, no pressure is needed. When the
406
OTOLOGIC SURGERY
implant is placed, the engine stops automatically at the set torque. The surgeon should now turn the handpiece about 5 degrees counterclockwise, put his index finger close to the abutment, and remove the abutment inserter from the abutment; this is to reduce any counter lever effect on the implant (see Fig. 33-7J).
Suturing the Flap The thin flap is positioned over the coupling, and using a biopsy punch, a hole is made in the flap. The coupling is pulled through the hole, and the edges of the flap are sutured down to the periosteum with 5-0 nylon monofilament (see Fig. 33-7K). A healing cap is snapped onto the coupling, and a piece of 1 cm wide cotton gauze, soaked in antibiotic ointment, is very loosely packed under the cap to avoid swelling and hematoma (see Fig. 33-7L). An ordinary mastoid dressing is applied, which is removed after 1 to 2 days.
Three weeks after surgery, the patient is back for the next postoperative check, and, in most cases, a time for fitting the Baha is decided. Most of our patients get their Baha about 6 weeks after surgery (Fig. 33-9). This timing is mainly due to practical and logistical issues and could probably be shortened. During the first 2 weeks, after the healing cap has been removed, the patient should use a mild ointment to keep the implant area clean and soft. After this time, the patient is told to use some mild soap or shampoo every day. The situation 3 months after surgery is shown in Figure 33-10 with the Baha coupling in place. Cleaning is crucial, and a high level of personal hygiene is imperative. Relatives are often asked to participate in the postoperative visits to be instructed on how to help the patient. After this, the patient comes back for follow-up at 6 to 12 month intervals. The Baha is fitted by an audiologist who selects and adjusts the device according to the patient’s audiogram
POSTOPERATIVE MANAGEMENT In most cases, this surgery is performed as a day procedure. After a few hours in the recovery room, the patient is discharged and can go home. The patient is instructed to remove the mastoid dressing the following day. The patient should clean the area with ordinary soap and water, and leave it without any further dressing. The healing cap with the gauze is left for 7 to 10 days. At this time, the patient comes for the first postoperative visit. The healing cap is snapped off, and the gauze is removed (Fig. 33-8). Often some or all stitches can be removed. If the healing is delayed for some reason, the healing cap can be put back on, and fresh gauze can be loosely placed around the coupling for another few days. Sometimes a healing cap is put on without any gauze just to provide a little protection.
FIGURE 33-9. Implant area after 6 weeks.
FIGURE 33-8. Implant area 10 days after surgery.
FIGURE 33-10. Healing is complete 3 months after surgery.
Chapter 33 • The Bone-Anchored Cochlea Stimulator (Baha)
and preferences. In this fitting procedure, it is important to take into account the patient’s lifestyle. Six weeks later, the patient returns for what is often a final adjustment.
ALTERNATIVES IN SURGERY The above-described technique is the one that we use at the Implant Unit in Göteborg. It has been slowly refined over the 30 years that we have been using Baha. Some colleagues use different techniques, however. Regarding preparing the implant site, the differences are small. Some surgeons do not follow all the steps described previously, but the success rate for establishing osseointegration seems to be very high, at least in adults. Instead of the Baha dermatome, some surgeons use a U-shaped manual flap. One reason cited for this technique is that the size of the implant area could be adjusted to the amount of soft tissue present. The thinning of the flap is the same, however. A straight vertical incision has been suggested, and could probably be used, too, but making the adequate soft tissue reduction would likely be more difficult, especially in an overweight patient.
407
Adverse Skin Reactions Adverse skin reactions of any significance are very infrequent during the years after Baha surgery with the manual technique and with the dermatome. In a study with 4 to 8 years of follow-up, a blade was used to thin the flap manually.27 Using the classification system by Holgers and associates,28 only 2.8% of this group, with a grade 2 condition or worse, needed active treatment. With the Baha dermatome, the corresponding figure was 2%.29
PITFALLS AND COMPLICATIONS IN SURGERY AND DURING FOLLOW-UP For a properly trained otologic surgeon, Baha surgery is simple and presents very few risks. Exposure of the dura mater or the wall of the sigmoid sinus is of no significance. Damage to these structures would be handled the same way as in ordinary mastoid surgery. A pedicle periosteal flap would stop cerebrospinal fluid leakage and blood from the low-pressure sigmoid sinus. Entering air cells during drilling is easily handled. If there is bone trabecula to support the implant, it could be used; if not, a new implant site close by is selected. When surgery is performed under local anesthesia, the patient could experience some discomfort because of suctioning.
FIGURE 33-11. A patient with medium-sized flap necrosis and exposed bone 3 weeks after surgery.
Flap Necroses Using the technique described, the frequency of flap necroses during the first 6 postoperative weeks was found to be 3%.26 Only minor necroses (defined as <25% of the flap area) have been noted. Before the introduction of the Baha dermatome, the flap was thinned manually with a knife. With this method, the frequency of minor necroses was 9.2% and 1.3% for medium necroses. A flap necrosis, whether minimal or complete, has a very good prognosis with local treatment with mild ointment. Healing may take a long time, however, up to several months. A medium-sized flap necrosis is shown in Figure 33-11, and the final result after 3 months is shown in Figure 33-12.
FIGURE 33-12. Same patient as in Figure 31-11, 3 months later.
408
OTOLOGIC SURGERY
FIGURE 33-13. A patient with bilateral radical cavities has been wearing his Baha for more than 10 years.
Implant Stability The possibility to establish osseointegration for the Baha is high. In an early study with 149 Baha patients, we had a success rate of 97.3% over 1 to 4 years. During the following 4 years, the success rate declined to 94.7%.27 In a study with 144 implants, the success rate was 98.6% over 3 years.30
HANDLING OF COMPLICATIONS Related to Bone Pain when the Baha coupling is touched is very rare. In almost all cases, the pain is a sign of loss of integration, and the implant comes out in a couple of weeks. In children, such an implant may be left without any load, and reintegration may occur. In adults, reintegration would not occur. A new implant could be placed just a few millimeters from the previous implant site. New bone formation is common, especially in younger patients. If the bone formation is a problem, the coupling is temporarily removed, and the excess bone is chiseled or drilled away. The coupling is put in place again. Bone resorption is often seen under the flange in implants that have been in place for several years. This resorption is often restricted to the nonthreaded part of the implant just under the flange. This finding is very seldom of any clinical importance, and the patient can continue to use his or her Baha.
Related to Soft Tissue A slight redness around the implant, grade 1 according to the Holger classification,28 was noted in 4.3% of observations over a 4 to 8 year period.31 The first thing to check in a patient who has skin problems is if the implant is mobile. The internal screw keeping the abutment connected to the implant sometimes may come loose and
has to be tightened. Over the same period, 69.7% of the patients did not have one episode of adverse skin reaction, and 87.1% had only one episode. The handling of a grade 1 reaction is often simply to ask the patient to intensify the daily cleaning procedure and apply some mild ointment. Extra outpatient controls are seldom needed. Even in more extensive negative soft tissue reactions, there is seldom any pain. As mentioned earlier, pain is often a sign of loss of implant integration. In the situation in which there are grade 2 to 4 skin reactions, the surgeon has to be more active. Replacing the healing cap and applying gauze with an antibiotic ointment is often the first step. If irritation continues after 7 to 10 days, a regrafting may be necessary. A thin graft can be taken from the fold behind the ear. Temporary removal of the abutment is often the next step leaving the implant in place. In rare situations, the implant has to be removed. Removal can be done through unscrewing the implant or drilling it out with the guide drill. Culturing and oral antibiotics could be indicated if there are prolonged problems.
BAHA IN CHILDREN Most authorities agree that hearing is crucial for normal speech, language, and mental development.18 This also seems to be true for unilateral hearing impairment. If a hearing-impaired child cannot use a conventional air conduction hearing aid and is not a candidate for a cochlear implant, the Baha is a strong alternative.
Surgical Aspects The placement of the implant is of special concern in children. In children with a malformed ear, it is important to place the implant about 65 to 70 mm from an anticipated external ear canal opening. This position is often much more posterior than expected, but would not
Chapter 33 • The Bone-Anchored Cochlea Stimulator (Baha)
jeopardize autogenous external ear reconstruction later on. The dura mater of the middle cranial fossa is often low, and the facial nerve could have a more superficial route. The surgical procedure is basically the same as for adults with the important difference that a two-stage procedure is used. At the first stage, the implant is placed, but the Baha coupling is not attached. The implant is left to integrate for 4 to 6 months, after which the implant is exposed, and the Baha abutment is attached. In children, the dermatome is less suitable, and a U-shaped incision is made manually at the first and at the second stage. The soft tissue reduction is as important in children as in adults, even if the amount of subcutaneous tissue is much less. The bone in a young child is soft because of a low mineral and high water content compared with adult bone. In drilling, care should be taken not to damage the dura mater or the wall of the sigmoid sinus. It does not matter, however, if these structures are found at the bottom of the implant site. The spiral drill with its countersink edges should be used without much pressure because the sharp edges could easily cut too much soft bone. The implant site often is in bone marrow space, which is of no importance. The thickness of the temporal bone is crucial for implant integration. If a 3 mm long fixture were to be inserted, a minimum of 2.5 mm bone would be needed. The bone thickness at the implant site has been measured in 30 children.32 The mean thickness, at age of 5 years, was found to be only 2 mm, but with great variations. In patients younger than 4 years, the bone thickness is often reduced, and bone augmentation to be able to install an implant may be needed. A bone augmentation technique of this kind has been described earlier.33 With expanded polytetrafluoroethylene membranes, appositional bone can be directed to grow under the flange of the fixture. The technique seems biologically sound because it can be used together with standard fixture and abutment placement, and it does not change therapy planning or time needed for osseointegration. If possible, a 4 mm implant should also be used in young children because the failure rate for 3 mm implants is much higher than for 4 mm implants. If an expanded polytetrafluoroethylene membrane is used, this has to be removed at the second stage. The time between first and second stage of surgery has been extended in the youngest children to at least 6 months. The decision to extend the healing time period was based mainly on clinical parameters, including the thickness of the bone, the softness of the bone, other malformations of the temporal bone, and the primary stability of the implant. When a child is 3 years old, the standard 3 to 4 months between the first and second stage of surgery seems applicable. In a study of 160 children, the success rate for osseointegration was 94.4%—that is, a failure rate of 5.6%.34 If these figures are broken down, however, it becomes
409
evident that the implant losses for the 3 mm implants were 8%, and the corresponding figure for the 4 mm implant was 1.7%
SUMMARY The Baha is a good option in selected patients who are hard of hearing. Patients with conductive or mixed hearing loss are often very satisfied with the Baha if the cochlea function is not too low. Patients with unilateral cochlear deafness or single-sided deafness and normal hearing on the contralateral side constitute a relatively new group of patients in which experiences with the Baha are most promising. If the guidelines for the surgical procedure are followed, the possibility of establishing osseointegration is very good. There seems to be a consensus regarding the prerequisites for a lasting and reaction-free skin penetration. Where the skin is penetrated, it needs to be hairless and laid down to the periosteum to avoid any relative mobility between implant and skin.
REFERENCES 1. Brandt A : On sound transmission characteristics of the human skull in vivo. Thesis; Technical Report No. 61L. Göteborg. Sweden, School of Electrical Engineering, Chalmers University of Technology, 1989. 2. Håkansson B, Tjellström A, Carlsson P: Percutaneous vs. transcutaneous transducers for hearing by direct bone conduction. Otolaryngol Head Neck Surg 102:339-344, 1990. 3. Håkansson B, Tjellström A, Rosenhall U: Hearing thresholds with direct bone conduction versus conventional bone conduction. Scand Audiol 13:3, 1984. 4. Håkansson B, Carlsson P, Tjellström A : The mechanical point impedance of the human head, with and without skin penetration. J Acoust Soc Am 80:1065-1075, 1986. 5. Carlsson P, Håkansson B, Rosenhall U, Tjellström A : A speech reception threshold test in noise with the boneanchored hearing aid: A comparative study. Otolaryngol Head Neck Surg 94:421-426, 1986. 6. Håkansson B, Tjellström A, Rosenhall U: Acceleration levels and threshold with direct bone conduction versus conventional bone conduction. Acta Otolaryngol 100:240-252, 1985. 7. Kylén P, Arlinger S: Drill-generated noise levels in ear surgery. Acta Otolaryngol 82(5-6):402-409, 1976. 8. Stenfelt S : Hearing by bone conduction: Physical and physiological aspects. Technical Report No. 358. Göteborg. Sweden, School of Electrical and Computer Engineering, Chalmers University of Technology, 1999. 9. Brånemark PI, Hansson B, Adell R : Osseointegration in the treatment of the edentulous jaw: Experience from a 10 year period. Scand J Plast Reconstr Surg 16:1-132, 1977. 10. Brånemark PI : Introduction to osseointegration. In Brånemark PI, Zarb G, Albrektsson T (eds): Tissue-Integrated Prostheses. Chicago, Quintessence, 1985, pp. 11-76.
410
OTOLOGIC SURGERY
11. Albrektsson T, Brånemark P-I, Hansson H -A, et al: The interface zone of inorganic implants in vivo: Titanium implants in bone. Ann Biomed Eng 11:1-27, 1983. 12. Tjellström A : Osseointegrated systems and their application in the head and neck. Adv Otolaryngol Head Neck Surg 3:39-70, 1989. 13. Kasemo B, Lausmaa J: Aspects of surface physics on titanium implants. Swed Dent J Suppl 28:19-36, 1985. 14. Devge C, Tjellström A, Nellström H : Magnetic resonance imaging in patients with dental implants: A clinical report. J Oral Maxillofac Implant 12:354-359, 1997. 15. Tjellström A, Granström G: One stage procedure to establish osseointegration: A zero to five years follow-up report. J Laryngol Otol 109:593-598, 1995. 16. Granström G: Osseointegration in irradiated cancer patients: An analysis with respect to implant failures. J Oral Maxillofac Surg 63:579-585, 2005. 17. Jahrsdoerfer R A, Yeakley JW, Aguilar E A, et al: Grading system for the selection of patients with congenital aural atresia. Am J Otol 13:6-12, 1992. 18. Cho Lieu JE: Speech-language and educational consequences of unilateral hearing loss in children. Arch Otolaryngol Head Neck Surg 130:524-530, 2004. 19. Sheehan PZ, Hans PS : UK and Ireland experience of bone anchored hearing aids (BAHA) in individuals with Down syndrome. Int J Pediatr Otorhinolaryngol 70:981986, 2006. 20. Snik A, Mylanus E, Cremers C, et al: Consensus statement on the BAHA system: Where do we stand 2004? Nijmegen Consensus Meeting. Ann Otol Rhinol Laryngol 114(Suppl 195):1-12, 2005. 21. Wazen JJ, Spitzer J B, Ghossaini sn, et al: Transcranial contralateral cochlear stimulation in unilateral deafness. Otol Head Neck Surg 129:248-254, 2003. 22. Niparko J K, Cox K M, Lustig L R : Comparison of the bone anchored hearing aid implantable device with contralateral routing of offside amplification in the rehabilitation of unilateral deafness. Otol Neurotol 24:73-78, 2003. 23. Hol M K S, Bosman A J, Snik A FM, et al: Bone-anchored hearing aid in unilateral inner ear deafness: A study of 20 patients. Audiol Neurotol 9:274-281, 2004.
24. Lin L -M, Bowditch S, Anderson M, et al: Amplification in the rehabilitation of unilateral deafness: Speech in noise and directional hearing effects with bone anchored hearing and contralateral routing of signal amplification. Otol Neurotol 27:172-182, 2006. 25. Vaneecloo FM, Ruzza I, Hanson J N, et al: Appareillage mono pseudo stereophonique par BAHA dans les cophoses unilateral: A propos de 29 patients. Rev Laryngol Otol Rhinol (Bord) 121:343-350, 2001. 26. Tjellström A, Granström G: How we do it: Frequency of skin necrosis after BAHA surgery. Clin Otolaryngol 31:216-220, 2006. 27. Reyes R , Tjellström A, Granström G: Evaluation of implant losses and skin reactions around extra-oral bone anchored implants: A zero to eight years follow-up report. Otolaryngol Head Neck Surg 122:272-276, 2000. 28. Holgers K M, Tjellström A, Erlandsson B E, et al: Soft tissue reactions around percutaneous implants: A clinical study of tissue conditions around skin-penetrating titanium implants for bone-anchored hearing aids. Am J Otol 9:56-59, 1988. 29. Stalfors J, Tjellström A : Skin reactions after BAHA surgery: a comparison between the U-graft technique and the BAHA dermatome. Otol Neurotol 29:1109-1114, 2008. 30. Tjellström A, Granström G, Odersjo M : Survival rate of self-tapping implants for bone-anchored hearing aids. J Laryngol Otol 121:101-104, 2007. 31. Jacobsson M, Tjellström A : Clinical application of percutaneous implants. In Szycher M (ed): High Performance Biomaterials: A Comprehensive Guide to Medical and Pharmaceutical Applications. Basel, Technomic Publishing, 1991, pp 207-229. 32. Granström G, Bergström K, Odersjö M, Tjellström A : Osseointegration in children: Experience from our first 100 patients. Otolaryngol Head Neck Surg 125:85-92, 2001. 33. Granström G, Tjellström A : Guided tissue-generation in the temporal bone. Ann Otol Rhinol Laryngol 108:349354, 1989. 34. Tjellström A, Håkansson B, Granström G: Bone a nchored hearing aids: Current status in adults and children. Otolaryngol Clin North Am 34:337-364, 2001.
34
Surgery of the Endolymphatic Sac Mark D. Packer and D. Bradley Welling Videos corresponding to this chapter are available online at www.expertconsult.com.
Meniere’s disease is a clinical syndrome consisting of fluctuating-progressive hearing loss, episodic vertigo lasting 20 minutes to 24 hours, tinnitus, and aural fullness that is diagnosed when other diagnoses have been excluded. Surgical treatment of Meniere’s disease has been controversial and a focus of debate ever since Portmann1 first proposed opening the endolymphatic sac in 1927. Despite controversy, however, surgical manipulation of the endolymphatic sac to alleviate the debilitating symptoms of Meniere’s disease has been a mainstay of surgical treatment when conservative therapy has failed. The mechanism of symptomatic relief from shunting, decompressing, or excising the endolymphatic sac is also controversial. For the purposes of this chapter, the definition of definite Meniere’s disease published by the American Academy of Otolaryngology–Head and Neck Surgery (AAOHNS) Committee on Hearing and Equilibrium in 19952 is used when referring to Meniere’s disease. The criteria given are as follows: Definite Meniere’s disease requires two or more definitive episodes of vertigo with hearing loss, tinnitus, or aural fullness. Certain Meniere’s disease has the same symptom complex as definite Meniere’s disease and requires histopathologic confirmation of endolymphatic hydrops. Probable Meniere’s disease requires only one definitive episode of vertigo and requires hearing loss, tinnitus, or aural fullness. Possible Meniere’s disease is defined as definitive vertigo without associated hearing loss, or hearing loss with nondefinitive dysequilibrium.2 (The former condition has been previously referred to as vestibular Meniere’s disease and the latter cochlear Meniere’s disease). Meniere’s disease is difficult to study because of its fluctuant nature and the minimum 2-year time course over which results of interventions must be documented.2 The absence of a definitive test for Meniere’s disease necessitates diagnosis based on historical data. Whether or not placebo effect is the cause of success in the endolymphatic sac decompression is also controversial. Before
discussing the techniques and outcomes of the endolymphatic sac procedures, we first consider the underlying anatomy and physiology.
ENDOLYMPHATIC ANATOMY AND EMBRYOLOGY Scarpa (1752-1832) discovered and described endolymph and the membranous labyrinth in 1789,3 72 years before Méniere4 ascribed symptoms of the clinical syndrome to the inner ear. Understanding the embryology of the endolymphatic system may provide some clues to understanding Meniere’s disease. During the fourth week of embryogenesis, three buds of the primordial otocyst appear representing the pars superior, pars inferior, and endolymphatic duct, which eventually develop into the utricle and the semicircular canals, the saccule and the cochlear duct, and the endolymphatic duct and sac.5 The endolymphatic duct leaves the medial vestibule and courses dorsally through the bony vestibular aqueduct to terminate on the posterior surface of the temporal bone, enveloped within dural folds of the posterior fossa. The short, straight endolymphatic duct acquires its mature hook shape configuration by the fourth year of life.6,7 Histologic and functional maturation of sac elements may predate complete anatomic maturity. Potassiumrich endolymph fills the endolymphatic sac and duct, the saccule and utricle, the membranous semicircular canals, and the cochlear duct or scala media. These structures are interconnected by the smaller utricular duct, saccular duct, and ductus reuniens. The membranous endolymphatic structures are surrounded by the sodium-rich perilymph that fills the periotic spaces within the bony labyrinth.8 The position of the endolymphatic sac along the posterior fossa dura is relatively constant, but its size and the amount of bony covering by the operculum are variable.9 In most patients, 50% of the sac lies outside of the temporal bone, and 50% of the sac is intraosseous. Approximately 10% of sacs are completely extraosseous along the posterior fossa dura.10 The sac can extend posterolaterally 411
412
OTOLOGIC SURGERY
to cover the lateral sinus.11 The morphology of the sac is also variable. The distal sac has a smooth open lumen within the dura mater. Its cuboidal epithelium contains light and dark cells.12 The lining of the intermediate portion of the sac shows more complex epithelial folds forming papillae and crypts of tall columnar light and dark cells. Cells of the intraosseous proximal rugose sac are intermediate between the taller, more distal cells and the squamous-to-cuboidal cells of the duct. The duct narrows at its isthmus to 0.1 to 0.2 mm in diameter. Luminal folding and transversely oriented tubules make the endolymphatic sac a more complex structure than it otherwise outwardly appears.13 The normal bony vestibular aqueduct is readily apparent on high-resolution computed tomography (CT) scanning of the temporal bone. It is funnel-shaped or tubular with the width of its external aperture averaging 6 mm.14,15 Radiographic observation of the affected ear in patients with Meniere’s disease showed a filiform narrowing of the external aperture averaging 2.2 mm.14 The amount of narrowing of the external aperture was also shown to be correlated with an increasing percentage of positive electrophysiologic measures in the affected ears of patients with Meniere’s disease.15 Statistically significant differences in the percentage of patients with enlarged summating potential-to-action potential (SP: AP) ratios by transtympanic electrocochleography were seen when correlated with the size of the external aperture of the vestibular aqueduct. An increased SP:AP ratio was noticed in 95% of ears with nonvisible external apertures, 91% when the aperture was less than 5 mm, 58% when the aperture was 5 to 7 mm, and 29% when the aperture was greater than 7 mm. The endolymphatic duct and sac can be seen on high-resolution fast spin echo magnetic resonance imaging (MRI). The endolymphatic sac and duct were seen on MRI of 20 temporal bones in healthy subjects using strongly T2-weighted sequences and postprocessing software.16 A retrospective review of 42 ears with MRI data that underwent endolymphatic sac surgery correlated surgical findings with the ability to image the endolymphatic sac and duct.17 Surgical findings were classified as normoplastic in 17, atrophic in 14, and invisible in 11. Proton density imaging and T2 sequencing positively identified the endolymphatic duct and sac in 14 patients. The endolymphatic sac and duct were shown by proton density imaging alone in 14; neither proton density imaging nor T2 sequencing showed images in the remaining 14 ears. Findings at surgery showed statistically significant correlation with the ability to identify structures on imaging. Normoplastic surgical anatomy was identified on both imaging modalities; however, atrophic sacs were rarely seen on T2 imaging. In another study using submillimeter MRI, the endolymphatic ducts were visualized in 29% of patients with Meniere’s disease and in 91% of healthy individuals. Temporal bone measurements between the posterior
semicircular canal and the subarachnoid space, and the vestibule and the subarachnoid space were shorter in Meniere’s disease patients than in healthy individuals. It was noted anecdotally that endolymphatic shunt surgery was more effective in the few patients with visualized ducts compared with patients with nonvisualized ducts.18 Currently, the practicality of imaging in Meniere’s disease or for endolymphatic surgery lies primarily in ruling out retrocochlear pathology. From a surgical perspective, the endolymphatic sac is generally approached through the mastoid bone, and is isolated on the lateral surface of the posterior fossa dura. When performing a suboccipital craniotomy, the sac can often be identified on the posterior petrous aspect of the temporal bone. Discussing the location of the endolymphatic sac, Gibson19 noted that the extraosseous portion is difficult to define surgically, appearing only as a thickened area of dura. He also noted that after splitting the layers of the endolymphatic sac no endolymph is seen, and usually no significant electrophysiologic changes occur. Huang20 attributed higher success in endolymphatic sac surgery to definite identification of the sac, with entry into the true sac lumen, and preservation of the sac anatomy. Amiratti and colleagues21 advocated preserving the integrity of the endolymphatic system, suggesting severe audiovestibular disturbances that may follow sac disturbance; however, our experience is that patients tolerate complete excision of the sac without hearing loss, whether incidental in other cranial base procedures, or intentional for complete endolymphatic sac ablation. Important topographic landmarks for identifying the endolymphatic sac exist from the transmastoid or posterior fossa approach. The transmastoid extradural landmarks for localization of the endolymphatic sac and for preservation of labyrinthine structures include Donaldson’s line, which is an imaginary line drawn posteriorly through the plane of the horizontal semicircular canal (see Fig. 34-1), and measurements delineating the hard angle. The endolymphatic sac is generally found along the posterior fossa dura inferior to Donaldson’s line. Caution must be used when identifying the sac to avoid damage to the facial nerve and the posterior semicircular canal. Anatomic variants of normal temporal bone anatomy have been associated with Meniere’s disease. An understanding of potentially altered anatomy is important for surgical planning of sac procedures.22 Hypoplasia of the mastoid air-cell system, hypocellularity of periaqueductal cells around the endolymphatic duct and sac, reduction of the aditus ad antrum, and hypoplasia of the facial recess all have been described. Intradural identification of the endolymphatic sac in relation to anatomic structures of the posterior fossa places the sac 10 to 15 mm lateral to the internal auditory meatus, and 11 to 17 mm posterosuperiorly to the eleventh cranial nerve in the jugular foramen.21 Typically, the thickening of the dura and the bony ledge of the operculum pinpoint the location of the sac.
Chapter 34 • Surgery of the Endolymphatic Sac
ENDOLYMPHATIC SAC PHYSIOLOGY Surgical shunting of the endolymphatic sac was proposed soon after the initial anatomic observations of hydrops in the endolymphatic compartments to alleviate inferred dysfunction.1,23 The presumed longitudinal flow of endolymph from the stria vascularis to the endolymphatic sac and the role of the sac as a primary resorptive organ have long been assumed.24 Radial flow of endolymph has also been shown, however.25 Endolymph is produced by dark cells located largely in the stria vascularis, but also in vestibular ampullae, within the maculae of the saccule and utricle, and along the endolymphatic duct. Maintenance of a potassium-rich endolymph produces the endocochlear potential, a DC voltage gradient, to drive the transduction process important in the detection of sound, motion, and position.26,27 This is a pH-sensitive process and is based on an active transport system in the vestibular dark cells by a sodium–hydrogen ion exchange system.27,28 Local production concentrated within strial dark cells and radial movement of endolymph with local chemical exchange throughout its course maintain a chemical balance and gradient that promote physiologic endolymphatic function, and may promote a slow linear flow toward the sac.25 Other theories of endolymphatic fluid homeostasis exist. Salt29 stated that direct measurements of the dispersal of markers in endolymph fail to support dynamic flow theories, and suggested that, in the normal state, there is negligible flow. The ionic component of endolymph is maintained through single cell transport of ions. This local control theory is overridden when endolymph volume is abnormally high or low, with the endolymphatic sac acting as a regulator of bidirectional flow in response to volume needs. The endolymphatic sac functions as the master volume regulator by numerous observed characteristics. In contrast to endolymph throughout the inner ear, a gradient exists along the duct leaving the electrolyte state in the endolymphatic sac high in sodium and low in potassium.30 An active equilibration of ions creates an osmotic potential that may influence the transepithelial flow of fluids. Higher concentrations of Na+,K+-ATPase are seen in the endolymphatic sac, but diminish proximally along the endolymphatic duct.31 Several other findings suggest an active role of the endolymphatic sac on endolymph fluid homeostasis.32-34 Aquaporin 2, vasopressin type 2 receptor, and transient receptor potential channel vanilloid (TRPCV), subfamily type 1 and 4, were found in the epithelial lining of the endolymphatic sac, but not in other extracellular tissues, although TRPCV 1 was seen in the surrounding vasculature. Similar findings are seen within the kidneys, suggesting a parallel role in fluid filtration and resorption.32 A unique protein, saccin, secreted by the endolymphatic sac, acts within the kidney as an endogenous inhibitor of sodium reabsorption. Intracellular morphology of the endolymphatic sac chief cells possesses organelles capable
413
of endocrine function, and cellular ultrastructure consistent with merocrine activity also is observed.33,34 Biochemical and cellular findings within the endolymphatic sac support its functional role in phagocytic, immune, and allergic responses of the inner ear.35-39 Volume may also be regulated by the functional mechanical entity of an extracellular matrix of interstitial cells that support the endolymphatic duct, and take part in the control of inner ear fluid dynamics and endolymph resorption.40 As new techniques are developed to detect and monitor minute fluid volumes and changes, and to evaluate the chemical composition and cellular characteristics of the endolymphatic sac, a better understanding of the physiologic role and pathophysiologic state of the endolymphatic system in Meniere’s disease will be achieved.26,41
ENDOLYMPHATIC HYDROPS: PATHOPHYSIOLOGY It was not until the histopathologic observation of Hallpike and Cairns in 193842 that a proposed malfunction of the endolymphatic system was correlated with the clinical syndrome. These authors showed dilated endolymphatic spaces in temporal bone specimens from two patients with the clinical symptoms of Meniere’s disease who died after neurotomy of CN VIII. These findings showed end organ changes of the inner ear in patients with hearing and balance symptoms. The hydropic state of the inner ear has been confirmed in other temporal bone studies and described as the primary pathologic correlate of Meniere’s disease.43-45 The underlying cause of the hydropic state is at present unknown, although many theories exist. Hydrops is seen more often in the cochlea and saccule, structures derived from the later developing pars inferior.43 Congenital insults or developmental aberrations later in the course of embryogenesis could presumptively account for this difference. A familial connection is seen in the history of 20% of patients clinically diagnosed with Meniere’s disease.44 This connection, in consideration of the embryologic and anatomic findings associated with Meniere’s disease, suggests a multifactorial predisposition to developing endolymphatic hydrops. Whether this precondition is genetic or related to shared environmental factors or insults is yet to be determined. The pathophysiologic state of the endolymphatic system has been partially modeled in mice by experimental destruction and obstruction of the sac or duct in attempts to pinpoint the underlying mechanism of the hydropic condition.46,47 Although these experiments were able to reproduce hydrops and audiometric findings similar to Meniere’s disease, vestibular dysfunction or vertigo was noticed only after placing the animals in a head-down position theoretically by inducing additional pressure within inner ear fluids.48
414
OTOLOGIC SURGERY
The predominant theories that could explain the symptom complex of Meniere’s disease are based on the observed pathologic and induced experimental evidence of hydrops. The various temporal bone findings are summarized succinctly by Costa and associates,49 and include ruptures of the membranous labyrinth, fistulas of the membranous labyrinth, collapse of the membranous labyrinth, obstruction of longitudinal flow, vestibular fibrosis, sensory lesions, and neural lesions. Schuknecht50 proposed the rupture theory. He reasoned that distention of the endolymphatic space with eventual membrane rupture could cross-contaminate perilymphatic spaces and toxify delicate sensory hair cells with the potassium-rich endolymph. In the distention theory, Paparella51 described decompensation of radial flow as the perilymphatic spaces ebb, leading to largely longitudinal flow along the hydropic membranous pathway with the saccule acting as a reservoir for the excess endolymph. As the dilated saccule encroaches on the confines of the vestibule, mechanical interference of cochlear and vestibular function occurs by inhibition of traveling waves and physical contact with the crista ampullaris. The drainage theory, as presented by Gibson and Arenberg,10 suggests that obstruction of a narrowed endolymphatic duct divests the sac of endolymph. The endolymphatic sac responds by secreting glycoproteins and saccin. Glycoproteins act osmotically by “pulling” endolymph toward the sac, and saccin stimulates secretion of endolymph from dark cells that distend the endolymphatic spaces “pushing” against the obstruction toward the sac. The obstructing debris ultimately and suddenly passes, and the resultant rapid flow of endolymph purportedly brings on an acute vertiginous episode. Gibson and Arenberg suggested that patients with Meniere’s disease with larger vestibular aqueducts could experience resolution of auditory symptoms as in Lermoyez’s syndrome after clearance of the obstructing debris. Patients with Meniere’s disease have been shown to have widening of the vestibular aqueduct aperture, but, although enticing to establish a mechanism of pathology, patients with Lermoyez’s syndrome have not been shown to have wider vestibular aqueducts than other patients with Meniere’s disease. Gibson and Arenberg also theorized that Tumarkin crises could be the effect of a membrane rupture in the overdistended endolymphatic space. Although Meniere’s disease has no known etiologic cause by definition, there are several etiologies of secondary Meniere’s syndromes. Meniere’s syndrome can be mimicked by disease processes in all categories— traumatic (inner ear fracture, perilymphatic fistula), infectious (viral, syphilis), inflammatory (autoimmune, allergy, Cogan’s syndrome, sarcoidosis), metabolic (diabetes), congenital/genetic (inner ear malformations, enlarged vestibular aqueduct), neoplastic (vestibular schwannoma), vascular (migraine, hemorrhage), and iatrogenic (stapes/mastoid surgery). These disease processes may lead to secondary endolymphatic hydrops by causing
scarring within the labyrinth; activating inflammatory, cytokine, and complement pathways, yielding edema and fibrosis and altered extracellular matrices; altering hemodynamics that may affect transcellular ionic exchange and alter endolymph homeostasis; creating obstructive immune complexes or cellular debris; and activating cellular responses through humoral messengers. The end result is distention or obstruction of the endolymphatic flow disturbing sensation of cochleovestibular signals and sending distorted messages of sound, station, and motion.
MENIERE’S DISEASE Epidemiology Arenberg and colleagues52 have suggested that incidence and prevalence estimations reported in Meniere’s disease are inaccurately low by not recognizing the early or atypical cases, or cases misdiagnosed by lengthy remissions drawing out the episodic nature. The definition of Meniere’s disease may alter prevalence and incidence numbers. The 1995 AAOHNS Committee on Hearing and Equilibrium recognized that the 1985 diagnostic criteria for Meniere’s disease were rigid to the point of precluding patients who were most likely Meniere’s cases. Current criteria are listed at the beginning of this chapter.2 These gradations may alter the epidemiologic accounting of Meniere’s disease because they allow for inclusion of patients who may be early in the course of their disease, and patients who may have milder symptoms of the disease. Several retrospective reviews have shown the prevalence of Meniere’s disease to be 10 to 20 per 100,000 in various populations around the world.49 Using 1995 committee criteria for diagnosis, a more recent study in the Finnish population showed 43 cases per 100,000 population, and an annual incidence of 4.3 per 100,000.53 Costa and associates49 described in detail the epidemiology of Meniere’s disease. It generally manifests in the fifth decade of life. The incidence in childhood is thought to be 1% to 7% of all Meniere’s cases.54 Of 14 children diagnosed with definite Meniere’s disease, 5 were shown to have secondary disease manifesting 5 to 11 years after a history of Haemophilus influenzae meningitis, mumps, temporal bone fracture, and congenital and embryopathic complications.55 Nine of the 14 children had idiopathic disease. These 14 children represented 1% of the combined Meniere’s population of four neurotologic clinics. There is a slight female preponderance. In women with definite Meniere’s disease who were pregnant, a clear decline in symptoms was associated with delivery.49 No socioeconomic, occupational, or racial effect has been consistently shown, although there are statistically increased numbers in married individuals and anxious individuals, and decreased numbers in obese individuals.
Chapter 34 • Surgery of the Endolymphatic Sac
Diagnosis A detailed history is the essence of diagnosis in Meniere’s disease. Classic symptoms were reaffirmed by the AAOHNS Committee on Hearing and Equilibrium in 1995. The symptoms include recurrent spontaneous episodic vertigo, definitive spells including spontaneous rotational vertigo lasting 20 minutes to 24 hours, often prostrating, accompanied by dysequilibrium that may last several days; nausea, commonly with vomiting or retching; but significantly no loss of consciousness. Horizontal or rotary nystagmus is always present during an acute episode. Auditory symptoms include hearing loss (����� fluctuating or �������������������������������������������������� not fluctuating) and aural fullness, tinnitus, or both. Patients who give a clear history of two or more definitive episodes of vertigo as detailed earlier with hearing loss and tinnitus or aural fullness without evidence of inciting illness, injury, or other inner ear pathology meet the diagnosis of definite Meniere’s disease. An audiometric evaluation is necessary, and may specify the side of disease. No other testing is necessary for diagnosis, although ancillary testing may help determine the side of disease and the extent of pathology, and help to identify primary causes of Meniere’s- like, symptoms such as syndromes, or other disease processes if suspected. Electronystagmography can establish vestibular dysfunction, and may delineate the side of the affected ear, but may also be normal early in the course of Meniere’s disease. MRI based on asymmetric hearing loss helps rule out retrocochlear pathology. Brainstem auditory evoked potentials in lieu of MRI are useful for patients with contraindications to or intolerance of MRI. Serology to detect otosyphilis or autoimmune disease is helpful as indicated. Allergy testing for evaluation of patients with seasonal or trigger-induced symptoms can be profitable. Empirical treatment or neurologic consultation for suspected vertebrobasilar or atypical migraine disease may be warranted, especially in patients with atypical symptoms. In children, CT scanning to evaluate inner ear anatomy is advisable. Posturography does not have diagnostic value in the diagnosis of Meniere’s disease, but may be useful in documenting rehabilitation progress. Electrophysiologic tests, including electrocochleography, cochlear microphonics, and vestibular evoked myogenic potentials, have been used as adjuncts in diagnosis, and to monitor the efficacy of treatment in Meniere’s disease.44,56 In 2002, Ge and Shea57 reported a 10-year experience using transtympanic electrocochleography. Transtympanic electrocochleography was performed in 2421 ears of 2140 patients with Meniere’s disease. These authors concluded that electrocochleography is a reliable test to detect the presence of endolymphatic hydrops in Meniere’s disease using parameters of an enlarged SP: AP ratio greater than 0.4, a broadened action potential waveform (>3 ms), and a prolonged action potential latency (>0.2 ms). Combined click and tone burst
415
responses yielded an enlarged SP:AP ratio in 81.7%, and a prolonged AP latency was found in 62.2% of ears with Meniere’s disease. An enlarged SP:AP ratio significantly correlated with stage and duration of disease. The SP:AP ratio was found to be elevated in 71% of stage 1 Meniere’s disease, 82% of stage 2, 85% of stage 3, and 90% of stage 4. The SP:AP ratio was elevated in 43% of patients during their first year of diagnosis and in 100% of patients with the disease for more than 30 years. Cochlear microphonics were also used to assess the presence of hair cell survival, and were found to be present in 69% of ears with pure tone averages greater than 40 dB. It has been suggested that large cochlear microphonics in patients with Meniere’s disease indicate hearing loss resulting from altered cochlear mechanics, whereas severe hearing loss with small cochlear microphonics represents a hearing deficit resulting from hair cell loss.58 Electrocochleography has been shown to have a low sensitivity (57%), but is specific for endolymphatic hydrops (94%).44 In patients with an elevated SP:AP ratio in the immediate preoperative period, a statistically signifi cant intraoperative reduction of the SP:AP ratio has been reported after endolymphatic sac incision and drainage.59 Vestibular evoked myogenic potential is used to assess the vestibulocollic or sacculocollic reflex. This test selectively assesses saccular function and integrity of the inferior vestibular nerve. It is being explored in Meniere’s disease for its potential in identifying endolymphatic or saccular hydrops.60 Vestibular evoked myogenic potential testing may aid in identification of active Meniere’s disease,60 and may help identify ears more prone to develop contralateral Meniere’s disease.62,63 Vestibular evoked myogenic potential may also offer information complementary to electronystagmography,64,65 and provide a way to measure severe saccular dysfunction, a finding associated with Tumarkin crisis.66 These tests show promising diagnostic potential. Further testing experience and validation are necessary.
Treatment Medical Treatment Of patients with Meniere’s disease, 75% to 93% are managed effectively with dietary and medical therapy allowing them control over their vertigo. Education regarding the role of stress is also useful. Dietary measures include reduction or restriction of caffeine, alcohol, and salt. Medical therapy is aimed at affecting fluid dynamics through diuretic therapy. Although many nutritional, vitamin, and medical therapies have been advocated for the treatment of Meniere’s disease, the combination diuretic Dyazide (triamterene and hydrochlorothiazide) is currently the only medical treatment that has shown in a randomized, placebo-controlled trial a statistical decrease in the vestibular symptoms of Meniere’s disease.70 Breakthrough episodes are managed symptomatically with a choice of several options of different classes
416
OTOLOGIC SURGERY
of vestibulosuppressants. In-depth medical treatment of Meniere’s disease is addressed elsewhere. The work-up and treatment of other processes that secondarily produce Meniere’s syndrome, alone or in conjunction with standard medical treatment of endolymphatic hydrops, are usually successful at alleviating symptoms.
Intratympanic Treatment For patients with Meniere’s disease who continue to manifest debilitating episodic vertigo despite strict dietary and medical measures, several surgical options aimed primarily at alleviating the vertigo associated with Meniere’s disease offer highly effective results. Surgical treatments can be categorized as minimally invasive or invasive, and destructive or nondestructive. Pressure equalization tubes have been shown by some investigators to decrease the frequency of vertigo. They may also be used for the instillation of topical medications, or for the adjunctive use of micropressure devices, such as the Meniette device. A brief discussion of the intratympanic administration of medicines follows. Minimally invasive transtympanic application of the aminoglycoside antibiotic gentamicin produces a titration effect, rather than a chemical labyrinthectomy, if administered judiciously, but has an associated risk of hearing loss. Gentamicin is toxic to strial and vestibular dark cells, by reducing production of endolymph to alleviate hydrops, and by ablating sensory hair cells diminishing aberrant end organ signaling. Chia and colleagues71 performed a meta-analysis of 980 patients in 27 studies comparing five different delivery methods. Studies were grouped by the delivery method of the gentamicin and fell into five categories: (1) a multiple daily dosing group, (2) a weekly dosing group, (3) a low-dose group, (4) a continuous microcatheter group, and (5) a titration group. Considering all groups together, complete control of vertigo was achieved in 73.6%, with a significantly greater complete vertigo control rate of 81.7% achieved by the titration method. Overall effective vertigo control was achieved in 90.2%, with a significantly higher rate seen in the titration group (96.3%). The low-dosing method showed significantly lower complete and effective vertigo control rates, whereas the remaining methods showed no significant difference in control. Hearing loss was seen in 25.1% of patients, with profound loss occurring in 0.066% overall. The multiple daily dosing regimens showed significantly higher rates of hearing loss (34.7%), and the weekly method trended to lower rates of hearing loss (13.1%), whereas the remaining groups did not show statistically significant differences in hearing loss rates. Intratympanic gentamicin delivery is also an effective therapy for patients who undergo unsuccessful endolymphatic sac procedures, greatly reducing the need for vestibular neurotomy.72,73 Transtympanic instillation of steroids for control of vertigo in Meniere’s disease is receiving considerable attention because it is nonablative, but studies show
mixed results.74-76 Surveys of the American Otologic Society and the American Neurotologic Society showed that 80% of respondents list steroid use in their protocols for treating Meniere’s disease. As with transtympanic techniques for gentamicin delivery, a multitude of theoretical applications and dosing methods exist and need to be sorted through to determine legitimate outcomes. Injection or instillation of dexamethasone into proximity with the round window allows effective diffusion of the medicine through the round window membrane into the inner ear. The steroid acts intracellularly, reducing inflammatory or immune-mediated responses, and suppressing destructive cytokines that theoretically may alter endolymph homeostasis. Several prospective randomized controlled studies of intratympanic steroids are under way, although their outcomes are not yet available.
Surgical Treatment In 1981, Thomsen and coworkers,67 in a well-designed, placebo-controlled surgical trial of endolymphatic sac decompression, attributed the control of Meniere’s disease symptoms to placebo effect. Pillsbury68 and Welling and Nagaraja69 in separate evaluations of the data found fault with the statistical analysis, and noted statistically significant improvements in five key parameters, including control of vertigo, in the actively treated group over the control mastoidectomy group. Despite this longstanding controversy, endolymphatic sac surgery remains the most common primary surgical treatment for medically recalcitrant Meniere’s disease. Retrospective studies report high efficacy in achieving reliable long-term control of episodic symptoms of vertigo.20,44,69,77-79 The weighted overall control of vertigo achieved by various endolymphatic sac procedures between 1986 and 1996 reached 86%.78 Other surgical procedures that are ablative, such as labyrinthectomy or vestibular nerve sections, have a 90% or greater rate of vertigo resolution; however, the risk of hearing loss and the 25% to 52% risk of developing hydrops in the contralateral ear are of concern. Bilateral vestibular ablation, whether by surgical intent or advanced bilateral Meniere’s disease, is complicated by severely debilitating bobbing oscillopsia or Dandy-Walker syndrome. Until the development and clinical application of vestibular prosthesis, this is a devastating condition. We recommend avoiding destructive surgical treatment of Meniere’s disease as the primary treatment option. According to a more recent survey of members of the American Neurotologic Society and the American Otologic Society, endolymphatic sac surgery was the most commonly employed initial surgical intervention for Meniere’s disease used by 50% of respondents, followed by the primary application of intratympanic gentamicin in 38%.77 Many different manipulations of the endolymphatic sac have been recommended. The sac has been decompressed, destroyed, shunted to the intradural and mastoid cavities, excised, and treated in conjunction with other
Chapter 34 • Surgery of the Endolymphatic Sac
surgical procedures or medical applications in attempts to enhance resorption, promote drainage, or otherwise favorably alter the homeostasis of the endolymphatic system.19,24,80-88 Adjuncts to surgery, such as exposing the opened sac to gentamicin, steroid, or mitomycin C, have been used in attempts to improve success.20,44,89 A range of success from 77% to 100% class A or class B results from these procedures with a weighted average result of 86% has been reported.78 No clear advantage has been shown so far for any particular procedure to the endolymphatic sac.
ENDOLYMPHATIC SAC SURGERY Patient Selection Patients with persistent debilitating vertigo, despite an adequate trial of dietary and medical management, are candidates for endolymphatic sac surgery. Counseling includes a discussion of the natural history of Meniere’s disease, including the possibility of developing bilateral disease, and review of other destructive and nondestructive surgical options. Because of a poor level of medical evidence in the literature, the patient ultimately selects the treatment option most comfortable to him or her. Patients who develop Meniere’s disease in an only hearing ear are not candidates for a destructive or high-risk procedure involving that ear. Endolymphatic sac surgery has been successfully employed to manage these patients.20,44 Preservation of cochlear nerve integrity is important for the rare few patients who end up with nonserviceable hearing as a result of the disease process or its treatment.44 Cochlear implantation has successfully restored hearing for patients in these categories. Patients who have had endolymphatic sac surgery with initial good results may experience recurrence after several years of quiescent disease. Paparella90 revised 5% of his patients who had good results after primary endolymphatic sac enhancement. Revision endolymphatic sac surgery offers results similar to the good results expected with primary sac procedures, showing significant improvement in 76% to 95% of patients.90-92 Similarly, patients with delayed onset endolymphatic hydrops, or secondary Meniere’s syndrome, have been treated with endolymphatic sac surgery to control intractable symptoms of vertigo and aural pressure. Endolymphatic surgery is not recommended for patients with atypical Meniere’s disease. Endolymphatic shunt surgery may be used for, and is well tolerated by, elderly patients.44
Endolymphatic Shunt Procedure The endolymphatic sac and duct are approached through a postauricular incision and a complete mastoidectomy. Facial nerve monitoring is not routinely used, but care is taken to find and protect the facial nerve. A curvilinear incision is created approximately 1 cm behind the
417
osterior auricular sulcus. The incision is carried through p the subcuticular tissues and postauricular musculature to the plane of the superficial layer of the deep temporal fascia and mastoid periosteum. A secondary incision in the deep tissues is created from above the external auditory canal and carried along the linea temporalis posteriorly to just beyond the skin incision. A T-incision is created from the midpoint of this line to the tip of the mastoid bone. The periosteum is elevated anteriorly toward the canal wall exposing the spine of Henle, and the temporalis muscle is elevated superiorly to offer full exposure of the mastoid cortex. This exposure is maintained with self-retaining retractors. After exposure of the mastoid bone, a complete mastoidectomy is done. With a large (Nos. 6 to 8) cutting burr and a large suction-irrigator, the cortical bone is removed and widely saucerized, maintaining the deepest level of dissection in the anterosuperior quadrant below McEwan’s triangle. The tegmen mastoideum, the sigmoid sinus, and the posterior external auditory canal are identified and thinned, and Körner’s septum is removed exposing the aditus ad antrum. The aditus is widened exposing first the horizontal semicircular canal, and next the short process of the incus. Care is taken not to touch the incus with the burr, and to avoid leaving bone dust in the middle ear. The facial nerve is identified at its outer genu and followed down into the vertical mastoid segment (see Video Clip 34-1). This is more readily accomplished after the posterior external auditory canal has been appropriately thinned, and with the patient rotated slightly away from the surgeon. A No. 4 diamond burr is used to remove bone in a line paralleling the descent of the facial nerve. The short process of the incus points to the outer or second genu of the facial nerve as it makes its bend along the inferior edge of the horizontal semicircular canal. Irrigation enhances visualization of the nerve through the bone and prevents heat injuries. Visualization of the entire vertical segment of the facial nerve while maintaining bony integrity of the fallopian canal helps prevent injury while searching for the endolymphatic sac. When the mastoid segment of the facial nerve is safely identified, the bone overlying the sigmoid sinus and the posterior fossa dural plate are thinned by removing retrofacial/infralabyrinthine air cells (Fig. 34-1). Bone removal inferior to Donaldson’s line and posteroinferior to the hard angle helps preserve the posterior semicircular canal. Rarely, blue-lining the posterior canal is necessary to identify an atrophic endolymphatic sac and duct. As the bone anterior to the sigmoid sinus is removed, the posterior fossa dura can be elevated from the medial surface of the dural plate. The bony plate overlying the endolymphatic sac is completely decompressed (Fig. 34-2). The endolymphatic sac is seen as a thickening of the posterior fossa dura, and the duct running anterolaterally tents the dura in the direction of the posterior canal (Fig. 34-3A). When the endolymphatic sac and duct are exposed, the sac and duct are opened along their posterolateral
418
OTOLOGIC SURGERY
6. 1. 2. 3.
7.
4. 8.
5.
FIGURE 34-1. Completed mastoidectomy showing the head of the incus (1), second genu of facial nerve (2), horizontal semicircular canal and plane of Donaldson’s line and triangle outlining the hard angle (3), position of endolymphatic sac (4), sigmoid sinus (5), retrofacial air cell tract (6), posterior semicircular canal (7), and posterior fossa dural plate (8).
surface (Fig. 34-3B). The dural leaves are left intact. The duct is probed with a sickle knife feeling for its entry into the posterior aspect of the petrous bone along the operculum. A wedge of thin silicone elastomer (Silastic) is cut to fit along the duct, and this is placed into the sac and slid anterosuperiorly into the duct with a sickle knife (Fig. 34-4). The surgical site is covered with absorbable gelatin sponge (Gelfoam). The periosteum is reapproximated, and the subcuticular space is closed with interrupted absorbable suture. The incision is covered with antibiotic ointment and a nonadherent Telfa dressing. Most patients are discharged on the day of surgery. We instruct them to use a Glasscock mastoid pressure dressing for 24 hours. We also explain that audiovestibular symptoms may persist or exacerbate during the first several weeks after surgery, although most patients achieve quicker stabilization. The wound is checked at 3 weeks, and an audiogram is obtained at 3 months.
COMPLICATIONS Complications after endolymphatic surgery are rare. Profound hearing loss can occur in 2% of cases, possibly related to labyrinthitis secondary to an exuberant healing process or activation of a latent virus.87,93 When postoperative
hearing loss follows an uneventful surgical procedure, oral or transtympanic steroids are considered, although the efficacy of steroids is unknown. Early postoperative vertigo is seen in approximately 10% to 20% of patients. These patients who do not respond to an initial endolymphatic sac operation, as shown by recurrent vertigo, are candidates for other ablative procedures. After an initially successful vestibular response to endolymphatic sac surgery, recurrence of symptoms after several years of quiescence can be seen, and these patients are offered revision endolymphatic sac surgery. They achieve satisfactory results again in 80% of cases in our experience. If symptoms or vestibular testing indicates involvement of the contralateral ear, the possibility of endolymphatic sac surgery for the newly involved side is discussed.
VESTIBULAR OUTCOMES Because the primary indication for endolymphatic sac surgery is ongoing episodic vertigo despite appropriate medical attempts to control Meniere’s disease, the primary outcomes measure of such surgery is the response of the vestibular system to the surgical manipulation of the sac. As noted earlier, in 1995, the AAOHNS Committee on Hearing and Equilibrium re-established guidelines to
Chapter 34 • Surgery of the Endolymphatic Sac
419
4.
5.
Thinning bone of P.F.D.
FIGURE 34-2. Removal of the posterior fossa dural plate overlying the sigmoid sinus, posterior fossa dura (PFD), and endolymphatic sac and duct. 4, position of endolymphatic sac; 5, sigmoid sinus.
standardize reporting of the outcomes of Meniere’s disease interventions (Tables 34-1, 34-2, and 34-3).2 These guidelines were an attempt to obtain some objectivity in a fluctuating disease process.20,44,94 Retrospective studies following large numbers of patients have collectively shown results with class A and B outcomes in 77% to 100% of patients treated (by the 1995 Committee on Hearing and Equilibrium guidelines).20,44,78,79,84-89 In a study of more than 3000 cases, Huang20 stated that the vestibular results of endolymphatic sac surgery can be expected to yield complete (class A) or substantial (class B) short-term (2 to 3 year) control of vertigo approximately 90% of the time in cases where the sac is definitely delineated, the real sac lumen is entered, the sac’s integrity is preserved, and a sac enlargement technique is employed. Huang20 added that little can likely be done to improve that rate, but that improved technique and earlier treatment may benefit patients’ long-term outcomes. Other authors report that the effect of surgery on the endolymphatic system is nonspecific, unproven, of no or doubtful value, and perpetuated simply out of emotional ties and training.83,95,96 Pensak and Friedman78 divided reports of endolymphatic sac procedures into three groups: (1) reports of mollified or abolished vertigo in 91% to 100% of cases, (2) reports with a lower percentage (77% to 84%) of class A and B results, and (3) reports dissatisfied with the procedure showing 46% to 67% success.78 Grant and Welling’s analysis97 of the
combined results of endolymphatic sac procedures retrospectively reported between 1986 and 1996 collectively showed class A or B outcomes in 86%. Since the AAOHNS Committee on Hearing and Equilibrium reconvened in 1995, there is still much interest and some skepticism in surgery of the endolymphatic system. Weighing the overall combined success of thousands of cases with advantages of a nondestructive outpatient procedure with low morbidity and low risk of complications, and considering the disadvantages of the higher risk and morbidity of destructive procedures in a process with high potential bilaterality, endolymphatic sac surgery is the most common initial treatment for Meniere’s disease when medical management has failed. The assessment of endolymphatic sac procedures is based on observation from four perspectives: (1) studies researching the efficacy of specific treatments, (2) surveys of clinically active members of surgical societies, (3) questionnaires polling patients who have been treated by certain means, and (4) comparison of the documented CPT entries of the different treatments for specific diagnosis.
Endolymphatic Sac Procedures 1995 to Present Since 1995, 33 studies discussing the effects of procedures on the endolymphatic system have been published. Six of these studies were outcomes-based questionnaires looking at the
420
OTOLOGIC SURGERY ELS fully decompressed
1. 2.
7.
4. 8.
5.
A
B FIGURE 34-3. A, Endolymphatic sac (ELS) is decompressed with the duct extending anterosuperiorly under the posterior semicircular canal. 1, head of the incus; 2, second genu of facial nerve; 4, position of ELS; 5, sigmoid sinus; 7, posterior semicircular canal; 8, posterior fossa dural plate. B, Opening ELS with a sickle knife.
impact of endolymphatic surgery on the quality of life.101-106 Six were reported in foreign language journals, and one dealt solely with immediate postoperative recovery.107-113 The remaining 20 studies attempt to classify results with
respect to the 1995 AAOHNS Committee on Hearing and Equilibrium guidelines; however, many still report results in nonstandardized fashions.19,20,24,78,83-85,87,89,91,96,114-122 Three studies used prospective designs.24,78,119
Chapter 34 • Surgery of the Endolymphatic Sac
421
Gelfoam
Silastic
4.
4. 8.
A
B
8.
FIGURE 34-4. A, Triangular shunt fashioned out of thin Silastic is placed within opened endolymphatic sac. B, Gelfoam placed within opened endolymphatic sac dilating the sac lumen. May be used as a chemical reservoir for delivery of medication. 4, Position of ELS; 8, posterior fossa dural plate.
TABLE 34-1 Summary of Reporting Guidelines
TABLE 34-2 Staging of Hearing*
Numerical Value
Class
Stage
Four-Tone Average (dB)
A B C D E F
1 2 3 4
≤25 26-40 41-70 ≥71
0 (complete control) 1-40 41-80 81-120 >120 Secondary treatment initiated owing to disability from vertigo
Numerical value = (X/Y) × 100, rounded to the nearest whole number, where X is the average number of definitive spells per month for the 6 months 18 to 24 months after therapy, and Y is the average number of definitive spells per month for the 6 months before therapy.
*Staging
is based on the four-tone average (arithmetic mean rounded to the nearest whole number) of the pure tone thresholds at 0.5 kHz, 1 kHz, 2 kHz, and 3 kHz of the worst audiogram during the interval 6 months before treatment. This is the same audiogram that is used as the baseline evaluation to determine hearing outcome from treatment. Staging should be applied only to cases of definite or certain Meniere’s disease.
TABLE 34-3 Functional Level Scale The patient completes the following: Regarding my current state of overall function, not just during attacks (check the one that best applies) 1. My dizziness has no effect on my activities at all. 2. When I am dizzy, I have to stop what I am doing for a while, but it soon passes, and I can resume activities. I continue to work, drive, and engage in any activity I choose without restriction. I have not changed any plans or activities to accommodate my dizziness. 3. When I am dizzy, I have to stop what I am doing for a while, but it does pass, and I can resume activities. I continue to work, drive, and engage in most activities I choose, but I have had to change some plans and make some allowance for my dizziness. 4. I am able to work, drive, travel, take care of a family, or engage in most essential activities, but I must exert a great deal of effort to do so. I must constantly make adjustments in my activities and budget my energies. I am barely making it. 5. I am unable to work, drive, or take care of a family. I am unable to do most of the active things that I used to. Even essential activities must be limited. I am disabled. 6. I have been disabled for ≥1 year and/or I receive compensation (money) because of my dizziness or balance problem.
Table 34-4 summarizes studies from 1995 to the present that report according to the 1995 AAOHNS guidelines. Generally, they tell a similar tale—that procedures on the endolymphatic sac offer variable, 53% to 100%, class A and B control of vertigo, but achieve a weighted average of 87%
when all studies are considered, and that when outcomes since 1985 are included, the average is consistent, 86%. Studies in which the minimal length of follow-up was less than 2 years, as recommended by the 1995 guidelines, were excluded. Need for revision has been documented in 5% to 7%
422 OTOLOGIC SURGERY
TABLE 34-4 Endolymphatic Sac Surgical Reports 1995-2007* Vertigocontrol (%) Author
Year
No.
Huang Welling Quaranta Moffat Gianoli Sajjadi
1995 1996 1997 1997 1998 1998
57 10 20 100 37 27
8y 2y 5y 2y 2y 2y
Quaranta Pensak Huang Ostrowski Yu Brinson
1998 1998 2002 2003 2007 2007
Pensak Kitahara
2007 2007
20 96 109 68 18 54 40 226 100
7y 5y 2.5y 2.5y 2y 2y 2y 5.5y 2y 7y 2y 7y 2y >2y
50 Wetmore Total
2007
83 1095
F/U
Class A/B 91 67 85 79 100 85 85 91 92 81 94 67 66 88 100* 100* 100** 100** 77 87%
Hearing (%)
Tinnitus (%)
F
Improved
Stable
Worse
4
18 11 18 15 60 19
54 56 47 56 22 52
28 33 35 29 18 30
0 0 4
Improved/ Worse
35/9 52/1 56/18
1 6 16
17
13 18 33 21 16
70 64 22 41 44
17 18 45 38 40
49 37
44
7
TechniqueS Arenberg shunt rev ES excision - Pro EMS
61/0 38/14 41/14
ELSVD 6%-> surg ESE - >65y 20EMS/18NH EMS 28% bilat - Pro ESBS 1rev Tumarkin ELSVD FLS 4-2 ES drainage EMS FLS EMD EMS 24% bilateral dz ESS + steroid ESS – steroid
19
8.3 - 69
-
EMS incl 16 rev 6.5% weighted average
*Reporting by the American Academy of Otolaryngology Guidelines *2y class A results 88%, 7y class A results 79%; **2y class A results 85%, 7y class A results 79% EMS = Endolymphatic mastoid shunt; ESS = Endolymphatic sac surgery; EMD = edolymphatic mastoid decompression; ESE = endolymphatic sac enhancement; ELSVD = Endolymphatic sac vein decompression; ESBS = Endolymphatic sac balloon surgery; NH = Natural History; ICTI = internal capillary tube insertion
Chapter 34 • Surgery of the Endolymphatic Sac
of larger studies.20,22 Class F outcomes were reported in 7% of cases considering all reports. The ultimate need for neurodestructive procedures is seen in only 7% of surgically treated cases, and less than 2% of all patients with Meniere’s disease. If the length of the follow-up is not considered, surgical success from reported studies would show greater than 87% class A and B results. Kato and colleagues104 looked at the effect of length of follow-up comparing results of 46 patients assessed at less than 18 months with 119 patients who had greater than 18 months of followup. Both groups showed statistically significant improvement in quality of life with the longer follow-up trending to increased improvement. Goldenberg and Justus123 followed a group of 24 patients showing that the 81% class A and B results of endolymphatic sac surgery seen at 1 to 5 years remained (83%) at 7 to 11 years, concluding that shorter follow-up apparently was valid in predicting longer term results. Quaranta and colleagues120 showed that endolymphatic shunts provided a statistically significant improvement over a natural history cohort at 2 and 4 years, but the improvement was not statistically different at 7 years, concluding that intervention hastens the symptomatic recovery that the natural course of Meniere’s disease provides with time. Several centers have followed large cohorts of patients after endolymphatic surgery for many years. Results over 7 to 12 years remain satisfactory in 63% to 100% of patients.20,22,124-126 Some patients with immediate class A or B results experience recurrent symptoms after several years. These patients also do well after revision sac surgery. Indications for revision sac surgery vary. Revision rates range from 4% to 37%.44,84,87,91 Data comparing revision surgery with primary surgery of the endolymphatic sac show that class A and B outcomes are achieved in similar numbers ranging from 65% to 100%.44,91,125,126 Most authors agree that revision sac surgery is more effective when the response to the initial sac intervention provided positive results. There is some controversy regarding the minimal length of the symptom-free interval between the primary surgery and the recurrence of symptoms that qualifies as a positive response. Some authors believe that 6 months of symptomatic relief warrant revision sac surgery, and others require 3 years of high functionality to offer revision surgery.124,125 Schwager and colleagues91 indicated that findings of new bone growth in the region of the primary surgery constricting the sac portend favorable revision outcomes when decompressed, although we have not seen such a regrowth in our revision cases. Further evidence-based research regarding timing of revision surgery is needed to provide adequate data to guide practice decisions. Endolymphatic sac surgery and revision sac surgery have reduced the number of patients with Meniere’s disease who ultimately require ablative surgical procedures. Ablative or titration interventions for failure of
423
e ndolymphatic sac operations or revisions are undertaken in 0 to 27% of patients, with average 6% of combined reporting.85,119,124 Destructive procedures are validated options for patients with poor hearing and proven unilateral disease, and for patients who fail more conservative endolymphatic sac procedures. Transtympanic gentamicin injections after failed endolymphatic sac surgery are effective.73
Patient’s Perspective—Outcomes Questionnaires Many retrospective reviews have included functional level scoring based on 1995 AAOHNS Committee on Hearing and Equilibrium guidelines. Besides the functional level scale (see Table 34-3), six studies have directly assessed the effect of treatment on the patient’s perceived disability and resultant quality of life, and function of disease-specific symptoms.101-106 Changes in preoperative and postoperative scoring on the Meniere’s Disease Outcomes Measure Questionnaire (MDOQ), the Medical Outcomes Short-Form 36 health survey (SF-36), the Vertigo Symptom Scale, the Hearing Disability Handicap Scale, the Tinnitus Severity Questionnaire, and the Sense of Coherence Scales were used according to AAOHNS reporting guidelines. In all quality-of-life measures, perception of disability improved after endolymphatic sac surgery in 79% to 100% of patients, and decreased in 0 to 12%. Tyagi and colleagues102 showed that quality-oflife measures improved most when preoperative functional levels were 4 or greater. This finding coincides with the bias seen in measuring surgical outcomes against natural history cohorts that are offered surgery but opt out, as patients choosing surgery likely have more severe disease. Durland and coworkers86 showed a significantly improved perception of physical health and physical and social functioning after endolymphatic surgery, although subjective assessment of mental health was not altered. Patients scored significantly lower than normal subjects in 6 of 10 categories on the SF-36 preoperatively, but scored below normal subjects only in the category of general health postoperatively. The total number of vertigo spells decreased on average from 8.3 to 2.6 per month. De la Cruz and colleagues103 also showed that although disability and imbalance improved significantly in all patients, imbalance and some vertigo remained. After 2 years of follow-up, their survey of subjects who had undergone endolymphatic sac surgery, vestibular neurectomy, and labyrinthectomy showed that current vertigo characteristics did not differ significantly between surgical groups; however, frequency, severity, and interference of balance did differ, with the endolymphatic sac group having the best ratings and the labyrinthectomy group the poorest. Vertigo was shown to resolve within two months in 75%. Kitahara111 showed symptomatic relief
424
OTOLOGIC SURGERY
and noted that recovery was quicker in patients with residual vestibular function.
Physician’s Practice Surveys In 2003, 2005, and 2007, members of the American Neurotologic Society and the American Otologic Society were queried about the way they treat patients with medically recalcitrant Meniere’s disease.77,98,99 Respondents replied at rates of 20% to 68% of all polled. The results of these surveys showed a continued frequent use of endolymphatic sac procedures as a treatment for Meniere’s disease. The use of intratympanic gentamicin in the 1990s showed a sharp increase, and a report from England showed a decline in the total number of endolymphatic sac surgical procedures pe������������������������� r������������������������ formed between 1989 and 2005. Practice habits within the United States show continued predominant use of endolymphatic sac surgery by total numbers, and as a percentage of surgeons’ firstline choice for failed medical treatment.77,98,100 Between 1990 and 1999, 7228 endolymphatic sac procedures were performed followed by 4091 intratympanic gentamicin injections, 3545 vestibular neurectomies, and 2197 labyrinthectomies as reported by the nearly 20% of member surgeon respondents.98 Trends over this period showed a rapid increase in intratympanic gentamicin use, an increase in endolymphatic sac procedures, consistent use of labyrinthectomy, and a decline in the number of nerve sections. Comparing their data on surgical treatment of vertigo, De la Cruz and colleagues103 used the endolymphatic sac operation for 75% of all cases over the last 3 decades. This pattern of use declined in the most recent decade to 62%, and was explained by an increased number of surgeons performing shunts elsewhere. Fifty percent of member surgeons preferred endolymphatic sac procedures as the first-line treatment of medically refractory Meniere’s disease; this was followed by 34% to 39% who chose intratympanic gentamicin, and 9% who would offer use of a micropressure device such as the Meniette device.80,98 When a patient with active Meniere’s disease in an only hearing ear was considered, the first-choice treatment for Meniere’s disease when medical management failed was a micropressure device for 55 surgeons, intratympanic steroids for 48 surgeons, and endolymphatic mastoid shunt for 33 surgeons.99 Peterson and Isaacson99 noted that practice surveys do not provide objective scientific data. Many factors influence response rate and the value of the data. Surveys are subject to biases held by the authors and the respondents, and the results are based on opinion and memory, not hard data. They define current practice patterns of the respondents only to the best of their recall. With this in mind, approximately 85% of responding members of the American Neurotologic Society and the American Otologic Society continue to perform surgery on the endolymphatic sac. Endolymphatic sac procedures
continue to be the most popular first-line surgical choice when medical management has failed in Meniere’s disease, even when contemplating management of an only hearing ear.80,98,99 There was no significant difference related to the duration in practice, or the geographical location of the respondents.
Hearing Outcomes Hearing status is a main factor in the decision process when selecting a surgical line of treatment. A good hearing outcome is desirable, but is currently a secondary measure behind alleviation of the disabling symptoms of vertigo. Patient satisfaction with procedures is determined more by alleviation of vertigo than preservation of hearing.108,125 When vertigo is adequately treated, disability from hearing loss likely becomes more relevant to patients. Hearing outcomes and quantitative analysis of hearing loss have been less well reported after surgical interventions. The 1995 AAOHNS Committee on Hearing and Equilibrium has standardized reporting of hearing outcomes to improve these observations. Gianoli and colleagues121 reported improvement in 60% of patients at 2 years using an endolymphatic sac vein decompression technique. Kitahara and associates89 showed 92% improvement in short-term hearing by placing high-dose steroids within the endolymphatic sac. The prospect of hearing improvement is enhanced in patients treated early in the disease process, and there is a higher chance of improving hearing in patients who exhibit a positive dehydration test or fluctuating hearing loss as opposed to progressive and stable hearing deficits.20,44 Huang and colleagues127 followed hearing results of 723 patients who underwent endolymphatic shunt for 12 years and found that hearing of patients with class A results was stable for the duration of ������� the study, whereas ������������������������������������������������� patients with class B-D hearing loss progressively lost more hearing slowly over the follow up period. Silverstein and colleagues98 determined that endolymphatic sac surgery resulted in a lower incidence of postoperative hearing loss than either vestibular neurectomy or intratympanic gentamicin. Hearing loss of 0 to 10% was seen in 78% of patients after sac surgery, 11% to 20% in 6% of patients, and greater than 21% in less than 1% of patients. After vestibular neurectomy and intratympanic gentamicin, 28% and 32% of patients had 11% to 30% hearing loss, and 2% and 10% had greater than 40% hearing loss. Adjunct treatments or patient profiling may lead to better long-term results, or enhanced hearing preservation. The endolymphatic sac and duct system has been shown to be an adequate viaduct to the inner ear for chemical delivery.128 Steroids, gentamicin, mitomycin, and other antibiotics have been used in endolymphatic sac surgery.22,89,116,129 To date, no long-term data regarding the application of mitomycin or gentamicin to
Chapter 34 • Surgery of the Endolymphatic Sac
the opened endolymphatic system have been reported. Kitahara and associates129 presented the long-term effects of addition of high-dose dexamethasone in conjunction with endolymphatic sac surgery. They showed excellent class A and B control of vertigo (100%), with significant (49%) improvement in hearing at 2 years that decreased only slightly at the 7-year evaluation (37%). These authors had previously reported encouraging short-term effects with significant to complete control of vertigo in 94% of 12 patients, and improved hearing in 92%, and decreased tinnitus in 92%.109 Exposure of the opened sac to 1 mg/mL solution of mitomycin for 5 minutes followed by irrigation with a cephalosporin antibiotic offered 90% class A and B response, similar to 92% response for a similar procedure without mitomycin.94 Hearing improved in 30% of the mitomycin group, however, compared with 12% improvement without mitomycin. Improved outcomes from saturation or perfusion of the opened endolymphatic sac with gentamicin with exposure of the round window to gentamicin for 7 to 10 minutes reportedly improved vestibular response in patients with severe hearing loss.44 In cases with severe hearing loss, a transtympanic injection alone would likely be as beneficial. The surgical treatment of Meniere’s disease is necessary in only a few patients. Although no consensus on the secondary management of Meniere’s disease exists, endolymphatic sac surgery remains the most popular treatment option for intractable cases because of its respectable track record over time, and its status as a nondestructive procedure in a potentially bilateral disease. It may optimistically limit the need for ablative surgery to about 2% of patients with Meniere’s disease. The use of the endolymphatic sac as a reservoir for adjunct medical treatment and as a possible route to treat hearing loss may make this method of surgery even more appealing in the future.
REFERENCES 1. Portmann G: Vertigo surgical treatment by the opening the saccus endolymphaticus. Arch Otolaryngol 6:309, 1927. 2. Committee on Hearing and Equilibrium: Méniere’s disease: Guidelines for the diagnosis and evaluation of therapy for reporting. Otolaryngol Head Neck Surg 113:181-185, 1995. 3. Canalis R F, Mira E, Bonandrini L , Hinojosa R : Antonio Scarpa and the discovery of the membranous inner ear. Otol Neurotol 22:105-112, 2001. 4. Méniere P: Maladies de l’oreille interne offrant les symptoms de la congestion cerebrale apoplectiforme. Gaz Med de Paris 16:88, 239, 379, 1861. 5. Watzke D, Bast TH: The development and structure of the otic (endolymphatic) sac. Anat Rec 106:361-379, 1950.
425
6. Fujita S, Sando I : Three-dimensional course of the vestibular aqueduct. Eur Arch Otorhinolaryngol 253:122125, 2004. 7. Ng M : Postnatal maturation of the human endolymphatic sac. Laryngoscope 110:1452-1456, 2000. 8. Dobie R A, Snyder J M, Donaldson J A : Electronystagmographic and audiologic findings in patients with Méniere’s disease. Acta Otolaryngol (Stockh) 94:19-27, 1982. 9. Friberg U, Jansson B, Rask-Anderson H, BaggerSjoback D: Variations in surgical anatomy of the endolymphatic sac. Arch Otolaryngol 114:389-394, 1988. 10. Gibson WPR , Arenberg K : Pathophysiologic theories in the etiology of Méniere’s disease. Otolaryngol Clin North Am 30:961, 1997. 11. Anson B J, Donaldson J A : Surgical Anatomy of the Temporal Bone, 2nd ed. Philadelphia, Saunders, 1973. 12. Lundquist PG: Aspects of endolymphatic sac morphology and function. Arch Otorhinolaryngol 212:231-240, 1976. 13. Bagger-Sjoback D, Jansson B, Friberg U: Rask Andersen H: Three dimensional anatomy of the human endolymphatic sac. Arch Otolaryngol 116:345-349, 1990. 14. Takeda T, Sawada S, Kakigi A, Saito H : Computed radiographic measurement of the dimensions of the vestibular aqueduct in Méniere’s disease. Acta Otolaryngol Suppl 528:80-84, 1997. 15. Shea JJ Jr, Ge X, Warner R M, Orchik DJ: External aperture of the vestibular aqueduct in Méniere’s disease. Am J Otol 21:351-355, 2000. 16. Eberhardt K EW, Hollenbach H P, Deimling M, et al: High-resolution magnetic resonance imaging of the endolymphatic duct and sac. MAGMA 3:77-81, 1995. 17. Kobayashi M, Fukay T, Noda M : The endolymphatic sac in patients with Méniere’s disease: Correlation between MRI and the surgical findings. Acta Otolaryngol 120:955-959, 2000. 18. Welling D B, Clarkson MW, Miles B A, et al: Submillimeter magnetic resonance imaging of the temporal bone in Méniere’s disease. Laryngoscope 106:1359-1364, 1996. 19. Gibson WPR : The effect of surgical removal of the extraosseous portion of the endolymphatic sac in patients suffering from Méniere’s disease. J Laryngol Otol 110:1008-1011, 1996. 20. Huang TS : Endolymphatic sac surgery for Méniere’s disease: Experience with over 3000 cases. Otolaryngol Clin N Am 35:591-606, 2002. 21. Amiratti M, Spallone A, Feghali J, et al: The endolymphatic sac: Microsurgical topographic anatomy. Neurosurgery 36:416-419, 1995. 22. Paparella M M : Revision of endolymphatic sac surgery for recurrent Méniere’s disease. Otolaryngol Clin North Am 35:607-619, 2002. 23. Guild S : The circulation of endolymph. Am J Anat 39:57-81, 1927. 24. Welling D B, Pasha R , Roth L J, Barin K : The effect of endolymphatic sac excision in Méniere’s disease. Am J Otol 17:278-282, 1996. 25. Lawrence M : The flow of endolymph—a unique concept. Otolaryngol Clin North Am 13:577, 1980.
426
OTOLOGIC SURGERY
26. Ando M, Takeuchi S, Kakigi A, et al: Acute ischemia causes ‘dark cell’ change of strial marginal cells in gerbil cochlea. Cell Tissue Res 309:229-335, 2002. 27. Marcus DC, Shipley A M : Potassium secretion by vestibular dark cell epithelium demonstrated by vibrating probe. Biophys J 66:1939-1942, 1994. 28. Wangemann P, Liu J, Shiga N: Vestibular dark cells contain the Na+/H+ exchanger NHE-1 in the basolateral membrane. Hear Res 94:94-106, 1996. 29. Salt A : Regulation of endolymphatic fluid volume. Ann N Y Acad Sci 942:306-312, 2001. 30. Miyamoto H, Morgenstern C : Potassium level in endolymphatic sac of guinea pigs in vivo. Arch Otorhinolaryngol 222:77-78, 1979. 31. Ichiyama I, Adams JC, Kimura R S : Immunolocalization of Na+/K+-ATPase, Ca2+-ATPase, calciumbinding proteins and carbonic anhydrase in the guinea pig inner ear. Acta Otolaryngol 114:167-176, 1994. 32. Taguchi D, Takeda T, Kakigi A, et al: Expressions of aquaporin-2, vasopressin type 2 receptor, transient receptor potential channel vanilloid (TRPV)1, and TRPV4 in the human endolymphatic sac. Laryngoscope 117:695-698, 2007. 33. Qvortrup K, Rostgaard J, Holstein-Rathlou N-H, Bretlau P: The endolymphatic sac: A potential endocrine gland? Acta Otolaryngol 119:194-199, 1999. 34. Rask-Anderson H, Danchkwardt-Lilliestrom N, Linthicum FH, House WF: Ultrastructural evidence of a merocrine secretion in the human endolymphatic sac. Ann Otol Rhinol Laryngol 100:148-156, 1991. 35. Fukuzawa K, Sakagami M, Matsunaga T, Fujita H : Endocytic activity of the free floating cells and epithelial cells in the endolymphatic sac: an electron microscopic study. Anat Rec 230:425-433, 1991. 36. Ruckenstein M J: Immunologic aspects of Méniere’s disease. Am J Otolaryngol 20:161-165, 1999. 37. Welling D B, Daniels R , Brainard J, et al: Detection of viral DNA in endolymphatic sac tissue from Méniere’s disease patients. Am J Otol 15:639-643, 1994. 38. Derebery J: Allergic management of Méniere’s disease: An outcome study. Otolaryngol Head Neck Surg 122:174-182, 2000. 39. Takumida M, Barbara M, Bagger-Sjoback D, RaskAndersen H : Lectin detection of carbohydrates in the endolymphatic sac. Eur Arch Otorhinolaryngol 246:8993, 1989. 40. Hultgard-Ekwall A H, Couloigner V, Rubin K, RaskAnderson H : Network organization of interstitial connective tissue cells in the human endolymphatic duct. J Histochem Cytochem 51:1491-1500, 2003. 41. Thalmann I, Hughes I, Tong B D, et al: Microscale analysis of proteins in inner ear tissues and fluids with emphasis on endolymphatic sac, otoconia, and organ of Corti. Electrophoresis 27:1598-1608, 2006. 42. Hallpike C S, Cairns H : Observation on pathology of Méniere’s syndrome. J Laryngol Otol 53:624-654, 1938. 43. Okuno T, Sando I : Localization, frequency, and severity of endolymphatic hydrops and the pathology of the labyrinthine membrane in Méniere’s disease. Ann Otol Rhinol Laryngol 96:438-445, 1987.
44. Paparella M M, Fina M : Endolymphatic sac enhancement: Reversal of pathogenesis. Otolaryngol Clin N Am 35:621-637, 2002. 45. Schuknecht H F, Ruther A : Blockage of longitudinal flow in endolymphatic hydrops. Eur Arch Otorhinolaryngol 248:209-217, 1991. 46. Kimura RS: Experimental blockage of the endolymphatic sac and duct and its effect on the inner ear of the guinea pig. Ann Otol Rhinol Laryngol 76:4664-4687, 1967. 47. Yazawa Y, Shea JJ, Kitahara M : Endolymphatic sac in guinea pig after cauterizing the sac with silver nitrate. Arch Otolaryngol 111:301-304, 1985. 48. Andrews JC, Strelioff D: Modulation of inner ear pressure in experimental endolymphatic hydrops. Otolaryngol Head Neck Surg 112:78-83, 1995. 49. Costa S S, Sousa LC A, Pizza M RT: Méniere’s disease: Overview, epidemiology, and natural history. Otolaryngol Clin North Am 35:455-495, 2002. 50. Schuknecht H F: Correlation of pathology with symptoms of Méniere’s disease. Otolaryngol Clin North Am 1:433-438, 1968. 51. Paparella M M : Pathology of Méniere’s disease. Ann Otol Rhinolaryngol 93(Suppl 112):31-35, 1984. 52. Arenberg I K, Balkany TJ, Goldman G, Pillsbury HC III: The incidence and prevalence of Méniere’s disease—a statistical analysis of limits. Otolaryngol Clin North Am 13:597-601, 1980. 53. Kotimaki J, Sorri M, Aantaa E, Nuutinen J: Prevalence of Méniere’s disease in Finland. Laryngoscope 109:748753, 1999. 54. Hance S E : Méniere’s disease in childhood: Implications for management in the school environment. Language, Speech, and Hearing Services in Schools 21:132-134, 1990. 55. Hausler R , Toupet M, Guidetti G, et al: Méniere’s disease in childhood. Am J Otolaryngol 8:187-193, 1987. 56. Orchik DJ, Shea JJ Jr, Ge N: Summating potential and action potential ratio in Méniere’s disease before and after treatment. Am J Otol 19:478-483, 1998. 57. Ge X, Shea JJ Jr: Transtympanic electrocochleography: A 10-year experience. Otol Neurotol 23:799-805, 2002. 58. Ge N N, Shea JJ Jr, Orchik DJ: Cochlear microphonics in Méniere’s disease. Am J Otol 18:58-66, 1997. 59. Huang TS, Hsu JC, Lee FP: Electrocochleographic monitoring in endolymphatic sac surgery for Méniere’s disease. Arch Otolaryngol Head Neck Surg 120:522529, 1994. 60. Young YH, Wu CC, Wu C H : Augmentation of vestibular evoked myogenic potentials: An indication for distended saccular hydrops. Laryngoscope 112:509-512, 2002. 61. Seo T, Node M, Yukimasa A, Sakagami M : Furosemide loading vestibular evoked myogenic potential for unilateral Méniere’s disease. Otol Neurotol 24:283-288, 2003. 62. Rauch S D, Zhou G, Kujawa SG, et al: Vestibular evoked myogenic potentials show altered tuning in patients with Méniere’s disease. Otol Neurotol 25: 333-338, 2004. 63. Ribeiro S, Almeida R R , Cauvilla H H, Gananca M M : Vestibular evoked myogenic potentials in affected
Chapter 34 • Surgery of the Endolymphatic Sac and asymptomatic ears in unilateral Méniere’s disease. Rev Bras Otorrinolaringol (Engl Ed) 71:60-66, 2005. 64. Rauch S D, Silveira M B, Zhou G, et al: Vestibular evoked myogenic potentials versus vestibular test battery in patients with Méniere’s disease. Otol Neurotol 25:981-986, 2004. 65. Iwasaki S, Takai Y, Ito K, Murofushi T: Abnormal vestibular evoked myogenic potentials in the presence of normal caloric responses. Otol Neurotol 26:1196-1199, 2005. 66. Timmer FC, Zhou G, Guinan JJ, et al: Vestibular evoked myogenic potential (VEMP) in patients with Méniere’s disease with drop attacks. Laryngoscope 116:776-779, 2006. 67. Thomsen J, Tos M, Johnson NJ: Placebo effect in surgery for Méniere’s disease. Arch Otolaryngol 107:271277, 1981. 68. Pillsbury HC : Endolymphatic sac surgery: The Dannish Sham study: An alternative analysis. Otolaryngol Clin North Am 16:123-127, 1983. 69. Welling D B, Nagaraja H N: Endolymphatic mastoid shunt: A reevaluation of efficacy. Otolaryngol Head Neck Surg 122:340-345, 2000. 70. Van Deelen GW, Huizing E H : Use of a diuretic ( Dyazide) in the treatment of Méniere’s disease: A double-blind cross-over placebo-controlled study. ORL J Otorhinolaryngol Relat Spec 48:287-292, 1986. 71. Chia S H, Gamst AC, Anderson J P, Harris J P: Intratympanic gentamicin therapy for Méniere’s disease: A meta-analysis. Otol Neurotol 25:544-552, 2004. 72. Marzo S J, Leonetti J P: Intratympanic gentamicin therapy for persistent vertigo after endolymphatic sac surgery. Otolaryngol Head Neck Surg 126:31-33, 2002. 73. Gouveris H, Lange G, Mann WJ: Intratympanic gentamicin treatment after endolymphatic sac surgery. Acta Otolaryngol 125:1180-1183, 2005. 74. Arriaga M A, Goldman S : Hearing results of intratympanic steroid treatment of endolymphatic hydrops. Laryngoscope 108:1682-1685, 1998. 75. Silverstein H, Isaacson J E, Olds M J, et al: Dexamethasone inner ear perfusion for the treatment of Méniere’s disease: A prospective, randomized, double-blind, crossover trial. Am J Otol 19:196-201, 1998. 76. Garduno-Anaya M A, Toledo HC, Hinojosa-Gonzalez R , et al: Dexamethasone inner ear perfusion by intratympanic injection in unilateral Méniere’s disease: A two-year prospective, placebo-controlled, doubleblind, randomized trial. Otolaryngol Head Neck Surg 133:285-294, 2005. 77. Kim H H, Wiet R J, Batista R A : Trends in the diagnosis and the management of Méniere’s disease: Results of a survey. Otolaryngol Head Neck Surg 132:722-725, 2005. 78. Pensak M L , Friedman R A : The role of endolymphatic mastoid shunt surgery in the managed care era. Am J Otol 19:337-340, 1998. 79. Kitahara M : Endolymphatic sac surgery for Méniere’s disease: Eighteen years experience with the Kitahara sac operation. Am J Otol 8:283-286, 1987. 80. Welling D B, Martyn M D, Miles B A, et al: Endolymphatic sac occlusion for the enlarged vestibular aqueduct syndrome. Am J Otol 19:145-151, 1998.
427
81. Huang TS, Lin CC : Endolymphatic sac ballooning surgery for Méniere’s disease. Ann Otol Rhinol Laryngol 103:389-394, 1994. 82. Cohen E J, Mattox D E : Histologic and ultrastructural features of explanted Arenberg shunts. Arch Otolaryngol Head Neck Surg 120:326-332, 1994. 83. Jackson CG, Dickins J R E, McMenomey S A, et al: Endolymphatic system shunting: A long-term profile of the Denver inner ear shunt. Am J Otol 17:85-88, 1996. 84. Huang TS : Three new surgeries for treatment of intractable Méniere’s disease. Am J Otol 20:233-237, 1999. 85. Moffat D A : Endolymphatic mastoid shunt surgery in unilateral Méniere’s disease. Ear Nose Throat J 9:642651, 1997. 86. Durland WF Jr, Pyle G M, Conner N P: Endolymphatic sac decompression as a treatment for Méniere’s disease. Laryngoscope 115:144-157, 2005. 87. Ostrowski VB, Kartush J M : Endolymphatic sac-vein decompression for intractable Méniere’s disease: Long term treatment results. Otolaryngol Head Neck Surg 128:550-559, 2003. 88. Luetje C M : A critical comparison of results of endolymphatic subarachnoid shunt and endolymphatic sac incision operations. Am J Otol 9:95-101, 1988. 89. Kitahara T, Takeda N, Mishiro Y, et al: Effects of exposing the opened endolymphatic sac to large doses of steroids to treat intractable Méniere’s disease. Ann Otol Rhinol Laryngol 110:109-112, 2001. 90. Paparella M M : Revision of endolymphatic sac surgery for recurrent Méniere’s disease. Otolaryngol Clin North Am 35:607-619, 2002. 91. Schwager K, Baier G, Nour El-Din M, et al: Revision surgery after saccotomy for Méniere’s disease: Does it make sense? Eur Arch Otorhinolaryngol 259:239-242, 2002. 92. Huang TS, Lin CC : Revision endolymphatic sac surgery for recurrent Méniere’s disease. Acta Otolaryngol (Stockh) Suppl 485:131-144, 1991. 93. Paparella M M, Sajjadi H : Endolymphatic sac enhancement principles of diagnosis and treatment. Am J Otol 8:294-300, 1987. 94. Silverstein H, Smouha E, Jones R : Natural history vs. surgery for Méniere’s disease. Otolaryngol Head Neck Surg 100:6-16, 1989. 95. Thomsen J, Kerr A, Bretlau P, et al: Endolymphatic sac surgery: Why we do not do it. The non-specific effect of sac surgery. Clin Otolaryngol Allied Sci 21:208-211, 1996. 96. Kerr AG: Emotional investments in surgical decision making. J Laryngol Otol 116:575-579, 2002. 97. Grant I L , Welling D B : The treatment of hearing loss in Méniere’s disease. Otolaryngol Clin North Am 30:11231144, 1997. 98. Silverstein H, Lewis WB, Jackson L E, et al: Changing trends in the surgical treatment of Méniere’s disease: Results of a 10-year survey. Ear Nose Throat J 82:185194, 2003. 99. Peterson WM, Isaacson J E : Current management of Méniere’s disease in an only hearing ear. Otol Neurotol 28:696-699, 2007. 100. Hari C K, Powell R , Weiner G M : Time trend analysis of otological procedures performed in England, 19892005. J Laryngol Otol 2:1-5, 2007.
428
OTOLOGIC SURGERY
101. Convert C, Franco-Vidal V, Bebear J P, Darrouzet V: Outcome-based assessment of endolymphatic sac decompression for Méniere’s disease using the Méniere’s disease outcome questionnaire: A review of 90 patients. Otol Neurotol 27:687-696, 2006. 102. Tyagi I, Goyal A, Syal R : Sac surgery results as a function of preoperative distress level. Otol Neurotol 27:951955, 2006. 103. De la Cruz A: Teufert KB, Berliner KI: Surgical treatment for vertigo, imbalance, and time course for recovery. Otolaryngol Head Neck Surg 135:541-548, 2006. 104. Kato B M, LaRouere M J, Bojrab D I, Michaelides E M : Evaluating quality of life after endolymphatic sac surgery: The Méniere’s disease outcomes questionnaire. Otol Neurotol 25:339-344, 2004. 105. Soderman AC H, Bergenius J, Bagger-Sjoback D, et al: Patients’ subjective evaluations of quality of life related to disease-specific symptoms, sense of coherence, and treatment in Méniere’s disease. Otol Neurotol 22:526533, 2001. 106. Pal’chun VT: Levina IuV: Dissection of endolymphatic duct in Méniere’s disease. Vestn Otorinolaringol 3:4-6, 2003. 107. Lu F, Dong R , Zhou W, Lu Y: Comparison of curative effect between decompression and incision of endolymphatic sac. Lin Chuang Er Bi Yan Hou Ke Az Zhi 16:334-335, 2002. 108. Yin S, Ke G, Gu N, et al: Long-term results of surgical treatment of intractable Méniere’s disease for control of vertigo. Lin Chuang Er Bi Yan Hou Ke Za Zhi 13:291292, 1999. 109. Kitahara T, Takeda N, Kondoh K, et al: Endolymphatic sac drainage and steroid-instillation surgery (EDSS) for intractable Méniere’s disease. Nippon Jibiinkoka Gakkai Kaiho 104:728-734, 2001. 110. Wilschowitz M, Sanchez-Hanke M, Ussmuller J: The value of saccotomy in Méniere disease: A long-term analysis of 42 cases. Head Neck Otolaryngol 49:180-187, 2001. 111. Kitahara T, Takeda N, Mishiro Y, et al: Vestibular symptoms and ENG findings during periods of convalescence after endolymphatic sac drainage and steroidinstillation surgery (EDSS). Nippon Jibiinkoka Gakkai Kaiho 103:1255-1262, 2000. 112. Kitahara T, Kondoh K, Morihana T, et al: Surgical management of special cases of intractable Méniere’s disease: Unilateral cases with intact canals and bilateral cases. Ann Otol Rhinol Laryngol 113:399-403, 2004. 113. Yu YP, Yang S M, Han DY, Yang WY: Long-term results of endolymphatic sac drainage for Méniere disease. Zhonghua Er Bi Yan Hou Tou Jing Wai Ke Za Zhi 42:173-176, 2007. 114. Brinson G M, Chen D A, Arriaga M A : Endolymphatic mastoid shunt versus endolymphatic sac decompression for Méniere’s disease. Otolaryngol Head Neck Surg 136:415-421, 2007. 115. Huang TS, Lin CC : A further critical assessment of the efficacy of endolymphatic sac surgery. Acta Otolaryngol Suppl 520:263-269, 1995.
116. Huang TS : Topical mitomycin C and cephalosporin in endolymphatic sac surgery. Laryngoscope 112:243-247, 2002. 117. Quaranta A, Onofri M, Sallustio V, Iurato S : Comparison of long-term hearing results after vestibular neurectomy, endolymphatic mastoid shunt, and medical t herapy. Am J Otol 18:444-448, 1997. 118. Sajjadi H, Paparella M M, Williams T: Endolymphatic sac enhancement surgery in the elderly patients with Méniere’s disease. Ear Nose Throat J 77:975-982, 1998. 119. Thomsen J, Bonding P, Becker B, et al: The nonspecific effect of endolymphatic sac surgery in treatment of Méniere’s disease: A prospective, randomized controlled study comparing “classic” endolymphatic sac surgery with the insertion of a ventilating tube in the tympanic membrane. Acta Otolaryngol 118:769-773, 1998. 120. Quaranta A, Marini F, Sallustio V: Long-term outcome of Méniere’s disease: Endolymphatic mastoid shunt versus natural history. Audiol Neurotol 3:54-60, 1998. 121. Gianoli GJ, Larouere M J, Kartush J M, Wayman J: Sacvein decompression for intractable Méniere’s disease: 2-year treatment results. Otolaryngol Head Neck Surg 118:22-29, 1998. 122. Smith D R , Pyle G M : Outcome-based assessment of endolymphatic sac surgery for Méniere’s disease. Laryngoscope 107:1210-1216, 1997. 123. Goldenberg R A, Justus M A : Endolymphatic mastoid shunt for Méniere’s disease: Do results change over time? Laryngoscope 100:141-145, 1990. 124. Telischi FF, Luxford WM : Long-term efficacy of endolymphatic sac surgery for vertigo in Méniere’s disease. Otolaryngol Head Neck Surg 109:83-87, 1993. 125. Pensak M, Samy R: Contemporary role of endolymphatic mastoid shunt surgery in the era of transtympanic perfusion strategies. Presented at the 26th Politzer Society meeting, Cleveland, OH, 2007. 126. Wetmore S: Endolymphatic sac surgery for Méniere’s disease: Long-term results after primary and revision surgery. Presented at the 26th Politzer Society meeting, Cleveland, OH, 2007. 127. Huang TS, Lin CC, Chang YL: Endolymphatic sac surgery for Méniere’s disease: A cumulative study of twelve years’ experience. Acta Otolaryngol Suppl 485:145-154, 1991. 128. Yamasoba T, Yagi M, Roessler BJ, et al: Inner ear transgene expression after adenoviral vector inoculation in the endolymphatic sac. Hum Gene Ther 10:769-774, 1999. 129. Kitahara T, Okumura SI, Kubo T: Effects of intraendolymphatic sac application of large doses of steroids for intractable Méniere’s disease: A randomized controlled trial. Presented at the 26th Politzer Society meeting, Cleveland, OH, 2007.
35
Middle Cranial Fossa—Vestibular Neurectomy Ugo Fisch and Joseph M. Chen
The treatment of Meniere’s disease continues to evoke controversy, as evidenced by a multitude of treatment modalities and their claims of efficacy. Medical management in an attempt to alter or stall the course of this condition has proved ineffective, with the exception of symptomatic relief of vertiginous attacks with the use of pharmacologic agents. In recent years, in light of the questionable efficacy of endolymphatic sac operation and the undesirable sequelae of myriad ablative procedures, vestibular neurectomy has been accepted as the most effective means to manage recalcitrant and disabling Meniere’s disease. With either the middle fossa or the posterior fossa approach, ablation of vestibular functions and vertigo is reported to be 85% to 99%,1-4 greatly exceeding the results of other treatment protocols. Technical difficulties have been cited by many authors as the major reason for abandoning the middle cranial fossa approach in favor of a predominantly neurosurgical approach from the posterior, especially as collaborative efforts between otoneurology and neurosurgery increase. Many centers in Europe and South America continue to use the middle fossa approach, however, with great success.3-6 The middle cranial fossa vestibular neurectomy performed at the University of Zurich is also known as the transtemporal-supralabyrinthine approach. In contrast to the middle fossa approach of House,7 which involves significant elevation of the middle fossa dura and retraction of the temporal lobe, the transtemporal-supralabyrinthine access to the internal auditory canal (IAC) is gained through bony reduction, with only minimal retraction of the dura.
PATIENT SELECTION Patients with unilateral Meniere’s disease who have incapacitating attacks of vertigo of at least 6 months’ duration despite maximal medical therapy are candidates for vestibular neurectomy. These patients usually have residual and fluctuant hearing. Patients with severe to profound hearing loss and extremely poor speech discrimination are better managed by translabyrinthine cochleovestibular neurectomy. The severity of the vertiginous attacks indicative of surgical management
is subjective, and depends more on the patient’s functional capacity than on the frequency of attacks. In bilateral Meniere’s disease, surgery could still be contemplated if a dominant side can be identified. Four patients at the University of Zurich underwent bilateral vestibular neurectomies in stages (separated by at least 1 year after good vestibular compensation); postoperative vestibular compensation was shorter and easier after the second operation. These patients have remained symptom-free for 16 to 20 years since surgery. Patients with Meniere’s disease suitable for a vestibular neurectomy account for approximately 10% of the whole group. Unilateral peripheral vertigo without the full spectrum of Meniere’s disease may also benefit from vestibular neurectomy.8 In these patients, it is important to confirm the side of pathology with vestibular function tests if there is no hearing loss to provide a lateralizing sign. Other indications for vestibular neurectomy are rare. Labyrinthine trauma with residual hearing and disabling vertigo after successful stapedectomy with good hearing could be considered. Contraindications of surgery include an only hearing ear, signs of central vestibular dysfunction, and poor medical condition. Age older than 70 years is a relative contraindication subject to individual assessment.
PREOPERATIVE EVALUATION AND COUNSELING Before surgery, the patient generally undergoes a full auditory-vestibular evaluation that includes full audiometry, electronystagmography, and auditory brainstem response. Electrocochleography and dehydration tests are not part of the test battery performed at the University of Zurich. A high-resolution computed tomography (CT) scan and magnetic resonance imaging (MRI) are obtained to rule out a cerebellopontine angle lesion, and to determine perilabyrinthine pneumatization. A Stenvers view is routinely obtained to show the contours of the floor of the middle cranial fossa and the relationship of the meatal plane and superior semicircular canal (SCC) with respect to the arcuate eminence. Immunologic evaluation is obtained if autoimmune disorders or 429
430
OTOLOGIC SURGERY
significant allergies are suspected. Immunologic evaluation is not a standard part of the test battery. Patients are usually referred from other otologists and have been managed conservatively for a prolonged period without relief. They are made aware of the efficacy of vestibular neurectomy and its related complications, particularly complications pertaining to hearing and facial nerve.
PREOPERATIVE MEDICATION The patient is premedicated with clonidine, metoclopramide, and midazolam before surgery. A perioperative antibiotic, ceftriaxone (Rocephin), 2 g intravenously for 24 hours, is started at the time of surgery and continued for the duration of the intravenous infusion, usually for 5 days.
SURGICAL SITE PREPARATION, POSITIONING, AND DRAPING The surgical site is prepared in the operating room after the induction of anesthesia. Hair over the temporal region is shaved 9 cm above and 5 cm behind the pinna. The skin is washed with povidone-iodine (Betadine). The patient is secured on the Fisch operating table (Fig. 35-1) in supine position with the head turned to the side. Draping is standard except for a large-reservoir plastic bag, which is at the head of the table to catch excess irrigation fluid and blood.
INTRAOPERATIVE MONITORING AND ANESTHETIC CONCERNS Intraoperative facial nerve monitoring using the Xomed nerve integrity monitor with percutaneous electromyography needles is standard. Unipolar and bipolar stimulating forceps are available. Intracranial pressure is controlled by deep anesthesia induced intravenously
before introducing inhalation anesthetic. The partial pressure of carbon dioxide is maintained between 30 mm Hg and 40 mm Hg. Pharmacologic manipulation with dexamethasone (Decadron), 4 mg every 8 hours perioperatively for 4 days, and mannitol, 0.5 mg/kg intravenously intraoperatively, are also standard. Furosemide is added when necessary. Lumbar cerebrospinal fluid drainage is not routinely performed. Hypotensive anesthesia with nitroglycerin or clonidine (Catapres) or both is used in most cases to maintain a systolic blood pressure between 80 mm Hg and 90 mm Hg.
SPECIAL INSTRUMENTS 1. Fisch operating table (Contraves/Zeiss): Motorized table with the center of rotation at the temporal bone, allowing easy remote control for optimal positional changes that can be operated by the scrub nurse or the anesthesiologist (see Fig. 35-1) 2. Articulated middle fossa retractor (Fischer-13-18-200): Self-retaining retractor that allows adjustments in the angle of retraction, tilting, and sideways shifting of the retractor blade (Fig. 35-2) 3. Temporalis muscle retractor (Fischer-13-18-200): Strong, self-retaining retractor with a wide opening span, essential for the exposure of the zygomatic root (Fig. 35-3) 4. Angled microraspatory (Fischer-13-18-308 [left], 13-18306 [right]): Specially designed to facilitate dural elevation; its shoulder retracts the dura away, and the tip is used to separate dural attachments to bone (Fig. 35-4)
SURGICAL TECHNIQUE The objective of the transtemporal-supralabyrinthine approach for vestibular neurectomy is to gain access to the IAC through the exenteration of the supralabyrinthine bone, while dural elevation and retraction are limited to no more than 1.5 cm (Fig. 35-5).
Skin Incision A preauricular incision is made from approximately the lower edge of the zygomatic root and extended to the temporal area at an angle for about 7 cm (Fig. 35-6). The depth of the incision is made to the temporalis fascia, and branches of the superficial temporal artery are divided and clamped. Retraction of the skin edges is provided by securing arterial clamps to the drapes.
Temporal Muscle Flap FIGURE 35-1. Fisch table. (From Fisch U, Mattox DE: Microsurgery of the Skull Base. New York, Thieme, 1988, p 17.)
After some undermining, the temporalis muscle is well exposed with a self-retaining retractor. Five muscle flaps (Fig. 35-7) are developed, elevated from the temporal
Chapter 35 • Middle Cranial Fossa—Vestibular Neurectomy
431
FIGURE 35-2. Middle fossa retractor. (From Fisch U, Mattox DE: Microsurgery of the Skull Base. New York, Thieme, 1988, p 426.) FIGURE 35-3. Temporal muscle retractor. (From Fisch U, Mattox DE: Microsurgery of the Skull Base. New York, Thieme, 1988, p 426.) FIGURE 35-4. Angled raspatory. (From Fisch U, Mattox DE: Microsurgery of the Skull Base. New York, Thieme, 1988, p 427.) FIGURE 35-5. A, Middle cranial fossa (MCF) topography. B and C, Correct (B) and incorrect (C) approaches. (Redrawn from Fisch U, Mattox DE: Microsurgery of the Skull Base. New York, Thieme, 1988, p 430.)
squama, and retracted away with stay sutures. The temporal squama should be exposed from the root of the zygoma to the parietosquamous suture line. A temporalis muscle retractor should be used for the inferior exposure, where the identification of the zygomatic root is essential in the accurate positioning of the craniotomy.
Craniotomy A 2 × 3 cm craniotomy (Fig. 35-8) is made perpendicular to, and at least 1 cm above, the temporal line, centered over the zygomatic root. This craniotomy is performed with a 5 mm cutting burr on a straight handpiece, and when dura is blue-lined, a 4 mm diamond burr is used.
432
OTOLOGIC SURGERY
Chapter 35 • Middle Cranial Fossa—Vestibular Neurectomy
433
FIGURE 35-6. Skin incision. FIGURE 35-7. A and B, Muscle flaps. FIGURE 35-8. Craniotomy.
Care is taken to avoid injuring the dura and branches of the middle meningeal artery, which usually cross the undersurface of the bone flap in its midportion, but could be variable. The bone flap is elevated from the dura with a dura raspatory and is kept in Ringer’s solution for later use. The dura is elevated from the edges of the craniotomy to facilitate the placement of the middle fossa retractor. In the region of the middle meningeal arterial branches, this step must be done with care. Sharp edges are removed with a small rongeur to prevent dural laceration. The craniotomy is extended inferiorly toward the zygomatic arch to the floor of the middle cranial fossa. Lateral extensions of 1 cm on each side provide better visibility for the next step (Fig. 35-9).
Dural Elevation Dural elevation is performed with the microscope. It is perhaps the most delicate part of the operation; if it is not done properly, troublesome hemorrhage can occur. It is important to keep the dural elevation to a minimum and to avoid the region of the middle meningeal artery anteriorly, where significant vascular channels between dura and cranium can be found around the foramen spinosum. In addition to pharmacologic reduction of the intracranial pressure with mannitol and dexamethasone, cerebrospinal fluid decompression through a small dural incision facilitates dural elevation further; this should be done by first coagulating a small central portion of the dura and then elevating this area with a hook before making an incision (Fig. 35-10). With the use of a curved suction and angled microraspatory, dural elevation is performed from posteriorly forward. Vascular channels are coagulated and cut close to bone. Persistent bleeding from bone can be controlled by drilling over it with a diamond burr. Oozing from dural vessels at the corners can be controlled with oxidized cellulose (Oxycel) packed beneath a Cottonoid. The dural attachment to the petrosquamous suture is also coagulated and cut.
Exposure of the Meatal Plane and Arcuate Eminence Dural elevation is extended to the superior petrosal sulcus and over the arcuate eminence. Moving anteriorly, the meatal plane is reached; this is an area bound by the arcuate eminence, superior petrosal sulcus, and facial hiatus
(Fig. 35-11). When the meatal plane is not well defined because of a flat arcuate eminence, it can be identified after blue-lining the superior SCC. If the attachment of the greater superficial petrosal nerve limits the exposure of the meatal plane, it can be separated from dura gently to avoid traction injury of the facial nerve. Meticulous hemostasis in this region is important, and direct coagulation is avoided because of the proximity of the facial nerve.
Introduction of the Middle Cranial Fossa Retractor The middle cranial fossa retractor can be introduced at this point without the use of a microscope. The selfretaining jaws are firmly attached to the edges of the craniotomy first; the multidirectional stage is slid over the self-retaining portion and loosely placed. The dura is retracted with a suction to reveal the meatal plane to allow the accurate placement of the retractor blade, which is positioned parallel to the superior petrosal sulcus, its tip just beyond the arcuate eminence (Fig. 35-12). When the desired position is attained, screws on the multidirectional stage are tightened. Fine adjustments subsequently are done under the microscope.
Bony Exenteration and Blue Lining of the Superior Semicircular Canal The operating table is now placed in Trendelenburg position to visualize better the area over the retractor blade. Bone posterior and lateral to the arcuate eminence is removed with a cutting burr (5 mm), enabling wider access for the final exposure and the progressive identification of the superior SCC. Because of the variable relationship of the superior SCC to the arcuate eminence, its identification is best accomplished from a posterolateral approach through the pneumatic cells; the yellow compact bone of the SCC can be readily exposed. To blueline the SCC, removal of bone should be done with small diamond burrs (2.3 mm, 3.1 mm, 4.0 mm) in a wide rotatory fashion (Fig. 35-13).
Exposure of the Internal Auditory Canal When the blue-line of the superior SCC is identified, using a 60-degree angle centered over the superior canal ampulla, the area of the meatal plane overlying the IAC can be defined. If the surgeon stays within this angle, there is no danger of damaging the facial nerve or the
434
OTOLOGIC SURGERY
Chapter 35 • Middle Cranial Fossa—Vestibular Neurectomy
435
FIGURE 35-9. A and B, Craniotomy extension. FIGURE 35-10. Cerebrospinal fluid decompression. FIGURE 35-11. A and B, Dural elevation.
basal turn of the cochlea. The initial drilling is focused over the meatal plane, staying as close as possible to the blue-lined SCC. The axis of the drill is directed medioinferiorly, toward the superior lip of the porus and the medial roof of the IAC. These areas are progressively thinned out, until they are blue-lined (Fig. 35-14). The lateral exposure of the IAC involves the removal of bone over the meatal fundus and tegmen tympani (Fig. 35-15). This small triangular area is bordered by the ampulla of the superior SCC, and the facial nerve in its labyrinthine and tympanic portions; only 3 mm separates the superior ampulla from the genu of the facial nerve. The tegmen is opened if necessary to expose the malleus head and incus for better orientation. Bone over the meatal foramen and the distal superior vestibular nerve (SVN) is removed to identify the vertical crest (Bill’s bar), which gives the fundus its inverted-W shape. If this is not well defined, the SVN proximal to the ampulla should be exposed. The meatal foramen of the facial nerve should not be unroofed to avoid facial nerve injury. At least one third of the upper circumference of the IAC must be exposed to perform a vestibular neurectomy properly. In particular, the posterior edge of the meatal roof deep in front of the superior SCC must be removed to expose the SVN adequately.
Vestibular Neurectomy When the IAC is unroofed from the fundus to the porus, dura over the SVN is opened with a 1 mm hook. The nerve is identified and cut sharply with a neurectomy knife distal to the vestibular crest (Fig. 35-16). Avulsion of the SVN with a hook is discouraged because of the high incidence of deafness associated with this maneuver as a result of either traction or vascular injury to the cochlear nerve. The meatal dura is incised along the posterior edge toward the porus. The cut end of the SVN is retracted with a microsuction to expose the saccular and singular branches of the inferior vestibular nerve, which are sectioned with a neurectomy knife. The entire vestibular nerve is stabilized with the suction while the vestibulofacial anastomoses are cut sharply with a neurectomy scissor. The vestibular nerve is now everted with Scarpa’s ganglion in full view, which is slightly darker and has a pronounced vascular pattern over it. The vessels are coagulated, and the nerve is resected proximal to the ganglion with neurectomy scissors (Fig. 35-17). The facial nerve should be exposed only partially, retaining most of its dural cover for protection. The cochlear
nerve is hidden beneath the facial nerve and need not be exposed.
Repair of the Floor of the Middle Cranial Fossa The IAC is covered with a free muscle plug and stabilized with fibrin glue. The tegmen defect is reconstructed with the thinner half of the craniotomy bone flap (wrapped in a gauze and fractured with a rongeur). This bony fragment usually straddles the tegmen defect perfectly, and is fixed in position with fibrin glue (Fig. 35-18). The middle fossa retractor is now removed.
Wound Closure The dura is elevated with a dural hook and suspended to the adjacent temporalis muscle with 4-0 polyglactin 910 (Vicryl) sutures (Fig. 35-19). This action obliterates the dead space between dura and bone, and prevents the formation of an epidural hematoma. The supralabyrinthine cavity is obliterated with the long, anteriorly based muscle flap (No. 1), which is sutured to its opposing flap (No. 5). The craniotomy bone flap is placed over the upper bony defect, and the remaining muscle flaps are sutured over it (Fig. 35-20). A single 3 mm suction drain is placed over the temporalis muscle and brought out through a separate stab incision posteriorly. The skin is closed in two layers with 2-0 catgut and 3-0 nylon sutures.
DRESSING AND POSTOPERATIVE CARE A compression dressing based over the surgical site is applied after skin closure and is left for 5 days. The patient is kept in the postanesthesia care unit for at least 24 hours after surgery, with close monitoring of routine and neurologic vital signs every 30 to 60 minutes. Adequate analgesics and antiemetics are ordered to keep the patient comfortably at bed rest. Ambulation and oral intake are started slowly in 2 or 3 days. Subcutaneous heparin is often given in the early convalescent period. Intravenous antibiotic (ceftriaxone) is discontinued when the patient no longer requires intravenous infusion. Oral sulfamethoxazole and trimethoprim (Bactrim Forte) or ciprofloxacin (Ciproxin) is prescribed for at least 5 days. The drain is removed when the daily drainage is less than 10 mL, and scalp sutures are removed after 10 days.
436
OTOLOGIC SURGERY
FIGURE 35-12. Retractor in position. FIGURE 35-13. Blue-lining superior semicircular canal (SCC). FIGURE 35-14. A and B, Medial internal auditory canal (IAC) exposure. FIGURE 35-15. Lateral internal auditory canal exposure.
Chapter 35 • Middle Cranial Fossa—Vestibular Neurectomy
437
FIGURE 35-16. A-D, Superior vestibular nerve (SVN) section. IVN, inferior vestibular nerve.
TIPS AND PITFALLS • The root of the zygoma must be well identified to guide
the placement of the craniotomy accurately. The craniotomy should be perpendicular to the temporal line, its width not exceeding 2 cm, because of the limited opening span of the middle fossa retractor. • The floor of the middle fossa is usually at the level of the temporal line. • The elevation of the dura should not exceed 1 cm. The necessary access is gained by removing bone at the floor of the middle fossa. • If the arcuate eminence is not prominent (flattened middle fossa), the dura should be elevated in the area where it is anticipated, that is, in line with the external auditory canal. The pneumatic cells are removed posterior to this region to identify the superior SCC. • The meatal plane should be identified before the retractor is placed. It is often more medial than expected, and may be hidden by a shelf of bone. • The blue-lined superior SCC is the only essential landmark for the identification of the IAC, but the surgeon should not hesitate to open the tegmen tympani to identify the malleus and incus or the tympanic portion of the facial nerve if more landmarks are needed.
• Adequate exposure of the IAC requires lowering of
bone between the superior ampulla and the geniculum of the facial nerve (genicular crest). This procedure must be performed with the utmost care. • When attempting to identify the blue-line of the IAC, the surgeon must work medially toward the porus, rather than laterally to avoid injuring the labyrinthine segment. • The vertical crest (Bill’s bar) should be clearly identified by thinning the bone over the meatal foramen. The meatal dura should not be opened before the completion of bone removal because the pressure of the cerebrospinal fluid suspends the facial and vestibular nerves against the dura and facilitates identification. • The intrameatal facial nerve often appears to impinge on and partially cover the SVN. The posterior roof of the IAC close to the superior SCC must be removed to expose adequately and identify the cleavage plane between the nerves for the subsequent neurectomy. • The surgeon should always cut the vestibular nerve with a neurectomy knife, rather than avulse the nerves with a hook. This prevents inadvertent traction on the cochlear nerve and its vascular supply, resulting in deafness. • One should look for a loop of the anterior inferior cerebellar artery before cutting the nerve close to the porus.
438
OTOLOGIC SURGERY
Chapter 35 • Middle Cranial Fossa—Vestibular Neurectomy
439
FIGURE 35-17. A and B, Vestibular neurectomy. FIGURE 35-18. Bone flap reconstruction. FIGURE 35-19. Dural suspension. FIGURE 35-20. Muscle closure.
• Lateral suspension of the dura decreases the dead space
and prevents epidural hematoma formation. • To avoid perforation, overzealous blue-lining of the superior SCC and its ampulla is discouraged. If perforation does occur, the area of opening should be immediately covered with a piece of fascia, with bone dust pâté or wax applied over it, and stabilized with fibrin glue. • The basal turn of the cochlea should not be encountered if the “60 rule” (see Fig. 35-14) is followed. One must be cautious, however, when a change of color is observed in the bone while working anteriorly within the 60 degree angle. Working deep in the bone without an adequate view around the tip of the burr can be avoided by removing the lateral bone overhang between the superior ampulla, labyrinthine, and tympanic segments of the facial nerve.
after rapid head movements in the first postoperative year. Two percent experience incomplete vestibular compensation and continue to experience disabling vertigo. Complications specifically associated with this approach include a sensorineural hearing loss in 2%. This complication is most likely a result of traction injury of the cochlear nerve and unrecognized perforation of the SCC. Transient facial paresis is seen in 3.2% of cases and occurs 5 to 7 days after surgery. Recovery is expected within 1 to 3 months. Transient cerebrospinal fluid rhinorrhea during the first 5 to 7 days occurs in 6% of patients, all successfully treated with conservative measures. One case of epidural hematoma was noted and required evacuation with no sequela. No meningitis, temporal lobe epilepsy, or atrophy was associated with this technique.
RESULTS
ALTERNATIVE TECHNIQUES
Follow-up of 3 to 15 years was available in 281 vestibular neurectomy patients operated on at the University of Zurich from 1967 to 1988.8 Of these, 218 of the operations were for Meniere’s disease, including four patients who underwent staged operations for bilateral disease, and 63 operations were for other forms of peripheral vestibular disorder. The success rate of transtemporal vestibular neurectomy to alleviate intractable vertigo was 98.2% for patients with Meniere’s disease, and 96.8% for patients with peripheral vestibular disorder. In 61% of the patients with Meniere’s disease, evidence suggests a stabilizing effect of the surgery on residual hearing. An initial hearing improvement of more than 15 dB was seen in 12% of patients, with some improvement lasting 7 years, but invariably hearing deterioration ensued. Although difficult to explain, this hearing improvement could be due to (1) the division of cochlear efferents traveling with the vestibular nerve, (2) the reduction in the rate of endolymph production resulting from the devascularization and destruction of the dark cell areas, or (3) a change in the parasympathetic innervation of the inner ear subsequent to the sectioning of the olivocochlear bundle in the vestibulofacial anastomosis.9
The indication for endolymphatic sac surgery seems to be hearing preservation only in patients with early-stage disease. The long-term results at the University of Zurich have been as disappointing as the results reported by others.10-12 Endolymph-perilymph shunting procedures and peripheral labyrinthine ablation by either medical or surgical means continue to find support in many centers.13-17 Many procedures are of historical interest only, whereas others, such as intratympanic injection of vestibulotoxic medications, show some promise; however, the results in the control of vertigo and in prevention of hearing loss make them undesirable alternatives. Vestibular neurectomy and neurotomy (nerve section) seem to offer the best control of intractable vertigo refractory to medical therapy. Retrolabyrinthine and retrosigmoid vestibular nerve sections1,2,11 have been proposed more recently as alternatives to the middle fossa (transtemporalsupralabyrinthine) vestibular neurectomy. It is, however, more logical to divide the vestibular nerve fibers in the distal IAC where fibers are distinct to achieve a complete section, while preserving the cochlear nerve. Also, the resection of the vestibular nerve, including Scarpa’s ganglion, ensures that regeneration does not occur. One is less likely to encounter large vessels in the distal portion of the IAC. These three important anatomic considerations are inadequately addressed with the posterior fossa approaches and must be kept in mind when comparing long-term results of the various surgical options. Although the surgical access to the IAC in the transtemporal-supralabyrinthine approach is narrower
COMPLICATIONS Vestibular compensation usually takes a few weeks, and most patients return to work in a few months. Twenty percent of patients complain of a mild, transient dizziness
440
OTOLOGIC SURGERY
compared with the access of the standard middle fossa approach, the dangers associated with temporal lobe retraction are practically eliminated. The posterior fossa approach entails more risks of severe cerebrospinal fluid leaks, meningitis, deafness, permanent facial paralysis, and intracranial hemorrhage, not to mention the common sequela of permanent headache2 that is often not emphasized. We concur that the transtemporal-supralabyrinthine approach is challenging and demands a thorough knowledge of the temporal bone; however, these challenges are similar to those of other neurotologic procedures.
REFERENCES 1. House JW, Hitselberger WE, Mc Elveen J, et al: Retrolabyrinthine section of the vestibular nerve. Otolaryngol Head Neck Surg 92:212-215, 1984. 2. Silverstein H, Norrell H : Microsurgical posterior fossa vestibular neurectomy: An evolution in technique. Skull Base Surg 1:16-25, 1991. 3. Fisch U, Mattox D: Microsurgery of the Skull Base. New York, Thieme, 1988. 4. Garcia-Ibanez E, Garcia-Ibanez J L : Middle fossa vestibular neurectomy: A report of 373 cases. Otolaryngol Head Neck Surg 88:486-490, 1980. 5. Castro D: Transtemporal-supralabyrinthine vestibular neurectomy for Ménière’s disease. In Fisch U, Yasargil MG (eds): Neurological Surgery of the Ear and Skull Base. Berkeley, Kugler & Ghedini Publications, 1989. 6. Portmann M, Sterkers J M, Charachon R , Chouard C H : Le Conduit Auditif Interne-Anatomie, Pathologie, Chirurgie. Paris, Librairie Arnette, 1973, pp 102-117.
7. House WF: Surgical exposure of the internal auditory canal and its contents through the middle cranial fossa. Laryngoscope 71:1363-1365, 1961. 8. Kronenberg J, Fisch U, Dillier N: Long-term evaluation of hearing after transtemporal supralabyrinthine vestibular neurectomy. In Nadol J B (ed): The Second International Symposium on Ménière’s Disease. Amsterdam, Kugler & Ghedini Publications, 1989, pp 481-488. 9. Chouard C H : Acousticofacial anastomosis in Ménière’s disorder. Arch Otolaryngol Head Neck Surg 101:296300, 1975. 10. Bretlau P, Thomsen J, Tos M, Johnsen NJ: Placebo effect in surgery for Ménière’s disease: Nine-year follow-up. Am J Otol 10:259-261, 1989. 11. Glasscock M E, Jackson CG, Poe DS, Johnson G D: What I think of sac surgery. Am J Otol 10:230-233, 1989. 12. Brown J S : A ten-year statistical follow-up of 245 consecutive cases of endolymphatic shunt and decompression with 328 consecutive cases of labyrinthectomy. Laryngoscope 93:1419-1424, 1983. 13. Schuknecht H F: Cochleosacculotomy for Ménière’s disease: Theory, technique, and results. Laryngoscope 92:853-854, 1982. 14. Pennington C L , Stevens E L , Griffin WL : The use of ultrasound in the treatment of Ménière’s disease. Laryngoscope 80:578-581, 1980. 15. Wolfson R J: Labyrinthine cryosurgery for Ménière’s disease-present status. Otolaryngol Head Neck Surg 92:221227, 1984. 16. Shea J: Perfusion of the inner ear with streptomycin. Am J Otol 10:150-155, 1989. 17. Moller C, Odkvist L M, Thell J, et al: Vestibular and audiologic functions in gentamicin-treated Ménière’s disease. Am J Otol 9:383-391, 1989.
36
Retrolabyrinthine and Retrosigmoid Vestibular Neurectomy Steven A. Telian and P. Daniel Ward Videos corresponding to this chapter are available online at www.expertconsult.com.
Surgical intervention is indicated when medical treat ments and dietary adjustments fail to control spontaneous episodic vertigo of labyrinthine origin. The most common diagnosis in such patients is Meniere’s disease. Delayed secondary endolymphatic hydrops may result from other types of inner ear disorders or injuries. Sometimes, ongoing vestibular symptoms after vestibular neuritis or labyrinthitis may be treated surgically, although surgical results in these settings are less predictable.1 Many of these patients have associated ipsilateral hearing loss, which allows the physician to offer treatment without regard to hearing, such as labyrinthectomy or intra tympanic gentamicin injections. In cases in which preserva tion of the residual hearing is desirable, selective section of the vestibular nerve with preservation of the auditory nerve fibers is preferred. Although surgery for unilateral ver tigo dates back more than a century, considerable debate remains regarding the value of each procedure, the value of attempted hearing preservation, and the ideal approach.
HISTORICAL BACKGROUND The first vestibulocochlear nerve section was performed in 1898 by Krause,2 and the first posterior fossa ves tibulocochlear nerve section for the treatment of ver tigo in a patient with Meniere’s disease was performed by Frazier in 1904.3 McKenzie4 was the first to perform selective division of the superior half of the vestibuloco chlear nerve in 1931. This procedure was popularized by Dandy, who first sectioned the vestibulocochlear nerve in 1925, and later began performing selective vestibular neurectomies in 1932. Dandy eventually performed 624 vestibular neurectomies; half of these patients underwent selective vestibular neurectomy.5 This series of patients is impressive not only because of the large number of patients, but also because the procedures were performed without microscopic assistance and modern antibiotics. With Dandy’s death in 1946, vestibular neurectomy was largely replaced by peripherally destructive procedures. Although highly successful in treating vertigo, residual hearing was compromised with these procedures.
The first microsurgical division of the vestibular nerve was performed in 1961 by House6 through a mid dle cranial fossa approach; this approach was modified further by Fisch7 and Glasscock.8 The middle cranial fossa approach allows the surgeon to identify and section selectively the vestibular nerve within the internal audi tory canal (IAC). The procedure is technically challeng ing, however, and has a high incidence of postoperative facial nerve weakness and deafness. The retrolabyrinthine approach for division of the tri geminal nerve was described by Hitselberger and Pulec in 1972,9 and in 1978, Brackmann and Hitselberger reported treatment of vertigo and tic douloureux by selective division of the vestibular and trigeminal nerves via a retrolabyrinthine approach.10 Silverstein and Nor rell11 noted a clear cleavage plane between the cochlear and vestibular nerve fibers while resecting a ninth cranial nerve neurilemmoma via a retrolabyrinthine approach, and suggested that selective vestibular neurectomy was feasible in this fashion. Silverstein’s subsequent reports12,13 on the success of retrolabyrinthine vestibular nerve section in eliminating vertigo, while avoiding facial nerve damage and hearing loss, ushered in a new era for selective vestibular neurectomy. The success and increasing popularity of the retro labyrinthine approach to vestibular neurectomy in the 1980s led to efforts to improve the technique further. In an effort to prevent cerebrospinal fluid leak, which was caused by difficulty reapproximating the dura anterior to the sigmoid sinus, many surgeons adopted a retrosig moid approach for vestibular neurectomy. This approach also allows division of the nerve more laterally in the cerebellopontine angle cistern, where identification of the vestibulocochlear cleavage plane may be easier. The retrosigmoid procedure can be extended in some cases to include drilling of the posterior IAC, allowing visual ization of the intracanalicular portion of the nerve; this may be helpful in cases where the nerve bundles cannot be distinguished within the posterior fossa. In an effort to avoid problems with postoperative headache, which was a common complication after retrosigmoid vestibular neurectomy, Silverstein and colleagues14-16 subsequently 441
442
OTOLOGIC SURGERY
proposed a combined retrosigmoid-retrolabyrinthine approach.
PREOPERATIVE EVALUATION Important preoperative considerations include the etiology of vertigo; the degree of disability; previous therapies; and the patient’s age, overall health, hearing status, and motivation. Surgical intervention for vertigo should be considered only in patients with incapacitat ing vertigo who have failed maximal medical therapy and are believed to be able to compensate centrally for the expected postoperative unilateral vestibular deficit. In addition, reliable identification of the offending ear is a crucial component of the preoperative evaluation. This identification is accomplished by audiometry, vestibu lar testing, and, most importantly, the patient’s history. Historical clues to diagnosis of laterality in vestibular disease include ipsilateral hearing loss, aural fullness, and tinnitus associated with the vertigo spells. If the patient has auditory symptoms in the contralateral ear, one must suspect that the disease may have become bilateral, and surgery on the poorer ear is relatively contraindicated in this setting. Patients are often desperate for relief of debilitat ing symptoms, and are anxious to try any treatment that offers a chance at cure. Patients should be counseled that there is no promise of a cure with any treatment, and that often the best course is a conservative one. This is especially true for patients with vertigo that is not due to Meniere’s disease because surgery for these patients is much less likely to succeed.1 If maximal medical therapy has failed, especially in classic unilateral Meniere’s disease, the patient may be a suitable candidate for surgery. Patients considering vestibular neurectomy should be counseled that this form of treatment does not address the underlying disease, but simply endeavors to eliminate one major symptom of the disease. The Meniere’s disease process continues to affect the inner ear even after surgery, and further hear ing loss is anticipated, unless remission occurs. In addi tion, patients should know that they are trading a partial unilateral vestibular lesion for a total loss of vestibular function on that side, and that postoperative vestibular rehabilitation is crucial to the central nervous system compensation process. Preoperative documentation of hearing with audio metric testing is obtained in all patients. Because the audiogram measures only the patient’s current hear ing ability, however, the utility of the patient’s future hearing should also be considered. Because the disease process may cause a continued deterioration in hearing postoperatively, a practical approach to determining sur gical candidacy takes this future hearing loss into con sideration. If residual hearing is of borderline usefulness to the patient, and future hearing loss is anticipated, a
rocedure that is less risky than vestibular neurectomy, p such as labyrinthectomy or intratympanic gentamicin treatments, may be a wiser choice. Radiographic evaluation to exclude retrocochlear pathology is mandatory before considering vestibular surgery. Generally, magnetic resonance imaging (MRI) of CN VIII with administration of paramagnetic contrast material is preferred. Computed tomography (CT) scan ning may also be useful to evaluate the bony anatomy of the temporal bone, but is not routinely required preop eratively. When a patient has been determined to be a candidate for surgery from an otologic standpoint, the patient must be determined to be medically fit for intracranial surgery. Any procedure performed for vertigo is elective, and every effort should be made to optimize the patient’s medical condition before surgery. Before proceeding with surgical intervention, the patient should understand the risks and benefits of the procedure. Providing patients with reason able expectations regarding their postoperative outcome is also important.
PERTINENT NEUROANATOMY Within the IAC lie the superior and inferior vestibular nerves, the cochlear nerve, the facial nerve, the singular (posterior ampullary) nerve, and the nervus intermedius. The anatomy of these nerves and their relationships with each other are described in this section for the right ear as if the patient is supine with the head turned away from the surgeon (Fig. 36-1). The transverse (falciform) crest divides the lateral IAC into superior and inferior halves. The superior half is divided into anterior and posterior quadrants by Bill’s bar. The facial nerve and nervus intermedius occupy the anterosuperior quadrant, and the superior vestibular nerve, which innervates the superior semicircular canal, the horizontal semicircular canal, the utricle, and a por tion of the saccule, occupies the posterosuperior quad rant. The inferior half of the IAC is occupied by the cochlear nerve anteriorly and the inferior vestibular nerve posteriorly. The inferior vestibular innervates the saccule, and is joined by the singular nerve, which innervates the posterior semicircular canal. This junction occurs where the singular canal emerges from its canal into the pos teroinferior quadrant of the IAC approximately 2 mm medial to the transverse crest. The cochlear nerve, which was initially directly ante rior to the inferior vestibular nerve in the fundus of the internal canal, rotates 90 degrees inferiorly so that it is directly inferior to the superior and inferior vestibular nerves at the level of the porus acusticus. The superior and inferior vestibular nerve fibers begin to merge just medial to the transverse crest. A gross separation of the cochlear and vestibular nerve fibers usually exists medial to the porus acusticus17; however, in 25% of cases, the
Chapter 36 • Retrolabyrinthine and Retrosigmoid Vestibular Neurectomy
443
Endolymphatic sac Superior vestibular n. (Divided to view facial n.)
Ampulla PSCC PAN
Facial n.
Saccular n. PAN
Superior vestibular nerve
F
Co
SV
IV
Inferior vestibular n.
Vestibular n.
Co F
Co VN F
VN
Co
F F
Cochleovestibular cleavage plane
VIII
FIGURE 36-1. Anatomy of CN VII and VIII from the brainstem to the internal auditory canal as viewed from a position posterior to the nerves. The facial nerve begins posteroinferior to the cochleovestibular nerve and rotates around the cochleovestibular nerve until it reaches the anterosu perior quadrant of the internal auditory canal. The cochlear nerve fibers begin to separate from the vestibular nerve fibers in the cerebellar cistern. The cross sections show the rotation of the nerves. The superior vestibular nerve is divided to show the facial nerve anteriorly. Co, cochlear nerve; F, facial nerve; IV, inferior vestibular nerve; PAN, posterior ampullary nerve; PSCC, posterior semicircular canal; SV, superior vestibular nerve; VN, vestibular nerve.
separation is evident only histologically. Even in cases of a visible cleavage plane, there can be a high degree of nerve fiber overlap (Fig. 36-2).18 The cochlear nerve fibers enter the brainstem slightly inferior and posterior to the vestibular nerve fibers. A landmark that may be useful in identification of the cochleovestibular cleavage plane is the nervus inter medius. This nerve contains the afferent and parasym pathetic fibers of the facial nerve, and runs between the facial and cochleovestibular nerves along their entire length. It is often located on the anterior portion of the cochleovestibular cleavage plane. It enters the brainstem near the cochleovestibular nerve, and its fibers terminate in the salivatory and glossopharyngeal nuclei. The facial nerve runs proximally from its antero superior position in the lateral IAC to a position that is inferior and slightly anterior to the cochleovestibular nerve at the brainstem. Because of its ventral position,
it is hidden by the cochleovestibular nerve along most of its intracisternal course; however, it may be visualized by gentle retraction of the cochleovestibular nerve or with the use of an endoscope. It is slightly grayer in color than the cochleovestibular nerve. CN V is located superior to the facial and cochleo vestibular nerves and can often be identified by its char acteristic striations that are perpendicular to its course. CN IX, X, and XI are located inferior to the facial and cochleovestibular nerves (Fig. 36-3).
RETROLABYRINTHINE APPROACH The patient is positioned in the standard fashion for mastoidectomy after removing approximately 3 to 5 cm of hair. The planned surgical site is marked and injected with a mixture of local anesthetic and epinephrine, and
444
OTOLOGIC SURGERY 39-271 L
39-271 L
38-1234 L
P.I.
P.I.
Vest.
Coch. P.I.
C
B
Coch.
Vest.
A P.I.
P.I. Vest.
Pia
D
39-28 R
E
39-371 R
F
38-893 R
FIGURE 36-2. A-F, Variability in the separation of the cochlear and vestibular fibers. The presence of a cleavage plane does not correspond to a complete segregation of cochlear and vestibular fibers. (Modified from Rasmussen A: Studies of the VIIIth cranial nerve of man. Laryngoscope 50:667, 1940. Used by permission from Wolters-Kluwer.)
ap
al fl
Dur
7.
. S.S
8. 5.
9.10. 11.
Ce.
FIGURE 36-3. Anatomy of the cerebellopontine angle as viewed through a large retrosigmoid craniectomy showing CN V, VII, VIII, IX, and X. SS, sigmoid sinus; ce, cerebellum; 5, cranial nerve V; 6, �������� cranial nerve VI; 7, cranial nerve VII; 8, cranial nerve VIII; 9, cranial nerve IX; 10, cranial nerve X; 11, cranial nerve XI.
the scalp is prepared and draped in the standard fashion (Fig. 36-4A). The facial nerve is monitored electromyo graphically throughout the procedure. Auditory function is generally monitored by recording intraoperative audi tory brainstem responses, although near-field CN VIII electrodes are preferred by some surgeons after the nerve is exposed. A wide postauricular scalp flap is elevated and turned forward, exposing the temporal bone. The authors prefer distinct scalp and musculoperiosteal incisions, each of which can be individually closed at the end of the case. During the initial drilling, a fragment of bone is acquired for later use in wound closure. A large tem poralis fascia graft should be harvested and set aside to dry. A complete mastoidectomy is performed, and the sigmoid sinus is carefully outlined and skeletonized (Fig. 36-4B). Posterior fossa dura is decompressed at least 1.5 cm behind the sigmoid sinus. Some surgeons leave an island of bone over the sinus, whereas others prefer to remove all bone. Regardless, enough bone overlying the sigmoid sinus should be removed to allow retraction of the sinus into the posterior aspect of the surgical defect so that the CN VIII complex can be easily visualized in the cerebellopontine angle cistern. The antrum is opened, but care should be taken not to dissect too far anteriorly to protect the epitympanum and ossicle heads. All presig moid bone and air cells are removed posterior to the pos terior semicircular canal, including the perilabyrinthine
445
Chapter 36 • Retrolabyrinthine and Retrosigmoid Vestibular Neurectomy
Endolymphatic sac VII
4 cm
5 cm
Jugular bulb
A
Incision Sigmoid sinus Posterior fossa dura removed posterior to sigmoid sinus
B Dural flap
Endolymphatic sac VIII
VII
IX X
V
XI
VN
Anterior inferior cerebellar a. Branch of anterior inferior cerebellar a.
D VN
CN
VII
C
E FIGURE 36-4. Retrolabyrinthine approach. A, Skin incision. B, Drilling the mastoid cavity with exposure of sigmoid sinus. C, Retraction of the dura and sigmoid sinus to identify facial and vestibulocochlear nerves. D, CN V, VII, VIII, and XI exposed. Note the relationship of the branch of the anterior inferior cerebellar artery to CN VII and VIII. This relationship is highly variable, and the vessel does not always run between the nerves as illustrated here. E, Division of the vestibular nerve after separation from the cochlear nerve fibers.
446
OTOLOGIC SURGERY
air cells, retrolabyrinthine air cells, and retrofacial tract air cells, which overlie the endolymphatic sac. The supe rior margin of dissection is the superior petrosal sinus, and the inferior margin is the jugular bulb. When this bone is all removed, the posterior fossa dura has been exposed. All retrofacial, facial recess, and zygomatic root air cells should be obliterated with bone wax. The wound is copiously rinsed to remove bone dust. The dura is incised parallel to the sigmoid sinus mar gin, and the cerebellum is lightly retracted. When some cerebrospinal fluid is evacuated, the tentorium cerebelli and the trigeminal nerve are identified superiorly. The vestibulocochlear nerve can be visualized with gentle retraction of the cerebellum, medial to the bone of the posterior semicircular canal (Fig. 36-4C). The facial nerve is identified positively by electric stimulation after gentle retraction of the CN VIII complex. If a cleavage plane is evident in CN VIII, this is developed by blunt dissection with a microscopic instrument (Fig. 36-4D). If no cleavage plane is present, the surgeon must cre ate an arbitrary plane by sharp dissection (Fig. 36-4E). The superior portion of the nerve, containing the vestibu lar fibers, is divided sharply, taking care not to injure the facial nerve or the nervus intermedius. When the nerve section is complete, the cut ends of the nerve physically separate. It is unnecessary to resect a portion of the nerve. Functional integrity of the facial nerve is confirmed by proximal stimulation after the vestibular nerve section. Care is also taken to preserve the nervus intermedius fibers, which run between CN VII and VIII. The wound is irrigated with bacitracin solution and is ready to be closed. An autogenous graft of abdominal fat is harvested to obliterate the temporal bone defect. The bone graft harvested previously is trimmed to size and placed over the aditus ad antrum to prevent prolapse of fat into the epitympanum. The temporalis fascia is placed anteriorly to isolate the retrolabyrinthine defect from the anterior temporal bone; this prevents the outflow of cere brospinal fluid into the eustachian tube. Rarely, a water tight closure of the presigmoid dura is possible. Fat is placed into the defect, and the scalp wound is closed in two or three layers. The retrolabyrinthine procedure is shown in the video accompanying this chapter.
RETROSIGMOID OR SUBOCCIPITAL APPROACH The patient is positioned in the standard fashion for mastoidectomy or in the lateral “park-bench” position, at the discretion of the surgeon. The head is rotated contralaterally with the neck slightly flexed. A lumbar drain is placed electively by some surgeons to allow for posterior fossa decompression. Electrodes for intraop erative brainstem auditory evoked response and facial nerve monitoring are placed. The planned surgical site parallel to the hairline is marked and injected with
local anesthetic containing 1:100,000 epinephrine (Fig. 36-5A). Placement of the incision avoiding the greater and lesser occipital nerves can be accomplished by using a curvilinear or sigmoidal incision along the hairline. This incision should extend from the top of the auricle down to the upper neck, just posteroinferior to the mastoid tip. The surgical site is prepared and draped in the stan dard fashion. Some surgeons routinely employ a lumbar drain for these cases. If this drain has been placed, 60 to 90 mL of cerebrospinal fluid is drained at the time of skin incision; this decompresses the posterior fossa and mini mizes the need for cerebellar retraction.19 The skull is exposed, and a 2 cm diameter retrosigmoid craniectomy is made. Any posterior mastoid air cells entered in this process must be obliterated. All drilling is completed before opening the dura to minimize entry of bone dust into the subarachnoid space. Emissary veins are a potential source of bleeding that can be controlled with cautery or bone wax. The dural inci sion may be either curvilinear or trifurcate. After opening the dura to expose the posterior fossa, the cerebellopon tine angle is approached by gentle retraction of the cer ebellum, preferably from the lower portion of the wound (Fig. 36-5B). Medial traction of CN VIII is dangerous to the hearing and is frequently associated with cerebel lar retraction. Auditory brainstem responses should be monitored closely during this portion of the procedure. This is particularly true if a cerebellar retractor is used. The cerebellum is elevated superiorly, separating it from the temporal bone. The operculum and the endolym phatic sac can be helpful landmarks because the endo lymphatic duct exits the temporal bone at the level of the IAC and is only a few millimeters posterior to the IAC. The cranial nerves in this region are exposed by sepa rating the arachnoid mater and the vessels. CN V and VII-XI can be visualized, with CN V being superior to the facial and vestibulocochlear nerve complex, and CN IX-XI exiting inferiorly (see Fig. 36-3). CN V can be identified by its characteristic striations, which are per pendicular to the long axis of the nerve. Identification of CN IX-XI is confirmed with identification of the jugu lodural fold, a condensation of dura that extends across the sigmoid sinus from the posterior aspect of the tem poral bone. Approximately 1 cm anterior to the fold, the cochleovestibular nerve enters the IAC. The facial nerve is located anteroinferior to the vestibulocochlear nerve medially along the brainstem, and continues to move superiorly and anterior to CN VIII until it enters its loca tion in the anterosuperior portion of the IAC (see Fig. 36-1). Identification of the facial nerve can be confirmed with use of the facial nerve stimulator. At the level of the porus, the cochlear and vestibular fibers are usually separated by a cleavage plane, allowing for sharp dissection of the cleavage plane and selective transection of the vestibular fibers. The vestibular nerve is slightly grayer in appearance than the cochlear nerve because of the more dense arrangement of the cochlear
Chapter 36 • Retrolabyrinthine and Retrosigmoid Vestibular Neurectomy
fibers. The nervus intermedius runs along the anterior aspect of the cleavage plane and may be identified with endoscopic assistance. There is also a small arteriole that frequently runs along the posterior cleavage plane. In cases where a definite cleavage plane cannot be identi fied, which occurs in approximately 25% of cases,20 the nerve may be arbitrarily divided as described previously,
or drilling of the posterior portion of the IAC may be performed (see Fig. 36-5C). If drilling of the posterior IAC is required, the drilling extends no further laterally than the singular canal, which carries the posterior ampullary nerve and enters the IAC approximately 2 mm medial to the falciform crest in the posteroinferior quadrant.21 When adequate exposure has
ES
Posterior wall of IAC removed
Planned surgical incision
447
9, 10, 11
Sigmoid sinus
Greater occipital n.
Lesser occipital nerve
SVN U-shaped dural flap
Transverse sinus
A
VN
B FN
FN
SVN
VN
PAN
Saccular n.
Co N
FN
C FIGURE 36-5. Retrosigmoid approach. A, Skin incision is designed to avoid injury to greater or lesser occipital nerves. B, View of nerves after craniectomy. C, Closer view of nerves with removal of a portion of the internal auditory canal (IAC). Co N, cochlear nerve; ES, endolymphatic sac; FN, facial nerve; PAN, posterior ampullary nerve; SVN, superior vestibular nerve; VN, vestibular nerve; 9, cranial nerve IX; 10, cranial nerve X; 11, cranial nerve XI.
448
OTOLOGIC SURGERY
been obtained, the superior vestibular and the singular nerves are sectioned. The inferior vestibular nerve should not be sectioned laterally because of its close relationship to CN VIII and the cochlear blood supply. Careful irrigation is performed, and the dura is reap proximated. Because this approach is associated with a higher incidence of postoperative headache, many authors recommend performing a cranioplasty after the retrosigmoid approach to repair the craniectomy defect.22-24 An alternative technique described by Silver man and coworkers,19 using a sandwich of absorbable gelatin sponge (Gelfoam), absorbable gelatin film (Gel film), and Gelfoam in the craniectomy defect, may pre vent adhesion of cervical musculature to the dura. This Gelfoam-Gelfilm-Gelfoam sandwich has been shown to minimize the incidence of postoperative headache, and may decrease reactive chemical meningitis related to for eign cranioplasty materials. The muscle, fascia, and skin are closed in layers.
. P.F.D
. S.S flap ral Du ision inc
COMBINED RETROLABYRINTHINE/ RETROSIGMOID APPROACH Silverstein�14 proposed a modification of the abovedescribed techniques, termed the combined retrolabyrinthineretrosigmoid approach. This approach involves a limited mastoidectomy with exposure of the sigmoid sinus and posterior fossa dura. A dural incision is made approxi mately 2 to 3 mm posterior to the sigmoid sinus, and the sinus and dura are retracted anteriorly (Fig. 36-6) to allow evaluation of the cerebellopontine angle to determine whether an identifiable cleavage plane exists between the cochlear and vestibular nerve fibers. If not, dura is reflected off the temporal bone, and the IAC is opened. The superior vestibular and posterior ampullary nerves are divided within the IAC, as in the retrosigmoid approach. This division allows performance of the neurectomy with less bone removal and less cerebellar retraction.
ENDOSCOPIC-ASSISTED VESTIBULAR NEURECTOMY Several authors have reported on the use of endoscopic assistance for vestibular neurectomy procedures.25-27 The potential advantages include a magnified view of relevant structures, and the ability to use angled or flex ible scopes to allow visualization of structures not seen well with the microscope, such as the cleavage plane between the cochlear and vestibular nerves, the nervus intermedius, the facial nerve, and branches of the ante rior inferior cerebellar artery. Potential disadvantages include difficulty in visualization because of fluids on the scope, thermal injury to important structures from the heat of the endoscope, lack of instruments specifi cally designed for endoscopic neurotology, and difficulty
FIGURE 36-6. Dural incision used in combined retrolabyrinthine/ retrosigmoid approach. PFD, posterior fossa dura; SS, sigmoid sinus.
because of lack of depth perception when viewing images in two dimensions.27 Any of these factors could lead to injury of structures not generally at risk when performing the procedure with binocular microscopy. The primary advantage of an endoscope may be to allow visualization of any mastoid air cells violated while drilling the IAC, so that these can be properly obliterated to prevent cerebro spinal fluid leak.
POSTOPERATIVE COURSE A mastoid dressing is applied postoperatively. Patients are typically monitored overnight in the neurosurgical intensive care unit, and the urinary catheter is removed on postoperative day 1. If a fat graft was taken from the abdomen, an abdominal drain is placed in the donor site, and is removed when it produces less than 30 mL over a 24 hour period. Routine use of sequential compression devices is necessary to assist in preventing deep venous thrombosis while the patient is immobile. Patients at high risk for thrombosis may receive subcutaneous hepa rin postoperatively. Vestibular rehabilitation activities are begun on postoperative day 1, and physical therapy is often consulted to encourage patient activity and an early return to independent ambulation. Postoperative antibiotics are not given routinely. Patients are typically
Chapter 36 • Retrolabyrinthine and Retrosigmoid Vestibular Neurectomy
hospitalized for 3 to 5 days. A postoperative audiogram is obtained at 1 month.
EFFICACY The gold standard for the surgical control of vertigo is the transmastoid labyrinthectomy. Assuming the offending ear is properly lateralized and the operation is completed successfully, the success rate for control of
episodic vertigo with labyrinthectomy should be 100%. Series report success rates of greater than 95%; poor out comes are most likely secondary to incomplete vestibu lar compensation.28 A review of the available literature (Table 36-1) shows that the results for the retrolabyrin thine and retrosigmoid techniques are favorable, with vertigo being improved in 91% to 100% of patients with Meniere’s disease, and 68% to 100% of patients with vertigo from etiologies not related to Meniere’s dis ease. The combined retrolabyrinthine and retrosigmoid
TABLE 36-1 Vertigo Resolution Rates* Retrolabyrinthine Vestibular Nerve Section RESOLUTION OF VERTIGO Meniere’s Disease Author, Year 198838
McElveen et al, Monsell et al, 198839 Boyce et al, 198840 Zini et al, 198841 Kemink et al, 19911 Glasscock et al, 199142 Teixido and Wiet, 199234 Ortiz Armenta, 199243 Nguyen et al, 199244 Aristegui et al, 199745 Silverstein and Jackson, 200246
Other Etiologies
No. Patients
Improved (%)
Cured (%)
Improved (%)
Cured (%)
115 31 39 38 90 42 25
97 94 96 97 96 97 91
76 NA 78 NA 94 83 67
68 NA 69 NA 69 100 NA
28 NA 31 NA 23 100 NA
45 143 35 78
99 93 100 95
92 82 97 88
87 74 NA NA
70 52 NA NA
Retrosigmoid Vestibular Nerve Section RESOLUTION OF VERTIGO Meniere’s Disease Author, Year Glasscock et al, 199142 Molony, 199647 Pareschi et al, 200248 Fukuhara et al, 200249 Miyazaki et al, 200526
Other Etiologies
No. Patients
Improved (%)
Cured (%)
Improved (%)
Cured (%)
44
100
77
100
33
27 58 28
95 97 93
82 90 75
0 NA NA
0 NA NA
331
96
NA
NA
NA
Combined Retrolabyrinthine/Retrosigmoid Vestibular Nerve Section RESOLUTION OF VERTIGO Meniere’s Disease Author, Year Silverstein and Jackson, 200246 Goksu et al, 200550
No. Patients
Improved (%)
449
Other Etiologies Cured (%)
Improved (%)
Cured (%)
126
92
85
NA
NA
210
95
90
NA
NA
*Series with >25 patients; only nonoverlapping data from a single group or institution are included. NA, not applicable.
450
OTOLOGIC SURGERY
TABLE 36-2 Effect on Postoperative Hearing Loss, Tinnitus, and Aural Fullness* Retrolabyrinthine Vestibular Nerve Section Author, Year McElveen et al, 198838 Monsell et al, 198839 Boyce et al, 198840 Zini et al, 198841 Kemink et al, 19911 Glasscock et al, 199142 Teixido and Wiet, 199234 Ortiz Armenta, 199243 Nguyen et al, 199244 Aristegui et al, 199745 Silverstein and Jackson, 200246
Follow-up Time (mo)
Patients with Hearing Loss (%)
Tinnitus (% Improved)
Fullness (% Improved)
6-60 18-24 6-36 >6 >24 >24 3-44 12 1-48 6-76 1
27 36 58 26 14 43 44 7 43 48 40
35 50 34 NA NA 36 NA NA 27 16 NA
54 66 64 NA NA 34 NA NA 36 16 NA
Retrosigmoid Vestibular Nerve Section Author, Year Glasscock et al, 199142 Molony, 199647 Pareschi et al, 200248 Fukuhara et al, 200249 Miyazaki et al, 200526
Follow-up Time (mo)
Patients with Hearing Loss (%)
Tinnitus (% Improved)
Fullness (% Improved)
>24 24-42 >48 19-65 3-132
45 10 7 7 0
32 NA 10 NA NA
39 NA 21 NA NA
Combined Retrolabyrinthine/Retrosigmoid Vestibular Nerve Section Author, Year Silverstein and Jackson, 200246 Goksu et al, 200550
Follow-up Time (mo)
Patients with Hearing Loss (%)
Tinnitus (% Improved)
Fullness (% Improved)
1
20
NA
NA
18-24
10
NA
NA
NA, not applicable. *Series with >25 patients; only nonoverlapping data from a single group or institution are included.
approach is reported to have a success rate of 92% to 95% for patients with Meniere’s disease. No data are available for patients with other causes of vertigo. A challenge in interpretation of these data is the natural course of Meniere’s disease, which is often a progres sion to spontaneous complete relief of vertigo even if no treatment is provided.29-32 The advantage of selective vestibular neurectomy over destructive labyrinthine procedures is division of the vestibular fibers with preservation of the cochlear fibers. The ability of the procedure to preserve hearing is an important consideration. As mentioned earlier, ves tibular neurectomy does not change the course of the hearing loss, but is designed only to treat the immedi ately debilitating symptom of vertigo. In addition, ves tibular neurectomy itself introduces a risk of hearing loss secondary to the surgery. Lack of a distinct cleavage plane between the vestibular and cochlear nerve fibers is the primary cause of damage to auditory fibers. This risk may be theoretically decreased using the retrosigmoid
approach, owing to surgical division of the vestibular nerve more laterally, where the cleavage plane becomes more distinct. Medial retraction of the auditory nerve may also lead to a sensorineural hearing loss. In addition, a conductive hearing loss may occur later from fixation of the ossicular chain caused by bone dust that accu mulates in the oval window during the procedure.33,34 A study by Teixido and Wiet34 showed that the most common audiometric abnormality after retrolabyrin thine vestibular neurectomy is low-frequency conductive hearing loss thought to be due to ossicular fixation from this mechanism. As shown in Table 36-2, reports of hearing loss after vestibular neurectomy state that retrolabyrinthine patients have a 7% to 58% incidence of hearing loss, whereas retrosigmoid patients have a 7% to 45% inci dence of hearing loss. These data must be interpreted with caution because the postoperative time points at which the hearing losses were evaluated vary among studies, and are no doubt contaminated by the continued
Chapter 36 • Retrolabyrinthine and Retrosigmoid Vestibular Neurectomy
451
TABLE 36-3 Common Complications* Retrolabyrinthine Vestibular Neurectomy Author, Year McElveen et al, 198838 Monsell et al, 198839 Boyce et al, 198840 Zini et al, 198841 Kemink et al, 19911 Glasscock et al, 199142 Teixido and Wiet, 199234 Ortiz Armenta, 199243 Nguyen et al, 199244 Aristegui et al, 199745 Silverstein and Jackson, 200246
CSF Leak (%)
Infection (%)
Meningitis (%)
Hearing Loss (%)
4 3 6 8 4 12 NA 7 3 0 10
1 0 0 0 6 2 NA 0 4 0 NA
1 0 4 3 0 0 NA 0 0 0 0
0 0 0 3 0 2 0 7 0 0 0
Retrosigmoid Vestibular Nerve Section Author, Year Glasscock et al, 199142 Molony, 199647 Pareschi et al, 200248 Fukuhara et al, 200249 Miyazaki et al, 200526
CSF Leak (%)
Infection (%)
Headache (%)
Hearing Loss (%)
0 0 0 0 4
5 0 0 0 0
9 7 0 7 0
0 4 0 7 0
Combined Retrolabyrinthine/Retrosigmoid Vestibular Nerve Section Author, Year Silverstein and Jackson, 200246 Goksu et al, 200550
CSF Leak (%)
Infection (%)
Headache (%)
Hearing Loss (%)
NA
NA
NA
0
1
0
0
10
CSF, cerebrospinal fluid; NA, not applicable. *Series with >25 patients; only nonoverlapping data from a single group or institution are included.
decline of hearing because of the natural progression of Meniere’s disease. A study by Wazen and coworkers35 showed that serviceable hearing decreased from 81% to 43% of patients at an average of 4 years after vestibular neurectomy. A reasonable conclusion may be drawn that hearing loss continues after vestibular neurectomy, just as it does with untreated Meniere’s disease,29-31 but per haps at a slightly slower rate. The other classic symptoms of Meniere’s disease, tin nitus and aural fullness, also occasionally improve after vestibular neurectomy. As shown in Table 36-2, tinnitus was reduced in half of patients, and aural fullness was reduced in two thirds of patients after the retrolabyrin thine approach.
COMPLICATIONS Table 36-3 summarizes the complications encoun tered with posterior fossa vestibular neurectomy. For the retrolabyrinthine approach, the most common
c omplication is cerebrospinal fluid leak, which occurs in up to 12% of patients. Wound infection and meningitis are also common complications, occurring in up ��������� to���� 6% and 4% of patients. The most commonly cited complica tion with retrosigmoid vestibular neurectomy is signifi cant postoperative headache, sometimes disabling, which can persist for years after surgery. The reported incidence of cerebrospinal leak and wound infection are much lower than for the retrolabyrinthine approach at 4%, and the wound infection rate is similar at 5%. Although facial nerve paresis is a risk, this occurs rarely with the poste rior fossa approaches and is not generally reported in the literature. Other complications reported in the literature for these operations include hydrocephalus, continued imbalance, tinnitus, aseptic meningitis, and abdominal hematoma if an abdominal fat graft is used. Review of the literature shows that the incidence of headache after retrosigmoid vestibular neurectomy approaches 1 in 10. The etiology for such headache syn dromes is still a matter of debate in the otology and neu rosurgery literature. Numerous explanations have been
452
OTOLOGIC SURGERY
proposed, and many strategies have been presented in an effort to reduce this problem. Adhesion of cervical musculature to the dura that is exposed during the cra niectomy can subsequently result in traction on the dura with head movement. Injury to the greater and lesser occipital nerves during the procedure, either by retraction or during the incision itself, has been frequently cited as a possible contributing factor. A third explanation may be that bone dust can be trapped in the subarachnoid space, which can cause chemical meningitis or interfere with cerebrospinal fluid resorption and cause increased intracranial pressure. Other factors including dural tension, cerebellar retraction, and aseptic meningitis have been posited as contributing to postoperative headache. Efforts to prevent the attachment of the deep neck musculature to exposed dura, designing skin incisions to avoid the occipital nerves, and the avoidance or minimization of intradural bone drilling all have been shown to decrease the incidence of postoperative head ache.19 The prevention of cervical muscle adherence to dura has been described using cranioplasty and wound closure modifications. Cranioplasty to replace missing bone volume was initially advocated to decrease the incidence of postoperative headache. Cranioplasty increases operative time, however, and more recent studies have shown that the foreign materials such as methylmethacrylate may lead to a higher incidence of headache, owing to increased local tissue reaction.36 In addition, the data suggest that there is no decrease in the frequency of postoperative headache in cranioplasty patients.36,37
REFERENCES 1. Kemink J L , Telian S A, el-Kashlan H, et al: Retrolaby rinthine vestibular nerve section: efficacy in disorders other than Meniere’s disease. Laryngoscope 101(5):523528, 1991. 2. Jackler R K, Whinney D: A century of eighth nerve sur gery. Otol Neurotol 22(3):401-416, 2001. 3. Frazier C : Intracranial division of the auditory nerve for persistent aural vertigo. Surgery Gynecology and Obstet rics 15:524-529, 1912. 4. McKenzie K : Intracranial division of the vestibular por tion of the auditory nerve for Meniere’s disease. Cana dian Medical Association Journal 34:1127-1152, 1936. 5. Green R E, Douglass CC : Intracranial division of the eighth nerve for Meniere’s disease; a follow-up study of patients operated on by Dr. Walter E. Dandy. Ann Otol Rhinol Laryngol 60(3):610-621, 1951. 6. House WF: Surgical exposure of the internal auditory canal and its contents through the middle, cranial fossa. Laryngoscope 71:1363-1385, 1961. 7. Fisch U: Vestibular and cochlear neurectomy. Trans Am Acad Ophthalmol Otolaryngol 78(4):ORL252-ORL255, 1974.
8. Glasscock M E, 3rd: Vestibular nerve section. Middle fossa and translabyrinthine. Arch Otolaryngol 97(2):112114, 1973. 9. Hitselberger WE, Pulec J L : Trigeminal nerve (posterior root) retrolabyrinthine selective section. Operative pro cedure for intractable pain. Arch Otolaryngol 96(5):412415, 1972. 10. Badke M B, Pyle G M, Shea T, et al: Outcomes in vestib ular ablative procedures. Otol Neurotol 23(4):504-509, 2002. 11. Silverstein H, Norrell H : Retrolabyrinthine surgery: a di rect approach to the cerebellopontine angle. Otolaryngol Head Neck Surg 88(4):462-469, 1980. 12. Silverstein H, Norrell H : Retrolabyrinthine vestibular neurectomy. Otolaryngol Head Neck Surg 90(6):778782, 1982. 13. Silverstein H, McDaniel A, Wazen J, et al: Retrolabyr inthine vestibular neurectomy with simultaneous moni toring of eighth nerve and brain stem auditory evoked potentials. Otolaryngol Head Neck Surg 93(6):736-742, 1985. 14. Silverstein H, Nichols M L , Rosenberg S, et al: Combined retrolabyrinthine-retrosigmoid approach for improved ex posure of the posterior fossa without cerebellar retraction. Skull Base Surg 5(3):177-180, 1995. 15. Silverstein H, Norrell H, Smouha E, et al: Combined retrolab-retrosigmoid vestibular neurectomy. An evolu tion in approach. Am J Otol 10(3):166-169, 1989. 16. Silverstein H, Norrell H, Smouha E, et al: An evolution of approach in vestibular neurectomy. Otolaryngol Head Neck Surg 102(4):374-381, 1990. 17. Silverstein H : Cochlear and vestibular gross and histo logic anatomy (as seen from postauricular approach). Otolaryngol Head Neck Surg 92(2):207-211, 1984. 18. Rasmussen A : Studies of the VIIIth cranial nerve of man. Laryngoscope 50:667, 1940. 19. Silverman D A, Hughes G B, Kinney S E, et al: Techni cal modifications of suboccipital craniectomy for preven tion of postoperative headache. Skull Base 14(2):77-84, 2004. 20. Megerian C A, Hanekamp J S, Cosenza M J, et al: Selec tive retrosigmoid vestibular neurectomy without internal auditory canal drill-out: an anatomic study. Otol Neuro tol 23(2):218-223, 2002. 21. Kartush J M, Telian S A, Graham M D, et al: Anatomic basis for labyrinthine preservation during posterior fossa acoustic tumor surgery. Laryngoscope 96(9 Pt 1):10241028, 1986. 22. Wazen JJ, Sisti M, Lam S M : Cranioplasty in acous tic neuroma surgery. Laryngoscope 110(8):1294-1297, 2000. 23. Feghali JG, Elowitz E H : Split calvarial graft cranioplasty for the prevention of headache after retrosigmoid resec tion of acoustic neuromas. Laryngoscope 108(10):14501452, 1998. 24. Soumekh B, Levine SC, Haines S J, et al: Retrospective study of postcraniotomy headaches in suboccipital ap proach: diagnosis and management. Am J Otol 17(4):617619, 1996. 25. Ozluoglu L N, Akbasak A : Video endoscopy-assisted ves tibular neurectomy: a new approach to the eighth cranial nerve. Skull Base Surg 6(4):215-219, 1996.
Chapter 36 • Retrolabyrinthine and Retrosigmoid Vestibular Neurectomy 26. Miyazaki H, Deveze A, Magnan J: Neuro-otologic sur gery through minimally invasive retrosigmoid approach: endoscope assisted microvascular decompression, ves tibular neurotomy, and tumor removal. Laryngoscope 115(9):1612-1617, 2005. 27. Wackym PA, King WA, Barker FG, et al: Endoscopeassisted vestibular neurectomy. Laryngoscope 108 (12):1787-1793, 1998. 28. Eisenman DJ, Speers R , Telian S A : Labyrinthectomy versus vestibular neurectomy: long-term physiologic and clinical outcomes. Otol Neurotol 22(4):539-548, 2001. 29. Quaranta A, Marini F, Sallustio V: Long-term outcome of Meniere’s disease: endolymphatic mastoid shunt versus natural history. Audiol Neurootol 3(1):54-60, 1998. 30. Quaranta A, Onofri M, Sallustio V, et al: Comparison of long-term hearing results after vestibular neurectomy, endolymphatic mastoid shunt, and medical therapy. Am J Otol 18(4):444-448, 1997. 31. Silverstein H, Smouha E, Jones R : Natural history vs. surgery for Meniere’s disease. Otolaryngol Head Neck Surg 100(1):6-16, 1989. 32. Filipo R , Barbara M : Natural course of Meniere’s dis ease in surgically-selected patients. Ear Nose Throat J 73(4):254-257, 1994. 33. Parikh A A, Brookes G B : Conductive hearing loss fol lowing retrolabyrinthine surgery. Arch Otolaryngol Head Neck Surg 122(8):841-843, 1996. 34. Teixido M, Wiet R J: Hearing results in retrolabyrinthine vestibular neurectomy. Laryngoscope 102(1):33-38, 1992. 35. Wazen JJ, Spitzer J, Kasper C, et al: Long-term hearing results following vestibular surgery in Meniere’s disease. Laryngoscope 108(10):1470-1473, 1998. 36. Lovely TJ, Lowry DW, Jannetta PJ: Functional outcome and the effect of cranioplasty after retromastoid craniec tomy for microvascular decompression. Surg Neurol 51(2):191-197, 1999. 37. Catalano PJ, Jacobowitz O, Post K D: Prevention of head ache after retrosigmoid removal of acoustic tumors. Am J Otol 17(6):904-908, 1996. 38. McElveen JT Jr, Shelton C, Hitselberger WE, et al: Retrolabyrinthine vestibular neurectomy: a reevaluation. Laryngoscope 98(5):502-506, 1988.
453
39. Monsell E M, Wiet R J, Young N M, et al: Surgical treat ment of vertigo with retrolabyrinthine vestibular neurec tomy. Laryngoscope 98(8 Pt 1):835-839, 1988. 40. Boyce S E, Mischke R E, Goin DW: Hearing results and control of vertigo after retrolabyrinthine vestibular nerve section. Laryngoscope 98(3):257-261, 1988. 41. Zini C, Mazzoni A, Gandolfi A, et al: Retrolabyrinthine versus middle fossa vestibular neurectomy. Am J Otol 9(6):448-450, 1988. 42. Glasscock M E 3rd, Thedinger B A, Cueva R A, et al: An analysis of the retrolabyrinthine vs. the retrosigmoid vestibular nerve section. Otolaryngol Head Neck Surg 104(1):88-95, 1991. 43. Ortiz Armenta A.: Retrolabyrinthine vestibular neurec tomy. 10 years’ experience. Rev Laryngol Otol Rhinol (Bord) 113(5):413-417, 1992. 44. Nguyen C D, Brackmann D E, Crane RT, et al: Retro labyrinthine vestibular nerve section: evaluation of tech nical modification in 143 cases. Am J Otol 13(4):328-332, 1992. 45. Aristegui M, Canalis R F, Naguib M, et al: Retrolaby rinthine vestibular nerve section: a current appraisal. Ear Nose Throat J 76(8):578-583, 1997. 46. Silverstein H, Jackson L E : Vestibular nerve section. Otolaryngol Clin North Am 35(3):655-673, 2002. 47. Molony TB : Decision making in vestibular neurectomy. Am J Otol 17(3):421-424, 1996. 48. Pareschi R , Destito D, Falco Raucci A, et al: Posterior fossa vestibular neurotomy as primary surgical treatment of Meniere’s disease: a re-evaluation. J Laryngol Otol 116(8):593-596, 2002. 49. Fukuhara T, Silverman D A, Hughes G B, et al: Vestibular nerve sectioning for intractable vertigo: efficacy of simpli fied retrosigmoid approach. Otol Neurotol 23(1):67-72, 2002. 50. Goksu N, Yilmaz M, Bayramoglu I, et al: Combined retrosigmoid retrolabyrinthine vestibular nerve section: results of our experience over 10 years. Otol Neurotol 26(3):481-483, 2005.
37
Translabyrinthine Vestibular Neurectomy Craig A. Buchman and Oliver F. Adunka
Deafferentation of the peripheral vestibular system continues to play a role in the management of patients with fluctuating or poorly compensated vestibulopathy that remains refractory to medical therapies or vestibular rehabilitation. In various peripheral vestibular disorders, abnormal spontaneous or motion-induced inputs that conflict with normal contralateral responses can create symptoms of dizziness, imbalance, vertigo, motion intolerance, and visual instability (i.e., oscillopsia) with a resulting negative impact on quality of life.1 Removing these dynamic vestibular signals generated from the abnormal ear can create a static vestibular lesion, for which the brain is more easily able to compensate.2 Deafferentation of the peripheral vestibular system can be produced through surgical or chemical labyrinthectomy, vestibular nerve section (i.e., neurectomy or neurotomy), or a combination thereof. Although chemical labyrinthectomy may produce either a partial or a total loss of peripheral vestibular function from the affected ear,3,4 surgical labyrinthectomy or vestibular nerve section reliably produces a complete lesion.5 Chemical labyrinthectomy destroys the sensory hair cells of the semicircular canals (SSC) and otolithic organs, whereas surgical labyrinthectomy comprehensively removes the contents of the vestibular labyrinth. Conversely, vestibular nerve section more proximally creates deafferentation by interrupting the transduction of the abnormal neural impulses from the labyrinth to the brainstem. Both surgical procedures allow for direct pathologic assessment of tissues, which may be necessary in certain instances. Each of these approaches has merits and disadvantages, and must be chosen on an individualized basis for the patient and his or her condition. This chapter focuses on translabyrinthine vestibular nerve section. This procedure, by necessity, combines the advantages of labyrinthectomy and vestibular nerve section, including (1) dual deafferentation of the peripheral vestibular end organs and (2) direct pathologic assessment of the contents of the vestibular labyrinth and the internal auditory canal (IAC). By removing preganglionic and postganglionic neural elements, a more complete vestibular lesion may be produced, especially in
cases where previous labyrinthectomy or vestibular nerve section attempts have failed.6 Examination of the tissues may reveal inflammatory or neoplastic processes that require further medical attention.6,7
HISTORY Charcot, in 1874, and later Frazier8 initially described neurectomy of the vestibulocochlear nerve. Subsequently, in 1928, Dandy9 strongly advocated intracranial sectioning of the eighth cranial nerve for paroxysmal vertigo symptoms. Although he noted that the procedure often was successful for relieving troubling vertigo, patients typically had a complete hearing loss in the operated ear. McKenzie,10 in 1931, developed a selective neurectomy technique, preserving auditory function in some cases. Some 30 years later, House11 developed the middle fossa and translabyrinthine approaches for vestibular nerve sectioning. Later modifications by Fisch and Glasscock and colleagues12-14 refined these approaches. The importance of preserving not only the cochlear nerve, but also the labyrinthine blood supply as a critical premise for hearing preservation was realized. Silverstein and colleagues15,16 later described in detail the complex interrelationship of the vestibular and cochlear nerve fibers in the cerebellopontine angle, allowing for even more selective neurectomy near the root entry zone in the brainstem. Over the last 50 years, Hitselberger and his colleagues can be credited with performing vestibular nerve section procedures on hundreds of patients with peripheral vestibular disorders.17 Today, successful vestibular nerve section relies on the anatomic and surgical principles developed by these pioneers.
DIAGNOSTIC CONSIDERATIONS Selecting an appropriate surgical intervention for patients with peripheral vestibular disorders is challenging. In considering such therapy, a diagnosis must first be rendered or, minimally, an affected ear must be identified. 455
456
OTOLOGIC SURGERY
Although a detailed discussion regarding the diagnosis and management of vestibular disorders is beyond the scope of this chapter, an understanding of the various vestibular disorders is crucial to identifying a correct diagnosis and making an adequate treatment recommendation.18,19 Establishing the correct diagnosis may be difficult sometimes, and can require the input from various professionals in differing disciplines. Vertigo is the cardinal symptom of a vestibular system disorder. It is also important to recognize that many cases of dizziness are not related to the vestibular system at all, and dizziness without vertigo is common. When vertigo is absent, the diagnosis of vestibular disorder should be carefully scrutinized. Although the presence of vertigo can be of either central or peripheral origin, most cases of vertigo arise from the peripheral vestibular apparatus (SSC, otolithic organs, or vestibular nerves). Although most of these peripheral vestibular disorders manifest to the clinician with a clear clinical picture, occasionally differentiation from the common central causes of vertigo is difficult. Differentiation among the peripheral and central varieties of vertigo takes on even greater significance when ear-specific therapies are being contemplated. When peripheral deafferentation is considered in the treatment regimen, defining the peripheral cause and the affected labyrinth is crucial. The most common causes of peripheral (or ear-related) vertigo in adult patients include benign paroxysmal positional vertigo (BPPV), Meniere’s disease, and vestibular neuronitis. More recently, superior semicircular canal dehiscence (SSCD) syndrome has become more frequently recognized.20,21 Common central causes of vertigo include migraine-related dizziness (also known as migraine vestibulopathy22), transient ischemic attack, and demyelinating disease such as multiple sclerosis and stroke. In nearly all patients with vertigo, a careful history, physical examination, and appropriate laboratory and radiologic testing can reveal the correct diagnosis. A minimal test battery usually includes comprehensive audiometric testing, imaging of the temporal bone and brain, and laboratory evaluations, which may include selective vestibular function testing (see section on preoperative testing). Meniere’s disease is characterized by episodic vertigo that lasts minutes to hours with associated, fluctuating aural symptoms, including hearing loss, tinnitus, and pressure. Audiometric testing characteristically reveals a low-frequency sensorineural hearing impairment. In contrast, BPPV is characterized by brief (<20 seconds), position-induced attacks of vertigo in the absence of other auditory symptoms. Symptoms in these patients can be reproduced with the Dix-Hallpike maneuver, which often reveals a torsional nystagmus when the head is positioned with the affected ear in a gravity-dependent position. By contrast, vestibular neuronitis typically is considered when patients experience a single, prolonged (i.e., days) attack of vertigo, without associated auditory
symptoms, which is followed by a period (i.e. days to weeks) of motion intolerance or BPPV or both. SSCD syndrome is characterized by symptoms of vertigo or dizziness induced by sound or pressure applied to the affected ear or via a Valsalva maneuver.20,21 Typically, auditory symptoms include hearing loss, pressure, and occasionally pulsatile tinnitus. In this disorder, audiometric testing commonly reveals a conductive hearing loss with preserved acoustic reflexes. Occasionally, the previously described peripheral vestibular disorders are not easily recognized, and further testing is needed, usually vestibular function testing and imaging studies. Vestibular function testing in the form of electronystagmography (ENG) can help identify (1) nystagmus associated with BPPV, (2) pressureinduced nystagmus in SSCD syndrome, and (3) reduced responsiveness to caloric irrigations after vestibular neuronitis or prolonged exposure to Meniere’s disease. Rotational chair testing can be useful in assessing a patient’s level of central compensation after a peripheral vestibular lesion, but is seldom helpful in localizing an affected ear. Vestibular-evoked myogenic potential (VEMP) testing can help identify inferior vestibular nerve dysfunction in the setting of a normal caloric response on ENG. VEMP testing can also be useful in patients with suspected SSCD syndrome in which thresholds are reduced in the affected ear. Although these tests can be a useful adjunct in making the diagnosis, it remains crucial to rule out central vestibular disorders, especially when deafferentation is being contemplated. In addition to a careful history and physical examination, magnetic resonance imaging (MRI) is often useful to help exclude multiple sclerosis, stroke, and tumors such as an acoustic neuroma. Migraine-related dizziness is common and can masquerade as any of the aforementioned vestibular disorders.22,23 It is characterized by episodic vertigo or dizziness that can last seconds to hours or even days. It may or may not be associated with headache, visual, or auditory symptoms. When retro-orbital headache and visual scotoma are associated with vertigo attacks, the diagnosis is seldom difficult. Symptoms often may abate following sleep, adding credence to the diagnosis. Patients with migraine-related dizziness frequently relate a history of motion or sound intolerance, however, in periods between attacks similar to individuals with BPPV and SSCD syndrome. Migraine-related vertigo may also be associated with a classic Meniere’s disease history, which can make distinguishing these disorders extremely difficult, especially when there is no hearing loss. Patients with migraine-related dizziness can have ENG abnormalities and can respond to diuretic therapies similar to patients with Meniere’s disease.24 All otologists who manage patients with vertigo should have a very high index of suspicion for this disorder. When patients have episodic vertigo in the absence of objective evidence of hearing loss, migraine-related dizziness must be ruled out before considering ear-specific therapy.
Chapter 37 • Translabyrinthine Vestibular Neurectomy
TREATMENT OF VESTIBULAR DISORDERS: AN OVERVIEW When the correct diagnosis has been made, medical management should be used in an attempt to control unwanted symptoms before considering ear-specific therapies, destructive therapies, or both. “Control of symptoms” is a personal judgment made by the patient and the physician, and can differ among individuals. In one patient, an occasional attack of vertigo may be unacceptable, whereas in another individual, frequent attacks may have little consequence to their daily life. The severity of attacks may be a more compelling reason for intervention than just the frequency of the attacks. Physicians are left to contemplate these interventions with their patients in the context of this varying degree of disability. When diagnostic uncertainty persists, further medical evaluations are often necessary before considering ear-specific therapies. For Meniere’s disease, medical therapies can be conveniently divided into either prophylactic or abortive treatments. Prophylactic interventions generally include dietary modifications and pharmacologic therapies with diuretics, vasodilators, antihistamines, and immune modu lators. Abortive treatments usually include some regimen of vestibular suppressants. These various medical therapies are usually efficacious in controlling episodic, spontaneous vertigo attacks in most patients with only a few requiring further treatment. Vestibular physical therapy can be a useful adjunct for inducing compensation in individuals with significant motion intolerance resulting from vestibular loss. Vestibular physical therapy does not treat or prevent spontaneous attacks of vertigo in patients with Meniere’s disease. When medical interventions fail to control symptoms, ear-specific therapies should be contemplated. Ear-specific therapies include destructive and nondestructive therapies. Destructive therapies are therapies that induce either partial or total vestibular deafferentation with or without hearing loss. Nondestructive therapies include endolymphatic sac procedures, intratympanic steroids,25 and transtympanic pressure therapy26 in patients with Meniere’s disease. For patients with SSCD syndrome, tympanostomy tube placement to control pressure-induced symptoms could also be considered as a nondestructive therapy. Canalith repositioning maneuvers are ear-specific therapies that often resolve unwanted positional vertigo in patients with BPPV. As a last resort, destructive or deafferentation procedures are considered for patients with symptoms that remain refractory to the aforementioned approaches. Deafferentation can be accomplished by either chemical or surgical means, and can be directed at the receptors of the inner ear, the vestibular afferent fibers to the brain, or some combination thereof. When the labyrinth remains intact, these procedures can be undertaken with the intent of hearing preservation. Hearing-preserving total labyrinthine deafferentation procedures include
457
vestibular nerve sectioning5,27 and intratympanic gentamicin.28 For patients with BPPV, partial deafferentation can be accomplished by partitioning of the affected posterior SCC while maintaining hearing.29 Likewise, patients with SSCD can achieve control of symptoms with surgery directed at occlusion of the affected superior SCC, often with improvement in hearing.30 When hearing is not an issue, transmastoid labyrinthectomy or translabyrinthine vestibular nerve sectioning can reliably provide comprehensive vestibular deafferentation.
Decision Making The selection of a procedure for a particular patient depends on the frequency and severity of symptoms and the associated effect on quality of life, hearing status, and patient and otologist preference. Nondestructive procedures should be considered before destructive ones because the consequences of deafferentation using either chemical or surgical means include acute, post-treatment dizziness or vertigo or both with associated imbalance. Although compensation generally occurs after weeks to months, supplementary vestibular rehabilitation therapy is often needed.31 Despite all efforts, patients who have undergone deafferentation are often left with some residual imbalance that is refractory to physical therapy.5,32 These patients can be notably discouraged by such an outcome. Before undertaking deafferentation, patients should understand what it can accomplish—control of unwanted spontaneous or motion-induced attacks of vertigo. This control frequently comes at the expense of having some permanent degree of imbalance and motion intolerance, at least in sensory-deprived situations, such as during ambulation in the dark or while on an unstable surface or in periods of relative weightlessness. When considering a destructive procedure, knowing what constitutes useful hearing is also crucial. Classically, serviceable hearing was defined as a pure tone average equal or better than 50 dB HL and a speech reception score (W22 word lists) of less or equal than 50% correct.33 In this regard, patients with hearing in an affected ear worse than this definition were considered for hearing sacrifice, whereas patients with better hearing were considered for a hearing-preserving approach. Subsequently, a classification scheme was developed by Shelton and Hitselberger,34 and later modified and adopted by the American Academy of Otolaryngology–Head and Neck Surgery.35 This scheme defines good hearing (class A, pure tone average 30 dB, standard deviation score 70%), serviceable hearing (class B, pure tone average 50 dB, standard deviation score 50%), measurable hearing (class C, any measurable hearing loss), and a “dead ear” (class D, absence of measurable hearing). Although this classification system provides only a rough guide regarding the quality of a patient’s hearing, this system is useful for reporting results of treatment. In a manner similar to vertigo control, the definition of what
458
OTOLOGIC SURGERY
constitutes good or useful hearing might differ from individual to individual. Two people with moderate to severe hearing loss in the affected ear might perceive anacusis quite differently when the contralateral ear is normal, or when it has a significant hearing loss. Conversely, patients with poor discrimination abilities in the affected ear may note improved hearing abilities after a hearingsacrificing procedure because unwanted, distorted signals are removed. Hearing-sacrificing procedures should be undertaken after a careful discussion between the treating physician and the patient.
Transmastoid Labyrinthectomy versus Translabyrinthine Vestibular Nerve Section for Deafferentation Transmastoid labyrinthectomy, and to a lesser extent transcanal labyrinthectomy, removes the contents of the vestibular labyrinth, including the otolithic organs of the saccule and utricle and the crista ampullaris of the SSC. In this regard, labyrinthectomy accomplishes postganglionic (i.e., distal to Scarpa’s ganglion) deafferentation without opening the cerebrospinal fluid–containing spaces. Transmastoid labyrinthectomy allows for direct visualization of the contents of the vestibule and pathologic assessment of the tissues. It can be performed as an outpatient procedure if postoperative vertigo is not severely disabling. Similar to transmastoid labyrinthectomy, translabyrinthine vestibular nerve section (TLVNS) results in a complete labyrinthectomy. In addition, preganglionic nerve sectioning, proximal to Scarpa’s ganglion, is accomplished.6,36 By necessity, TLVNS opens the IAC, exposing the patient to potential cerebrospinal fluid leakage and other intracranial complications. When should TLVNS be undertaken? This question is open to some debate, and might represent differing philosophical approaches among neurotologists. The merits of TLVNS over transmastoid labyrinthectomy are dual denervation (i.e., preganglionic and postganglionic) of the vestibular system and pathologic assessment of the IAC and cerebellopontine angle tissues. If preoperative MRI identifies IAC enhancement, pathologic assessment might be warranted to rule out inflammatory or neoplastic conditions. TLVNS might be useful in cases of previously failed labyrinthectomy or vestibular nerve section in which partial removal of the neuroepithelium or partial sectioning of the vestibular nerves was accomplished. Previous investigators have suggested that failed labyrinthectomy or vestibular nerve section might also occur because of the development of a postsurgical neuroma in the labyrinth or at the cut ends of the vestibular nerves.6,7,36 Traumatic neuromas can result in spontaneous neural impulses through either ephaptic transmission or neural crosstalk.37 These spontaneous impulses can presumably result in conflicting information at the level of the brainstem nuclei, creating vertigo and dizziness. Finally,
some neurotologists might consider the dual denervation accomplished by TLVNS as the most reliable method for eliminating ear-related vertigo in patients with poor hearing, and would use this as their primary approach in selected cases. The use of TLVNS as a primary form of deafferentation must be balanced against the additional risks of dural opening in an individual patient.
PREOPERATIVE TESTING Before vestibular denervation procedures are considered, a thorough functional and anatomic evaluation of the patient’s auditory and vestibular systems is indicated.
Functional Vestibular System Assessment Electronystagmography ENG provides an objective assessment of the oculomotor and vestibular systems. The standard ENG battery consists of three parts: oculomotor evaluation, positioning/ positional testing, and caloric stimulation of the vestibular system. Results of each subtest can help determine whether a balance disorder is of central or peripheral origin. Caloric stimulation predominantly assesses the functional status of the superior vestibular nerve and the horizontal SCC. Before vestibular deafferentation, the ENG test battery serves to delineate the peripheral vestibular reactivity of each ear. This delineation may help determine the affected side if this is not clinically obvious. In cases of known laterality, ENG can rule out an unsuspected, contralateral peripheral vestibular deficit, preventing unwanted bilateral, vestibular hypofunction.5
Vestibular Evoked Myogenic Potentials The clinical utility of VEMP testing is primarily for documenting inferior vestibular nerve function or dysfunction, and assisting in the diagnosis of SSCD syndrome. For patients with SSCD syndrome, VEMP thresholds are classically reduced compared with unaffected ears.20 For patients contemplating deafferentation, the usefulness of the results of VEMP testing remains to be determined. It seems that the clinical utility of VEMP testing can be thought of as being similar to ENG testing. That is, VEMP testing might be useful to determine the affected side if this is not clinically obvious. In cases of known laterality, VEMP might help rule out an unsuspected, contralateral peripheral vestibular deficit, preventing unwanted bilateral, vestibular hypofunction.
Platform Posturography Platform posturography can be used to determine the overall functional impact of a particular vestibular disorder on balance. This test can provide some information
Chapter 37 • Translabyrinthine Vestibular Neurectomy
regarding visual, vestibular, proprioceptive, and motor contributions to balance function.38 Although this test may not be helpful preoperatively for decision making, serial posturography may allow for a better assessment of the effects of postoperative vestibular rehabilitation efforts.
Functional Auditory System Assessment Because TLVNS results in anacusis in the operated ear, a critical assessment of the patient’s hearing abilities is indicated before surgery. This assessment usually includes diagnostic behavioral audiometry using pure tone and speech stimuli. Acoustic reflex testing and tympanometry can also be useful to assess further the acoustic-facial reflex pathways (acoustic reflex testing) and middle ear status (tympanometry).
Anatomic Vestibular and Auditory System Assessment Magnetic Resonance Imaging MRI should be part of the diagnostic work-up for patients with persistent balance and equilibrium problems. Specifically, MRI should be used to assess possible central and retrocochlear pathology.39 An adequate study should be furnished as a high-resolution study with at least 1.5 T and should include T1-weighted and T2-weighted studies. Also, a contrast-enhanced T1-weighted series should be obtained. A continuous interference in a steady state (CISS) sequence might help to delineate each nerve within the IAC, and is generally part of a CN VIII protocol used in most institutions.
459
sterile saline are on the field. Monopolar and bipolar electrocautery are available. Today, CN VII monitoring equipment and the operating microscope are used for every case. The facial nerve monitoring electrodes are inserted into the lateral portions of the orbicularis oris and orbicularis oculi muscles and covered with a sterile, impermeable drape. Also, a small abdominal area left of the umbilicus is prepared for harvesting an abdominal fat graft. Informed consent, audiogram, and ENG are reviewed to confirm the correct patient and laterality of the procedure. Generally, a first-generation or second-generation cephalosporin is given before incision for broad antimicrobial prophylaxis. In cases of β-lactam allergy, other agents may be used. An incision is made approximately 1 cm above and behind the postauricular crease and follows the contour of the auricle (Fig. 37-1). A plane is established in the galea aponeurotica lateral to the temporalis muscle, and the auricle is turned forward. This layer is best established directly over the temporalis fascia superiorly and over the mastoid periosteum inferiorly. The musculoperiosteal flap is incised separately and elevated off the mastoid cortex with the help of a periosteal elevator or monopolar electrocautery. This flap is moved forward to the level of the posterior external auditory canal and mastoid tip, and subsequently fixed with a large selfretaining retractor. Care is taken not to create an opening in the external auditory canal skin because this can provide a route for cerebrospinal fluid leakage postoperatively. The high-speed drill with a large cutting burr and constant suction-irrigation is used to perform a cortical mastoidectomy. The posterior aspect of the external
High-Resolution Computed Tomography Some authors supplement MRI with a high-resolution computed tomography (CT) scan of the temporal bones. High-resolution CT provides visualization of bony structures and can delineate middle ear and mastoid anatomy. High-resolution CT remains the gold standard to determine the anatomic integrity of the superior SCC and should be used to rule out superior SSCD.40
SURGICAL PROCEDURE Surgery is performed under general anesthesia. The patient is placed on the operating table in a supine position; the table is rotated to allow for table manipulations and to permit the surgeon to sit comfortably with knees under the table. A postauricular sterile area of about 3 cm is prepared and draped off. This area should be extended slightly superior to the auricle to provide adequate exposure of the middle fossa plate. The highspeed drill with various burrs and suction-irrigation with
FIGURE 37-1. Postauricular skin incision about 2 to 3 mm behind the aural crease with an extension superior to the auricle.
460
OTOLOGIC SURGERY
FIGURE 37-2. Surgical view after cortical mastoidectomy has been performed. Facial nerve is visualized in its entire mastoid segment. Facial recess is opened (not shown). Sigmoid sinus and bony middle fossa plate have been visualized. Bony labyrinthine capsule has been exposed.
auditory canal wall and the bone overlying the inferior temporal lobe dura is thinned, and the sigmoid sinus is skeletonized (Fig. 37-2). The mastoid antrum, horizontal SCC, and short process of the incus are brought into clear view. For the sigmoid sinus, an eggshell of bone is usually left over the vein wall to protect it from manipulation during deeper dissection. Some surgeons leave a larger bony island over the sinus to compress it medially and gain better exposure of the medial temporal bone. The sinodural angle is opened as far posteriorly on the cortex as possible. Because the vestibule lies medial to the facial nerve, a tangential view is facilitated by this additional exposure. At this point, the descending mastoid segment of the facial nerve is identified, and the presigmoid posterior fossa dura can be thinned. Generally, the nerve is identified while opening the facial recess. This access to the middle ear is used later for eustachian tube closure to prevent cerebrospinal fluid egress. The incudostapedial joint is divided, and the incus is removed. Care should be taken not to disrupt the stapediovestibular ligament because this may provide another route for cerebrospinal fluid leakage into the middle ear. It also may be useful to transect the tensor tympani tendon for
better access to the eustachian tube lumen. The mastoid segment of the facial nerve is followed inferiorly toward the stylomastoid foramen. This step is necessary for wide access to the ampullated end of the posterior SCC and, more medially, the inferior trough below the IAC. The three SSCs of the vestibular labyrinth are skeletonized. Before labyrinthectomy and subsequent permanent hearing loss, the operative consent, audiogram, and ENG are reviewed again as an additional level of safety. The labyrinthectomy is initially carried out using a 3 or 4 mm cutting burr. First, the horizontal SCC is half-opened on its superior border from the ampulla anteriorly to the intersection with the posterior SCC. The horizontal SCC is initially only half-opened to protect the external genu of the facial nerve inferiorly. Next, the posterior SCC is opened and traced superiorly to its confluence with the superior SCC (i.e., common crus) (Fig. 37-3). The common crus may be followed directly toward the vestibule (Fig. 37-4). The superior SCC is also opened along the entire path to the ampulla. During this maneuver, care is taken not to damage the temporal lobe dura, which usually lies in very close approximation to the superior SCC lumen.
Chapter 37 • Translabyrinthine Vestibular Neurectomy
The superior and posterior surfaces of the external genu of the facial nerve are now carefully identified to facilitate dissection of the posterior SCC ampulla, which lies more medially. Inferior dissection in the retrofacial air cell tract beyond the posterior SCC ampulla is unnecessary in this procedure, and might reveal the jugular bulb. When the three SCC ampullae and the common crus have been opened to the vestibule, the spherical and elliptical recesses of the vestibule and the endolymphatic duct can be appreciated. At this point, all soft tissue elements of the membranous labyrinth are removed. This step would be the normal end point for postganglionic, transmastoid labyrinthectomy (Fig. 37-5). A key factor for successful TLVNS is thorough dural decompression of the IAC from the porus acusticus to vestibule with precise identification of the facial nerve at the meatal foramen and in its labyrinthine segment. Beyond the labyrinthectomy, the approach to the IAC is aligned by recognizing that the lateral IAC boundary is the vestibule and the three SCC ampullae. The subarcuate artery courses through the arch of the superior SCC within the petromastoid canal, typically superior to the IAC. Also, the IAC lies in a similar plane as the external auditory canal. With these anatomic relationships in mind, the presigmoid posterior fossa dura is decompressed medially until the porus acusticus dura is identified. When the IAC dura is identified medially, further anterior dissection superior and inferior to the canal allows for the creation of “troughs.” These troughs should create approximately 270 degrees of dural exposure over the medial two thirds of the IAC (Fig. 37-6). The superior trough is bounded superiorly by the temporal lobe dura and inferiorly by the IAC dura. The inferior trough is bounded superiorly by the IAC dura and inferiorly by the cochlear aqueduct and jugular bulb. Frequently, air cells are encountered in the superior trough above the IAC. In the inferior trough, dissection usually reveals the cochlear aqueduct as a fibrous tissue–containing tract that elutes cerebrospinal fluid. Inferior dissection below the aqueduct should be avoided because the lower cranial nerves can be at risk in this region. Finally, dural decompression of the lateral aspect of the IAC should proceed to identify the vertical and the horizontal crests of the IAC, and subsequently the facial nerve in its labyrinthine segment. The vertical crest (or Bill’s bar) is identified by following the superior IAC dura laterally to the vestibule. The superior vestibular nerve can also be traced to the ampulla of the superior SCC. The vertical crest lies anterior to the superior vestibular nerve. When the vertical crest is identified, emanating medially off the central portion of the superolateral end of the IAC, it is anticipated that the labyrinthine segment of the facial nerve will be just anterior to this structure. The labyrinthine segment is partially identified for confirmation. When bisecting the
461
posterior wall of the IAC in its lateral aspect, the horizontal or transverse crest can be identified. This crest separates the superior and inferior vestibular nerve bundles as they exit the labyrinth. After drilling is complete, bone chips are removed, and the dura is opened. Dural opening is accomplished via sharp dissection. The safest way to identify the intradural facial nerve is by careful palpation of Bill’s bar with visualization of the labyrinthine segment. The superior vestibular nerve is avulsed from its labyrinthine attachment, and the soft tissue plane between the vestibular and facial nerves is developed. The inferior vestibular nerve is transected similarly and dissected medially (Fig. 37-7). Scarpa’s ganglion lies approximately midway within the IAC portion of the nerve. If pathologic assessment is needed, a portion of the nerve is removed and sent to the laboratory (Fig. 37-8). Although the cochlear nerve may also be sectioned, this action would preclude its use in future cochlear implantation if that opportunity arises.41 Closure is accomplished by first plugging the eustachian tube and then closing the dural defect and wound. The eustachian tube is addressed through the facial recess by placing a piece of absorbable knitted fab ric (Surgicel) deeply in to the tubal lumen. Secondarily, pieces of muscle usually from the temporalis are packed into the eustachian tube and middle ear up to the facial recess opening; another sheet of Surgicel holds this in place. The dural opening is closed by careful packing with strips of abdominal fat placed partly through the opening.42 With such a small defect, a watertight closure is usually attained with a few carefully placed pieces. A titanium mesh placed over the mastoid cortex opening is often helpful to bolster the fat in place.43 The wound is closed in layers with an interrupted absorbable suture, and a firm mastoid dressing is placed.
POSTOPERATIVE CARE AND FOLLOW-UP The patient is watched in the intensive care unit for a day with hourly neurologic checks and then is moved to a step-down room when neurologic stability is ensured. If an abdominal wound drain was used, this is removed on postoperative day 1. Antibiotics are discontinued after 24 hours so as not to promote infection with resistant organisms. Patients are encouraged to sit up and dangle their legs off the side of the bed on the 1st or 2nd postoperative day, and to begin ambulation with help as soon as possible. Ambulation and activity are usually limited by the degree of vertigo, nausea, and vomiting. These symptoms are usually controlled medically with vestibular suppressants, such as droperidol, diazepam, meclizine, and dimenhydrinate. When static compensation is sufficient, ambulation is usually possible with some assistance. At this point, head movements still elicit significant vertigo
462
OTOLOGIC SURGERY
Chapter 37 • Translabyrinthine Vestibular Neurectomy
463
FIGURE 37-3. Lateral and posterior semicircular canals have been opened. Posterior canal can be traced superiorly and medially to common crus with superior semicircular canal. FIGURE 37-4. Superior semicircular canal has been opened. Also, common crus of superior and posterior semicircular canals is now visible. Vestibule has been carefully opened, and soft tissue can be stripped out using a right angle hook. FIGURE 37-5. Vestibule has been opened. Care must be taken when drilling the ampullated part of the posterior semicircular canal because it is located medial to the mastoid segment of the facial nerve. FIGURE 37-6. Skeletonizing the bony internal auditory canal (IAC). FIGURE 37-7. Establishing the contents of internal auditory canal. Superior and inferior vestibular nerves have been avulsed. Vertical crest (Bill’s bar) can be seen with the facial nerve underneath. The cochlear nerve can be found inferior and anterior within internal auditory canal. FIGURE 37-8. Both branches of vestibular portion of CN VIII have been avulsed. Cochlear and facial nerves are visible in anterior compartment of internal auditory canal.
464
OTOLOGIC SURGERY
and are usually avoided. Discharge is possible when the patient is able to ambulate and tolerate a diet; however, driving is contraindicated until head movement does not cause dizziness. This level of activity usually requires dynamic compensation to occur. This process can take weeks to occur, and may require a course of vestibular rehabilitation therapy to become complete.31 The rate of static compensation for vertigo while in the hospital is predicted by the severity of the postoperative nystagmus. Horizontal spontaneous nystagmus is evident in all three directions of gaze (third-degree nystagmus) on the 1st or 2nd postoperative days. According to Alexander’s law, as the level of static compensation increases as a result of alterations in the vestibular nuclei tonic discharge rate, nystagmus becomes evident only in central and contralateral gazes (second-degree nystagmus). When further compensation occurs, nystagmus may be evident only on contralateral gaze (first-degree nystagmus). This level of nystagmus usually coincides with the patient’s ability to ambulate and readiness for discharge around the 5th to 7th postoperative day.18
COMPLICATIONS Complications seen with TLVNS may include cerebrospinal fluid leak, meningitis, and facial paralysis.6,36,44 Facial paralysis can be considered the consequence of working in an area in which the nerve is anatomically at risk. Avoidance is the best solution to this problem: the surgeon should be certain of the nerve’s location and treat it with respect.45 Postoperative steroids or limited decompression of the nerve, particularly in the labyrinthine segment, may be useful in preventing sequelae when the nerve is known to have been traumatized. When fever and meningismus occur, a lumbar puncture is performed, and appropriate antibiotic coverage is instituted to treat bacterial meningitis. This complication is rare. Cerebrospinal fluid leakage occurs in less than 5% of TLVNS cases using the previously described techniques. This low incidence is likely due to the limited dural opening needed for this procedure and careful closure. If wound leakage occurs, this can be managed with simple oversewing at the bedside. If rhinorrhea occurs, this is managed with a firm pressure bandage, head of bed elevation, and lumbar subarachnoid drainage for 5 days.46 If resolution of the leak is not seen after aggressive drainage, the wound is explored, and the adipose plug is readjusted.
RESULTS AND LITERATURE OVERVIEW TLVNS is successful at relieving spontaneous attacks of vertigo in 75% to 95% of patients with Meniere’s disease; it is substantially less successful in patients with vertigo
not related to Meniere’s disease.36,47 These findings suggest that our ability to diagnose the disorder correctly as one of the peripheral vestibular apparatus or ability to induce stable and complete central vestibular compensation remains imperfect.5,6,32
CONCLUSION TLVNS is a highly effective approach for achieving complete surgical deafferentation in patients with refractory vertigo from Meniere’s disease and other peripheral labyrinthine disorders. Patients should be carefully selected for this procedure and be willing to give up their residual hearing in the affected ear for vertigo control. They should be aware that deafferentation may be a tradeoff between relief of spontaneous vertigo at a cost of unsteadiness and imbalance. Although compensation is common among young and healthy patients, elderly patients may not be as tolerant. When carefully performed, TLVNS is the gold standard procedure for peripheral deafferentation, and can be used in patients who have failed previous medical and surgical attempts at controlling symptoms. Although complications can occur, they are uncommon and mostly not serious.
REFERENCES 1. Green J D Jr, Verrall A, Gates G A : Quality of life instruments in Meniere’s disease. Laryngoscope 117:1622-1628, 2007. 2. Igarashi M : Vestibular compensation: An overview. Acta Otolaryngol 406:78-82, 1984. 3. Hoffmann K K, Silverstein H : Inner ear perfusion: Indications and applications. Curr Opin Otolaryngol Head Neck Surg 11:334-339, 2003. 4. Blakley BW: Clinical forum: A review of intratympanic therapy. Am J Otol 18:520-526, 1997. 5. Eisenman DJ, Speers R , Telian S A : Labyrinthectomy versus vestibular neurectomy: Long-term physiologic and clinical outcomes. Otol Neurotol 22:539-548, 2001. 6. Monsell E M, Brackmann D E, Linthicum FH Jr: Why do vestibular destructive procedures sometimes fail? Otolaryngol Head Neck Surg 99:472-479, 1988. 7. Linthicum FH Jr, Alonso A, Denia A : Traumatic neuroma: A complication of transcanal labyrinthectomy. Arch Otolaryngol 105:654-655, 1979. 8. Frazier C H : Intracranial division of the auditory nerve for persistent aural vertigo. Surg Gynecol Obstet 15:524529, 1912. 9. Dandy WE : Intracranial division of the vestibular portion of the auditory nerve for Meniere’s disease. Arch Surg 16:1127-1152, 1928. 10. McKenzie KG: Intracranial division of the vestibular portion of the auditory nerve for Meniere’s disease. Can Med Assoc J 34:369-381, 1936. 11. House WF: Surgical exposure of the internal auditory canal and its contents through the middle, cranial fossa. Laryngoscope 71:1363-1385, 1961.
Chapter 37 • Translabyrinthine Vestibular Neurectomy 12. Fisch U: Vestibular and cochlear neurectomy. Trans Am Acad Ophthalmol Otolaryngol 78:ORL252ORL255, 1974. 13. Glasscock M E III, Kveton J F, Christiansen SG: Middle fossa vestibular neurectomy: An update. Otolaryngol Head Neck Surg 92:216-220, 1984. 14. Glasscock M E III: The surgical treatment of vertigo. South Med J 66:526-530, 1973. 15. Silverstein H, Norrell H, Haberkamp T, et al: The unrecognized rotation of the vestibular and cochlear nerves from the labyrinth to the brain stem: Its implications to surgery of the eighth cranial nerve. Otolaryngol Head Neck Surg 95:543-549, 1986. 16. Silverstein H : Cochlear and vestibular gross and histologic anatomy (as seen from postauricular approach). Otolaryngol Head Neck Surg 92:207-211, 1984. 17. Nguyen C D, Brackmann D E, Crane RT, et al: Retrolabyrinthine vestibular nerve section: Evaluation of technical modification in 143 cases. Am J Otol 13:328-332, 1992. 18. Baloh RW, Honrubia V: Clinical Neurophysiology of the Vestibular System, 3rd ed. New York, Oxford University Press, 2001. 19. Baloh RW, Halmagyi GM: Disorders of the Vestibular System, 1st ed. New York, Oxford University Press, 1996. 20. Minor LB, Carey JP, Cremer PD, et al: Dehiscence of bone overlying the superior canal as a cause of apparent conductive hearing loss. Otol Neurotol 24:270-278, 2003. 21. Minor L B, Solomon D, Zinreich J S, et al: Sound- and/ or pressure-induced vertigo due to bone dehiscence of the superior semicircular canal. Arch Otolaryngol Head Neck Surg 124:249-258, 1998. 22. Brantberg K, Trees N, Baloh RW: Migraine-associated vertigo. Acta Otolaryngol 125:276-279, 2005. 23. Shepard NT: Differentiation of Meniere’s disease and migraine-associated dizziness: A review. J Am Acad Audiol 17:69-80, 2006. 24. Furman J M, Sparto PJ, Soso M, et al: Vestibular function in migraine-related dizziness: A pilot study. J Vestib Res 15:327-332, 2005. 25. Boleas-Aguirre M S, Lin FR , Della Santina CC, et al: Longitudinal results with intratympanic dexamethasone in the treatment of Meniere’s disease. Otol Neurotol 29:33-38, 2008. 26. Gates G A, Verrall A, Green J D Jr, et al: Meniett clinical trial: Long-term follow-up. Arch Otolaryngology Head Neck Surg 132:1311-1316, 2006. 27. Rosenberg S I, Silverstein H, Hoffer M E, et al: Hearing results after posterior fossa vestibular neurectomy. Otolaryngol Head Neck Surg 114:32-37, 1996. 28. Cohen-Kerem R , Kisilevsky V, Einarson TR , et al: Intratympanic gentamicin for Meniere’s disease: A metaanalysis. Laryngoscope 114:2085-2091, 2004. 29. Parnes L S, McClure J A : Posterior semicircular canal occlusion in the normal hearing ear. Otolaryngol Head Neck Surg 104:52-57, 1991. 30. Limb C J, Carey J P, Srireddy S, et al: Auditory function in patients with surgically treated superior semicircular canal dehiscence. Otol Neurotol 27:969-980, 2006.
465
31. Whitney S L , Rossi M M : Efficacy of vestibular rehabilitation. Otolaryngol Clin North Am 33:659-672, 2000. 32. Teufert KB, Berliner KI: De la Cruz A: Persistent dizziness after surgical treatment of vertigo: An exploratory study of prognostic factors. Otol Neurotol 28:1056-1062, 2007. 33. De la Cruz A: Teufert KB, Berliner KI: Surgical treatment for vertigo: Patient survey of vertigo, imbalance, and time course for recovery. Otolaryngol Head Neck Surg 135:541-548, 2006. 34. Shelton C, Hitselberger WE : The treatment of small acoustic tumors: Now or later? Laryngoscope 101:925928, 1991. 35. Monsell E M: New and revised reporting guidelines from the Committee on Hearing and Equilibrium. American Academy of Otolaryngology–Head and Neck Surgery Foundation, Inc. Otolaryngol Head Neck Surg 113:176-178, 1995. 36. De La Cruz A, Teufert K B, Berliner K I : Transmastoid labyrinthectomy versus translabyrinthine vestibular nerve section: Does cutting the vestibular nerve make a difference in outcome? Otol Neurotol 2007 Sep; 28(6):801-8. 37. Rasminsky M : Spontaneous activity and cross-talk in pathological nerve fibers. Assoc Res Nerv Mental Dis 65:39-49, 1987. 38. Monsell E M, Furman J M, Herdman SJ, et al: Computerized dynamic platform posturography. Otolaryngol Head Neck Surg 117:394-398, 1997. 39. Sedwick J D, Gajewski B J, Prevatt A R , et al: Magnetic resonance imaging in the search for retrocochlear pathology. Otolaryngol Head Neck Surg 124:652-655, 2001. 40. Minor L B : Superior canal dehiscence syndrome. Am J Otol 21:9-19, 2000. 41. Facer GW, Facer ML, Fowler CM, et al: Cochlear implantation after labyrinthectomy. Am J Otol 21:336-340, 2000. 42. House J L , Hitselberger WE, House WF: Wound closure and cerebrospinal fluid leak after translabyrinthine surgery. Am J Otol 4:126-128, 1982. 43. Fayad JN, Schwartz MS, Slattery WH, et al: Prevention and treatment of cerebrospinal fluid leak after translabyrinthine acoustic tumor removal. Otol Neurotol 28:387-390, 2007. 44. Langman AW, Lindeman RC : Surgery for vertigo in the nonserviceable hearing ear: Transmastoid labyrinthectomy or translabyrinthine vestibular nerve section. Laryngoscope 103:1321-1325, 1993. 45. Dew L A, Shelton C : Iatrogenic facial nerve injury: Prevalence and predisposing factors. Ear Nose Throat J 75:724-729, 1996. 46. Fishman A J, Marrinan M S, Golfinos JG, et al: Prevention and management of cerebrospinal fluid leak following vestibular schwannoma surgery. Laryngoscope 114:501-505, 2004. 47. Nelson RA: Translabyrinthine vestibular neurectomy. In Brackmann DE, Shelton C, Arriaga MA (eds): Otologic Surgery, 2nd ed Philadelphia, Saunders, 2001, pp 433-439.
38
Posterior Semicircular Canal Occlusion for Benign Paroxysmal Positional Vertigo Lorne S. Parnes Videos corresponding to this chapter are available online at www.expertconsult.com.
Benign paroxysmal positional vertigo (BPPV) is the most common vestibular end organ disorder; in one busy vestibular clinic, BPPV accounted for 17% of all diagnoses.1 Patients complain of brief vertigo spells, often accompanied by nausea, but rarely vomiting. The actual duration of the spells (5 to 15 seconds) is usually much shorter than what the patients describe. Spells are induced by characteristic head movements, such as rolling to the affected side while in bed or extending the neck while upright. Less common precipitating movements include bending forward, arising from a supine position, and rotating the head. When the disease is very active, in addition to the brief positional vertigo episodes, patients may complain of protracted, nonspecific imbalance and dizziness accompanied by mild lassitude. BPPV is most often an idiopathic disorder. The most common identifiable cause is head or temporal bone trauma.2 Other, less common causes include viral labyrinthitis, vestibular neuronitis, stapedectomy, perilymph fistula, Meniere’s disease, and chronic otitis media.3-8 Various combinations of rotatory, vertical, and oblique nystagmus may be seen in response to the Dix-Hallpike maneuver, depending on the position of the globe within the orbit during nystagmus. The nystagmus profile correlates, however, with the known neuromuscular pathways arising from the crista of the undermost posterior canal.9-12 The classic diagnostic maneuver as described by Dix and Hallpike13 has the patient laying back from a sitting to a head-hanging position. Before lying the patient back, the head is turned 45 degrees toward the side being tested such that when lying back, that side is directed down toward the floor. In some patients, it may be impossible to extend the neck back into a “head-hanging” position because of arthritis, kyphoscoliosis, or vascular disease. For pure diagnostic testing purposes, the lack of hyperextension should not preclude a positive diagnostic test, but it becomes a factor later on when attempting the therapeutic repositioning maneuver (see later). The Dix-Hallpike maneuver serves to rotate the undermost posterior semicircular canal (SCC) in the earth’s vertical plane. This rotation results in an endolymph current
and secondary cupular displacement, which produces the characteristic oculomotor response, primarily a rotatory nystagmus. From the examiner’s viewpoint, this nystagmus has its fast phase beating clockwise with the left ear down and counterclockwise with the right ear down. There is a brief latent period (usually 2 to 5 seconds, but as long as 10 seconds or rarely longer) between the time the patient assumes the head-hanging position and the onset of the nystagmus. Along with the nystagmus, the patient complains of an accompanying vertigo, but rarely nausea. The vertigo and nystagmus briefly crescendo, plateau, and then gradually decrescendo with a typical limited total duration of 10 to 20 seconds. When the nystagmus stops, the patient is returned to the sitting position, where after a short latent period, a milder reverse-direction nystagmus occurs with a less intense sensation of vertigo. Fatigability occurs whereby the nystagmus and vertigo responses decrease in intensity and duration with each repeated maneuver at the same sitting. Because standard electronystagmography or videonystagmography does not record rotational eye movements, normal electronystagmography or videonystagmography positional testing should not preclude the diagnosis of BPPV. The necessity for direct visualization of the eyes during the Dix-Hallpike maneuver cannot be overstated. The examiner can use the naked eye or video goggles. BPPV must be differentiated from other causes of vertigo and nystagmus. The history is usually typical, and a positive response to the Dix-Hallpike maneuver is virtually diagnostic, assuming that all features are present. Typical posterior canal BPPV, which produces a positive response to the Dix-Hallpike maneuver, must be differentiated from the much less common lateral canal BPPV14 and the extremely rare anterior canal BPPV. Lateral canal BPPV gives rise to more severe and prolonged vertigo spells. It is brought on by head movements that produce gravitational forces on the affected lateral canal and is diagnosed by rolling the patient from one lateral supine position to the other and observing the eyes for horizontal nystagmus. Vertigo usually resolves within days to weeks of onset, but similar to the classic posterior canal variant, 467
468
OTOLOGIC SURGERY
it can also be treated with various repositioning maneuvers.15 The remainder of this discussion deals solely with the management of the classic posterior canal BPPV. Untreated posterior canal BPPV has three clinical courses. Most commonly, patients experience a selflimited single episode that subsides spontaneously over weeks to months. A second group of patients experiences remissions and recurrences ranging from weeks to years. A third, smaller group has the more chronic form of this disorder. In one busy vestibular clinic, about 30% of untreated patients had symptoms lasting longer than 1 year.8
PATHOPHYSIOLOGY Under normal physiologic conditions, the cupula has the same density as the surrounding endolymph. The SCCs are normally not sensitive to linear acceleration (e.g., gravity). A fixed cupular deposit would render the posterior canal crista sensitive to gravity, however.16 Rotation of the canal in the earth’s vertical plane during the DixHallpike maneuver would produce cupular displacement through the gravitational pull on the deposit, resulting in nystagmus and vertigo. This condition, so-called cupulolithiasis, may represent the extremely rare, more chronic form of this disorder. Cupulolithiasis also seems to be more prevalent in lateral canal BPPV.15 The more common, self-limited variant and the variant with remissions and recurrences likely have a different pathophysiologic mechanism. Free-floating endolymph particles within the posterior canal produce a Dix-Hallpike response identical to that of a fixed cupular deposit.17 Because the posterior canal is the most gravitydependent part of the vestibular labyrinth, free-floating endolymph particles have a predilection for settling in the posterior canal endolymph. With the head upright, the most dependent part of the canal is the area just posterior and inferior to the ampulla on the side of the cupula opposite the utricle. As the posterior canal rotates during the Dix-Hallpike maneuver, the particles initially rotate upward because of their inertia. After a short latent period, gravity pulls them down and away from the cupula (utriculofugal) to a more dependent position. Their hydrodynamic drag creates an endolymph current in the same direction, displacing the cupula away from the utricle. Utriculofugal displacement of the posterior canal cupula increases the resting discharge rate of the hair cells and first-order vestibular neurons. As known from previous animal studies, this action produces excitation of the ipsilateral superior oblique and contralateral inferior rectus muscles,9 which causes counterclockwise eye rotation with stimulation of the left posterior crista and clockwise rotation with right-sided stimulation. The compensatory fast component of the induced nystagmus is in the opposite direction, however, corresponding with the clinical findings of typical BPPV.
These free-floating posterior canal particles have been identified in vivo in patients undergoing surgery for BPPV.18,19 This theory of free-floating particles, also referred to as canalithiasis, is an important concept because it relates to the treatment of this condition.
PREOPERATIVE PATIENT COUNSELING AND CONSERVATIVE MANAGEMENT First, the patient must be reassured that BPPV is an inner ear disorder that is relatively benign and most often selflimited. Medical management for BPPV is mostly ineffective.20 The most efficacious means of vertigo control is avoidance of the specific provocative head movements that induce the attacks. Most patients already use this approach by not lying on the affected side and by not extending the neck to look upward. Patients who stringently avoid these movements may have more prolonged courses because the absence of provocative movements prevents movement within and subsequent dispersement of the particles from the canal. Most cases of BPPV resolve spontaneously over weeks to months without any treatment. Brandt and Daroff21 recommended a rigorous course of physiotherapy under heavy sedation during several days of hospitalization. They believed that the exercises shook free the otolithic debris from the cupula. Other clinicians could not reproduce their good results. Semont and associates22 reported excellent results using a technique termed the liberatory maneuver. They theorized that this technique liberated deposits from the cupula and reported a 92% success rate following two maneuvers. The liberatory maneuver is difficult to perform in elderly, frail patients. The particle repositioning maneuver23-25 provides the same benefits as the liberatory maneuver, but is simpler to accomplish. It is based on the free-floating particle (canalithiasis) pathophysiologic theory of BPPV and is adapted from Epley’s canalith repositioning procedure.26 For the purpose of this discussion, it is important to remember that the cupula forms a complete barrier across the ampullated end of the canal that is impermeable to endolymph and free-floating particles. Free-floating posterior canal endolymph particles can enter and exit the canal only through the common crus. The current particle repositioning technique (Fig. 38-1) begins with the patient seated lengthwise on the examining table. The first part of the Dix-Hallpike maneuver is performed by rotation of the posterior SCC of the undermost (affected) ear in the earth’s vertical axis (see Fig. 38-1B). The examiner should observe the classic nystagmus response, which confirms the diagnosis, and then reassure the patient as the vertigo subsides. The patient maintains this position for 30 to 60 seconds after resolution of the nystagmus, allowing the particles to settle in their new dependent position closer to the
Chapter 38 • Posterior Semicircular Canal Occlusion for Benign Paroxysmal Positional Vertigo
469
FIGURE 38-1. Particle-repositioning maneuver, four positions, right ear. Schematic representation of patient and concurrent movement of labyrinth, specifically posterior and superior semicircular canals. In each position, dark oval represents the new position of the particle conglomerate in the most dependent part of posterior canal, and open oval represents the previous position. A, Patient seated. B, Patient in Dix-Hallpike head position. Particles gravitate in ampullofugal direction, causing counterclockwise rotatory nystagmus (right ear). Position is maintained for 2 to 3 minutes. C, Midposition. D, Final position of second stage of maneuver performed in one steady continuous motion. Particles continue gravitating in ampullofugal direction through common crus into utricle. Eyes are observed for nystagmus response. Position is maintained for 1 to 2 minutes, then patient sits up. D, direction of view of labyrinth.
c ommon crus. In the second stage, the patient rolls laterally through position C into position D onto the opposite side with the head turned 45 degrees downward. This stage is performed in a smooth, continuous motion, and the neck is kept extended throughout. This method rotates the affected posterior canal 180 degrees in the plane of gravity, allowing the free-floating particles to follow the natural curve of the canal and continue their relative course through the common crus into the utricle, presumably from whence they came. While the examiner supports the patient’s head in position D, a secondary nystagmus response is usually noted, again following a short latent period. A nystagmus response that replicates the initial nystagmus during the Dix-Hallpike maneuver can result only from further passage of the particles in the same ampullofugal direction. Such passage would lead the particles through the common crus into the utricle, where they would no longer induce a pathologic response. Conversely, a secondary nystagmus that reverses direction from that of the initial head-hanging position (position B) may occur through two possible mechanisms. In one, the particles reverse their direction of movement because of an improperly performed maneuver, resulting in a utriculopetal endolymph current. This
reversal usually occurs when the neck is not hyperextended enough during the roll. Cupulolithiasis is the other possible mechanism underlying reversal nystagmus. The gravitational effect on a fixed cupular deposit results in utriculofugal cupular deflection during the DixHallpike maneuver (see Fig. 38-1B), as would be seen with free-floating particles. The position assumed during the second stage of the particle-repositioning maneuver effectively rotates the posterior canal 180 degrees in the earth’s vertical plane (see Fig. 38-1D). This action serves to flip the whole posterior canal and its cupula upside down, reversing the gravitational pull on the cupula and resulting in utriculopetal cupular displacement and a reversal of the nystagmus response. After another 30 to 60 seconds, the patient is brought back up to the sitting position, and the eyes are observed for nystagmus. With a successful maneuver, there should be no nystagmus when the patient returns to the upright position because the particles would have been removed from the posterior canal. This result is in contrast with that of the conventional Dix-Hallpike maneuver, in which one notes a reversal of the nystagmus when the patient sits back up. No specific postmaneuver instructions to limit head and neck movements are necessary.15 Patients should be reassessed 1 to 4 weeks after the maneuver,
470
OTOLOGIC SURGERY
and if the Dix-Hallpike test is still positive, the repositioning maneuver should be repeated. When appropriately administered, repositioning maneuvers should alleviate BPPV in greater than 95% of cases.27 Several other clinical findings help support the freefloating particle theory as the mechanism underlying most cases of BPPV. In some patients with classic histories of BPPV, an initial Dix-Hallpike maneuver often fails to induce a positive response. In many of these patients, a vigorous head shake or application of a skull oscillator may often elicit a latent response. In these instances, it is postulated that the vibration overcomes the particle conglomerate’s inertia or its minor adherence to the membranous canal wall, and allows for its mobilization. Free-floating particles may also explain the fatigability of a conventional Dix-Hallpike maneuver. Each maneuver likely causes increased endolymph dispersion of the particle conglomerate, resulting in a smaller mass effect with each subsequent maneuver and reducing the degree of hydrodynamic drag and endolymph current. In several patients, a repeat Dix-Hallpike maneuver 30 to 60 minutes after a fatigued response often elicits the same maximal response seen during the initial Dix-Hallpike maneuver. Theoretically, the particles have had time to reassemble into the large conglomerate mass within the endolymph of the posterior canal.
SURGICAL PATIENT SELECTION Operative intervention is offered for intractable cases in which symptoms are severe enough to affect the patient’s occupation or lifestyle significantly, and when the disorder fails to respond to repeated repositioning maneuvers. Rarely, surgery is offered to patients with severe and frequent recurrences. Until the advent of posterior canal occlusion, transection of the posterior ampullary nerve (singular neurectomy) was the gold standard of operative treatment.28-30 Singular neurectomy is technically difficult, is performed by very few surgeons, and yields variable rates of failure and sensorineural hearing loss.31,32 Ohmichi and colleagues33 showed that the singular nerve would be inaccessible through a tympanotomy approach in 14% of human temporal bones. In a 2007 literature review and temporal bone dissection study comparing posterior canal occlusion with transection of the posterior ampullary nerve, Leveque and associates34 found that posterior canal occlusion presented far fewer technical issues and a far lower risk of hearing loss compared with singular neurectomy. Posterior SCC occlusion evolved to circumvent the technical challenges and high risk of hearing loss associated with transection of the posterior ampullary nerve. Money and Scott35 initially used this technique in feline vestibular physiology experiments. Plugging individual SCCs blocked their receptivity to angular acceleration without influencing the responses of the other ipsilateral
vestibular receptors. Although posterior canal occlusion was at first a theoretical remedy for BPPV, the main concern in applying this technique to humans was its possible detrimental effect on hearing. This problem was not addressed in the original cat studies. Parnes and McClure36 carried out a study in guinea pigs to measure the effect of canal occlusion on hearing using brainstem auditory evoked responses. The hearing responses remained unchanged during follow-up periods lasting 6 months. The hypothesis that canal occlusion would abolish BPPV was tested when two patients presented with intractable BPPV in ears with coexisting profound sensorineural hearing losses. With no hearing to lose, both patients agreed to undergo what at that time was an experimental procedure. Both patients were relieved of their BPPV and remained symptom-free for at least 4 years, when they were lost to follow-up.37,38 In addition, both patients maintained postoperative lateral SCC function as measured by caloric responses. This important finding indicated that the procedure’s success resulted from the isolated defunctioning of the posterior canal and not from a generalized destructive process of the vestibular labyrinth. The theoretical intent of the procedure is to compress the membranous labyrinth closed against the opposite bony wall, creating a closed, fluid-filled (endolymph) space between the plug and cupula, both of which are impermeable to endolymph. Because fluid cannot expand or compress without a change in temperature, this action eliminates all endolymph movement within the posterior canal, effectively fixing the cupula. The canal’s end organ receptor no longer responds to free-floating endolymph particles even if they remain in the partitioned posterior canal (canalithiasis); it likewise does not respond to a fixed cupular deposit (cupulolithiasis), permanently eliminating the BPPV. In addition, the canal no longer responds to physiologic angular acceleration. Because the deficit is constant and permanent, however, gradual compensation occurs through central adaptation, while the corresponding contralateral anterior SCC assumes control of response detection in that particular plane. Singular neurectomy eliminates the resting discharge from the posterior SCC crista, creating a static vestibular asymmetry39 between the two posterior canals. This effect results in immediate postoperative vertigo at rest and spontaneous rotatory nystagmus. Posterior canal occlusion does not disturb the resting neuronal discharge from the occluded canal, however; most patients do not have spontaneous postoperative vertigo or nystagmus at rest, unless complicated by other factors. Singular neurectomy and posterior canal occlusion result in a dynamic vestibular asymmetry,39 itself resulting in motion sensitivity. The dynamic asymmetry gradually resolves with central adaptation, which can be hastened with vestibular physiotherapy.
Chapter 38 • Posterior Semicircular Canal Occlusion for Benign Paroxysmal Positional Vertigo
PREOPERATIVE EVALUATION Preoperative evaluation includes a routine audiogram. The procedure is not recommended in an only or significantly better hearing ear. Preoperative electronystagmography or videonystagmography helps to ensure there is a normal vestibular response from the contralateral ear. A high-resolution computed tomography (CT) scan of the temporal bone defines the anatomy and ensures that the posterior canal is accessible through a transmastoid approach. Preoperative imaging with either CT or magnetic resonance imaging (MRI) rules out central lesions that may mimic BPPV, a very rare occurrence.40
SURGICAL TECHNIQUE The use of perioperative broad-spectrum antibiotic coverage is debatable, but antibiotics should be used in ears with a past history of otitis media. The procedure is contraindicated during acute or subacute episodes of otitis media. The procedure is also relatively contraindicated in much better hearing or vestibular functioning ears. Under general anesthesia, the patient is placed in the supine position with the head turned 45 degrees toward the opposite side. The surgical site preparation and draping are performed in a routine fashion. Intraoperative facial nerve monitoring is standard as in all mastoid surgery, but auditory nerve monitoring is not required. A limited mastoidectomy with a fluted ball drill and suction-irrigation is performed through a standard postauricular incision (Fig. 38-2A). The antrum is opened, providing exposure of the lateral SCC. In most instances, identification of the tegmen and digastric ridge is unnecessary. The bony sigmoid sinus prominence is identified, but in most cases it is unnecessary to bare the sigmoid sinus dura. Bone removal proceeds anteriorly from the sigmoid sinus along the cerebellar plate toward the posterior canal. When the posterior canal otic capsule is identified, the bone is blue-lined (thinned down so much that the transmitted light through the canal’s perilymph creates a dark appearance) with progressively smaller diamond burrs and copious suction irrigation. The target zone for the occlusion is the area at, or just inferior to, a line extending posteriorly from the lateral SCC (Donaldson’s line). Sometimes there are abundant air cells between the posterior canal otic capsule bone and the dura, and sometimes the bony canal directly contacts the dura. The preoperative CT scan helps with navigation through this area. After blue-lining the canal, a 1 mm diamond burr is used to skeletonize a 3 mm segment of the canal 180 degrees around the outer circumference down to endosteum, creating a 1 × 3 mm endosteal island (Fig. 38-2B). Bone removal should proceed evenly along the circumference so that when the endosteum is violated and perilymph is exposed, all drilling can cease. The endosteal
471
island is removed with a fine 90 degree pick to expose the perilymph (Fig. 38-3). Great care must be taken not to suction directly on the perilymph and especially the membranous labyrinth. At this stage, the exact outline and limits of the membranous labyrinth are usually not clearly discernible. Although not essential, perilymph may be gently “wicked” away with a Cottonoid, or gently suctioned with a No. 24 suction with the thumb off the hole to expose the membranous labyrinth, at which time the membranous duct is seen to collapse. Various materials have been used to occlude the canals, including bone dust alone, bone wax, fascia or other tissue, and bone pâté made from either bone dust and blood or bone dust and fibrin glue. I use dry bone chips gathered from the mastoidectomy mixed with a couple of drops of two-component, fast-acting human fibrinogen glue. Once set (about 30 seconds), it forms an easily workable, but malleable plug with a firm consistency (Fig. 38-4). The plug is gently and firmly inserted through the fenestra with the intention of completely filling the canal lumen and compressing the membranous labyrinth closed (Fig. 38-5). A blunt 45 degree probe works well to pack the plug in the canal tightly. The membranous labyrinth is resistant to shearing and tearing. As has been learned from partial labyrinthectomies in skull base surgery (see next section), controlled resection of the membranous duct is not risky or dangerous to the other inner ear receptors (i.e., hearing), unless the membranous labyrinth is roughly avulsed. The bone chips within the plug lead to intracanal ossification resulting in complete permanent occlusion of the canal. After completing the plug insertion, the fenestra and surrounding bone are covered with a piece of temporalis fascia, which is kept in place by several more drops of fibrinogen glue (Fig. 38-6). A good tissue seal is necessary to prevent a postoperative perilymph fistula. The incision is closed in two layers; a standard mastoid dressing may be applied. A drain is unnecessary. Patients are admitted to the hospital typically for 1 to 3 days, mostly because of the postoperative dysequilibrium and motion sensitivity, the recovery of which can be hastened with vestibular physiotherapy. In variations of this technique, Anthony41 successfully treated BPPV by applying an HGM argon laser to the blue-lined posterior canal. The laser burns were purported to create fibrous bands within the canal, leading to obstruction of the membranous duct. Kartush and Sargent42 used a CO2 laser–assisted occlusion technique.
RESULTS To date, 97 posterior canal occlusions have been reported in the literature. Many more have been done and gone unreported, such as in one busy Australian clinic that has successfully done more than 70 cases (Pohl DV, personal communication). To date 2008, I have performed
472
OTOLOGIC SURGERY
Chapter 38 • Posterior Semicircular Canal Occlusion for Benign Paroxysmal Positional Vertigo
473
FIGURE 38-2. A, Exposing right posterior semicircular canal otic capsule. B, Creating the 1 × 3 mm endosteal island with small diamond burr. C, Cross-sectional view. FIGURE 38-3. A, Lifting out endosteal island with a fine 90 degree pick. B, Magnified lateral view. FIGURE 38-4. Creating plug with two-component fibrinogen glue and mastoid cortex bone chips. FIGURE 38-5. A, Tamping plug through fenestra into canal. B, Cross section of canal shows intact but occluded membranous canal. FIGURE 38-6. Covering fenestra and surrounding bone with fascia and glue.
55 posterior canal occlusions in 52 patients; 3 patients underwent sequential bilateral occlusions. The average age at surgery was 61 years (range 25 to 83 years). There were 43 women and 12 men, and 34 left ears and 21 right ears. The duration of symptoms ranged from 1 to 35 years, with a mean of 6.8 years and median of 3 years. The follow-up times range from 1 to 20 years. In approximately 30% of ears, after the canal has been fenestrated, and the membranous labyrinth has been exposed, particles can be seen freely floating within the endolymph.19 In one previously reported case, the membranous labyrinth was resected with the particles and was subjected to scanning electron microscopy. In this particular case, the particles proved to be degenerating otoconia.19 All 55 ears remain completely free of BPPV. Four ears had profound preoperative deafness. Of the 51 ears with normal preoperative hearing, most had an initial postoperative mixed hearing loss that in all cases recovered to the preoperative level. One patient had a delayed sudden hearing loss accompanied by a 7 day vertigo spell during the 4th postoperative month. The symptoms were preceded by an intense 2 day headache. This patient did not return for follow-up until 1 month later, at which time her audiogram showed a 70 dB sensorineural hearing loss with only 32% word recognition. This hearing loss persisted on subsequent audiograms. The working diagnosis was labyrinthitis, but the exact causal relationship to the canal occlusion was unknown. The patient had undergone two previous unsuccessful attempts at singular neurectomy in the same ear. One other patient had a 20 dB decrease in bone conduction levels at her 1 year follow-up audiogram and no change in word recog nition score. All 55 cases were followed by an initial 1 to 4 week period of imbalance, giddiness, and motion sensitivity. Six cases had more protracted courses, and five were thought to have had labyrinthitis. Four of these were thought to have had aseptic labyrinthitis; the other was thought to have postoperative otitis media/mastoiditis. This patient had a history of chronic otitis media and had undergone a prior cortical mastoidectomy; her sensorineural hearing loss recovered after antibiotic therapy. The sixth patient developed BPPV in the contralateral ear, but there also seemed to be some functional overlay. Four other cases warrant brief mention. As noted previously, three patients underwent sequential bilateral posterior canal occlusions, ranging from 1 to 8 years between
procedures All three patients remain symptom-free with no other adverse effects from the surgery. As expected, the intact contralateral superior SCC provided the complementary vestibular input of the posterior canals that were rendered nonfunctional by surgical occlusion. The other notable patient had her canal occluded under local anesthesia because of multiple other medical problems. This patient had no intraoperative complaints of vertigo or dizziness. The postoperative hospitalization was 1 to 5 days in uncomplicated cases. As expected, older patients had longer recovery periods and tended to require longer hospital stays. The degree of postoperative motion sensitivity usually determined the length of stay. Although postoperative vestibular suppressants may provide early shortterm relief, their use was discouraged because they tend to prolong the overall recovery. All patients were provided with early postoperative vestibular physiotherapy.
FURTHER APPLICATIONS OF CANAL OCCLUSION SURGERY Several modifications of posterior canal occlusion have ensued for various other inner ear and skull base conditions. There are only two reported cases of canal occlusion for intractable BPPV of the other canals: one of a single lateral canal occlusion using an argon laser on a blue-lined canal,43 and another case report of an anterior canal plugging.44 McElveen and colleagues45 described three patients who each underwent a successful hearingpreserving modified translabyrinthine acoustic neuroma resection in which all three SCCs were occluded. Molony and associates46 reported two patients who underwent extended middle fossa approaches and one who underwent an extended suboccipital approach for acoustic tumor removal. Hearing was preserved in all three patients after occluding and resecting the superior SCC and posterior SCC. Hirsch and colleagues47 first reported on the partial labyrinthectomy, in which the superior and posterior canals of the same labyrinth were occluded and resected for access to various skull base lesions, with consistent hearing preservation. Minor and coworkers48 published their seminal work on the superior SCC dehiscence syndrome in 1998, in which several patients underwent superior SCC occlusion through the middle fossa approach. Several subsequent reports have ensued from their center
474
OTOLOGIC SURGERY
and other centers showing good hearing preservation rates. Further positive reports on partial labyrinthectomy by Sekhar and associates49 and Kaylie and colleagues50 substantiated the utility and success of labyrinthine partition with multiple canal occlusions, with predictably good hearing preservation rates. Dehiscence of the posterior canal can also occur (rarely), and it too can be treated successfully with canal occlusion.51 Agrawal and Parnes52 reported their successful results using the transmastoid approach for superior SCC occlusion in three cases of superior SCC dehiscence. In one patient, hearing remained unchanged, whereas in the other two patients hearing improved with the resolution of the conductive component of the hearing loss.
SUMMARY Most typical BPPV cases seem to result from free-floating posterior SCC endolymph particles. Most of these patients may be cured with the particle-repositioning maneuver. Intractable cases are rare. For this small subset of patients, posterior SCC occlusion is a safe, curative procedure. Occlusion of all three ipsilateral SCCs alone or in combination can be safely performed for various other disorders, with predictably good hearing results.
REFERENCES 1. Nedzelski J M, Barber HO, McIlmoyl L : Diagnoses in a dizziness unit. J Otolaryngol 15:101-104, 1986. 2. Barber HO, Leigh R J: Benign (and not so benign) postural vertigo: Diagnosis and treatment. In Barber HO, Sharpe A (eds): Vestibular Disorders. Boca Raton, FL, CRC Press, 1988, pp 215-232. 3. Lindsay J R , Hemenway WG: Postural vertigo due to unilateral sudden partial loss of vestibular function. Ann Otol Rhinol Laryngol 65:692-706, 1956. 4. Spector M : Positional vertigo after stapedectomy. Ann Otol Rhinol Laryngol 70:251-254, 1961. 5. Stahle J, Terins J: Paroxysmal positional nystagmus. Ann Otol Rhinol Laryngol 74:69-83, 1965. 6. Barber HO: Positional vertigo and nystagmus. Otolaryngol Clin North Am 6:169-187, 1973. 7. McClure J A, Rounthwaite J: Vestibular dysfunction associated with benign paroxysmal vertigo. Laryngoscope 87:1434-1442, 1977. 8. Baloh RW, Honrubia V, Jacobson K : Benign positional vertigo: Clinical and oculographic features in 240 cases. Neurology 37:371-378, 1987. 9. Cohen B, Suzuki J, Bender M B : Nystagmus induced by electrical stimulation of ampullary nerves. Acta Otolaryngol (Stockh) 60:422-436, 1965. 10. Harbert F: Benign paroxysmal positional vertigo. Arch Ophthalmol Head Neck Surg 84:298-302, 1970. 11. Baloh RW, Sakala S, Honrubia V: The mechanism of benign paroxysmal positional nystagmus. Adv Otorhinolaryngol 25:161-166, 1979.
12. Katsarkas A, Outerbridge J S : Nystagmus of paroxysmal positional vertigo. Ann Otol Rhinol Laryngol 92: 146-150, 1983. 13. Dix M R , Hallpike C S : Pathology, symptomatology and diagnosis of certain disorders of the vestibular system. Proc R Soc Med 45:341, 1952. 14. Fife TD: Recognition and management of horizontal canal benign positional vertigo. Am J Otol 19:345-351, 1998. 15. Parnes L S, Agrawal S K, Atlas J: Diagnosis and management of benign paroxysmal positional vertigo (BPPV). Can Med Assoc J 169:681-693, 2003. 16. Schuknecht H F: Pathology of the Ear. Cambridge, MA, Harvard University Press, 1974. 17. Hall S F, Ruby R R F, McClure J A : The mechanics of benign paroxysmal vertigo. J Otolaryngol 8:151-158, 1979. 18. Parnes L S, McClure J A : Free-floating endolymph particles: A new operative finding during posterior semicircular canal occlusion. Laryngoscope 12:988-992, 1992. 19. Welling D P, Parnes L S, O’Brien B, et al: Particulate matter in the posterior semicircular canal. Laryngoscope 107:90-94, 1997. 20. McClure J A, Willett J M : Lorazepam and diazepam in the treatment of benign paroxysmal vertigo. J Otolaryngol 9:472-477, 1980. 21. Brandt T, Daroff R B : Physical therapy for benign paroxysmal positional vertigo. Arch Otolaryngol Head Neck Surg 106:484-485, 1980. 22. Semont A, Freyss G, Vitte E : Curing the BPPV with a liberatory maneuver. Adv Otorhinolaryngol 42:290-293, 1988. 23. Parnes L S, Price-Jones G: Particle-repositioning maneuver for benign paroxysmal positional vertigo. Ann Otol Rhinol Laryngol 102:325-331, 1993. 24. Parnes L S, Robichaud J: Further observations during the particle repositioning maneuver for benign paroxysmal positional vertigo. Otolaryngol Head Neck Surg 116:238-243, 1997. 25. Fung K, Hall S F: Particle-repositioning maneuver: Effective treatment for benign paroxysmal positional vertigo. J Otolaryngol 25:243-248, 1996. 26. Epley J M : The canalith-repositioning procedure: For treatment of benign paroxysmal positional vertigo. Otolaryngol Head Neck Surg 107:399-404, 1992. 27. Epley J M : Positional vertigo related to semicircular canalithiasis. Otolaryngol Head Neck Surg 112:154-161, 1995. 28. Gacek R : Transection of the posterior ampullary nerve for the relief of benign paroxysmal positional nystagmus. Ann Otol Rhinol Laryngol 83:596-605, 1974. 29. Epley J M : Singular neurectomy: Hypotympanotomy approach. Otolaryngol Head Neck Surg 88:304-309, 1980. 30. Gacek R : Singular neurectomy update. Ann Otol Rhinol Laryngol 91:469-473, 1982. 31. Silverstein H : Singular neurectomy: A treatment for benign positional vertigo. In Brackmann D E (ed): Neurological Surgery of the Ear and Skull Base. New York, Raven Press, 1982, pp 331-335. 32. Meyerhoff WL : Surgical section of the posterior ampullary nerve. Laryngoscope 95:933-935, 1985. 33. Ohmichi T, Rutka J, Hawke M : Histopathologic consequences of surgical approaches to the singular nerve. Laryngoscope 99:963-970, 1989.
Chapter 38 • Posterior Semicircular Canal Occlusion for Benign Paroxysmal Positional Vertigo 34. Leveque M, Labrousse M, Seidermann L , Chays A : Surgical therapy in intractable benign paroxysmal positional vertigo. Otolaryngol Head Neck Surg 136:693-698, 2007. 35. Money K E, Scott JW: Functions of separate sensory receptors of nonauditory labyrinth of the cat. Am J Physiol 202:1211-1220, 1962. 36. Parnes L S, McClure J A : Effect on brainstem auditory evoked responses of posterior semicircular canal occlusion in guinea pigs. J Otolaryngol 14:145-150, 1985. 37. Parnes L S, McClure J A : Posterior semicircular canal occlusion for intractable benign paroxysmal positional vertigo. Ann Otol Rhinol Laryngol 99:330-334, 1990. 38. Parnes L S, McClure J A : Posterior semicircular canal occlusion in the normal hearing ear. Otolaryngol Head Neck Surg 104:52-57, 1991. 39. McClure J A, Lycett P: Vestibular asymmetry. Arch Otolaryngol Head Neck Surg 109:682-687, 1983. 40. Shoman N, Longridge N: Cerebellar vermis lesions and tumours of the fourth ventricle in patients with positional and positioning vertigo and nystagmus. J Laryngol Otol 121:166-169, 2007. 41. Anthony P: Partitioning the labyrinth: Application in benign paroxysmal positional vertigo. Am J Otol 12:388393, 1991. 42. Kartush J M, Sargent EW: Posterior semicircular canal occlusion for benign paroxysmal positional vertigo-CO2 laser-assisted technique: Preliminary results. Laryngoscope 105:268-274, 1995. 43. Nomura Y: Argon laser irradiation of the semicircular canal in two patients with benign paroxysmal positional vertigo. J Laryngol Otol 116:723-725, 2002.
475
44. Brantberg K, Bergenius J: Treatment of anterior benign paroxysmal positional vertigo by canal plugging: A case report. Acta Otolaryngol 122:28-30, 2002. 45. McElveen JT Jr, Wilkins R H, Erwin AC, Wolford R D: Modifying the translabyrinthine approach to preserve hearing during acoustic tumour surgery. J Laryngol Otol 105:34-37, 1991. 46. Molony TB, Kwartler J A, House WF, Hitselberger WE : Extended middle fossa and retrolabyrinthine approaches in acoustic neuroma surgery: Case reports. Am J Otol 13:360-363, 1992. 47. Hirsch B E, Cass S P, Sekhar L N, Wright DC : Translabyrinthine approach to skull base tumors with hearing preservation. Am J Otol 14:533-543, 1993. 48. Minor L B, Solomon D, Zinreich J S, Zee DS : Soundand/or pressure-induced vertigo due to bone dehiscence of the superior semicircular canal. Arch Otolaryngol Head Neck Surg 124:249-258, 1998. 49. Sekhar L N, Schessel D A, Bucur S D, et al: Partial labyrinthectomy petrous apicectomy approach to neoplastic and vascular lesions of the petroclival area. Neurosurgery 44:537-550, 1999. 50. Kaylie D M, Horgan M A, Delashaw J B, McMenomey SO: Hearing preservation with the transcrural approach to the petroclival region. Otol Neurotol 25:594-598, 2004. 51. Mikulec A A, Poe DS : Operative management of a posterior semicircular canal dehiscence. Laryngoscope 116:375-378, 2006. 52. Agrawal S K, Parnes L S : Transmastoid superior semicircular canal occlusion. Otol Neurotol 29:363-367, 2008.
39
Cochleosacculotomy Harold F. Schuknecht,* Michael J. McKenna, and K. Paul Boyev
The surgical treatment for Meniere’s disease can be classified into two groups, according to mode of action: (1) procedures that have the objective of total or partial ablation of vestibular function, and (2) procedures that are intended to enhance the drainage of endolymph by fistulization of the membranous labyrinth and decompression of the endolymphatic sac. Endolymphatic drainage procedures can be divided further into external shunts that attempt to drain excessive endolymph from the endolymphatic sac into the mastoid or subarachnoid space, and internal shunts that attempt to drain excessive endolymph into the perilymphatic space. The cochleosacculotomy operation falls into the latter group—an internal shunt procedure.
PHYSIOLOGIC, ANATOMIC, AND PATHOLOGIC RATIONALE Meniere’s disease is characterized pathologically by progressive endolymphatic hydrops that is probably related to a disturbance in endolymphatic sac function. This condition must be differentiated from nonprogressive endolymphatic hydrops, in which the hydrops is the result of a single traumatic or inflammatory insult to the labyrinth, causing a permanent but not progressive endolymphatic hydrops.1 The symptoms of progressive endolymphatic hydrops can be correlated with two principal types of pathologic change: (1) distentions and ruptures of the endolymphatic system,2,3 and (2) alterations in the cytoarchitecture of the auditory and vestibular sense organs, sometimes accompanied by atrophic changes. Coincident with rupture, there is sudden contamination of the perilymphatic fluid with neurotoxic endolymph (140 mEq/L of potassium) that causes paralysis of the sensory and neural structures and is expressed clinically as episodic vertigo, fluctuating hearing loss, or both. The American Academy of Otolaryngology–Head and Neck Surgery (AAOHNS)4 recommended that these episodes be designated the “definitive” symptoms of Meniere’s disease.
*Deceased
As the disease progresses, there are changes in the cytoarchitecture of the sense organs that consist of distortion and atrophy of the sensory cells and supporting cells, and disruption and deformation of their gelatinous aprons. These alterations impair the motion mechanics of the sense organs, resulting in permanent functional deficits. The symptoms for the auditory system are hearing loss and tinnitus, and for the vestibular system are constant or recurring sensations of unsteadiness, described as being off-balance, floating, tilting, falling, or spinning, and are often aggravated by head movement. The AAOHNS recommended that they be known as “adjunctive” symptoms. To be successful, surgical procedures based on facilitating drainage of endolymph should alleviate definitive symptoms and arrest the progression of adjunctive symptoms. Internal shunt procedures include the sacculotomy of Fick,5,6 the tack operation of Cody,7,8 the otic-perotic shunt of Pulec and House,9 and the cochleosacculotomy.10 In the sacculotomy and tack procedures, picks are introduced through the footplate of the stapes to puncture the saccule with the hope of producing a permanent fistula in the saccular wall by which excessive endolymph can drain into the perilymphatic space. This approach fails to consider, however, the histopathologic observation that in Meniere’s disease, the distended saccule often fills the vestibule; its distended wall is adherent to the footplate and could not be fistulized into the perilymphatic space by these maneuvers.11 The otic-perotic shunt, as conceived by House and Pulec, involves the placement of a platinum tube through the basilar membrane to connect the scala media and scala tympani; however, the procedure is not surgically feasible because of the small size of the cochlear duct. The cochleosacculotomy operation consists of creating a fracture-disruption by impaling the osseous spiral lamina and cochlear duct with a pick introduced through the round window. The rationale is supported by two histopathologic observations and muted by a third: 1. Histologic study of the temporal bones from patients with Meniere’s disease shows that the distended membranous labyrinth may fistulize permanently in 477
478
OTOLOGIC SURGERY
any area. Such spontaneous fistulization may account for the long remissions and even permanent arrest of symptoms of episodic vertigo and fluctuating hearing loss in some patients. 2. Animal experiments have shown that surgical disruption of Reissner’s membrane12 or the walls of the utricle, saccule, or semicircular canals13 results in prompt healing of the fistulas. It has been shown in experimental studies on cats14-16 and guinea pigs,17 however, that fracture-disruption of the osseous spiral lamina and cochlear duct can sometimes result in a permanent communication between the endolymphatic and perilymphatic spaces. Those experiments show that such fistulas exist without impairing the hearing for frequencies other than those tonotopically located immediately adjacent to the fistulas. 3. A histopathologic finding in human temporal bones of patients with Meniere’s disease that militates against the success of internal shunts is that the distended membranes in some cases block the flow of endolymph toward such a fistula.18 For this reason, a successful cochleosacculotomy fistula does not relieve symptoms in all instances.
PATIENT SELECTION Some otolaryngologists believe that surgery is never indicated for the relief of symptoms of Meniere’s disease because in the normal course of the disease, the vertigo eventually subsides. This approach has merit if the patient is not unduly handicapped. In many cases, however, the dysequilibrium erodes occupational efficiency and recreational and family lifestyle to the extent that invasive therapy is justified. This approach applies to patients having frequent and severe vertiginous episodes unrelieved by medication and patients having falling attacks (Tumarkin’s otolithic catastrophe).19 Some general considerations need to be addressed before cochleosacculotomy is recommended. The diagnosis of Meniere’s disease must be unequivocal. Other disorders that cause fluctuating hearing loss and episodic vertigo, such as otosyphilis, inner ear autoimmune disease, perilymph fistula, demyelinating diseases, and intracranial neoplasms, must be ruled out. The operation is not recommended for patients who present with audiovestibular symptoms that are atypical for progressive endolymphatic hydrops. There should be no evidence of involvement of Meniere’s disease or other disorders that might be progressive in the opposite ear. It is prudent to require a duration of symptoms of at least 1 year. The opposite ear becomes involved in almost 50% of cases, sometimes many years after the onset of symptoms in the first ear. Cochleosacculotomy is the procedure of choice for patients who for health reasons are at risk for the stress of postoperative vertigo, and who should not have general anesthesia, and for elderly patients who often compensate
poorly to procedures that ablate vestibular function. It has the advantage of being technically simple to perform, is almost totally free of morbidity, and carries little or no risk of mortality. Labyrinthectomy is more certain than cochleosacculotomy to relieve disabling episodic vertigo; however, it has the disadvantage of producing severe postoperative vertigo, permanent hearing loss, and, in some cases, prolonged dysequilibrium. For these reasons, some patients may choose to have a cochleosacculotomy as a first attempt to resolve their vertigo problem.
SURGICAL TECHNIQUE The ear is prepared and draped similarly to any transcanal procedure. The operation is performed under local anesthesia. With the aid of a nasal speculum, the fibrocartilaginous external auditory canal is infiltrated circumferentially in four positions with 1% lidocaine (Xylocaine) containing 1:100,000 epinephrine, using a 27 gauge, 1.5 inch needle. The ear canal is dilated with the nasal speculum, which also serves to diffuse the anesthetic solution uniformly into the tissues. An ear speculum of the surgeon’s preference is introduced into the ear canal and locked in an adjustable speculum holder, freeing both of the surgeon’s hands. The posterior aspect of the bony canal is exposed, and the skin of the bony canal is infiltrated at two sites with 1% lidocaine containing 1:1000 epinephrine, using a 30 gauge, 1.5 inch needle. The beveled opening of the needle must be introduced flush against the bone at an obtuse angle, and the blanching area of infiltration should extend to the tympanic annulus from the 6 to 12 o’clock position. Adjustments of the speculum and speculum holder are made as often as necessary to achieve optimal exposure of the surgical field. Incisions are made, and a triangular skin flap is elevated to the tympanic annulus (Fig. 39-1A). Bleeding must be meticulously controlled at all times. The tympanomeatal flap is elevated by lifting the tympanic annulus from its sulcus and folding the flap into the anterior tympanomeatal angle of the ear canal (Fig. 39-1B). The chorda tympani nerve and the ossicles are not disturbed. The ear speculum is locked into a position that exposes the round window niche and posterior aspect of the hypotympanum. In rare cases, the round window niche is partly hidden behind the posteroinferior part of the bony tympanic annulus, in which case a small burr is used to remove sufficient bony annulus to allow access to the round window niche. Usually, the round window niche accommodates a 3 mm right angled pick without removal of bone. The pick is advanced through the round window membrane, which may or may not be visible. The pick is guided in the direction of the oval window while hugging the lateral wall of the inner ear to ensure that the cochlear duct is traversed (Fig. 39-1C). When the pick has been introduced to its full 3 mm length, the end of the pick is located beneath
Chapter 39 • Cochleosacculotomy
479
FIGURE 39-1. A, Skin incisions are made in the posterior wall of the bony external auditory canal. B, Tympanomeatal flap is elevated and reflected into the anterior tympanomeatal angle. C, A 3 mm right angle pick is advanced through the round window membrane in the direction of the oval window. D, If niche does not accommodate a 3 mm right angle pick, the bony lip of the round window is removed, and a 2 mm right angle pick is used to accomplish the cochleosacculotomy. RW, round window.
480
OTOLOGIC SURGERY
the footplate of the stapes. Occasionally, the subiculum, which is a ridge of bone lying in the boundary between the round window niche and sinus tympani, interferes with introduction of the pick. It can readily be shaved down with a 2 mm burr. Occasionally, the overhanging
bony lip of the round window niche must be removed to accommodate the pick (Fig. 39-1D). In this case, a 2 mm (rather than a 3 mm) pick is used to avoid excessively deep penetration into the vestibule and possible injury to the utricular macula (Fig. 39-2). Rarely, a high jugular bulb
FIGURE 39-2. Vertical section of normal ear shows anatomic features of middle and inner ears that are relevant to cochleosacculotomy. Histologic sections represent similar anatomic characteristics as the drawing. CD, cochlear duct; Dr, ductus reuniens; RW, round window; S, saccule; U, utricle.
FIGURE 39-3. Vertical section of normal ear shows high jugular bulb abutting tympanic annulus and encroaching on round window niche. Other sections show reduced orifice leading to a small niche. It is not feasible to attempt cochleosacculotomy in such cases. IAC, internal auditory canal; OW, oval window; RW, round window; SS, sigmoid sinus.
Chapter 39 • Cochleosacculotomy
blocks access to the round window niche, in which case it may be necessary to abort the operation (Fig. 39-3). Occasionally, a slight loss of resistance is felt as the pick passes through the cochlear partition and causes the planned fracture-disruption of the osseous spiral lamina and cochlear duct (Fig. 39-4). Usually, patients experience no sensation as the pick is advanced, but a few have noted momentary vertigo, and several have reported hearing a “click.” The maneuver does not produce vertigo, presumably because the vestibular sense organs are not mechanically disturbed, and the endolymph from the fistulized area drains into the scala tympani rather than into the perilymphatic space of the vestibule. The pick is withdrawn, and the perforation in the round window membrane is sealed by a tissue graft of perichondrium or adipose tissue. The operation is terminated by returning the tympanomeatal flap to its original position. Strips of silk cloth are laid over the incisions, and a round synthetic rubber sponge of appropriate size is placed in the canal to maintain slight pressure on the skin flap. Cotton is placed in the ear canal. A few patients note slight unsteadiness for 1 or 2 days; however, all feel well enough to be discharged from the hospital the next day. The packing is removed 1 week later, at which time the tympanic membrane and canal wall skin should be well healed. Serial hearing tests at weekly intervals show a sensorineural hearing loss for 2 to 3 weeks, followed by recovery
481
in most cases (Ginsberg I, personal communication). The authors first test hearing 6 weeks after surgery, when the hearing has stabilized. The complications of cochleosacculotomy are similar to the complications for stapedectomy and include perforation of the tympanic membrane, tears of the jugular bulb, postoperative otitis media, sensorineural hearing loss, facial nerve injury, and perilymph fistula. The only significant complication in cochleosacculotomies performed by the authors, other than sensorineural hearing loss, was one case of otitis media that resulted in profound sensorineural hearing loss in the infected ear.
RESULTS The Massachusetts Eye and Ear Infirmary experience currently consists of 142 cochleosacculotomies performed since April 1979.20-22 Follow-up times average 6.16 years, ranging from 1 month to 7.4 years. Definitive control of vertigo has been achieved in 68.3% of patients. Hearing was made worse in 35.2% (as defined by AAOHNS criteria4 of either 15 dB loss in pure tone average or 15% loss in speech discrimination). Postoperative profound sensorineural hearing loss occurred in 11%. The success rates reported for either internal or external endolymphatic shunt procedures should be viewed
FIGURE 39-4. A and B, A 3 mm right angle pick is shown penetrating dilated cochlear duct and dilated saccule. Dr, ductus reuniens; M, macula; OSL, osseous spiral lamina; RW, round window; S, saccule; ST, scala tympani; SV, scala vestibuli; U, utricle.
482
OTOLOGIC SURGERY
with some caution. In a review of 834 articles published on Meniere’s disease between 1952 and 1975, Torok23 found that almost without exception, advocates of either medical or surgical treatment reported success rates of 60% to 80%. Not only is there a strong placebo effect, but also sudden prolonged remission of symptoms are characteristic of Meniere’s disease. Some patients who had vertiginous attacks on a weekly basis (or more often) had a total cessation of attacks after cochleosacculotomy. In the management of disabling Meniere’s disease, cochleosacculotomy is a useful alternative to vestibular nerve section and labyrinthectomy in selected cases.
REFERENCES 1. Schuknecht H F, Gulya A J: Endolymphatic hydrops: An overview and classification. Ann Otol Rhinol Laryngol 92(Suppl 106):1-20, 1983. 2. Schuknecht H F: Ménière’s disease: A correlation of symptomatology and pathology. Laryngoscope 73:651665, 1963. 3. Dohlman G F: On the mechanism of the Ménière attack. Arch Otorhinolaryngol 212:301-307, 1976. 4. Committee on Hearing and Equilibrium: Report of Subcommittee on Equilibrium and its Measurement: Ménière’s disease: Criteria for diagnosis and evaluation of therapy for reporting. Trans Am Acad Ophthalmol Otolaryngol 76:1462-1464, 1972. 5. Fick I A : van N: Decompression of the labyrinth: A new surgical procedure for Ménière’s disease. Arch Otolaryngol 79:447-458, 1964. 6. Fick I A : van N: Ménière’s disease: Aetiology and a new surgical approach-sacculotomy. J Laryngol Otol 80:288306, 1966. 7. Cody DTR , Simonton K M, Hallberg O E : Automatic repetitive decompression of the saccule in endolymphatic hydrops (tack operation): Preliminary report. Laryngoscope 77:1480-1501, 1967. 8. Cody DTR : The tack operation for endolymphatic hydrops. Laryngoscope 79:1737-1744, 1969.
9. Pulec J L : Ménière’s disease: The otic-perotic shunt. Otolaryngol Clin North Am 1:643-648, 1968. 10. Schuknecht H F: Cochleosacculotomy for Ménière’s disease: Theory, technique, and results. Laryngoscope 92:853-858, 1982. 11. Schuknecht H F: Pathology of Ménière’s disease as it relates to the sac and tack procedures. Ann Otol Rhinol Laryngol 86:677-682, 1977. 12. Duval A J III, Rhodes VT: Ultrastructure of the organ of Corti following intermixing of cochlear fluids. Ann Otol Rhinol Laryngol 76:688-708, 1967. 13. Kimura R S, Schuknecht H F: Effect of fistulae on endolymphatic hydrops. Ann Otol Rhinol Laryngol 84:271286, 1975. 14. Schuknecht H F, Neff WD: Hearing losses after apical lesions in the cochlea. Acta Otolaryngol (Stockh) 42:263274, 1952. 15. Schuknecht H F, Sutton S : Hearing losses after experimental lesions in basal coil of cochlea. Arch Otolaryngol 57:129-142, 1953. 16. Schuknecht H F: Seifi A El: Experimental observations on the fluid physiology of the inner ear. Ann Otol Rhinol Laryngol 72:687-721, 1963. 17. Kimura R S, Schuknecht H F, Ota CY, Jones D D: Experimental study of sacculotomy in endolymphatic hydrops. Arch Otorhinolaryngol 217:123-137, 1977. 18. Schuknecht H F, Rüther A : Blockage of longitudinal flow in endolymphatic hydrops. Eur Arch Otorhinolaryngol 248:209-217, 1991. 19. Tumarkin A : Otolithic catastrophe: A new syndrome. BMJ 2:175-177, 1936. 20. Schuknecht H F: Cochlear endolymphatic shunt. Am J Otol 5:546-548, 1984. 21. Schuknecht H F, Bartley M : Cochlear endolymphatic shunt for Ménière’s disease. Am J Otol 6(Suppl):20-22, 1985. 22. Schuknecht H F: Cochleosacculotomy for Ménière’s disease: Internal endolymphatic shunt. Op Tech Otolaryngol Head Neck Surg 2:35-37, 1991. 23. Torok N: Old and new in Ménière’s disease. Laryngoscope 87:1870-1877, 1977.
40
Transcanal Labyrinthectomy Joseph B. Nadol, Jr., and Michael J. McKenna
Labyrinthectomy is an effective surgical procedure for the management of unremitting or poorly compensated unilateral peripheral vestibular dysfunction in the presence of ipsilateral, profound, or severe sensorineural hearing loss. The physiologic rationale is that central vestibular compensation is more rapid and complete for unilateral absence of peripheral vestibular function than for unilateral abnormal function, either episodic or chronic.1 Unilateral vestibular ablation has been advocated for more than 6 decades. Selective or total eighth cranial nerve transection by the suboccipital approach was introduced by Dandy in 1928.2 Destruction of the peripheral end organs of the vestibular labyrinth was introduced by Jansen3 in 1895 for complications of suppurative labyrinthitis. This technique was applied to unilateral peripheral vestibular disturbance by Milligan4 and by Lake5 in 1904, and was reintroduced by Cawthorne6 in 1943 as a canal wall up technique. In his original description, Cawthorne apparently ablated only the lateral semicircular canal. In its current form, complete vestibular ablation is accomplished by exenteration of all three of the semicircular canals and both maculae. The earliest report of a transcanal procedure for vertigo is credited to Crockett,7 who in 1903 described removal of the stapes as an effective treatment for vertigo. Lempert8 described an endaural transmeatal approach to the oval and round windows for Meniere’s disease. In this procedure, the stapes was removed, and the round window was punctured to “decompress” the membranous labyrinth. There was no mention, however, of the importance of destruction of the vestibular end organs. The modern transcanal labyrinthectomy for unilateral peripheral vestibular dysfunction was introduced by Schuknecht in 19569 and by Cawthorne in 1957.10 In a series of articles, Schuknecht’s technique evolved to emphasize the importance of destruction of all five vestibular end organs.11-13 Armstrong14 and Ariagno15 also emphasized the importance of total ablation of peripheral vestibular function.
PATIENT SELECTION The modern complete transcanal labyrinthectomy is an extremely effective treatment option for unilateral peripheral vestibular dysfunction. Rates of control of vertigo of 95% to 99% have been achieved by several authors. The modified Cawthorne transmastoid labyrinthectomy and the translabyrinthine vestibular or eighth cranial nerve section are equally effective options for ablation of peripheral vestibular dysfunction. The transcanal labyrinthectomy has the advantages of a more direct approach to the vestibular end organs, a shorter operating time, and a lower morbidity, particularly for postoperative facial nerve dysfunction and cerebrospinal fluid leak. Medical management appropriate to the unilateral vestibular disorder, including vestibular suppressants and diuretics for Meniere’s disease, should be attempted before consideration of labyrinthectomy. These forms of medical management are less successful for poorly compensated peripheral vestibular dysfunction, such as the sequelae of vestibular neuronitis, labyrinthitis, or trauma. In these cases, rehabilitative vestibular physical therapy should be attempted before labyrinthectomy. Labyrinthectomy should be performed only when it has been shown that the vestibular dysfunction is unilateral, and when the ipsilateral hearing loss is severe or profound. Although the published indications for labyrinthectomy have included hearing levels poorer than a 50 dB speech reception threshold and a 50% discrimination score, in view of the incidence of bilateral Meniere’s dis ease of 10% to 40%, as reported by Greven and Oosterveld16 and Paparella and Griebie,17 labyrinthectomy should be reserved for cases in which the hearing loss is severe to profound, generally with a speech reception threshold of 75 dB or worse and a speech discrimination score of less than or equal to 20%. This threshold for labyrinthectomy should be increased if hearing in the contralateral ear is not in the normal or near-normal range. Because of the acute and often protracted vestibular disturbance after labyrinthectomy, this procedure should 483
484
OTOLOGIC SURGERY
be done only for debilitating peripheral vestibular dysfunction. That is, a patient with only mild or infrequent attacks may be best treated nonoperatively. The definition of handicapping vertigo also depends on many other clinical factors, such as age, intercurrent disease, and occupation of the patient. A successful labyrinthectomy depends not only on total ablation of peripheral vestibular dysfunction, but also on compensation for this unilateral vestibular loss. Negative indicators for successful vestibular compensation generally include increased age, visual disturbances, obesity, sedentary lifestyle, arthritis or other lower limb dysfunction, dependent personality, or clear indication of secondary gain.
PREOPERATIVE EVALUATION A complete history and otolaryngologic–head and neck examination should be performed. Bilateral behavioral audiometry, including pure tone thresholds for air and bone conduction and speech discrimination, is necessary. Vestibular testing should include at least bilateral caloric function, best done by electronystagmography. This assessment is necessary to evaluate the possibility of bilateral vestibular dysfunction, and to confirm vestibular dysfunction in the affected ear based on audiometry and history. Hallpike’s positional testing and evaluation for the presence of the fistula and Hennebert’s signs should be done.18 A neurologic examination should be done to rule out concurrent cranial nerve, cerebellar, or other neurologic dysfunction that would belie the working diagnosis of a peripheral unilateral vestibular dysfunction. Radiographic assessment with computed tomography (CT) and magnetic resonance imaging (MRI) is not essential in every case. The symptoms and findings of long-standing unilateral Meniere’s disease may be similar, however, to the symptoms and findings caused by lesions of the posterior fossa. MRI with gadolinium enhancement is useful to rule out cerebellopontine angle or other tumors and demyelinating lesions. The ideal candidate for labyrinthectomy is an individual with unremitting or uncompensated peripheral vestibular dysfunction with severe to profound unilateral sensorineural hearing loss, unilateral vestibular dysfunction on electronystagmography, and lack of neurologic and radiographic evidence of central neurologic disease. Generally, the functional outcome is better in patients with unilateral Meniere’s disease than in patients with other peripheral vestibular dysfunction. In some patients with Meniere’s disease, electronystagmography is normal. In such cases, labyrinthectomy is justified if the symptoms and signs are sufficiently localizing to be convincing of unilateral peripheral dysfunction. The presence of fluctuating or severe to profound sensorineural loss, ipsilateral tinnitus, and aural symptoms concurrent with an attack of Meniere’s disease is sufficient to warrant
labyrinthectomy, even in the presence of normal caloric function if other selection criteria are met. The patient should be aware that postoperative vertigo is more severe when preoperative function is normal or nearly so in the affected ear.
PREOPERATIVE PATIENT COUNSELING AND INFORMED CONSENT Preoperative counseling should include a discussion of the natural history of Meniere’s disease, including the spontaneous rate of remission of approximately 70% within 8 years and the 10% to 40% incidence of involvement of the second ear.19 In addition, the patient should be aware that all hearing will be lost in the ear receiving surgery, and that the effect on tinnitus is unpredictable. The patient must be aware that immediately postoperatively there is a period of vertigo similar to a typical attack, and that this episode lasts several days. In addition, a period of protracted dysequilibrium may occur, and in patients with negative indicators for compensation, there may be some degree of permanent disability that requires a rehabilitative program. A discussion of alternative treatments for the vestibular symptoms of Meniere’s disease should be well understood by the patient. The discussion should include medical regimens; alternative ablative techniques, including transmastoid or translabyrinthine approaches; and selective ablative techniques through the middle or posterior fossa to save residual hearing. Particularly in elderly patients or in patients with other negative indicators for compensation, a round window labyrinthotomy should be considered and discussed with the patient as a possible alternative to labyrinthectomy. This procedure has the advantage of not resulting in a protracted period of dysequilibrium, and does not preclude a labyrinthectomy, if necessary. The usual risks of ear surgery also should be discussed, including paresis or paralysis of the facial nerve, perforation of the tympanic membrane, dysgeusia, failure of the procedure to achieve the desired result, the possible need for revision or secondary procedures, cerebrospinal fluid leakage or meningitis, and the fact that harvesting of a fat graft may be necessary.
SURGICAL TECHNIQUE General anesthesia is required because of the violent vestibular response during removal of the vestibular end organs. One exception may be in a revision labyrinthectomy in an ear with minimal residual vestibular function. In such cases, local anesthesia may allow intraoperative confirmation that the site of residual vestibular function has been located. The patient is placed in a supine position in a head holder with the head positioned similar to any transcanal procedure. Hair is shaved 0.5 inch around
Chapter 40 • Transcanal Labyrinthectomy
the auricle and prepared with an antiseptic solution. Generally, systemic antibiotics and steroids are not required. Facial nerve monitoring is usually not done in primary labyrinthectomy, but may be useful in a revision case if there is considerable scarring in the oval window area. No special instruments are necessary, but an instrument to remove the utricle from the recesses of the vestibule and a probe to destroy mechanically the cristae of the three semicircular canals should be available. For these purposes, a 4 mm right angled hook or a whirlybird from the Austin middle ear instrument set is useful. A microdrill is necessary to widen the oval window or to connect the oval and round windows for exposure.
Incision and Exposure The procedure is done by the transcanal approach in most cases. Occasionally, with a very narrow meatus, an endaural incision or postauricular transcanal approach may be useful. An anteriorly based tympanomeatal flap, longer than that used for stapedectomy, is elevated to allow wide curettage in the oval and round window areas (Fig. 40-1). The horizontal segment of the facial nerve, the entire stapes footplate, and the entire round window niche all should be easily visible after elevation of the flap and curettage of the posterior aspect of the bony tympanic annulus.
Preparation for Opening the Vestibule The incus is removed. The stapedial tendon is sectioned, and the stapes is removed in a rocking motion in an anteroposterior direction to allow removal of the stapes without fracture. Every effort should be made to avoid aspiration of the vestibule at this time to avoid displacement of the utricle. To obtain access to the vestibular end organs, the oval window may simply be enlarged at its anterior and inferior aspects (Fig. 40-2), or the oval and round windows may be connected to remove a segment of the promontory (Fig. 40-3). At this juncture, an attempt may be made to expose the posterior ampullary nerve. It may be exposed near the posterior aspect of the round window niche (see Figs. 40-2 and 40-3), which affords the surgeon an opportunity to practice identification and section of the posterior ampullary nerve, and helps guarantee a more complete labyrinthectomy. In a study of labyrinthectomy in the cat, Schuknecht23 reported that subtotal destruction of the vestibular end organs occurred in 10 of 24 ears, and that the crista of the posterior semicircular canal was the end organ most commonly missed.
Removal of Vestibular End Organs Total mechanical destruction of the five vestibular end organs is the goal of this surgery. The normal position of the vestibular end organs and their relationship to the
485
oval and round windows are shown in Figure 40-4. Endolymphatic hydrops or intraoperative loss of perilymph may cause displacement of these end organs, however. During aspiration or loss of perilymphatic fluid, the utricle usually retracts superiorly to lie medial to the horizontal segment of the facial nerve. Before aspiration of the vestibule, the utricle should be removed with a 4 mm hook, whirlybird, or utricular hook in the superior aspect of the vestibule (Fig. 40-5). The utricle is substantial and can easily be seen under low power of the operating microscope. Avulsion of the utricle from the vestibule usually results in avulsion of the cristae of lateral and superior semicircular canals as well, but not that of the posterior semicircular canal. The saccule is destroyed mechanically by aspiration of the medial aspect of the vestibule in the area of the spherical recess. Manipulation of the medial aspect of the vestibule must be done with care to avoid fracture of the cribrose area, which would result in profuse leakage of cerebrospinal fluid from the internal auditory canal. Any residual neuroepithelium of the cristae of the three semicircular canals is destroyed by mechanical probing. The surgeon can feel the 4 mm hook drop into the ampullary ends of the bony canals (Fig. 40-6). This entire technique should be practiced in the temporal bone laboratory to gain familiarity with the anatomy and proficiency in this procedure. After destruction of the vestibular end organs, the vestibule is usually packed with absorbable gelatin sponge (Gelfoam) or, preferably, a small fat graft from the ear lobe (Fig. 40-7). Some surgeons recommend the use of gentamicin or streptomycin-soaked Gelfoam in the medial aspect of the vestibule to guarantee destruction of residual neuroepithelium. As seen in Figure 40-8, the absence of a tissue graft of the open oval window results in a pneumolabyrinth postoperatively. Generally, this result does not cause a problem, but a tissue seal with fat or fascia provides a better barrier between the middle ear and cerebrospinal fluid spaces. Leakage of cerebrospinal fluid after aspiration of the vestibule should be repaired with a tissue seal. The tympanomeatal flap is returned to the posterior canal wall and held in place with packing, and the procedure is terminated.
POSTOPERATIVE CARE The degree of postoperative vestibular disturbance is positively correlated with preoperative residual vestibular function. That is, individuals with normal or near-normal caloric response preoperatively have a more severe reaction than individuals with minimal or no vestibular response. Third-degree nystagmus, nausea, and vomiting can be expected. Control of the vestibular symptoms may be achieved with promethazine or droperidol. Vestibular suppressants should be tapered as quickly as possible, given the evidence that they may impair or slow central compensation for unilateral vestibular ablation. Rapid
486
OTOLOGIC SURGERY
FIGURE 40-1. Incisions for transcanal labyrinthectomy. Anteriorly based tympanomeatal flap should be slightly wider and longer than flap used for stapedectomy to allow wide curettage of bony tympanic annulus. Curettage should allow visualization of horizontal segment of the facial nerve and entire oval and round window niches. FIGURE 40-2. Access to vestibule may be achieved by widening oval window niche by removing a segment of promontory. At this time, posterior ampullary nerve may be located in the floor of the round window niche. This step is facilitated by removal of round window overhang.
FIGURE 40-3. As an alternative method to expose vestibule, the round and oval windows may be connected to remove lateral aspect of the promontory.
FIGURE 40-4. Normal anatomic positions of five vestibular end organs are superimposed on surgical exposure.
Chapter 40 • Transcanal Labyrinthectomy
487
FIGURE 40-5. Removal of utricle can be accomplished with a 4 mm hook (as shown), utricular hook, or whirlybird. Removal of utricle often results in simultaneous avulsion of ampullary ends of the lateral and superior canals, but not of the posterior canal. FIGURE 40-6. After removal of utricle and aspiration of saccule, any residual neuroepithelium of the three semicircular canals is destroyed by mechanical disruption using a 3 to 4 mm angled hook. FIGURE 40-7. After avulsion and destruction of vestibular end organs, vestibule is filled with Gelfoam soaked in gentamicin or streptomycin solution and a tissue graft, such as fat.
488
OTOLOGIC SURGERY
FIGURE 40-8. Vestibule (V) remains in communication with middle ear (ME) 14 years after transcanal labyrinthectomy.
advancement of physical activity should be encouraged, including sitting and ambulation with assistance in the first few days. The patient may be discharged when he or she is relatively self-sufficient and ambulating without assistance. This improvement may take 3 to 7 days. At the first postoperative visit at 1 week, the packing is removed, and progressive activity is encouraged. The patient is seen 1 month postoperatively to evaluate residual symptoms, progress, and compensation, and to ascertain healing of the tympanic membrane and ear canal. At this time, there is usually no residual spontaneous nystagmus, and a caloric test of the ear that had surgery, using 20 mL of ice water and Frenzel’s lenses, is used to ascertain the completeness of unilateral vestibular ablation. Further follow-up depends on the patient’s progress.
aid in retrieval. Alternatively, the utricle remains fixed to the lateral and superior semicircular canals, which may be palpated in the depths of the vestibule. A wide exposure achieved by removing the bone from the inferior aspects of the oval window or by connecting the round and oval windows improves access to the vestibule and its contents.
SURGICAL COMPLICATIONS AND MANAGEMENT
Postoperative Complications
Intraoperative Complications
The incidence of incomplete labyrinthectomy in many series is less than 5%. Persistent postoperative vestibular symptoms or failure to compensate after labyrinthectomy may signal residual neuroepithelium and vestibular function in the operated ear. A postoperative ice water caloric test is valuable, if the results are positive, to confirm the presence of functional neuroepithelium. Absence of induced symptoms or nystagmus on postoperative ice water caloric test does not indicate total destruction of the vestibular end organs, however. In the presence of persistent symptoms or poor compensation after surgery in the absence of contralateral disease, the possibility of an incomplete labyrinthectomy should be considered despite the absence of caloric response. Management of suspected incomplete labyrinthectomy should include revision surgery. One option is
Cerebrospinal Fluid Leakage Fracture of the cribrose area bone on the medial aspect of the vestibule results in profuse cerebrospinal fluid leakage from the internal auditory canal. The leakage can be controlled by a tissue graft using fascia or subcutaneous fat, or both, to seal the vestibule.
Failure to Find the Utricle Identification and removal of the utricle are essential to a complete labyrinthectomy. When perilymph has been aspirated from the vestibule, the utricle may retract superiorly under the horizontal segment of the facial nerve. The use of a modified right angled hook (utricular hook) (see Fig. 40-5) or a right angled No. 3 Fr suction may
Facial Nerve Injury The facial nerve may be damaged in its horizontal segment. Care must be taken, particularly in retrieval of the utricle, to avoid trauma to the medial aspect of the horizontal segment of the nerve. Treatment of facial nerve injury should follow the usual principles of management of iatrogenic injury. A delayed facial paresis occurs infrequently and may be managed expectantly.
Incomplete Labyrinthectomy
Chapter 40 • Transcanal Labyrinthectomy
489
revision transcanal labyrinthectomy with the patient under local anesthesia using the patient’s response during manipulation of the vestibule as a means of localizing residual neuroepithelium. This is frequently a difficult procedure, however, with fibrous tissue and, occasionally, new bone formation within the vestibule. A more certain means of ablating vestibular function in such cases is a transmastoid labyrinthectomy and translabyrinthine vestibular nerve section.
HISTOPATHOLOGY OF LABYRINTHECTOMY Postmortem histopathology of temporal bones from patients who, in life, underwent labyrinthectomy has been reported by Belal and Ylikoski,20 Linthicum and associates,21 Pulec,22 and Schuknecht,23 and all have reported examples of incomplete mechanical disruption of the vestibular end organ after transcanal labyrinthectomy. These results underscore the importance of proper exposure, removal of the utricle, and thorough probing of the ampullae. In an animal study of labyrinthectomy, Schuknecht23 reported that the neuroepithelium of the posterior semicircular canal persisted in 10 of 24 ears. This fact argues for wider exposure of the vestibule by connecting the oval and round windows, and for selective destruction of the posterior ampullary nerve, as recommended by Gacek,24 as an additional step to guarantee complete labyrinthectomy. Linthicum and associates21 and Belal and Ylikoski20 reported traumatic neuroma within the vestibule after labyrinthectomy (Fig. 40-9). Both groups interpret this finding as an indication of the superiority of the translabyrinthine vestibular neurectomy. In one case, a traumatic neuroma was also described after transmastoid labyrinthectomy and section of the superior vestibular nerve. In an experimental study in the cat, Schuknecht23 reported no evidence of regeneration of vestibular nerve fibers or formation of traumatic neuroma after labyrinthectomy. In temporal bone specimens from human subjects who had undergone transcanal labyrinthectomy during life, degeneration of the vestibular nerve was seen in one, and a proliferation of nerve fibers was identified in another (Fig. 40-10). No evidence exists, however, to suggest that nerve fibers, whether residual or regenerative, can contribute to afferent vestibular input if the vestibular neuroepithelium distal to it has been destroyed. Two patients are cited by Linthicum and associates21 as examples of failure of labyrinthectomy because of traumatic neuroma. In the first case, reported by Hilding and House,25 the neuroma was uncovered at revision labyrinthectomy, but there was no evidence that the persistent symptoms resulted from the neuroma, rather than from residual neuroepithelium. In the second case, reported by Pulec,22 a traumatic neuroma was identified by postmortem temporal bone histopathology in a patient with persistent vestibular
FIGURE 40-9. Histopathology of superior vestibular nerve 36 months after transcanal labyrinthectomy in a cat. A, Unoperated control ear at the level of the normal macular utriculi (MU). B, In the ear that had undergone labyrinthectomy, there was moderate degeneration of vestibular nerve (VN), new bone and fibrous tissue within the vestibule (V), and no evidence of proliferation of remaining vestibular nerves.
FIGURE 40-10. Histopathology 4 months after left labyrinthectomy. A, Macular utriculi (MU) of the operated ear shows atrophy. The s troma of the macula is intact, however, and there has been proliferation of nerve fibers (NF) within it. B, Superior vestibular nerve and MU of the unoperated side appear normal.
symptoms for 10 years after labyrinthotomy, not labyrinthectomy. In addition, despite no response on the premortem caloric testing, the ampulla of the posterior semicircular canal was normal. In this case, the persistent vestibular symptoms probably resulted from residual vestibular neuroepithelium rather than from the traumatic neuroma.
490
OTOLOGIC SURGERY
RESULTS OF SURGERY The reported results as measured by ablation of caloric function or cure of the patient have varied considerably. Linthicum and associates21 reported that only 17 (60%) of 25 patients who underwent labyrinthectomy were cured or improved by this approach; they advocated translabyrinthine nerve section as a more reliable method of ablating vestibular function. Ariagno15 found, however, a 98% success rate in controlling peripheral vestibular disorders through transcanal labyrinthectomy, emphasizing the need for total destruction of the vestibular end organs by joining the oval and round windows. Hammerschlag and Schuknecht26 reported a cure of episodic vertigo in 120 (96.8%) of 124 patients by transcanal labyrinthectomy. The remaining four patients had continuing dysequilibrium, and three with persistent vestibular response by postoperative ice water caloric tests were cured by revision transcanal labyrinthectomy, resulting in an overall cure rate of 99%.
SPECIAL CONSIDERATIONS The advent of cochlear implantation as a possibility for rehabilitation of profoundly deaf patients, and the fact that 10% to 40% of patients with Meniere’s disease have bilateral involvement require consideration of the implications for eventual implantation after any surgical procedure for management of vestibular disturbance. Chen and colleagues27 reported on three temporal bone cases from patients who in life had undergone labyrinthectomy, two by the transcanal route and one by the transmastoid approach. Based on the patency of the cochlear duct, remaining spiral ganglion, and neural elements, and maintenance of the organ of Corti as evaluated by histopathologic study, these authors predicted that labyrinthectomy would not preclude subsequent cochlear implantation. In six patients who had undergone previous unilateral transmastoid labyrinthectomy, Lambert and coworkers28 reported that round window electric stimulation resulted in a psychophysical response to stimulus in all patients and electrically evoked middle latency response in five of six patients. Kveton and associates29 reported an ear deafened by transmastoid labyrinthectomy with subsequent successful cochlear implantation resulting in speech comprehension comparable to that of patients deafened by other causes.
REFERENCES 1. Stockwell CW, Graham M D: Vestibular compensation following labyrinthectomy and vestibular neurectomy. In Nadol J B Jr (ed): Second International Symposium for Ménière’s Disease. Amsterdam, Kugler & Ghedini Publishers, 1989, pp 489-498.
2. Dandy WE : Ménière’s disease: Its diagnosis and a method of treatment. Arch Surg 16:1127-1152, 1928. 3. Jansen A : Referat uber die operationsmethoden bei den verschiedenen otitischen gehirukoneplikationen. Verh Dtsch Otol Gesell (Jena), 1895, p 96. 4. Milligan W: Ménière’s disease: A clinical and experimental inquiry. BMJ 2:1228, 1904. 5. Lake R : Removal of the semicircular canals in a case of unilateral aural vertigo. Lancet 1:1567-1568, 1904. 6. Cawthorne TE : The treatment of Ménière’s disease. J Laryngol Otol 58:63-71, 1943. 7. Crockett E A : The removal of the stapes for the relief of auditory vertigo. Ann Otol Rhinol Laryngol 12:67-72, 1903. 8. Lempert J: Lempert decompression operation for hydrops of the endolymphatic labyrinth in Ménière’s disease. Arch Otolaryngol Head Neck Surg 47:551-570, 1948. 9. Schuknecht H F: Ablation therapy for the relief of Ménière’s disease. Laryngoscope 66:859-870, 1956. 10. Cawthorne T: Membranous labyrinthectomy via the oval window for Ménière’s disease. J Laryngol Otol 71:524527, 1957. 11. Schuknecht H F: Ablation therapy in the management of Ménière’s disease. Acta Otolaryngol Suppl (Stockh) 132:1-42, 1957. 12. Schuknecht H F: Destructive therapy for Ménière’s disease. Arch Otolaryngol Head Neck Surg 71:562-572, 1960. 13. Schuknecht H F: Destructive labyrinthine surgery. Arch Otolaryngol Head Neck Surg 97:150-151, 1973. 14. Armstrong BW: Transtympanic vestibulotomy for Ménière’s disease. Laryngoscope 69:1071-1074, 1959. 15. Ariagno R P: Transtympanic labyrinthectomy. Arch Otolaryngol Head Neck Surg 80:282-286, 1964. 16. Greven A J, Oosterveld WJ: The contralateral ear in Ménière’s disease. Arch Otolaryngol Head Neck Surg 101:608-612, 1978. 17. Paparella M M, Griebie M S : Bilaterality of Ménière’s disease. Acta Otolaryngol (Stockh) 97:233-237, 1984. 18. Nadol J B : Jr: Positive “fistula sign” with an intact tympanic membrane. Arch Otolaryngol Head Neck Surg 100:273-278, 1974. 19. Silverstein H, Smouha E, Jones R : Natural history versus surgery for Ménière’s disease. In Nadol J B Jr (ed): Second International Symposium for Ménière’s Disease. A msterdam, Kugler & Ghedini Publishers, 1989, pp 543544. 20. Belal A, Ylikoski J: Pathology as it relates to ear surgery, II: Labyrinthectomy. J Laryngol Otol 97:1-10, 1983. 21. Linthicum FH, Alonso A, Denia A : Traumatic neuroma. Arch Otolaryngol Head Neck Surg 105:654-655, 1979. 22. Pulec J L : Labyrinthectomy: Indications, technique, and results. Laryngoscope 84:1552-1573, 1974. 23. Schuknecht H F: Behavior of the vestibular nerve following labyrinthectomy. Ann Otol Rhinol Laryngol 91(5 Suppl 97):16-32, 1982. 24. Gacek R R : Transection of the posterior ampullary nerve for the relief of benign paroxysmal positional vertigo. Ann Otol Rhinol Laryngol 83:596-605, 1974. 25. Hilding D A, House WF: Acoustic neuroma”: Comparison of traumatic and neoplastic. J Ultrastruct Res 12:611623, 1965.
Chapter 40 • Transcanal Labyrinthectomy 26. Hammerschlag PE, Schuknecht H F: Transcanal labyrinthectomy for intractable vertigo. Arch Otolaryngol Head Neck Surg 107:152-156, 1981. 27. Chen D A, Linthicum R H, Rizer FM : Cochlear histopathology in the labyrinthectomized ear: Implications for cochlear implantation. Laryngoscope 98:1170-1172, 1988.
491
28. Lambert PR , Ruth R A, Halpin C F: Promontory electrical stimulation in labyrinthectomized ears. Arch Otolaryngol Head Neck Surg 116:197-201, 1990. 29. Kveton J F, Abbott C, April M, et al: Cochlear implantation after transmastoid labyrinthectomy. Laryngoscope 99:610-613, 1989.
41
Chemical Treatment of the Labyrinth Edwin M. Monsell, Stephen P. Cass, Leonard P. Rybak, and Julian M. Nedzelski
Despite some successes, the medical treatment of inner ear conditions, such as Meniere’s disease and sudden sensorineural hearing loss, is often frustrating to patients and physicians. Molecular studies of inner ear function have revealed some promising approaches to therapy. Chemoprotection strategies for exposure to noise,1 cisplatin,2 and aminoglycosides3 have generated considerable interest. Other than aminoglycoside treatment for vertigo in Meniere’s disease, no intratympanic treatment strategy has gained widespread acceptance as standard therapy. We review the status of this treatment and of intratympanic treatment with corticosteroids. Most patients with Meniere’s disease can be managed satisfactorily with dietary salt restriction, diuretics, and vestibular suppressants. Only about 10% of patients become disabled in regard to work, domestic activities, travel, and the general enjoyment of life. The goals of treatment of Meniere’s disease are to control the definitive spells of vertigo and to preserve hearing.4 From the standpoint of the clinical skills required by the practitioner and the potential morbidity to patients, the administration of aminoglycosides for vertigo by any route of administration should be considered as great an undertaking as surgical treatment. Because not all the problems and pitfalls of any technique can be conveyed in a printed article, we advise that streptomycin treatment be learned under the guidance of a practitioner experienced in its use. In 1944, streptomycin was isolated from cultures of a soil organism, Streptomyces griseolus.5 This drug displayed broad-spectrum antibacterial activity and was the first found to be effective against tuberculosis. Because effective treatment of tuberculosis with streptomycin required prolonged therapy, ototoxicity became evident soon after introduction of the drug. In 1948, streptomycin was used to treat patients with unilateral Meniere’s disease specifically on the basis of its vestibulotoxic effects.6 Mechanisms of aminoglycoside ototoxicity have been reviewed more recently.7 Aminoglycosides exert toxic effects on the hair cells of the inner ear by several mechanisms. First, aminoglycosides bind to the plasma
membrane and displace calcium and magnesium. This event results in acute but reversible interference with calcium-dependent mechanicoelectric transduction channels.8 Second, dihydrostreptomycin enters the mouse outer hair cell through the mechanoelectric transduction channel.9,10 This channel can act as a one-way valve for aminoglycoside entry, promoting the accumulation of aminoglycoside within the cell. There seems to be a competition between the aminoglycoside and calcium for entry into the outer hair cell. The normal low endolymph concentration of calcium promotes intracellular accumulation of aminoglycosides.10 Third, aminoglycosides are transported into the cell by an energy-dependent process. Within the cell, the drug binds to phosphatidylinositol. This event is associated with progressive disruption of the plasma membrane and inhibition of the second messenger inositol triphosphate. With progressive disruption of the second messenger system and the plasma membrane, cell death occurs.11-13 The disruption of cell membranes and other intracellular components may be mediated by free radicals. More recent studies have shown that aminoglycosides form a complex with iron, and that this complex catalyzes the production of free radicals. Reactive oxygen species formation in vitro requires iron and polyunsaturated lipids, such as those found in cell membranes, as electron donors. Gentamicin and iron form ternary complexes with phospholipids. Oxidative damage to phospholipids can cause the release of arachidonic acid, which can also form complexes with gentamicin and iron. These complexes can lead to further lipid peroxidation, damaging membranes, proteins, and DNA to disrupt the function of the outer hair cell, leading to programmed cell death (apoptosis).14 Reactive oxygen species may promote the opening of the mitochondrial permeability pore.15 The Jun N-terminal kinase (JNK) pathway seems to play a role in the death from gentamicin of auditory and vestibular hair cells.16 The combination of iron chelators and free radical scavengers in animal experiments provides complete protection from gentamicin ototoxicity.17 Pretreatment with aspirin protected against amikacin ototoxicity in 493
494
OTOLOGIC SURGERY
animals.18 More recent clinical trials in China showed protection against gentamicin-induced hearing loss in patients pretreated with salicylates.19 Aminoglycosides do not become concentrated in cochlear fluids, although the elimination half-life increases with long-term administration. These observations support the notion that intracellular sequestration of the drug occurs.10,20 Aran and associates21 confirmed that aminoglycosides undergo rapid uptake by cochlear and vestibular hair cells, and slow clearance from these cells. Amikacin, dihydrostreptomycin, and kanamycin are primarily cochleotoxic, whereas gentamicin and streptomycin are primarily vestibulotoxic. At high doses, streptomycin is also cochleotoxic. Streptomycin, 25 mg/kg per day, administered systemically to cats resulted in loss of vestibular hair cells only, but at 100 mg/kg/day, vestibular and cochlear hair cells were lost.22 More recent animal experiments have tried to model the pharmacokinetics of intratympanic administration of gentamicin applied in a sustained-release vehicle of liquid fibrin glue. High levels of gentamicin were measured in perilymph within 8 hours of administration. These high levels persisted for at least 24 hours, then declined rapidly by 72 hours. The elimination rate for gentamicin was 1.04 mg/mL/hr.23 The hair cells of the cristae, the ampullae, and the cochlea degenerate to different degrees after the administration of aminoglycosides. The primary vestibular neurons, the cochlear nuclei, and the vestibular nuclei are not directly affected, even at high doses.22,24 The basal turn of the cochlea is the region most susceptible to permanent loss of hair cells, resulting in an initial loss of high-frequency hearing sensitivity. Although the mechanisms of this differential toxicity are incompletely understood, several contributing factors have been identified, including the route of administration, dose variables, and the specific aminoglycoside used. Mutations in the 12S ribosomal RNA gene have been identified in association with maternally inherited susceptibility to deafness from treatment with aminoglycosides.25 Affected individuals can sustain profound deafness from even a single dose administered systemically. Damage to vestibular dark cells, which are thought to play a role in the production of endolymph, has been reported after administration of doses of aminoglycoside below the threshold for damage to hair cells. It is possible, but unproven, that impaired function of dark cells is beneficial in Meniere’s disease.26,27
INTRAMUSCULAR APPLICATION OF STREPTOMYCIN Clinical Studies Between 1948 and 1980, eight investigators reported a total of 49 patients treated with intramuscular streptomycin for unilateral or bilateral Meniere’s disease.
In the first extensive studies of intramuscular streptomycin, Schuknecht28,29 administered 0.75 to 1.75 g intramuscularly every 12 hours. Treatment continued until there were no ice water caloric responses in the diseased ears. This treatment end point was assumed to represent total or near-total vestibular ablation. The total doses ranged from 13.5 to 89 g (mean 39 g). All patients became severely ataxic, and most had oscillopsia early in the course of treatment, but none experienced hearing loss; 35% had persistent ataxia, and 15% had persistent oscillopsia. Ninety-five percent of patients had no post-treatment vertigo. Hearing levels did not change or fluctuate in 90% of patients. When a totally ablative dose of intramuscular streptomycin was administered, vertigo was controlled, and hearing was preserved.30 Patients who undergo total vestibular ablation may still be disabled by chronic dysequilibrium, ataxia, and oscillopsia.31 To attempt to limit the chronic oscillopsia and ataxia that follow total bilateral vestibular ablation, subtotal or “titration” treatment protocols with streptomycin were developed.32,33 Rather than giving an ablative dose, streptomycin was administered until episodic vertigo was controlled.
Treatment Method for Subtotal Vestibulectomy with Intramuscular Streptomycin Our suggestions have been adapted from established protocols and modified by our own experience.32,33
Indications Intramuscular streptomycin may be considered for patients with disabling episodic vertigo caused by Meniere’s disease in the following situations: (1) simultaneously active Meniere’s disease in both ears, or when it is unclear from which ear the attacks of vertigo are arising; (2) Meniere’s disease in an only hearing ear; or (3) in the second ear after an ablative procedure on the opposite side, such as selective vestibular nerve section. It is essential to determine to what extent a patient may be disabled by the definitive episodic vertigo of Meniere’s disease as defined by the American Academy of Otolaryngology–Head and Neck Surgery (AAO-HNS).4 Patients disabled primarily by continuous dysequilibrium, ataxia, oscillopsia, or motion intolerance are generally not good candidates for vestibular destructive procedures of any type because their symptoms are due to a failure of central compensation, to the perception of dysequilibrium in the presence of normal balance performance, or both.34,35 It is important to identify patients with Cogan’s syndrome, luetic hydrops, and autoimmune disease of the ear because these patients may respond to nondestructive medical treatment, such as corticosteroids.36 Although reliable data are lacking, it is likely that the recognition of
Chapter 41 • Chemical Treatment of the Labyrinth
autoimmune inner ear disease and immunosuppressive treatment have resulted in reduction of the number of cases of bilateral Meniere’s disease treated with intramuscular streptomycin. Patients with markedly reduced caloric function before treatment with streptomycin should be managed with additional caution because they may develop permanent dysequilibrium or oscillopsia with the loss of additional vestibular function. Treating such patients with lower doses of streptomycin should be considered.
Pretreatment Evaluation and Patient Counseling The pretreatment evaluation requires a complete history and physical examination, including a neurologic examination, observation of eye movements, and vestibulospinal examination. The vestibulospinal examination consists of observation of gait, tandem gait, and Romberg and sharpened Romberg maneuvers. A simple, effective assessment of postural stability can be performed in the examination room or bedside by having a patient stand on 4 to 6 inches of dense foam with eyes closed. The foam reduces the reliability of proprioceptive input to postural control mechanisms, so the subjects must depend on vestibular inputs when eyes are closed. Normal individuals have no difficulty standing on the foam, but patients with bilateral loss of vestibular function or an acute vestibular loss fall when forced to rely solely on vestibular cues to maintain posture. Baseline audiometric and laboratory vestibular function tests are performed, including electronystagmography, rotational testing, and posturography. Renal function tests are performed as indicated. It is essential to distinguish clearly between two phenomena in patients with Meniere’s disease who are undergoing intramuscular treatment with aminoglycosides: (1) vertigo owing to the disease, and (2) the syndrome of acute bilateral vestibular loss caused by vestibulotoxicity. Vertigo is the hallucination of motion—the perception of motion when none is occurring. The vertigo of Meniere’s disease occurs without provocation, consists of a spinning or rotating sensation always with spontaneous nystagmus, lasts 15 minutes to several hours, and is accompanied by dysequilibrium and nausea that may last for hours.4 The syndrome of acute bilateral vestibular loss may include rotational vertigo, but the vertigo is related to treatment rather than being spontaneous. Commonly, patients experience discomfort associated with rapid head movements, a sense of disorientation in space, and ataxia. Patients may also manifest oscillopsia, a disturbing sensation of the visual field bobbling as the patient walks about or rides in a car. This phenomenon is due to impairment or loss of the vestibulo-ocular reflex, which helps to maintain a stable image on the retina during head movement. Oscillopsia may be temporary or permanent after bilateral loss of vestibular function.30,31,37
495
These distinctions are important because symptoms resulting from Meniere’s disease are indications to resume treatment, whereas symptoms resulting from vestibulotoxicity are indications to halt treatment. Occasionally, the physician may find it difficult to make this distinction from a patient’s history. In such cases, caution would dictate that treatment should be discontinued until the situation is clarified by repeated observations over time. Patient counseling is crucial. Patients are counseled that the purpose of the titration of streptomycin is to control the recurrent episodes of vertigo typical of Meniere’s disease, and that disability secondary to dysequilibrium may replace disability from episodic vertigo. Patients need to understand that treatment may be protracted, and that additional courses of streptomycin may be needed in the future. Other risks and complications include hearing loss, tenderness from the deep intramuscular injections, nausea, perioral or peripheral numbness or tingling, rash, and fever. Renal, visual, and hematologic effects have also been reported with prolonged treatment with aminoglycosides. Special discussion is required for patients with abnormal renal function because renal toxicity is a potential complication of aminoglycosides. Careful monitoring of renal function and consultation with a nephrologist are recommended to ensure safe treatment of a patient with impaired renal function. Occasionally, a patient reports allergy to aminoglycosides. In this case, the nature of the allergy is explored, and an intradermal test dose is considered.
Treatment Technique Therapy may start with streptomycin sulfate injected intramuscularly, 1 g twice a day for 5 days (10 g total cumulative dose). After this initial course, clinical reassessment is performed by interview, vestibulospinal physical examination, audiometry, and vestibular function tests. If studies remain unchanged, there is no clinical indication of decreased vestibular function, and the patient continues to experience vertigo, additional streptomycin may be given. A second course could consist of additional streptomycin sulfate injected intramuscularly, 1 g twice a day for 5 days (10 g in the second course and a total cumulative dose of 20 g). After administration of the first 20 g of total cumulative dose, we suggest that additional doses be smaller, such as streptomycin sulfate injected intramuscularly, 1 g once per day for 5 days (5 g per increment). Streptomycin injections should be stopped when any one of the following occurs: (1) episodic vertigo ceases or seems to abate; (2) the syndrome of acute vestibular loss appears or worsens; (3) worsening of balance performance on the vestibulospinal examination; (4) laboratory tests of vestibular function show loss of vestibular function, such as reduction of caloric responses, increase in
496
OTOLOGIC SURGERY
phase or decrease in gain of the vestibulo-ocular reflex on rotational testing, and a fall on posturography when forced to rely solely on vestibular cues; or (5) decline in hearing. It is important to proceed with treatment slowly and cautiously. The effects of streptomycin can be delayed. If the reduction of vestibular function continues to complete ablation, ataxia and oscillopsia may persist chronically. If episodic vertigo recurs, additional courses of streptomycin can be given as indicated.
Results The long-term results of intramuscular subtotal streptomycin therapy for bilateral Meniere’s disease are known from only one report.37 Nineteen patients were reviewed with follow-up from 2 to 9 years. Recurrent vertigo was controlled in 95% of patients within the first 6 to 18 months after treatment. At last follow-up, 63% of patients reported having had no recurrence of vertigo. Persistent mild dysequilibrium occurred in 60% of patients without recurrent vertigo, and oscillopsia persisted in 16% of patients.37
APPLICATION OF STREPTOMYCIN TO LATERAL SEMICIRCULAR CANAL Experimental Studies Kimura and colleagues38 reported that the application of gentamicin to the lateral semicircular canal of normal guinea pigs produced a selective vestibular lesion. There was sensory cell degeneration in the utricular maculae and all but one of the cristae of the superior, lateral, and posterior canals of 27 ears. The saccular maculae were less affected. All cochleas were normal except one that had a small lesion of outer hair cells at the basal turn. Fenestrated control inner ears were normal throughout. Kimura and colleagues39 repeated some of their experiments with streptomycin after producing experimental endolymphatic hydrops by occlusion of the endolymphatic duct. Hydropic ears sustained substantial cochlear and vestibular lesions when streptomycin was applied to the lateral semicircular canal. Fenestration of the lateral canal without drug application produced a significant cochlear lesion, but not a vestibular lesion.
Clinical Studies The application of streptomycin to the labyrinth through the lateral semicircular canal (labyrinthotomy with streptomycin infusion [LSI]) was introduced by Shea.40,41 The rationale of this procedure was that application to the vestibular labyrinth might cause more drug to reach the vestibular hair cells than cochlear hair cells, and produce a more selective lesion with a single treatment. Use of this route of administration requires the performance of a mastoidectomy and opening the bony otic capsule.
Surgical Method Shea and Norris40 described the technique of LSI. A simple mastoidectomy is performed, and the lateral semicircular canal is identified. The bone of the dome of the semicircular canal is gradually thinned with a diamond burr. A “double blue-line” technique is used to create a fenestration of the lateral semicircular canal. A small amount of fluid containing streptomycin is slowly infused into the perilymphatic space of the lateral semicircular canal. Opening the lateral semicircular canal in Meniere’s disease results in enhancement of the ratio of the summating potential to the action potential of the electrocochleographic recording, primarily by reduction in the amplitude of the action potential. The ratio does not change during infusion of streptomycin, but sometimes declines slightly after the fenestration is closed.42 The volume of fluid, the composition of the diluent (lactated Ringer solution or other physiologic solution), the amount of streptomycin in the solution delivered, the amount of time during which the fluid is introduced, and whether the streptomycin solution is followed by a “rinse” of a physiologic solution without streptomycin are important technical variables that could affect the amount of streptomycin administered and the amount of trauma to the inner ear. After the drug is infused, the fenestration is closed with a thick piece of temporalis muscle and fascia. The postauricular wound is closed in the usual manner. On the basis of animal experiments, Shea40,41 initially recommended puncturing the lateral membranous canal in hydropic ears to release endolymph, possibly decompressing endolymphatic hydrops acutely, but later withdrew that recommendation.
Results In 1989, a multicenter study was initiated to produce an independent study of LSI results for hearing preservation and control of vertigo. Preliminary data from this and other studies were reported.42,43 Preoperative pure tone averages ranged from 14 to 76 dB HL with a mean of 54 dB (SD 14). Postoperative pure tone averages ranged from 25 to 110 dB HL with a mean of 76 dB. Four patients (9%) had an early postoperative hearing result that was improved over the preoperative hearing level.44 In 11 patients (23%), the hearing was unchanged; in 32 patients (68%), hearing worsened. In 27 patients (57%), the postoperative pure tone level was 71 dB or worse (severe to profound hearing loss). Hearing outcome (change in pure tone average or word recognition) did not seem to be a function of patient age, sex, side of surgery, duration of hearing loss before surgery, or whether the patient had had a surgical procedure on the ear before the LSI, or which surgeon performed the procedure. A trend was identified suggesting that patients with milder preoperative losses may sustain less hearing damage from LSI. Opening the endolymphatic
Chapter 41 • Chemical Treatment of the Labyrinth
space of the lateral membranous canal resulted in more loss of hearing than when the membranous canal was not opened (P = .05). The postoperative follow-up time ranged from 1 to 18 months, with a mean of 10 months. Eight patients (17%) had secondary procedures for control of vertigo during the first few postoperative months.
Discussion Silverstein33 reported that patients with good hearing seemed to be more resistant to hearing loss from streptomycin applied intratympanically. There seemed to be less postoperative hearing loss from LSI when the preoperative pure tone average was 40 dB or better. Findings in patients with vestibular disorders seem to corroborate findings in the guinea pig model; that is, the hydropic ear shows more sensitivity to aminoglycoside ototoxicity than the normal ear, and there is less selectivity of lesions between vestibular and cochlear hair cells in hydropic ears.39 The long-term results (≥2 years) for hearing preservation and control of vertigo by LSI have not been reported. Because of the high risk of associated hearing loss and treatment failures, and because of generally better results with intratympanic gentamicin therapy, LSI has mostly fallen out of use.
INTRATYMPANIC GENTAMICIN THERAPY Experimental Studies Intratympanic injection of aminoglycosides allows treatment of unilateral Meniere’s disease without producing systemic toxicity or effects on the contralateral ear. Tracer studies have shown that the primary route of entry into the inner ear is through the round window membrane.45-49 A secondary route of entry into the inner ear may be through the annular ligament of the stapes.38,50 After the intratympanic application of streptomycin to guinea pigs, the cristae of the semicircular canals showed the most degeneration, followed by the utricle, then the saccule and the cochlea.51 The selectivity of the lesion for vestibular versus cochlear hair cells after intratympanic injection was reduced at higher dosages. The application of 8 mg of streptomycin to the round window membrane of the cat produced only vestibular toxicity, whereas 20 to 40 mg produced cochlear and vestibular toxicity.52 These experimental results emphasize the potential for cochlear toxicity when the primary route of entry into the inner ear is through the round window membrane.
Clinical Studies Schuknecht28,29 introduced the use of intratympanic aminoglycosides to induce a chemical labyrinthectomy for the treatment of unilateral Meniere’s disease. He
497
reported the results of eight patients given large daily doses of streptomycin (150 to 600 mg/day) for 1 to 7 days. The treatment end point was the onset of signs and symptoms of vestibular ablation. The treatment was successful in controlling vertigo in five of eight patients, all of whom lost substantial hearing in the treated ear. Although hearing was preserved in the remaining three of eight patients, persistent vertigo necessitated surgical labyrinthectomy. Schuknecht28,29 concluded that complete control of vertigo with intratympanic streptomycin required abolition of ice water caloric responses, and that when this was accomplished, hearing was also lost. Lange53,54 reported an extensive experience using various aminoglycosides to treat Meniere’s disease. In his group of 92 patients, Lange reported that 90% had no further episodes of severe vertigo, and hearing remained unchanged in 76% of patients. Many of the studies reported from 1956 to 1990 did not use standardized reporting methods (Table 41-1). In a study to characterize the delayed effects of intratympanic gentamicin, Magnusson and Padoan55 treated five patients with two doses of gentamicin (30 mg/mL, pH 6.4), given 12 hours apart. The first symptom of an ototoxic reaction noted by the patients was a sensation of unsteadiness occurring 2 to 5 days (mean 3.2 days) after the injections. Vertigo and nystagmus were noted 3 to 8 days (mean 5.1 days) after the injections. With 1 year of follow-up, vertigo was controlled (with a loss of caloric responsiveness in the treated ear), and hearing was preserved in all five patients.55 This preliminary study raised the question of whether very low intratympanic doses of gentamicin may be effective in controlling vertigo and preserving hearing. Because this technique did not totally ablate vestibular function, vertigo may recur, but hearing was usually preserved. Additional gentamicin could be used if vertigo recurred. Beck and Schmidt56 administered gentamicin, 30 mg/ day, and stopped after 6 days or at the slightest indication of ototoxicity. In 40 patients treated with their regimen, control of vertigo occurred in 92.5% of patients; 85% of patients maintained hearing. Following the work of Beck and Schmidt, several investigators developed the hypothesis that if a “titration” dosage schedule were adopted, it might be possible to achieve satisfactory rates of control of vertigo while preserving hearing in more patients. Much of the ongoing controversy about intratympanic gentamicin treatment has been over the issue of whether such approaches achieve this goal, and how various protocols compare with each other. Studies by several authors showed remarkably similar results for control of vertigo, ranging from 86% to 93% of patients treated (Table 41-2; see Table 41-1). Some patients required additional gentamicin injections or surgery. Authors have noted an association between loss of the ice water caloric response and complete control of vertigo.57,58
498
TABLE 41-1 Review of Literature on Intratympanic Aminoglycoside Therapy Aminoglycoside Used
No
63
63
37
Vestibular ablation
No
91
NR
42
First ototoxic reaction First ototoxic reaction
No No
92 90
NR NR
85 76
First ototoxic reaction First ototoxic reaction Ablative nystagmus or hearing loss Ablative nystagmus
No No No
93 86 89
100 51 28
66 70 67
Yes*
90
70
Improved—5
62
Unchanged—50 Worse—45 (profound—30) Improved—24
Beck and Schmidt56
43
Gentamicin
40 92 15 62 82
Gentamicin Streptomycin, tobra mycin, gentamicin Gentamicin 30 mg/mL Gentamicin 30 mg/mL Gentamicin 40 mg/mL
“6 doses planned” 60 mg/day “typically several days” 15-30 mg/dose ≤30 mg/day; 1-8 doses 5-40 mg/dose 1-2 days
20
Gentamicin 40 mg/mL
0.2 mL/day × 3 days, then qod for 3-12 doses
Kaplan et al87
Youssef and Poe57
Minor83
20
37
34
Gentamicin 26.7 mg/mL
Gentamicin 30 mg/mL
Gentamicin
3 times per day for 4 days
2-4 doses once/wk, then re assessment; 1-8 doses total
1-6 weekly doses
12 doses or first ototoxic reaction
Loss of hearing (priority) or relief of vertigo
First sign of vestibular hypofunction
Yes
93
Yes
A—41
Yes
B—46 C D Failed—14 A—74
23 (n = 26)
68
Gentamicin 26 mg/mL
1-4 weekly doses
4 doses or first ototoxic reaction
Yes*
D—6 A—84
20
*Not
93
Gentamicin 30 mg/mL
1-4 daily doses
Fixed doses
all patients had 24 month follow-up. for vertigo control were not reported. AAO-HNS, American Academy of Otolaryngology–Head and Neck Surgery; NR, not reported; qod, every other day. †Results
Yes†
D—9 NR
Improved—36 Unchanged—32 Worse—32 (profound—3)
45
B—6 C—1
Eklund et al84
Unchanged—50 Worse—26 Improved—14 Unchanged—23 Worse—63 (n = 35)
B—17 C—3
Atlas and Parnes58
Hearing Preserved (%)
Vestibular ablation
50-300 mg/dose; 350-600 mg total dose 30 mg/day
Moller et al79 Sala80 Blessing and Schlenter81 Laitakari82
Loss of Caloric Response (%)
End Point
8
Lange53
Control of Vertigo (%)
Dosage
Schuknecht29
Streptomycin
AAO-HNS Guidelines Used?
Improved—40 Unchanged—47 Worse—13 (profound—0) (n = 47)
NR
Improved—15 Unchanged—47 Worse—38 (profound—11)
OTOLOGIC SURGERY
Author
No. Patients Treated
Chapter 41 • Chemical Treatment of the Labyrinth
499
TABLE 41-2 Five Studies Using AAO-HNS REPORTING GUIDELINES (Minimum of N = 25) Study
N
Dose
Hirsch and Kammerer85 Youssef and Poe57 Atlas and Parnes58 Kaasinen et al86 Kaplan et al87
28 37 68 93 90
15 mg × 1-9 30 mg × 1-8 13-26 mg × 1-8 12-20 mg × 1-4 26.7 mg × 12 (17 retreated)
Protocol
Frequency of Treatment
Vertigo ontrol (%) C
T T T T F
Weekly/biweekly Weekly Weekly Daily 3 times daily
91 87 90 81 93
Hearing Loss (Deafened) (%) 31 (?) 28 (3) 17 (0) 40 (10) 23 (16)
AAO-HNS, American Academy of Otolaryngology–Head and Neck Surgery.
FIGURE 41-1. Components required for catheter assembly for instillation of gentamicin include butterfly intravenous tubing set, T-type tympanostomy tube, and intravenous catheter set. Two Pope wicks are used to stabilize the tubing in the external auditory canal.
Reports showed considerable variation in rates of hearing preservation, ranging from 55% to 85%. The results suggest that the administration schedule and total dosage may be significant factors in hearing preservation, and that low-dose methods may be associated with a lower incidence of hearing loss.59 The variability in reports may also be due to differences in case selection because some series have more elderly patients or patients with worse pretreatment hearing levels than other series. It is suspected, but has not been well established, that there may be a tradeoff between control of vertigo and preservation of hearing. One of the authors (J.M.N.) has developed a fixeddose protocol, which has been applied with consistent technique since 1988.60-62 A catheter assembly is constructed (Figs. 41-1 and 41-2). Gentamicin is buffered to a pH of 6.4 with sodium bicarbonate such that the final concentration is 26.7 mg/mL. The T-tube assembly is inserted into the middle ear so that the solution can easily access the round window niche. This procedure is done in the outpatient clinic under local anesthesia. If possible, the � T-tube is inserted into the posteroinferior quadrant of the tympanic membrane. Three doses of approximately 0.7 mL of gentamicin are instilled daily for a total of 12 doses over 4 days. A family member or friend is instructed how to evacuate the catheter tubing before instilling the solution each morning and evening. Patients are seen in the office daily during treatment. The
FIGURE 41-2. The� T-tube end of the assembly is inserted through the tympanic membrane. Tubing in external auditory canal is stabilized using Pope wicks, and tubing is anchored to skin of external meatus and upper neck using adhesive tape.
mid-day dose is administered during the office encounter, but only if there has been no been no history of vertigo, and no nystagmus is observed. The catheter and gentamicin solution produce a mild conductive hearing loss during the period of the instillations. Early in the series, a daily bone conduction audiogram was performed; however, this is now done at completion of the treatment. The patient is seen on the morning of the 5th day, the catheter assembly is removed, and an audiogram is performed. The treatment is generally carried out over a Monday-to-Friday interval. The patient is advised that vertigo rarely occurs sooner than 2 to 3 days after completion of treatment. The patient is also seen by a physiotherapist who has a special interest in vestibular rehabilitation, and a suitable exercise program is recommended before discharge on day 5. All patients have been studied prospectively using the AAO-HNS guidelines for reporting treatment results in Meniere’s disease.4,44 All patients have electronystagmography before treatment. An audiogram and an electronystagmogram are performed 1 month after treatment. Comparison of the caloric response before and after treatment is viewed as a “barometer” of peripheral vestibular loss. This protocol has produced consistent results. The first publication in 199262 has been followed by several others. In the initial cohort of 30 patients, complete control (vertigo class A) of vertigo was obtained in 83%, and
500
OTOLOGIC SURGERY
substantial control (vertigo class B) was obtained in 17%. Hearing worsened in 27%, and 13% of patients sustained a profound hearing loss. An ice water caloric response was abolished in 53% of individuals. A subsequent study60 consisting of 90 patients reported complete vertigo control in 85% and substantial control in 9%. Hearing loss occurred in 23%; 16% patients became deaf. The most recent study has shown the effectiveness of long-term vertigo control using this protocol. Patients followed for 5 years or longer (15 year mean interval) have continued to be consistently free of vertigo. One of the authors (S.P.C.) uses a low-dose titration method, which is described here.63 Before performing an ablative procedure, a full diagnostic assessment is recommended, including a complete history, physical examination, neurotologic examination, vestibular function testing (electronystagmography at a minimum, posturography and rotational chair testing if available), and
FIGURE 41-3. Patient positioning for intratympanic injection of gentamicin.
audiometric evaluation. The neurotologic examination should include a search for spontaneous nystagmus (eyes open in light and dark), nystagmus after head shaking, and the head thrust test. Retrocochlear and metabolic disorders should be excluded. Because of the low total dose of gentamicin used, the need to withhold treatment based on abnormal renal function is rare. Patients are counseled that the purpose of the intratympanic gentamicin injections is to control recurrent episodes of vertigo typical of Meniere’s disease. It is important to explain the expected time course of posttreatment unsteadiness and dysequilibrium typical of unilateral vestibular ablation. The possibility that additional courses of gentamicin may be needed in the future, or that surgical ablation may be required to control vertigo is reviewed. The possibility of increased hearing loss, including a profound loss of hearing that would be unaidable, should be explained. Low-dose titration chemical labyrinthectomy is performed by placing the patient in the supine position with the head turned 45 degrees away from the affected ear (Fig. 41-3). The eardrum is anesthetized with a small dot of topical phenol. Gentamicin (40 mg/mL) is injected using a tuberculin syringe with a 25 gauge 11⁄2 inch needle; approximately 0.2 to 0.5 mL is injected into the middle ear space (Fig. 41-4). The patient is asked to remain for 30 minutes in the supine position and to refrain from excessive swallowing, which may contribute to early evacuation of the drug through the eustachian tube. One week after the initial gentamicin injection, an interim evaluation is performed, focused on detecting the onset of signs and symptoms of an aminoglycosidemediated ablative effect. The patient is questioned about the new onset of movement-associated ataxia or dysequilibrium. It is usually quite easy to distinguish between episodes of vertigo caused by Meniere’s disease and the
FIGURE 41-4. Intratympanic injection of gentamicin through pinhole perforation. Gentamicin fills middle ear space and bathes round window (RW) membrane with antibiotic solution.
501
Chapter 41 • Chemical Treatment of the Labyrinth
movement-associated ataxia or dysequilibrium characteristic of the onset of vestibular ablation. Patients often report these symptoms as distinctive. A physical examination that includes a vestibuloocular and vestibulospinal examination and audiometric testing is performed. The history, physical examination, and auditory testing are used to assess whether or not signs and symptoms of aminoglycoside-mediated toxicity are present. The physical examination includes an observation for spontaneous vestibular nystagmus with eyes open in the dark using infrared video goggles and head thrust testing. By performing these tests before and during the process of chemical labyrinthectomy, the new onset of vestibular ototoxicity may be reliably detected. At the first follow-up visit, a treatment decision must be made. This decision is based on whether or not there are signs and symptoms of aminoglycoside-mediated vestibular toxicity, whether or not there are signs of early auditory toxicity, and whether or not the episodes of acute vertigo have ceased. Based on the answers to these three questions, a decision is made to continue treatment or to defer additional injections. This treatment decision is individualized. If a patient has poor hearing before treatment and is not concerned about additional auditory toxicity, the occurrence of auditory toxicity is not a contraindication to additional injections. Patients with preexisting severe hearing loss are more likely to be treated with more injections than a patient with good pretreatment hearing. The final arbitrator is usually whether or not there is control of acute episodic vertigo. Through this iterative process, small, intermittent increments of middle ear injections of gentamicin are “titrated” to control vertigo. At the completion of the initial treatment course, typically 6 weeks after the last injection, vestibular function and audiometric testing are repeated.
RESULTS A total of 66 patients have been treated with 2 years or greater follow-up. The average age of treated patients is 60 years (range 34 to 89 years). The most common number of injections given is two (Table 41-3). The wide span in the number of injections reflects the individualized nature of the titration therapy. Eighty percent of patients
have had complete control of vertigo; 94% have complete or substantial vertigo control (Table 41-4). Pretreatment and post-treatment caloric testing was available for analysis in 48 patients. An absent ice water caloric response was found in 88% of patients. Bithermal caloric responses or a positive ice water response was found in 12% of patients. Three of five retreated patients and all three patients who required surgery for salvage had positive ice water response on caloric testing. Hearing results are presented according to the pretreatment Gardner-Robinson Hearing Scale (Table 41-5).64 In the Gardner-Robinson scale, patients with grade A hearing have a speech reception threshold (SRT) less than 30 dB and speech recognition greater than 70%. Patients with grade B have SRT less than 50 dB and speech recognition greater than 50%. Patients with grade C have SRT greater than 50 dB but less than 70 dB and speech recognition greater than 30% but less than 50%. Patients with grade D hearing have SRT greater than 70 dB and speech recognition less than 30%. Overall, using 10 dB SRT and 15% discrimination criteria, hearing was worse in 45% of patients; change in Gardner-Robinson grade occurred in 35%; loss of serviceable hearing (50 dB SRT and 50% speech recognition) occurred in 25%; and profound loss of hearing occurred in 3%. Table 41-2 shows a comparison of the results from the fixed-dose protocol versus five studies following the clinical research reporting guidelines of the AAO-HNS for Meniere’s disease4 and with at least 25 subjects. The results are generally the same for control of vertigo among these studies, and are about the same as results for control of vertigo by selective vestibular nerve section.
TABLE 41-4 Vertigo Control Vertigo Control Grade (Numeric Value)
No. Patients (%)
A (complete) B (1-40) C (41-80) D (81-120) E (≥121) F (surgical treatment initiated) *5
53 (80%) 9 (14%)* 1 (1.5%) 0 0 3 (4.5%)
patients with class B results were retreated.
TABLE 41-5 Hearing Results TABLE 41-3 Gentamicin Injections No. Injections 1 2 3 4 ≥5
No. Patients 10 24 12 11 9
Pretreatment Grade A (n = 10) B (n = 30) C (n = 6) D (n = 20)
Post-Treatment Grade (N = 66) A
B
C
D
60% 10%
40% 57% 33% 10%
10% 66% 10%
23% 33% 80%
502
OTOLOGIC SURGERY
The results for hearing loss are also generally the same, although more patients may be deafened by the fixed dosage protocol. Most of the patients deafened had poor hearing before treatment. The advantage of the fixed dosage protocol is that it can be completed in less time and with fewer clinical visits for testing and evaluation than the titration protocols. Because intratympanic aminoglycoside therapy is a nonsurgical outpatient treatment, elderly patients and patients with significant surgical or anesthetic risks may be treated safely and are primary candidates. Intratympanic aminoglycoside therapy can also be used to avoid additional surgery in patients with recurrent vertigo and a return of caloric function after previous ablative vestibular surgery. Patients with profound hearing loss are good candidates for intensified treatment. The data on efficacy of intratympanic aminoglycoside therapy to control vertigo and its safety in preservation of hearing are almost exclusively from patients with Meniere’s disease. Most patients with non–Meniere’s disease vestibulopathy have normal hearing. The nonhydropic ear seems to be relatively resistant to the effects of aminoglycosides. For these reasons, we do not currently recommend intratympanic aminoglycoside therapy in the primary treatment of vertigo caused by disorders other than Meniere’s disease.
TREATMENT OF LABYRINTHINE DISORDERS WITH INTRATYMPANIC CORTICOSTEROIDS Experimental Studies Various components of the inflammatory and immune responses are inhibited by glucocorticoids. Macrophages, monocytes, and endothelial cells produce arachidonic acid and its metabolites, which include prostaglandins and leukotrienes. The production of these metabolites is inhibited partly by the induction of a protein, lipocortin, that is an inhibitor of phospholipase A2, the first enzyme in the cascade of prostaglandins and leukotrienes. Macrophages and monocytes also release cytokines, including interleukins and tumor necrosis factor-α. These substances have multiple effects on the inflammatory response, including the activation of T cells and stimulation of the growth of fibroblasts. The action of basophils is also affected by glucocorticoids. They inhibit the IgE-dependent release of histamine and leukotriene C4. The production of fibroblasts and the release of arachidonic acid derivatives are inhibited by glucocorticoids. Similar to macrophages and monocytes, lymphocytes produce cytokines and tumor necrosis factor-α, which are inhibited by steroids. These multiple effects all would serve to attenuate an inflammatory or autoimmune process in the middle and inner ear when glucocorticoids are administered intratympanically. The properties of glucocorticoids vary. Cortisone and prednisone have an 11-keto group, which must be
e nzymatically reduced in the liver before they are biologically active.53 Steroids such as prednisone and methylprednisolone have a shorter half-life than dexamethasone or betamethasone. In addition, the latter steroids are about seven times as potent as prednisone and prednisolone, and about five times as potent as methylprednisolone.53 Animal studies have shown that conditioning by sound to the harmful effects of noise exposure depends on the pituitary-hypothalamic-adrenal axis.65 Glucocorticoid and mineralocorticoid receptors have been identified in inner ear tissue.66 Endogenous steroids may stabilize junctions in the stria vascularis,67 and up-regulate the synthesis of glutathione, an endogenous antioxidant, in the spiral ganglion.68 Although clinical evidence is lacking, results have suggested the possibility that systemic steroids may protect against inner ear trauma during middle ear surgery and cochlear implantation.67,69
Clinical Studies There is substantial clinical evidence that some otologic conditions are autoimmune in etiology and responsive to systemic immunosuppressive medications.36 Systemic corticosteroids have been shown to improve the outcome for hearing in bacterial meningitis70 and autoimmune inner ear disease,71 and have become accepted treatment for these conditions. In contrast, a more recent review cast doubt on the efficacy of systemic corticosteroids in sudden hearing loss.72 Immunosuppressants have undesirable systemic side effects, which limit their use. Consequently, there is considerable interest in applying immunosuppressive drugs, principally corticosteroids, intratympanically. Much less is known about the pharmacokinetics and clinical effects of intratympanic corticosteroid administration than is known about aminoglycosides. It may be possible to achieve high concentrations in some otic tissues, but safety and efficacy in otologic disorders remain unproven. Intratympanic treatment of sudden hearing loss and Meniere’s disease with corticosteroids has been the subject of retrospective, uncontrolled studies and a few controlled studies with small numbers of subjects. Complications of infection, tympanic membrane perforation, and sensorineural hearing loss have been reported.73-75 One prospective, randomized, doubleblind, crossover trial showed no benefit compared with placebo in Meniere’s disease.76 Larger, prospective, controlled studies with more rigorous attention to study methods are needed to evaluate the potential benefit and harms of intratympanic corticosteroid treatment.77 Rescue trials of intratympanic treatment after failure of oral steroid therapy also must be controlled because recovery may occur several months after onset of sudden hearing loss following oral corticosteroid therapy.78 A large-scale clinical trial funded by the National Institutes of Health is currently under way to assess the effectiveness of intratympanic steroid
Chapter 41 • Chemical Treatment of the Labyrinth
t reatment in sudden sensorineural hearing loss (www. nidcd.nih.gov). Future studies may define safe and effective protocols for intratympanic treatment with corticosteroids and perhaps other medications in certain subsets of patients.
SUMMARY Subtotal bilateral chemical labyrinthectomy by intramuscular streptomycin administration is an accepted treatment for disabling vertigo resulting from bilaterally active Meniere’s disease or Meniere’s disease in an only hearing ear. Long-term results show preservation of hearing and control of vertigo in most cases, although chronic dysequilibrium and oscillopsia remain as problems for some patients. Consequently, this treatment should be reserved for patients who are disabled and meet criteria for treatment. Intratympanic application of gentamicin frequently results in treatment-related hearing loss, although episodic vertigo is usually well controlled. Control rates are comparable to control rates for selective vestibular nerve section. Patients with recurrent post-treatment vertigo can be treated with intratympanic gentamicin again. A current trend in intratympanic application is to administer one or two doses, rather than treating until ototoxicity is clinically evident. Just enough additional doses are applied to achieve control of attacks. The advantages of such titration protocols over fixed-dosage protocols have not been proven in long-term studies. Intratympanic gentamicin should be reserved for patients with classic Meniere’s disease who meet appropriate treatment criteria. Prolonged disabling dysequilibrium occurs about as frequently as after surgical labyrinthectomy or vestibular nerve section. Complete ablation and long-term control of vertigo are not achieved as reliably as with surgical labyrinthectomy.57 The safety and efficacy of corticosteroid intratympanic treatment of Meniere’s disease and other inner ear disorders are undefined at this time.
REFERENCES 1. Yamashita D, Jiang HY, Le Prell CG, et al: Postexposure treatment attenuates noise-induced hearing loss. Neuroscience 134:633-642, 2005. 2. Dickey DT, Muldoon L L , Doolittle N D, et al: Effect of N-acetylcysteine route of administration on chemoprotection against cisplatin-induced toxicity in rat models. Cancer Chemother Pharmacol 62:235-241, 2007. 3. Chen Y, Huang WG, Zha DJ, et al: Aspirin attenuates gentamicin ototoxicity: From the laboratory to the clinic. Hear Res 226:178-182, 2007. 4. Committee on Hearing and Equilibrium: Committee on Hearing and Equilibrium guidelines for the diagnosis and evaluation of therapy in Ménière’s disease. Otolaryngol Head Neck Surg 113:181-185, 1995.
503
5. Sande M , Mandell G : Antimicrobial agents. In G ilman A, Rall T, Nies A, Taylor P (eds): Goodman and Gilman’s the Pharmacological Basis of Therapeutics, 8th ed. New York, Pergamon Press, 1999, pp 1098–1116. 6. Fowler E : Streptomycin treatment of vertigo. Trans Am Acad Ophth Otolaryngol 52:239-301, 1948. 7. Selimoglu E : Aminoglycoside-induced ototoxicity. Curr Pharm Des 13:119-126, 2007. 8. Ohmori H : Mechano-electrical transduction currents in isolated vestibular hair cells of the chick. J Physiol 359:189-217, 1985. 9. Marcotti W, van Netten S, Kros C : The aminoglycoside antibiotic dihydrostreptomycin rapidly enters mouse outer hair cells through the mechano-electrical transducer channels. J Physiol 567:505-521, 2005. 10. Waguespack J, Ricci A : Aminoglycoside ototoxicity: Permeant drugs cause permanent hair cell loss. J Physiol 567:359-360, 2005. 11. Rybak L : Ototoxic mechanisms. In Altschuler R A, Hoffman DW, Bobbin R P (eds): Neurobiology of Hearing. New York, Raven Press, 1986, pp 441–454. 12. Williams S, Smith D, Schacht J: Characteristics of gentamicin uptake in the isolated crista ampullaris of the inner ear of the guinea pig. Biochem Pharmacol 36: 89-95, 1987. 13. Williams S, Zenner H, Schacht J: Three molecular steps of aminoglycoside ototoxicity demonstrated in outer hair cells. Hear Res 30:11-18, 1987. 14. Lesniak W, Pecoraro V, Schacht J: Ternary complexes of gentamicin with iron and lipid catalyze formation of reactive oxygen species. Chem Res Toxicol 18:357-364, 2005. 15. Jacotot E, Costantini P, Laboureau E, et al: Mitochondrial membrane permeabilization during the apoptotic process. Ann N Y Acad Sci 887:18-30, 1999. 16. Ylikoski J, Xing-Qun L , Virkkala J, Pirvola U: Blockade of c-JUN N-terminal kinase pathway attenuates gentamicin-induced cochlear and vestibular hair cell death. Hear Res 166:33-43, 2002. 17. Song B -B, Anderson D, Schacht J: Protection from gentamicin toxicity by iron chelaters in guinea pig in vivo. J Pharmacol Exp Ther 282:369-377, 1999. 18. Lecain E, Omri B, Behar-Cohen F, et al: The role of PKCzeta in amikacin-induced apoptosis in the cochlea: Prevention by aspirin. Apoptosis 12:333-342, 2007. 19. Sha S -H, Qiu J-H, Schacht J: Aspirin attenuates gentamicin-induced hearing loss. N Engl J Med 354:18561857, 2006. 20. Henley C I, Schacht J: Pharmacokinetics of aminoglycoside antibiotics in inner ear fluids and their relationship to ototoxicity. Audiology 27:137-147, 1988. 21. Aran J, Chappert C, Dulon D, et al: Uptake of amikacin by hair cells of the guinea pig cochlea and vestibule and ototoxicity in comparison with gentamicin. Hear Res 82:179-183, 1999. 22. McGee T, Olszewski J: Streptomycin sulfate and dihydrostreptomycin toxicity. Arch Otolaryngol 75:295-311, 1962. 23. Balough B, Hoffer M, Wester D, et al: Kinetics of gentamicin uptake in the inner ear of chinchilla langier after middle-ear administration in a sustained-release vehicle. Otolaryngol Head Neck Surg 119:427-431, 1998.
504
OTOLOGIC SURGERY
24. Berg K : The toxic effect of streptomycin on the vestibular and cochlear apparatus. Acta Otolaryngol 157:1-77, 1951. 25. Rodriguez-Ballesteros M, Olarte M, Aguirre L S, et al: Molecular and clinical characterisation of three Spanish families with maternally inherited non-syndromic hearing loss caused by the 1494C→T mutation in the mitochondrial 12S rRNA gene. J Med Genet 43:e54, 2006. 26. Park J, Cohen G: Vestibular ototoxicity in the chick: Effects of streptomycin on equilibrium and on ampullary dark cells. Am J Otolaryngol 6:117-127, 1982. 27. Pender D: Gentamicin tympanoclysis: Effects on the vestibular secretory cells. Am J Otolaryngol 6:358-367, 1985. 28. Schuknecht H : Ablation therapy for the relief of Ménière’s disease. Laryngoscope 66:859-870, 1956. 29. Schuknecht H : Ablation therapy in the management of Ménière’s disease. Acta Otolaryngol (Stockh) Suppl 132:1-42, 1957. 30. Wilson W, Schuknecht S : Update on the use of streptomycin therapy for Ménière’s disease. Am J Otolaryngol 2:108-111, 1980. 31. “J.C.”: Living without a balance mechanism. N Engl J Med 246:458-460, 1952. 32. Graham M, Kemink J: Titration streptomycin therapy for bilateral Ménière’s disease: A progress report. Otolaryngol Head Neck Surg 92:440-447, 1984. 33. Silverstein H : Streptomycin treatment for Ménière’s disease. Ann Otol Rhinol Laryngol Suppl 112:44-88, 1984. 34. Konrad H : Intractable vertigo—when not to operate. Trans Am Acad Ophthalmol Otolaryngol 95:482-484, 1986. 35. Monsell E, Brackmann D, Linthicum F: Why do vestibu lar destructive procedures sometimes fail? Otolaryngol Head Neck Surg 99:472-479, 1988. 36. Harris J, Tomiyama S : Immunology/virology of Ménière’s disease. In Harris J (ed): Ménière’s Disease. The Hague, Kugler Publications, 1999, pp 123–138. 37. Langman A, Kemink J, Graham M : Titration therapy for bilateral Ménière’s disease, follow-up report. Ann Otol Rhinol Laryngol 99:923-926, 1990. 38. Kimura R , Iverson N, Southard RE : Selective lesions of the vestibular labyrinth. Ann Otol Rhinol Laryngol PMID: 3264488 [PubMed-indexed for MEDLINE]; 97 (6 Pt 1):577-584, 1988 Nov-Dec. 39. Kimura R , Lee K -S, Nye C, Trehey J: Effects of systemic and lateral semicircular canal administration of aminoglycosides on normal and hydropic inner ears. Acta Otolaryngol (Stockh) 111:1021-1030, 1991. 40. Shea J, Norris C : Streptomycin perfusion of the labyrinth. In Nadol J B (ed): Second International Symposium on Ménière’s Disease. Cambridge, MA, Amsterdam, Kugler & Ghendini Publications, June 20-22, 1988, 1989. 41. Shea J: Perfusion of the inner ear with streptomycin. Am J Otol 10:150-155, 1989. 42. Monsell E : Electrocochleographic recording in patients undergoing labyrinthotomy with streptomycin infusion. In Arenberg I K (ed): Surgery of the Inner Ear, Proceedings of the Third International Symposium and Workshops on Surgery of the Inner Ear. Snowmass, CO, Amsterdam, Kulger & Ghedini, July 29-August 4, 1990, 1991.
43. Monsell E, Shelton C : Labyrinthotomy with streptomycin infusion: Early results of a multicenter study. Am J Otol 13:416-422, 1992. 44. Subcommittee on Equilibrium: Ménière’s disease: Criteria for diagnosis and evaluation of therapy for reporting. AAO-HNS Bull 7:6-7, 1985. 45. Smith B, Myers M: The penetration of gentamicin and neomycin into the perilymph across the round window membrane. Otolaryngol Head Neck Surg 87:888-891, 1978. 46. Saijo S, Kimura R : Distribution of HRP in the inner ear after injection into the middle ear cavity. Acta Otolaryngol (Stockh) 97:593-610, 1999. 47. Goycoolea M, Carpenter A, Muchow D: Ultrastructural studies of the round-window membrane of the cat. Arch Otolaryngol 113:617-624, 1987. 48. Kawauchi H, DeMaria T, Lim D: Endotoxin permeability through the round window. Acta Otolaryngol (Stockh) Suppl 457:100-115, 1988. 49. Lundman L , Bagger-Sjöbäck D, Holmquist L , Juhn S : Round window membrane permeability. Acta Otolaryngol (Stockh) Suppl 457:73-77, 1988. 50. Jahnke K : Transtympanic application of gentamicin with cochlea protection. In Nadol J B (ed): Second International Symposium on Ménière’s Disease. Cambridge, MA, Amsterdam, Kugler & Ghendini Publications, June 20-22, 1988, 1989. 51. Lindeman H : Regional differences in sensitivity of the vestibular sensory epithelia to ototoxic antibiotics. Acta Otolaryngol (Stockh) 67:177-189, 1969. 52. Cass S, Bouchard K, Graham M : Controlled application of streptomycin to the round window membrane of the cat. Otolaryngol Head Neck Surg 103:223, 1990. 53. Lange G: Gentamicin and other ototoxic antibiotics for the transtympanic treatment of Ménière’s disease. Arch Otorhinolaryngol 246:269-270, 1989. 54. Lange G: Isolierte Medikamentose Ausschaltungeines Gleichgewichtsorganes beim Morbus Ménière mit Streptomycin-Ozothin. Arch Klin Exp Ohren-NasenKehlkopfheilkd 191:545-549, 1999. 55. Magnusson M, Padoan S : Delayed onset of ototoxic effects of gentamicin in treatment of Ménière’s disease. Acta Otolaryngol 111:671-676, 1991. 56. Beck C, Schmidt C : Ten years of experience with intratympanically applied streptomycin (gentamicin) in the therapy of morbus Ménière. Arch Otorhinolaryngol 221:149-152, 1978. 57. Youssef T, Poe D: Intratympanic gentamicin injection for the treatment of Ménière’s disease. Am J Otol 19:435442, 1998. 58. Atlas J, Parnes L : Intratympanic gentamicin titration therapy for intractable Ménière’s disease. Am J Otol 20:357-363, 1999. 59. Chia S H, Gamst AC, Anderson J P, Harris J P: Intratympanic gentamicin therapy for Ménière’s disease: A metaanalysis. Otol Neurotol 25:544-552, 2004. 60. Commins D, Nedzelski J: Topical drugs in the treatment of Ménière’s disease. Curr Opin Otolaryngol Head Neck Surg 4:319-323, 1996. 61. Hone S, Nedzelski J: Selective chemical ablation as treatment for Ménière’s disease. In Harris J (ed): Ménière’s Disease. The Hague, Kugler Publications, 1999, pp 381–389.
Chapter 41 • Chemical Treatment of the Labyrinth 62. Nedzelski J, Schessel D, Bryce G, Pfleiderer A : Chemical labyrinthectomy: Local application for the treatment of unilateral Ménière’s disease. Am J Otol 13:18-22, 1992. 63. Cass S : Chemical labyrinthectomy using intratympanic gentamicin for treatment of disabling vertigo associated with Ménière’s disease. In Lim D (ed): Ménière’s Disease and Inner Ear Homeostasis Disorders. Los Angeles, House Ear Institute, 2005. 64. Gardner G, Robertson J H : Hearing preservation in unilateral acoustic neuroma surgery. Ann Otol Rhinol Laryngol 97:55-66, 1988. 65. Tahera Y, Meltser I, Johansson P, et al: Sound conditioning protects hearing by activating the hypothalamicpituitary-adrenal axis. Neurobiol Dis 25:189-197, 2007. 66. Trune D R , Kempton J B, Gross N D: Mineralocorticoid receptor mediates glucocorticoid treatment effects in the autoimmune mouse ear. Hear Res 212:22-32, 2006. 67. Hashimoto K, Seki M, Miyasaka H, Watanabe K : Effect of steroids on increased permeability of blood vessels of the stria vascularis after auditory ossicle vibration by a drill in otologic surgery. Ann Otol Rhinol Laryngol 115:769-774, 2006. 68. Nagashima R , Ogita K : Enhanced biosynthesis of glutathione in the spiral ganglion of the cochlea after in vivo treatment with dexamethasone in mice. Brain Res 117:101-108, 2006. 69. Ye Q, Tillein J, Hartmann R , et al: Application of a corticosteroid (Triamcinolon) protects inner ear function after surgical intervention. Ear Hear 28:361-369, 2007. 70. van de Beek D, de Gans J, McIntyre P, Prasad K : Corticosteroids for acute bacterial meningitis. Cochrane Database Syst Rev 1:CD004405, 2007. 71. Harris J P, Weisman M H, Derebery J M, et al: Treatment of corticosteroid-responsive autoimmune inner ear disease with methotrexate: A randomized controlled trial. JAMA 290:1875-1883, 2003. 72. Conlin A E, Parnes L S : Treatment of sudden sensorineural hearing loss, II: A meta-analysis. Arch Otolaryngol Head Neck Surg 133:582-586, 2007. 73. Shulman A, Goldstein B : Intratympanic drug therapy with steroids for tinnitus control. A preliminary report. Int Tinnitus J 6:10-20, 2000.
505
74. Herr B, Marzo S J: Intratympanic steroid perfusion for refractory sudden sensorineural hearing loss. Otolaryngol Head Neck Surg 132:527-531, 2005. 75. Spandow O, Hellstrom S, Anniko M : Impaired hearing following instillation of hydrocortisone into the middle ear. Acta Otolaryngol (Stockh) Suppl 455:90-93, 1988. 76. Silverstein H, Isaacson J E, Olds M J, et al: Dexamethasone inner ear perfusion for the treatment of Ménière’s disease: A prospective, randomized, double-blind, crossover trial. Am J Otol 19:196-201, 1998. 77. Doyle K J, Bauch C, Battista R , et al: Intratympanic steroid treatment: A review. Otol Neurotol 25:1034-1039, 2004. 78. Slattery WH, Fisher L M, Iqbal Z, Liu N: Oral steroid regimens for idiopathic sudden sensorineural hearing loss. Otolaryngol Head Neck Surg 132:5-10, 2005. 79. Moller C, Odkvist L , Thell J, et al: Vestibular and audiologic functions in gentamicin-treated Ménière’s disease. Am J Otolaryngol 9:383-391, 1988. 80. Sala T: Transtympanic administration of aminoglycosides in patients with Ménière’s disease. Arch Otorhinolaryngol 245:293-296, 1988. 81. Blessing R , Schlenter W: Langzeitergebnisse der Gentamicin-Therapie Des Morbus Ménière. Laryngorhinootology 68:657-660, 1989. 82. Laitakari K : Intratympanic gentamicin in severe Ménière’s disease. Clin Otolaryngol 15:545-548, 1990. 83. Minor L : Intratympanic gentamicin for control of vertigo in Ménière’s disease: Vestibular signs that specify completion of therapy. Am J Otol 20:209-219, 1999. 84. Eklund S, Pyykko I, Aalto H, et al: Effect of intratympanic gentamicin on hearing and tinnitus in Ménière’s disease. Am J Otol 20:356, 1999. 85. Hirsch B, Kammerer D: Intratympanic gentamicin therapy for Ménière’s disease. Am J Otol 18:44-51, 1997. 86. Kaasinen S, Pyykko I, Ishizaki H, Aalto H : Intratympanic gentamicin in Ménière’s disease. Acta Otolaryngol 118:294-298, 1998. 87. Kaplan D, Nedzelski J, Chen J, Shipp D: Intratympanic gentamicin for treatment of unilateral Ménière’s disease. Laryngoscope 110:1298-1305, 2000.
42
Superior Semicircular Canal Dehiscence Syndrome Benjamin T. Crane, John P. Carey, and Lloyd B. Minor
Superior canal dehiscence syndrome (SCDS) has only more recently been described, but during the decade since the initial description,1 the pathophysiology has been elucidated, and curative surgery is now regularly performed at several centers throughout the world. SCDS is characterized by the clinical findings of sound-induced vertigo and eye movements, chronic dysequilibrium, conductive hearing loss, and decreased hearing thresholds for bone conducted sounds, so that patients may experience autophony and may even hear their pulse or eye movements. Symptoms caused by abnormal openings into the labyrinth have been known for decades. Fenestration of the semicircular canals was known to produce eye movements in response to sound in animals 80 years ago.2 The Tullio phenomenon, or eye movements in response to loud sound, was initially identified in humans with advanced syphilis secondary to gummatous osteomyelitis and labyrinthine fistulas.3 In patients with the Tullio phenomenon, the Hennebert sign (eye movement induced by pressure in the external auditory canal) is also often present. Subsequent reports have identified the Tullio phenomenon in perilymphatic fistula,4 head trauma,5 and cholesteatoma with semicircular canal erosion and fenestration.6 Today it is recognized that all of these causes of the Tullio phenomenon are uncommon compared with SCDS. The incidence of superior canal dehiscence (SCD), meaning only the anatomic abnormality, regardless of the symptomatic status, has been estimated from a temporal bone library.7 This study of 1000 temporal bones revealed a 0.5% incidence of complete dehiscence of the superior canal into the middle fossa or superior petrosal sinus. In an additional 1.4% of specimens, the bone was 0.1 mm or thinner. The incidence of SCDS is not known with certainty, but it is likely that only a subset of patients with SCD actually experience symptoms.
DIAGNOSTIC EVALUATION With any patient presenting with complaints of dizziness, a good history is the most effective diagnostic tool. Patients with SCDS usually present with a primary
c omplaint of either autophony or dizziness. There are many possible causes of dizziness, and patient evaluation should focus on narrowing the diagnosis. Vertigo symptoms related to SCDS are usually induced by loud sound or pressure changes and are brief in duration. Dizziness or oscillopsia induced by loud sound are present in 90% of SCDS patients.8 Vestibular symptoms induced by pressure changes such as coughing or straining are present in 73% of patients, with 67% exhibiting pressure-related and sound-related symptoms.8 Chronic dysequilibrium symptoms and cognitive impairment (“brain fog”) may also be attributed to SCDS. Auditory symptoms are also a common feature of SCDS. Hyperacusis for bone-conducted sound9 is present in 52% of SCDS patients.8 Symptoms often include hearing one’s pulse, eye movements, or the impact of the feet during walking. More uncommon but quite dramatic conductive hyperacusis symptoms include being able to hear others’ bowel sounds when sitting in the same church pew or the motor of a faraway vehicle while sitting on a park bench. Autophony or the patient’s own voice sounding disturbing to them is present in varying degree in 60% of patients.8 Patients with SCDS occasionally can hear in the affected ear a 512 Hz tuning fork placed against the foot or ankle.10 Evoked eye movements in the plane of the superior canal are the hallmark of SCDS11; these are present in 60% of our patients. The eyes should be examined under Frenzel lenses, with infrared video goggles, or by some other means to eliminate the effect of visual fixation. Using an audiometer, pure tones at levels up to 110 dB normal hearing level (nHL) should be delivered in one ear at a time covering the frequency range of 125 to 4000 Hz. Sound-evoked eye movements at one or more frequencies were noted in 82% of SCDS patients using such stimuli.8 Eye movements can also be induced with Valsalva maneuvers (75%) or pressure in the external auditory canal (45%). Depending on the type of stimulus, either excitation or inhibition of the superior canal may occur as shown in Figure 42-1. Pressure-evoked or sound-evoked eye movements almost always occur in the plane of the superior canal as shown in Figure 42-2. 507
508
OTOLOGIC SURGERY -10 0 10 20 30 40 50 60 70 80
FIGURE 42-1. Route of excitatory and inhibitory pressure changes causing stimulation of superior canal ampulla in superior canal dehiscence syndrome. Superior canal excitation is caused by ampullofugal displacement of the cupula typically by positive external auditory canal pressure, nasal Valsalva maneuver, or sound. Superior canal inhibition is caused by ampullopetal displacement of the cupula from negative external auditory canal pressure or glottic Valsalva maneuver, which transiently increases intracranial pressure.
Right Superior Canal
90 100 110
250
500
1k
2k
4k
8k
FIGURE 42-3. Typical audiogram in a patient with right-sided superior canal dehiscence syndrome. Circles represent air conduction, and brackets represent bone conduction. There is a negative bone conduction threshold at 250 Hz and 500 Hz, and the air-bone gap is largest at low frequencies.
Left Superior Canal Center Gaze
Right Gaze
Left Gaze
FIGURE 42-2. Direction of slow phase of eye movements with superior canal excitation. Eye movement occurs in the plane of the superior canal regardless of direction of gaze. There are vertical and torsional components when the patient is looking directly ahead (center gaze). The torsional and vertical components can be separated by having the patient look to the right or left during stimulation.
In the case of larger dehiscences, eye movements may be shifted out of the superior canal plane12; however, if eye movements are not in this direction, the diagnosis of SCDS should be questioned, and alternative diagnoses of posterior canal dehiscence13 or horizontal canal fistula14 must be considered. Sound-evoked rotation of the head, which also tilts in the plane of the affected superior canal, occurred with tones in 14% of our SCDS patients. The audiogram (Fig. 42-3) is an important part of the SCDS evaluation, and a few patients have auditory symptoms in the absence of any vestibular signs
or symptoms.8,10,15,16 Conductive hearing loss is often largest at lower frequencies,15-17 and bone conduction thresholds are often less than 0 dB nHL (conductive hyperacusis). Because of the conductive hearing loss and normal appearance of the ear, some patients with primarily auditory symptoms have historically been misdiagnosed as having otosclerosis.10 The key differences are (1) that conductive hyperacusis does not occur in otosclerosis, and (2) that the acoustic stapedial reflex, which is often normal in SCD, should be absent in an ear affected with otosclerosis. Vestibular evoked myogenic potential (VEMP) thres holds are usually decreased in patients with SCDS. These potentials are most commonly measured from the sternocleidomastoid muscles using averaged electromyography in response to multiple loud clicks or tone bursts delivered to the ear (Fig. 42-4). The reflex is thought to be activated by sound transmitted through the stapes footplate to the saccule and inferior vestibular nerve.18 Decreased VEMP thresholds are indicative of SCDS. For air-conducted 500 Hz tone bursts, we have found that cervical VEMP thresholds were 80 to 95 dB sound pressure level (SPL) for 13 patients with SCDS (83.85 ± 1.40 dB SPL, mean ± SD), 20 to 30 dB lower than in normal control subjects (110.25 ± 1.28 dB SPL).19 It has been argued that VEMP is better than 90% sensitive and specific for SCD,20 whereas other series have found the sensitivity and specificity closer to 80%.21 The VEMP is not measurable in all patients and is especially likely to be absent in patients who have had previous middle ear surgery. The VEMP threshold may also be decreased in other conditions, such as enlarged vestibular aqueduct syndrome.22
Chapter 42 • Superior Semicircular Canal Dehiscence Syndrome CLICKS L–Threshold Study
509
CLICKS R–Threshold Study
FIGURE 42-4. Typical vestibular evoked myogenic potential (VEMP) results in a patient with right-sided superior canal dehiscence syndrome and an intact left side. VEMP is initially measured with clicks at 95 dB nHL, and stimulus amplitude is decreased until response is no longer measurable. In the left ear, the patient has VEMP response at 95 dB, but not with lower amplitude stimuli. In the right ear, the amplitude of VEMP is much larger at 95 dB, and the response continues to be detectable at amplitudes of 60 dB. In this example, VEMP threshold is 95 dB nHL on the left and 60 dB nHL on the right.
For the diagnosis of SCD to be considered, imaging of the temporal bone with computed tomography (CT) must show the absence of bone over the superior canal. If the superior canal appears surrounded with bone on CT, the diagnosis of SCDS is excluded; however, the appearance of a dehiscence on CT does not rule out thin bone covering the superior canal below the resolution of the scanner. CT is a highly sensitive test for SCD, but it is not specific.21 Optimal imaging uses high-resolution CT formatted in the plane of the superior canal.21,23 The term high-resolution has been applied to a wide variety of CT scanning parameters, which continue to change as technology is updated. In a review of temporal bone CT scans done in the general population, 9% of scans had apparent SCD with one observer identifying 12%.24 Many of these are likely false dehiscences caused by the limits of resolving thin bone because the incidence of SCD in a survey of temporal bones was only 0.7%.7 A properly done CT scan should have a resolution near 0.2 mm. Attaining this resolution requires attention to numerous parameters. The most important of these is slice thickness: Collimation of the x-ray beam to 0.5 mm allows the data to be represented by nearly isotropic voxels, so that the images can be reformatted in any plane without distortion. Helical CT scanning, in which the table moves along the z-axis while the gantry rotates and scans, may lead to some loss of resolution. The “step, scan, and repeat” mode is preferred. The field of view used to reconstruct the images of the inner ear should be of the smallest size possible, so that the labyrinth is displayed to maximal resolution over the fixed size of the image matrix (usually 512 × 512 pixels). Image filters should be set for bone edge detection because filters producing less “noisy” images are likely to filter out a thin layer of bone that might remain over the canal.
Images should be reconstructed in the plane of the superior canal and orthogonal to it so that any dehiscence can be definitively shown (Fig. 42-5). Even optimized scans are not without the risks of false-positive findings, however, so the diagnosis of SCD must never be based on a CT scan alone. A finding of SCD on CT should be considered in the context of findings on physical examination, VEMP, and audiogram, and the patient’s symptoms before concluding that the patient has SCDS.
DIFFERENTIAL DIAGNOSIS SCDS should be considered in the differential diagnosis of other disorders. When conductive hearing loss is present in a setting without trauma and with a normal otoscopic examination, SCDS should be considered along with otosclerosis. Autophony raises the possibility of a patulous eustachian tube, but SCDS can produce a similar sensation. Episodic vertigo evoked by intracranial or middle ear pressure changes could indicate a perilymphatic fistula, but SCDS should strongly be considered as an alternative diagnosis. We have seen several patients who have undergone previous surgical explorations for such presumed otologic disorders, only later to be found to have SCDS. The conductive hearing loss component of SCDS can appear similar to otosclerosis because both occur in adulthood in ears that appear normal on physical examination.10 The audiograms differ in that SCDS patients often have conductive hyperacusis (see Fig. 42-3), and if there is no previous history of middle ear surgery, the acoustic reflex is often intact. Otosclerosis is not associated with decreased VEMP thresholds, vertigo symptoms, or CT findings of SCD.
510
OTOLOGIC SURGERY
A
B
C
FIGURE 42-5. CT showing superior canal dehiscence. A, CT image is reformatted in plane of superior canal. Area of dehiscence between the superior canal and middle fossa is present. B, Orthogonal reconstructions are performed at 3 degree intervals for 180 degrees around superior canal. These planes of reconstruction are shown as white lines. C, Orthogonal reconstruction shows superior canal dehiscence. Region of reconstruction is shown in small view in lower left.
Meniere’s disease is characterized by the triad of low-frequency hearing loss, vertigo, aural fullness, and tinnitus.25 Although the hearing loss in Meniere’s disease is classically sensorineural, conductive hearing loss has also been described.26 The attacks of vertigo associated with Meniere’s disease usually are severe and last hours with normal periods between attacks. The dizziness associated with SCDS can be chronic, but there are also often short periods of vertigo associated with exposure to noise or pressure changes; a careful history usually differentiates these two conditions. Autophony is often the predominant symptom in patients with a patulous eustachian tube,27 but it can also be the most disturbing symptom in SCDS. One distinguishing feature between the two conditions is that patients with patulous eustachian tube typically have autophony for their own breath sounds, whereas patients with SCDS usually do not.27 Also, a history of vertigo symptoms and hyperacusis of bone-conducted sound are not typical of a patulous eustachian tube. The audiogram, VEMP, and CT typically differentiate a patulous eustachian tube and SCDS. A perilymphatic fistula is a leak of perilymph somewhere in the vestibular labyrinth, and the leak creates an abnormal compliance that allows fluid to move and stimulate the vestibular end organs in response to sound or pressure changes. Perilymph fistula along with fenestrations of other semicircular canals are often considered in the differential diagnosis of SCDS.28 The term perilymphatic fistula is usually used to describe a fistula into the middle ear through the round or oval window. The perilymphatic fistula diagnosis is most clear in the presence of recent stapes surgery, temporal bone fracture, or barotrauma injury. In these cases, acute vertigo is usually accompanied by a sensorineural hearing loss. A fistula in the horizontal canal can be acquired in cases of cholesteatoma or prior mastoidectomy.29 Spontaneous perilymphatic fistula is a controversial diagnosis, which if considered at all should be considered only after all other possible causes are excluded.30 Just as it is possible for patients with SCDS to have prior incorrect diagnoses, patients with other disorders are
occasionally diagnosed with SCDS. In recent years, the SCDS diagnosis has become well publicized with the disorder featured on nationally televised news programs and Internet forums. Radiologists are also aware of the diagnosis, and 12% of temporal bone CT scans of the general population may be read as showing SCD,24 even though the true prevalence of the disorder is likely less than 1%.7 The most common cause of spontaneous (nonpositional) vertigo is migraine-associated vertigo.31 The incidence of migraine is 17.6% of women and 5.7% of men,32 and approximately 25% of migraine patients report some vertigo.33 Migraine is much more common than SCDS, and inevitably we have found some patients with radiographically apparent SCD whose symptoms were nonspecific and better explained by migraine. Particularly challenging are patients who have specific symptoms of SCDS and migraine. It may be difficult to determine if their sound sensitivity is due to one more than the other. Chronic dysequilibrium may be related to migraine, or it may be due to the constant transmission of intracranial pressure pulsations through the dehiscence to the labyrinth. The physiologic disturbances of the labyrinth caused by SCDS could serve as triggers to exacerbate migraine in susceptible individuals. The neurotologist must also consider, however, that failure to recognize and treat coexistent migraine can lead to disappointing results in SCDS surgery, as it can with other causes of vertigo as well.
PREOPERATIVE DECISION MAKING The decision to undergo surgery for SCD plugging is often more difficult than settling on the diagnosis. The physician must help the patient weigh the severity of symptoms against the risks and benefits of surgery. In the authors’ institution, only about a third of patients with SCDS opt to have surgical superior canal plugging, with the remaining patients choosing to live with their symptoms or making lifestyle changes to avoid situations that exacerbate the symptoms, such as loud noise.
Chapter 42 • Superior Semicircular Canal Dehiscence Syndrome
In most patients with SCDS who present to a neurotology clinic, dizziness or vertigo symptoms are the predominant complaint. In some patients, these symptoms are very situational: One patient experienced symptoms only on her commute to work on the District of Columbia Metro system for a few seconds when passing through a low section of track when the ambient pressure increased slightly. In this case, the symptoms were managed by the patient making sure she was seated during this part of her journey, and surgery was not desired. Other patients are disabled by their symptoms, and surgery is the only viable option for them to return to their careers and have an acceptable quality of life. There are also patients in the middle zone where the decision to undergo surgery is less clear. The Dizziness Handicap Inventory (DHI)34 is an instrument that may be helpful in gauging vestibular symptom severity. This questionnaire grades dizziness symptoms on a scale from 0-100. It has previously been validated for surgical treatment of benign paroxysmal positional vertigo,35 acoustic neuroma surgery,36,37 and ablative procedures for Meniere’s disease.38 We measured the DHI in 19 patients with SCDS before they underwent SCD repair via a middle fossa approach. The average preoperative DHI score was 44 ± 24 (mean ± SD)39; this compares with the handicap caused by untreated primary benign paroxysmal positioning vertigo, in which the DHI score averaged 38.5 in one series,40 and with the handicap caused by active Meniere’s disease, in which the DHI score averaged 39.6 ± 21.1 in another series.41 The comparisons indicate a high degree of dizziness handicap for SCDS patients who seek surgical treatment. Patients with mild or no dizziness symptoms may still elect to undergo surgery for other symptoms such as autophony. The section on long-term results documents our experience with changes in the DHI after surgery to repair SCD. Auditory symptoms are the primary complaint in a significant fraction of SCDS patients.16 Autophony, or the abnormal sound of the one’s own voice, can often be quite disabling, especially in patients for whom singing or speaking is important. There is no medical treatment for autophony symptoms resulting from SCDS because the sound transmission is via bone, not the eustachian tube. For SCDS patients who are significantly disturbed by autophony, surgery is the only option for relief. With conductive hyperacusis or abnormally high sensitivity to other bone-conducted sounds, patients may hear sounds emanating from within their body, such as their pulse or the motion of their eyes. These symptoms can also be very annoying to many patients, and surgery is the only option for relief. Conductive hearing loss is a common symptom in SCDS,42 but because it is often limited to the low frequencies and usually affects only one ear, many patients do not have a significant disability because of it. In most patients, the conductive hearing loss improves with
511
surgery,42 and resolution of a large sensorineural hearing loss has been reported.43 Plugging of SCD also carries a risk of hearing loss, however, and this risk is greater in patients who have had previous inner ear surgery for the problem, including stapes surgery.42 For this reason, patients who have hearing loss as their primary symptom of SCDS should first be encouraged to consider nonsurgical options, such as a hearing aid. Patients should be advised on their likely postoperative course and possible complications of SCD plugging as part of the preoperative decision-making process. Although dizziness symptoms are often the motivation for surgery, it is common for imbalance symptoms to be worse during the immediate postoperative period. These symptoms improve as the patient adapts; in about one third of patients, a physical therapist is involved in vestibular rehabilitation. Plugging of the superior canal causes a vestibular sensory deficit owing to the hydrodynamic insufficiency of the canal. This deficit is permanent, as can be shown with head thrust testing in the superior canal plane (rotating the head quickly nose down and rolling toward the ipsilateral side in the superior canal plane).3 Patients can adapt very well to this single-canal insufficiency, however, because low-frequency, lowacceleration head movements still generate useful inhibitory signals from the contralateral posterior canal. These are silenced only for high-frequency, high-acceleration head movements, which are rarer. Vestibular physical therapy can take advantage of the function of the contralateral posterior canal and of other gaze-stabilizing mechanisms in promoting compensation for the loss caused by SCD plugging. In our experience, the compensated state after SCD plugging allows the patient to lead a much more active lifestyle than did the SCDS condition. Surgery for SCD plugging shares the risk of perioperative complications common to any middle fossa approach.44 Cerebrospinal fluid leak may occur if the dura is violated, especially if air cells into the mastoid are exposed during surgery, or if there is a tegmen dehiscence. Intracranial hematoma is a rare postoperative complication that can occur after any middle fossa surgery. The patient’s mental status should be closely monitored during the acute postoperative period, and the onset of unusually severe pain should be a warning sign. If this complication occurs, the patient must be returned to the operating room quickly for hematoma evacuation to prevent more serious sequelae. The risk of sensorineural hearing loss has previously been discussed, and is likely higher in patients with previous inner ear surgery.42 Patients should be carefully counseled about this risk. The age and general state of health of the patient should also be considered in the decision to undergo surgery. In older patients, it is more difficult to elevate the middle fossa dura without tearing the dura and causing cerebrospinal fluid leak.45 Language impairment caused by damage of the dominant temporal lobe must
512
OTOLOGIC SURGERY
be considered. Postoperative vestibular adaptation and recovery can be a longer and more difficult process in older patients.
BILATERAL DEHISCENCES At our institution, 28% of individuals diagnosed with SCDS have the appearance of bilateral SCD on highresolution CT scan. One side is usually responsible for most of the symptoms and can be readily identified by the patient. In some cases, symptoms and signs can be elicited from both ears, including decreased VEMP thresholds, conductive hyperacusis, and sound-induced or pressure-induced eye movements. In such patients who do have bilateral SCDS, every effort should be made to identify the more symptomatic ear and to operate on that side first. In most cases, symptoms either resolve after operating on the more symptomatic side or abate to the point that contralateral surgery is not required. In our series of 68 patients operated for SCD, only 1 patient has had bilateral surgery. We recommend that the second side should be considered for plugging surgery only after at least 6 months have passed since the initial operation. Plugging of both superior canals significantly impairs the ability to sense downward head rotation in the vertical plane, so these patients are at risk of developing vertical oscillopsia during ambulation.
OPERATIVE TECHNIQUE Two approaches to plugging the SCD have been described. The middle cranial fossa approach was described first,1 and this technique is detailed in the following paragraphs. An alternative approach that has been described more recently is SCD plugging via a transmastoid approach. Advocates of the transmastoid approach have noted that it avoids a craniotomy, involves no temporal lobe retraction, and may lead to better stability of the canal plug. Most otolaryngologists are more familiar with mastoidectomy.46,47 The transmastoid approach was initially described in two patients in 2001, and although these patients were relieved of vertigo symptoms, one patient experienced significant sensorineural hearing loss after surgery.48 More recently, additional reports of transmastoid SCD plugging have been published with minimal morbidity and improvement in symptoms.46,47,49 We favor the middle fossa approach over the transmastoid approach for several reasons. The transmastoid approach does not allow direct confirmation of the dehiscence, and transmastoid plugging of a superior canal that was later found to be intact has been described.46 The transmastoid approach may be impossible in patients with a low-hanging dura or extensive tegmen dehiscences.46 In the transmastoid approach, the plug is also placed closer to the sensory epithelia of the ampulla and the utricle;
this may be more traumatic to these structures, risking disturbance of their baseline firing rates. Opening the superior canal distal to the dehiscence may place the plug into the common crus, causing loss of sensory function of the posterior canal as well.50 Finally, the transmastoid approach requires drilling, irrigation, and suctioning on the bony canal. When the canal is opened, these manipulations could contaminate or remove perilymph from the canal and cause collapse of the membranous labyrinth or serous labyrinthitis. Our preferred technique for SCD repair is to plug the canal via a middle cranial fossa approach. We routinely use image guidance to minimize the risk of applying suction to the dehiscence when searching for it, and to identify the SCD in what is often a field of tegmen dehiscences also found in these patients. The day before surgery, the patient undergoes a CT scan with radiopaque surface fiducial markers in place on the face and scalp. These markers are used in the operating room to register the navigation system. We use the LandmarX system (Medtronic Corporation, Minneapolis, MN), which allows us to fuse the low-resolution, whole-head dataset with a high-resolution scan of the temporal bone. The latter is invaluable for precise localization of the dehiscence. Although use of such a system adds extra time at the beginning of the case and is not strictly required, in most cases use of such a system has proven beneficial for the reasons already cited. In addition, it allows placement of the craniotomy for optimal exposure to the superior canal, while avoiding mastoid air cells. On the day of surgery, after the anesthesiologist has intubated the patient and placed any necessary lines and monitors, the table is rotated 180 degrees so that the head faces the surgeon. Dexamethasone, 0.1 mg/kg, and appropriate prophylactic antibiotics are given intravenously; 0.5 g/kg of mannitol should be prepared to be administered just before making the craniotomy. An area of the scalp away from the area of the middle fossa approach incision is prepared and sterilely draped for placement of the reference frame. In positioning of the reference frame, the surgeon should anticipate the position of the eventual incision; the location of the surgeon’s hands during surgery; the location of the microscope; the location of the navigation system; and the patient’s anatomy, including the thickness of the bone and the location of the superior sagittal sinus. When the site is chosen, a 1 cm incision is made, a small patch of periosteum is cleared from the bone, and the reference frame is anchored (Fig. 42-6). The reference frame is registered with the fiducial markers to allow navigation during surgery. Typically, the precision of the navigation registration is 1 mm. The neuromonitoring team places the necessary sound probes and electrodes for facial nerve monitoring, somatosensory evoked potential monitoring, and auditory brainstem response. This placement is done after registration of the reference frame to limit any effect of
Chapter 42 • Superior Semicircular Canal Dehiscence Syndrome
513
FIGURE 42-7. Incision planning.
FIGURE 42-6. Placement of reference frame. Fiducial markers are shown on the scalp. Planned area of incision and craniotomy are indicated by dashed lines.
these manipulations on the location of the fiducial markers because this would adversely affect the accuracy of the registration. The incision is marked on the scalp extending from the helical root around the helix to a location over the external auditory canal, and then superiorly (Fig. 42-7). Hair around the area of the planned incision is shaved, and the area is infiltrated with 1% lidocaine with 1:100,000 epinephrine. The skin is sterilized widely enough to include the previously placed reference frame in the field and to be prepared for the rare case in which a craniotomy may need to be enlarged to control bleeding or evacuate a hematoma. After the skin incision is completed, bleeding is controlled using Raney clips along the skin edges. A piece of temporalis fascia is harvested for later use in plugging the superior canal, and for repair of any tegmen defects that may be encountered or cerebrospinal fluid leak that may occur (Fig. 42-8). Afterward, the temporalis muscle is divided, and the area of the craniotomy is exposed. The intraoperative navigation system is used to plan the craniotomy. The trajectory view mode is used to “sight” a line from the surface of the skull to the dehiscence, and the craniotomy is centered here on the skull. The lower border of the craniotomy is placed just high enough to avoid the mastoid air cells, also located with the navigation probe. If a navigation system is not used,
FIGURE 42-8. Harvest of temporalis fascia after incision is opened and Raney clips are applied.
the craniotomy should be centered on the external auditory canal. This is slightly different from the placement used for drilling of the internal auditory canal, where the craniotomy is placed with its center anterior to the external auditory canal because of the more anterior location of the internal auditory canal relative to the labyrinth. The width and height of the craniotomy is enough to accommodate the Fisch retractor, typically 3 cm wide × 4 cm
514
OTOLOGIC SURGERY
FIGURE 42-10. Elevation of dura using Fisch retractor to expose superior canal.
FIGURE 42-9. Craniotomy.
high (Fig. 42-9). Care is taken to ensure the anterior and posterior cuts of the craniotomy are parallel to facilitate stable placement of the Fisch retractor. After the craniotomy is marked, the bone is opened by drilling troughs around the borders using a 4 mm cutting burr and, closer to the dura, a 4 mm diamond burr, drilling until an eggshell layer of bone remains over the dura. This bone layer is fractured with a blunt instrument. The bone flap is carefully elevated away from the dura and preserved in saline for later use. Bleeding from branches of the middle meningeal artery, which traverses the field, often must be controlled with bipolar cautery. The dura is elevated slightly further from the edges of the craniotomy to accommodate the retractor. The sharp edges of the craniotomy are removed using 2 mm and 1 mm Kerrison rongeurs, and the bone chips created in this process are saved for later use as plugs for the superior canal. The Fisch middle cranial fossa retractor is placed and used to elevate the dura gently off the floor of the middle fossa (Fig. 42-10). Dura here can be very thin, especially if tegmen dehiscences are also present, and we find that large cotton balls soaked in saline are the least traumatic means for the dural elevation. A hemostatic agent such as dry microfibrillar collagen (Avitene) or absorbable gelatin sponge (Gelfoam) mixed as a paste with thrombin is generously applied in advance of the
cotton balls. The image navigation system is frequently useful during the exploration to identify the precise location of the superior canal and its dehiscence. The surgeon is careful to suction only on the cotton balls and not to suction the area of the dehiscence directly because of the risk that this poses for removing excessive perilymph or for tearing the membranous labyrinth, which could cause sensorineural hearing and vestibular loss. When the SCD has been identified, attention is immediately shifted toward plugging the dehiscence (Fig. 42-11). Small pieces are already prepared from the previously harvested temporalis fascia. These moist pieces of fascia slide into the two open lumina of the bony superior canal with gentle pressure from a curved pick. Several pieces are placed in each end to push the plugs several millimeters beyond the dehiscence; this is done to prevent a recurrence if further bone erosion occurs from the ends of the present dehiscence. Hydraulic pressure tends to push previously placed pieces of fascia out of one end of the dehiscence while the other is being packed. We look for this as the final confirmation that the correct holes are being plugged. Care must be taken that one end is not left open because its fascia is displaced. To prevent this, when the fascia is in place, bone chips matching the diameter of the canal are firmly lodged to “cork” each end of the dehiscence. The auditory brainstem response is carefully monitored during this process, and any degradation of the response serves as a warning that too much pressure may be built up within the inner ear. Closure is achieved by anchoring the previously harvested bone flap in place using titanium plates (Fig. 42-12). A burr may be used to recess the plates into
Chapter 42 • Superior Semicircular Canal Dehiscence Syndrome
515
Fascia and bone chips fill superior canal
Bone chips placed on top
Retracted dura Bone chip cover Fibrin glue
Fascia
Bone chips
FIGURE 42-11. Plugging of superior canal dehiscence. Area of superior canal is identified while dura is retracted. Fascia and bone chips are used to plug both ends of superior canal.
the bone so that they are not palpable postoperatively. The temporalis muscle is reapproximated with absorbable sutures, and the skin is closed with staples or nylon suture or both. A drain is not typically used.
POSTOPERATIVE CARE It is recommended that patients remain in a monitored bed with frequent neurologic checks overnight because of the risk of epidural hematoma in the immediate postoperative period. Postoperative patients are treated with intravenous dexamethasone, 0.1 mg/kg every 8 hours, with a taper beginning on 2nd postoperative day. Longer courses may be considered for patients who experience postoperative sensorineural hearing loss or loss of sensory function in the horizontal or posterior canals as manifested on head thrust testing. Patients are encouraged to be out of bed in a chair and ambulating starting on postoperative day 1. An oral diet can be started the day after surgery. The typical hospital length of stay is 2 or 3 days.
Patients frequently experience nausea during the initial hours after surgery. The nausea is best controlled using intravenous promethazine (Phenergan). Because of the risk of sedation, low doses should be given initially, starting at 6.25 mg, but these may be repeated as necessary up to 25 mg, then dosed again as needed in 4 to 6 hours. Many other medications are available to control nausea, some of which may be traditionally preferred in neurosurgical patients because of the risk of sedation associated with promethazine. For nausea related to simulation of the vestibular end organs, we have found superior results with promethazine, however. Postoperative pain is usually not severe and is localized to the area of the incision. The pain is mostly due to division of the temporalis muscle and is often worse with chewing. Routine postoperative analgesics are sufficient to control the pain. If the patient is experiencing intense pain, an epidural hematoma may be the cause, and an immediate head CT scan should be considered. Any change in mental status or consciousness should also raise concerns of intracranial bleeding.
516
OTOLOGIC SURGERY
had at least partial closure of the air-bone gap after surgery.42 In patients with previous middle cranial fossa or stapes surgery, the risk of hearing loss was high in this series. In a more recent study from our institution, the average patient experienced a 10 dB improvement in air conduction hearing, although individual results varied from a 45 dB gain to a 45 dB hearing loss.21 There has even been a report of improvement in sensorineural hearing loss after SCD surgery.43
SUMMARY
FIGURE 42-12. Closure of craniotomy. Bone flap is replaced using titanium plates.
LONG-TERM RESULTS In our experience, most patients are extremely satisfied with the surgery. Relief of dizzy symptoms has been documented by measuring the DHI34 before SCD plugging surgery and 3 months afterward. On average, DHI improved by 26 points, with patients with more severe dizziness (preoperative DHI ≥30) improving by an average of 39 points.39 This improvement is greater than the mean improvement seen after surgical labyrinthectomy for Meniere’s disease, which decreased DHI score by 17, and after vestibular neurectomy, which decreased DHI score by 16.38 Studies on improvement in autophony and hyperacusis after SCD surgery have not been published, although studies are currently in progress. We have found that when patients have significant autophony or hyperacusis, these symptoms are frequently much improved immediately after surgery. Occasionally, some autophony symptoms take time to resolve, which is likely due to fluid collecting in the middle ear during the immediate postoperative period and causing conductive hearing loss. The results for improving hearing with SCD surgery are less clear. Dramatic results have been observed in individual patients,43 but are uncommon. In a series of six patients with an air-bone gap before SCD plugging who had no previous history of ear surgery, four (66%)
The diagnosis of SCDS is based on an appropriate patient history; physical examination findings including eye movements in response to sound or pressure; and other supporting studies including the audiogram, VEMPs, and CT scanning. The spectrum and severity of SCDS symptoms varies significantly among patients, and one must carefully weigh the potential benefit of surgery against the risks and probability of success in each patient. A large portion of patients with SCDS do not opt for surgical treatment. SCD plugging may be performed via a middle fossa approach. Overall, patients experience an improvement in dizziness, autophony, and hyperacusis symptoms. Although there is often an improvement in hearing after surgery, this must be carefully weighed against the risk of hearing loss, which is significant in patients who have had previous middle fossa or stapes surgery.
REFERENCES 1. Minor L B, Solomon D, Zinreich J S, Zee DS : Soundand/or pressure-induced vertigo due to bone dehiscence of the superior semicircular canal. Arch Otolaryngol Head Neck Surg 124:249-258, 1998. 2. Tullio P: Das Ohr und die Entstehung de Sprache und Schrift. Berlin, Urban Scharzenberg, 1929. 3. Mayer O, Fraser J S : Pathological changes in the ear in late congenital syphilis. J Laryngol Otol 51:683-714, 1936. 4. Fox E J, Balkany TJ, Arenberg I K : The Tullio pheno menon and perilymph fistula. Otolaryngol Head Neck Surg 98:88-89, 1988. 5. Kacker S K, Hinchcliffe R : Unusual Tullio phenomena. J Laryngol Otol 84:155-166, 1970. 6. Ishizaki H, Pyykko I, Aalto H, Starck J: Tullio phenomenon and postural stability: Experimental study in normal subjects and patients with vertigo. Ann Otol Rhinol Laryngol 100:976-983, 1991. 7. Carey J P, Minor L B, Nager GT: Dehiscence or thinning of bone overlying the superior semicircular canal in a temporal bone survey. Arch Otolaryngol Head Neck Surg 126:137-147, 2000. 8. Minor L B : Clinical manifestations of superior semicircular canal dehiscence. Laryngoscope 115:1717-1727, 2005.
Chapter 42 • Superior Semicircular Canal Dehiscence Syndrome 9. Watson S R , Halmagyi G M, Colebatch JG: Vestibular hypersensitivity to sound (Tullio phenomenon): Structural and functional assessment. Neurology 54:722-728, 2000. 10. Halmagyi G M, Aw ST, McGarvie L A, et al: Superior semicircular canal dehiscence simulating otosclerosis. J Laryngol Otol 117:553-557, 2003. 11. Cremer PD, Minor L B, Carey J P, Della Santina CC : Eye movements in patients with superior canal dehiscence syndrome align with the abnormal canal. Neurology 55:1833-1841, 2000. 12. Minor L B, Cremer PD, Carey J P, et al: Symptoms and signs in superior canal dehiscence syndrome. Ann N Y Acad Sci 942:259-273, 2001. 13. Krombach G A, DiMartino E, Schmitz-Rode T, et al: Posterior semicircular canal dehiscence: A morphologic cause of vertigo similar to superior semicircular canal dehiscence. Eur Radiol 13:1444-1450, 2003. 14. Sheehy J L , Brackmann D E : Cholesteatoma surgery: Management of the labyrinthine fistula—a report of 97 cases. Laryngoscope 89:78-87, 1979. 15. Merchant S N, Rosowski JJ: Conductive hearing loss caused by third-window lesions of the inner ear. Otol Neurotol 29:282-289, 2008. 16. Mikulec A A, McKenna M J, Ramsey M J, et al: Superior semicircular canal dehiscence presenting as conductive hearing loss without vertigo. Otol Neurotol 25:121-129, 2004. 17. Songer J E, Rosowski JJ: A mechano-acoustic model of the effect of superior canal dehiscence on hearing in chinchilla. J Acoust Soc Am 122:943-951, 2007. 18. Welgampola M S, Colebatch JG: Characteristics and clinical applications of vestibular-evoked myogenic potentials. Neurology 28(7):920-6, 2005. 19. Welgampola M S, Myrie O A, Minor L B, Carey J P: Vestibular-evoked myogenic potential thresholds normalize on plugging superior canal dehiscence. Neurology 70:464-472, 2008. 20. Zhou G, Gopen Q, Poe DS: Clinical and diagnostic characterization of canal dehiscence syndrome: A great otologic mimicker. Otol Neurotol 28(7):920-926, 2007. 21. Crane BT, Minor L B, Carey J P: Three-dimensional computed tomography of superior canal dehiscence syndrome. Otol Neurotol 29(5):699-705, 2008. 22. Sheykholeslami K, Schmerber S, Habiby Kermany M, Kaga K: Vestibular-evoked myogenic potentials in three patients with large vestibular aqueduct. Hear Res 190: 161-168, 2004. 23. Belden C J, Weg N, Minor L B, Zinreich SJ: CT evaluation of bone dehiscence of the superior semicircular canal as a cause of sound- and/or pressure-induced vertigo. R adiology 226:337-343, 2003. 24. Williamson R A, Vrabec JT, Coker NJ, Sandlin M : Coronal computed tomography prevalence of superior semicircular canal dehiscence. Otolaryngol Head Neck Surg 129:481-489, 2003. 25. Minor L B : Meniere’s disease and migraine. Arch Otolaryngol Head Neck Surg 131:460, 2005. 26. Muchnik C, Hildesheimer M, Rubinstein M, Arenberg I K : Low frequency air-bone gap in Meniere’s disease without middle ear pathology: A preliminary report. Am J Otol 10:1-4, 1989.
517
27. Poe DS : Diagnosis and management of the patulous eustachian tube. Otol Neurotol 28:668-677, 2007. 28. Minor L B : Labyrinthine fistulae: Pathobiology and management. Curr Opin Otolaryngol Head Neck Surg 11:340-346, 2003. 29. Hakuba N, Hato N, Shinomori Y, et al: Labyrinthine fistula as a late complication of middle ear surgery using the canal wall down technique. Otol Neurotol 23:832-835, 2002. 30. Friedland D R , Wackym PA : A critical appraisal of spontaneous perilymphatic fistulas of the inner ear. Am J Otol 20:261-276, 1999. 31. Eggers S D: Migraine-related vertigo: Diagnosis and treatment. Curr Pain Headache Rep 11:217-226, 2007. 32. Tepper S J: A pivotal moment in 50 years of headache history: The first American migraine study. Headache 48:730-731, 2008. 33. Kayan A, Hood J D: Neuro-otological manifestations of migraine. Brain 107(Pt 4):1123-1142, 1984. 34. Jacobson G P, Newman CW: The development of the Dizziness Handicap Inventory. Arch Otolaryngol Head Neck Surg 116:424-427, 1990. 35. Shaia WT, Zappia JJ, Bojrab D I, et al: Success of posterior semicircular canal occlusion and application of the dizziness handicap inventory. Otolaryngol Head Neck Surg 134:424-430, 2006. 36. Tufarelli D, Meli A, Labini FS, et al: Balance impairment after acoustic neuroma surgery. Otol Neurotol 28:814-821, 2007. 37. Humphriss R L , Baguley D M, Moffat D A : Change in dizziness handicap after vestibular schwannoma excision. Otol Neurotol 24:661-665, 2003. 38. Badke M B, Pyle G M, Shea T, Miedaner J: Outcomes in vestibular ablative procedures. Otol Neurotol 23:504509, 2002. 39. Crane BT, Minor L B, Carey J P: Superior canal dehiscence plugging reduces dizziness handicap. Laryngoscope 118(10):1809-1813, 2008. 40. O’Reilly RC, Elford B, Slater R : Effectiveness of the particle repositioning maneuver in subtypes of benign paroxysmal positional vertigo. Laryngoscope 110:1385-1388, 2000. 41. Perez N, Martin E, Garcia-Tapia R : Dizziness: relating the severity of vertigo to the degree of handicap by measuring vestibular impairment. Otolaryngol Head Neck Surg 128:372-381, 2003. 42. Limb C J, Carey J P, Srireddy S, Minor L B : Auditory function in patients with surgically treated superior semicircular canal dehiscence. Otol Neurotol 27:969-980, 2006. 43. Wilkinson E P, Liu GC, Friedman R A : Correction of progressive hearing loss in superior canal dehiscence syndrome. Laryngoscope 118:10-13, 2008. 44. Sanna M, Taibah A, Russo A, et al: Perioperative complications in acoustic neuroma (vestibular schwannoma) surgery. Otol Neurotol 25:379-386, 2004. 45. Oghalai J S, Buxbaum J L , Pitts L H, Jackler R K : The effect of age on acoustic neuroma surgery outcomes. Otol Neurotol 24:473-477, 2003. 46. Agrawal S K, Parnes L S : Transmastoid superior semicircular canal occlusion. Otol Neurotol 29:363-367, 2008.
518
OTOLOGIC SURGERY
47. Crovetto M, Areitio E, Elexpuru J, Aguayo F: Transmastoid approach for resurfacing of superior semicircular canal dehiscence. Auris Nasus Larynx 35:247-249, 2008. 48. Brantberg K, Bergenius J, Mendel L , et al: Symptoms, findings and treatment in patients with dehiscence of the superior semicircular canal. Acta Otolaryngol 121:68-75, 2001.
49. Kirtane MV, Sharma A, Satwalekar D: Transmastoid repair of superior semicircular canal dehiscence. J Laryngol Otol 123(3):356-358, 2009. 50. Carey J P, Migliaccio A A, Minor L B : Semicircular canal function before and after surgery for superior canal dehiscence. Otol Neurotol 28:356-364, 2007.
43
Overview of Transtemporal Skull Base Surgery Moisés A. Arriaga
OBJECTIVE The objective of neurotologic skull base surgery is exposure of the skull base through precise management of the temporal bone. In subsequent chapters, procedures are presented that accomplish ample surgical exposure and minimize brain retraction in posterior, medial, and lateral skull base lesions. The modern era of neurotologic transtemporal skull base surgery began in 1961, when House introduced the operating microscope and multidisciplinary surgery for removal of acoustic neuromas. The conceptual advantage of this transtemporal technique was a wide exposure of the lesion with substantially less cerebellar retraction than the techniques available at that time, in addition to direct facial nerve preservation. With its low mortality rate and enhanced facial nerve preservation rate, House established the translabyrinthine procedure as a technique with which all other microsurgical approaches to the cerebellopontine angle are compared.1 The emphasis on functional preservation has increased over the years from the initial enthusiasm with ablative skull base approaches.2,3 As surgeons and patients are demanding better outcomes, such strategies as the fallopian bridge technique4 to avoid facial nerve mobilization, partial labyrinthine occlusion to preserve hearing,5-7 and endoscopic strategies8 to minimize incisions have decreased the morbidity of surgery. Neurotologic skull base surgery includes various techniques that permit the surgeon to tailor the procedure to a particular patient’s pathology and physiologic status. An array of neurotologic procedures provide safe exposure of the midbrain, clivus, cerebellopontine angle, vertebrobasilar junction, petrous apex, and infratemporal fossa. The modern skull base surgeon has an expanding armamentarium of treatments, including surgery, stereotactic radiosurgery, and advanced imaging. New strategies combining observation, surgery, and stereotactic radiation are part of modern patient management. This chapter presents an anatomic framework for organizing and planning transtemporal neurotologic skull base approaches. In addition, the difficulties of terminology
and classification of approaches are discussed. The emphasis is on anatomic descriptions rather than eponyms. Figure 43-1 presents an organizational framework for transtemporal surgery based on management of the otic capsule. The otic capsule is selected as the organizational center based on its function and location. Functionally, anatomic preservation of the otic capsule is the requirement for preservation of audiovestibular function (although exceptions to this principle are developing). Anatomically, the paired petrous pyramids encompass the center of lateral skull base exposure. The approaches presented in Figure 43-1 can be used individually; however, in certain cases, combinations of these approaches offer the ideal exposure. Approaches that traverse the otic capsule (transcapsular) permit wide exposure by sacrificing hearing: translabyrinthine (see Chapter 49), transcochlear (see Chapter 52), and transotic (see Chapter 51). The posterior approaches that spare the otic capsule (retrocapsular) provide varying degrees of cerebellopontine angle exposure with an opportunity for hearing preservation: retrolabyrinthine (see Chapter 36) and retrosigmoid (see Chapter 50). Superior approaches (supracapsular) permit unroofing the internal auditory canal with varying degrees of petrous apex exposure and an opportunity for hearing preservation: middle fossa (see Chapter 48) and middle fossa transpetrous (see Chapter 53). Combined approaches permit the widest transtemporal exposure with varying opportunities for preservation of neurologic function: retrolabyrinthine petrosal (see Chapter 56), translabyrinthine petrosal (see Chapter 56), and transcochlear petrosal (see Chapter 56). The inferior approaches (infracapsular) permit minimally invasive access for drainage of cystic lesions of the petrous apex: infracochlear and infralabyrinthine (see Chapter 45). The anterior approaches (precapsular), such as the infratemporal fossa (see Chapter 54) techniques, permit exposure to the middle skull base, including the region of the foramen ovale, foramen spinosum, foramen lacerum, pterygoid space, and avenues to the nasopharynx and paranasal sinuses. These lateral approaches can be 519
520
OTOLOGIC SURGERY
Supracapsular
Transcapsular
Retrocapsular
Precapsular
Infracapsular Transcapsular: TL, TO, TC, TP Retrocapsular: RL, RS, ERS Supracapsular: MF, MFT Infracapsular: IL, IC Precapsular: ITF
FIGURE 43-1. Transtemporal neurologic skull base approaches, based on management of the otic capsule. Transcapsular: translabyrinthine (TL), transotic (TO), transcochlear (TC), and transpetrous (TP). Retrocapsular: retrolabyrinthine (RL), retrosigmoid (RS), and extended retrosigmoid (ERS). Supracapsular: middle fossa (MF) and middle fossa transpetrous (MFT). Infracapsular: infralabyrinthine (IL) and infracochlear (IC). Precapsular: infratemporal fossa (ITF).
combined with facial disassembly and endoscopic sinus approaches in selected cases. Neurotologic skull base surgery is not a hodgepodge of unrelated techniques. Instead, when considered in the context of the management of the otic capsule, these approaches are a spectrum of techniques for threedimensional surgical exposure of the cranial base.
NOMENCLATURE There has been a rapid expansion of terminology describing skull base surgical approaches. The techniques and their applications have evolved extensively. In the context of this rapid expansion of application and techniques, there has been a conflicting development of terminology for these approaches. Not only are various eponyms attached to the approaches, but also the same terms are used for different surgical techniques. Because of the potential conflict and debate over attribution, eponyms for the description of surgical approaches generally should be avoided. Instead, anatomic terminology should be selected. Considering transtemporal surgical approaches to the skull base, even this concept becomes confusing. Conceptually, most of these transtemporal approaches involve management of the petrous bone. These approaches have all rightfully been described as petrosal approaches in various modifications at different times. The terminology in Figure 43-1 is anatomically
descriptive based on the structures of the otic capsule itself. Generally, we use the term petrosal approaches for combined posterior fossa and subtemporal surgical techniques that include division of the superior petrosal sinus (see Chapter 56). The following chapters summarize the current state of the art in neurotologic skull base surgery. Although the terminology is the same, many of these approaches have been modified from their original description. The standard translabyrinthine approach includes removal of bone posterior to the sigmoid sinus and along the tegmen mastoideum. As these techniques have evolved, it is unnecessary to refer to this as an extended translabyrinthine technique. Similarly, with the transcochlear approach, House originally described an anterior extension of the translabyrinthine approach without transection of the ear canal and removal of the middle ear contents. In the current context, the transcochlear approach usually includes transection of the ear canal, removal of the skin of the ear canal, removal of the tympanic membrane and ossicular chain, and cochlear removal.
COLLABORATION IN TRANSTEMPORAL SURGERY Multidisciplinary transtemporal approaches for posterior fossa skull base neoplasms are an adjunct and not a substitute for standard neurosurgical techniques in managing
Chapter 43 • Overview of Transtemporal Skull Base Surgery
these lesions. The neuro-otologist and neurosurgeon are truly cosurgeons in the management of these lesions, each having intimate familiarity with the other’s role.9 Many of the following chapters are coauthored by otolaryngologists and neurosurgeons reflecting the true collaborative nature of modern skull base surgery. Precise temporal bone management offers the operative team flexibility to tailor the management to a patient’s specific anatomic and functional needs. As a team considers an individual patient’s management, it is useful to consider each of the basic categories of surgical approaches systematically in terms of otic capsule management. A lesion can be managed in many different ways; however, systematic consideration of these approaches and their respective merits ensures that all surgical options are being considered.
SUMMARY The fundamental prerequisite for success in transtemporal skull base surgery is complete, three-dimensional understanding of temporal bone anatomy and its surgical and functional applications. The following chapters present the indications, contraindications, and technical details of neurotologic skull base surgery.
521
REFERENCES 1. Arriaga M : Anatomy of transtemporal surgery. In Janecka I P, Tiedemann K (eds): Skull Base: Anatomy, Biology and Technology. Philadelphia, Raven-Lippincott, 1997. 2. Sekhar L N, Janecka I P: Surgery of Cranial Base Tumors. New York, Raven Press, 1993. 3. Arriaga M A, Janecka I P: Facial translocation approach to the cranial base: The anatomical basis. Skull Base Surg 1:22-30, 1991. 4. Pensak M, Jackler R: Removal of jugular foramen tumors: The fallopian bridge technique. Otolaryngol Head Neck Surg 117:586-591, 1997. 5. Hirsch B E, Cass S P, Sekhar L N, et al: Translabyrinthine approach to skull base tumors with hearing preservation. Am J Otol 14:533-543, 1993. 6. McElveen JT, Wilkins R H, Erwin A, et al: Modifying the translabyrinthine approach to preserve hearing during acoustic tumor surgery. J Laryngol Otol 105:34-37, 1991. 7. Arriaga MA, Gorum M: Enhanced retrosigmoid exposure with posterior semicircular canal resection. Otolaryngol Head Neck Surg 115:46-48, 1996. 8. Zimmer L A, Hirsch B E, Kassam A, et al: Resection of a recurrent paraganglioma via an endoscopic transnasal approach to the jugular fossa. Otol Neurotol 27:398-402, 2006. 9. Arriaga M A, Day J: Neurosurgery and Otolaryngology: Principles and Practice of Collaborative Surgery. Philadelphia, Lippincott, 1999.
44
Operations for Vascular Compressive Syndromes Peter J. Jannetta and Raymond F. Sekula, Jr.
In recent years, microvascular decompression (MVD) has become the operative treatment of choice for many disorders of the cranial nerves, augmenting and replacing ablative procedures. These cranial rhizopathies have long defied categorization, clarification of pathophysiology, and definitive treatment despite efforts by clinicians in several disciplines. Although the clinical presentations of disorders such as trigeminal neuralgia, hemifacial spasm, glossopharyngeal neuralgia, tinnitus, spasmodic torticollis, disabling positional vertigo, and Meniere’s disease are quite different, these problems all share a common underlying pathology: vascular compression of the respective cranial nerve in the proximal nonfascicular region. Several studies have shown improved blood pressure control in medically refractory, severely hypertensive patients after MVD of the left lateral medulla oblongata.1,2 Beginning in 1966, a patient with trigeminal neuralgia was noted at operation to have compression of the trigeminal nerve by a small artery near the brainstem.3 The same year, a patient with hemifacial spasm was cured after coagulation of a vein that was distorting the facial nerve at the brainstem. As we performed operations to relieve vascular compression of the trigeminal and facial nerves, it became apparent that pulsatile, mechanical forces from a blood vessel were the pathophysiologic mechanism responsible for these cranial hyperactive syndromes. These syndromes later included glossopharyngeal neuralgia, tinnitus, and vertigo. All the cranial nerves are subject to the forces of vascular compression with arteriosclerosis, elongation, and brain sag as operant factors. This chapter describes the patient selection criteria, operative technique, perioperative management, results, and complications after more than 8000 MVDs for the aforementioned cranial rhizopathies.
PREOPERATIVE EVALUATION Before MVD for any cranial rhizopathy, all patients at our institution give a detailed history and undergo a physical examination. Preoperative testing for every patient involves audiometry (pure tone and speech), acoustic
middle ear reflexes, and brainstem auditory evoked potentials (BAEPs). All patients have preoperative magnetic resonance imaging (MRI) to rule out tumors, cysts, vascular anomalies (arteriovenous malformation), aneurysms, and congenital abnormalities, especially Chiari malformation. The audiometric tests are performed preoperatively to obtain a baseline for quantitatively determining deteriorations or improvement in hearing function. The pure tone audiogram often shows anomalies represented by small dips in the frequency range of 1500 to 2000 Hz.4 Additionally, preoperative BAEPs provide baseline information for the clinical neurophysiology team so that they may warn the surgeon of any deviations during monitoring of the intraoperative auditory evoked potentials. Typical changes observed in the BAEP are an increased interpeak latency between peaks I and III ipsilaterally or prolongation of interpeak latency between III and V contralaterally (Figs. 44-1 to 44-4). These findings are based on our knowledge of the neural generators of BAEPs.5-8 If there is any question of hearing loss, at operation, audiometric tests are repeated in late in the operative procedure. Some patients may experience transient conductive hearing loss because of middle ear effusions. They resolve spontaneously within 6 weeks. Audiometry should be repeated 3 months postoperatively in such cases.
TRIGEMINAL NEURALGIA Patient Selection Patients with trigeminal neuralgia refractory to medical management are selected for MVD on the basis of their symptoms, history, and ability to undergo a general anesthetic. Patients with contraindications to general anesthetics or intracranial procedures may be treated with ablation procedures, including percutaneous glycerol rhizotomy, percutaneous radiofrequency rhizotomy, trigeminal balloon microcompression, stereotactic radiosurgery, and peripheral neurectomy.9,10 A discussion of these other surgical techniques is imperative for understanding 523
524
OTOLOGIC SURGERY
FIGURE 44-1. Brainstem auditory evoked potential recordings from a patient with normal hearing and with hemifacial spasm of 8 years’ duration on the right side. Recordings from the left ear (LE) show a normal response, whereas recordings from the right ear (RE) show an increased interpeak latency I-III and a double-peak II.
FIGURE 44-2. Brainstem auditory evoked potential r ecordings preoperatively and 6 days postoperatively from the left ear (LE) in a patient with trigeminal neuralgia on the right side. Preoperatively, peak V fluctuates in latencies, indicating changes at the level of the contralateral lateral lemniscus. Postoperatively, peak V is normal.
Chapter 44 • Operations for Vascular Compressive Syndromes
525
FIGURE 44-3. Results of audiometry and recordings of brainstem auditory evoked potentials in a patient with a 10 year history of disabling positional vertigo (DPV) on the right side. There is a slight decrease in pure tone threshold in the right ear (RE), and bilateral increased thresholds of acoustic middle ear reflex responses. Brainstem auditory evoked potentials show low amplitude and increased interpeak latency I-III for the right ear. Initial vestibular tests showed canal paresis on the right side, and 4 years later directional preponderance to the right; tests before operation show normal electronystagmographic result. LE, left ear. (From Møller MB: Results of microvascular decompression of the eighth nerve as treatment for disabling positional vertigo. Ann Otol Rhinol Laryngol 99:724-729, 1990.)
why surgeons in many centers consider MVD to be the operative procedure of choice for most patients. Peripheral neurectomy, although less invasive, typically has a short duration of tic relief, and may be indicated for patients with a shortened life expectancy.11 Radiofrequency lesions, created by percutaneous passage of a needle into the gasserian ganglion, have an initial success rate of 90% or greater, with long-term pain relief ranging from 20% to 78% of patients.12-15 Approximately 20% of patients have facial dysesthesias after radiofrequency rhizotomy. The greater the sensory loss, the more likely the trigeminal neuralgia will be relieved, but it is also more likely that dysesthesias leading to frank anesthesia dolorosa will supervene. Percutaneous glycerol rhizotomy rarely causes facial dysesthesias or sensory loss when performed strictly according to the technique introduced by Hakanson in 1981.16 Although glycerol rhizotomies have initial success rates of 80% to 90%,17 median time to recurrence varies from 16 to 36 months.18-21 Balloon microcompression of the trigeminal ganglion has initial pain relief ranging from 78% to 100%,17,22 with mean time to recurrence of 3.5 years.23 Minor dysesthesias occur in approximately 20% of patients,23 and mild temporary masseter weakness is seen in most patients treated with balloon microcompression.22 The corneal reflex is usually present.
Patients who have undergone radiofrequency lesions and balloon compression are initially satisfied because the trigeminal neuralgia is gone, but the side effects of numbness gradually become bothersome. In large series, new or worsening numbness or paresthesias after stereotactic radiosurgery (i.e., Gamma Knife) for trigeminal neuralgia occurred in 10%,24 10%,25 and 19%26 of patients, and complete pain relief (without medication) was achieved in 40% (2-year median follow-up),24 48% (1-year follow-up),25 and 34% (3-year follow-up).26 Median time to response (achievement of at least 50% pain relief) after stereotactic radiosurgery was 60 days,24 10 days,25 and 24 days.26 In another series of patients undergoing repeat stereotactic radiosurgery for trigeminal neuralgia, the authors concluded that most patients (85%) achieved more than 50% pain relief. New or worsening paresthesias occurred in 13% of patients, however, and complete pain relief (without medication) was achieved in only 19% of patients.27 Too often, advocates of stereotactic radiosurgery for trigeminal neuralgia report high rates of “success” in patients despite continued medication use. Particularly in elderly patients, medication side effects, including imbalance, confusion, and lethargy, are debilitating. The goal of all treatments for trigeminal neuralgia should be complete pain relief without medication. Pain relief
526
OTOLOGIC SURGERY
FIGURE 44-4. Results of audiometry and recordings of rainstem auditory evoked potentials in a patient with severe b tinnitus, such as that occurring after trauma. Increased interpeak latency I-III occurs on the left side and there is a double-peak II. LE, left ear; RE, right ear.
with medication is not a good result. To our knowledge, follow-up of patients after stereotactic radiosurgery has been frequently inadequate. Of the first 15 patients we operated on who had failed gamma irradiation, only 1 had been followed. The choice between MVD and other procedures should be governed by patient preference and ability to tolerate a craniotomy under general anesthesia. Advantages of MVD include consistently higher long-term success rates with a minimal chance of facial sensory loss or dysesthesias. These advantages are quite substantial in that the average life expectancy of the patients treated by the authors was 32 years from the initial onset of symptoms.28 The classic symptom of typical trigeminal neuralgia is lancinating pain in one or more distributions of the trigeminal nerve. Talking, eating, shaving, and feeling the wind blowing often precipitate symptoms. Patients typically have an abrupt and memorable onset of symptoms, with a variable duration of remission. A burning, aching
pain without specific trigger points characterizes atypical trigeminal neuralgia,29 in contrast to typical trigeminal neuralgia. Although surgical success with these patients is less than with patients with typical symptoms, no other reasonable alternative treatments currently exist. Many patients have initial success with medical management using agents such carbamazepine, phenytoin, baclofen, valproic acid, clonazepam, and gabapentin. Carbamazepine historically has been most effective, but only 56% of patients maintain satisfactory pain relief after 10 years.30 Often, side effects, allergies, and the development of refractory symptoms to medications necessitate the need for early surgical intervention.
Operative Technique for Microvascular Trigeminal Decompression As in any operation, patient positioning is important. The patient is placed on the operating table with the head at the foot of the bed, providing maximal room for the surgeon
Chapter 44 • Operations for Vascular Compressive Syndromes
during the microscopic portion of the case. A three-point head holder is applied and the patient is placed in the lateral decubitus position. An axillary roll and padding to other pressure points are placed. The neck is minimally put on stretch and slightly flexed, always maintaining at least two fingerbreadths between the patient’s chin and sternum. The patient is secured to the bed using straps and adhesive tape across the hips. The head holder is fastened into position. The patient’s shoulder is taped caudally for maximal working room (Fig. 44-5).31,32 After satisfactory positioning, a 2 cm strip of hair behind the ear is shaved and prepared. The incision, approximately 3 to 5 cm long (longer in a large man, shorter in a small woman), is placed 0.5 cm posterior to the hairline, extending one fourth above the iniomeatal line and three fourths below.31,32 Electrocautery is used to dissect and clear soft tissue until the mastoid eminence is adequately visualized. Before beginning the craniectomy, the digastric groove should be visualized (5 cm behind the ear canal [4 cm in women] and 1 cm caudal to the lateral canthus, external ear canal line). The mastoid emissary vein, which is a good landmark for the junction between the sigmoid and transverse sinuses, may bleed. It should be waxed. The craniectomy is expanded until the junction of the transverse and sigmoid sinuses is definitively appreciated, with the apex of the triangular craniectomy pointed at this junction (variations are discussed with other cranial nerve syndromes) (Fig. 44-6). All air cells are waxed diligently, after which a curvilinear or T-shaped durotomy is performed, exposing a direct corridor along the petrotentorial bone down to the brainstem.31,32 When the dura is sutured back, the operating microscope is used to “turn the corner,” or expose the cerebellopontine angle. While gently retracting on the cerebellum with a Cottonoid on a sterile piece of latex (rubber dam), the surgeon must allow adequate cerebrospinal fluid to drain so that the cerebellum falls away, minimizing the need for much cerebellar retraction. Pene tration of the trigeminal cistern and sharp dissection of arachnoid adhesions greatly reduce the need for significant cerebellar retraction medially. The Cottonoid and a precisely flexible, self-retaining brain retractor are placed on the supralateral portion of the cerebellum for trigeminal exposures. The cerebellum is retracted up and medially. Often, the first vessels encountered are the petrosal vein complex, which are coagulated and divided after evaluating relationships to the nerve.31,32 Before performing MVD of the trigeminal nerve, the surgeon must be aware that the dorsal root exit zone of the trigeminal nerve (the junctional area of myelin) extends to the distal portion of the nerve. The nerve must be meticulously inspected from the brainstem to Meckel’s cave, and all offending vessels must be decompressed. When performing the actual decompression, all arachnoid over the nerve must be dissected away. Shredded polytef (Teflon®) felt is placed between the vessel and
527
the nerve. The most frequent vessel found is the superior cerebellar artery. Often multiple pieces of shredded Teflon felt are used to decompress looping arteries affecting more than one side of the trigeminal nerve.31,32 Before closing, a Valsalva maneuver is performed to ensure hemostasis. The dural opening is closed carefully with a watertight closure over a strip of absorbable gelatin sponge (Gelfoam) to prevent cerebrospinal fluid leakage. If a watertight closure cannot be obtained, cadaveric tissue, synthetic material, or muscle (if the leak is small) is used to patch the leak. The bone edges are waxed for a second time, and a cranioplasty is performed using titanium wire mesh placed over a piece of Gelfoam. Fascia, subcutaneous tissue, and skin are closed in standard manner.
Results for Trigeminal Microvascular Decompression The most frequent operative finding was trigeminal compression by the superior cerebellar artery in 76% of the patients (Fig. 44-7). Veins were causing neural compression in 68% of patients, but were the sole offending vessel in only 13%.38,39 In a report38 of 1204 patients who underwent initial MVD, 80% were pain-free at 1 year, and an additional 8% had greater than 75% relief, for a total success rate of 88%.38 At 10 years, 70% remained pain-free, and 4% had greater than 75% relief. Persistent mild postoperative facial numbness was noted in 17%, of which only 1% was severe. Repeat MVD was performed in 11% of patients for persistent or recurrent pain.38 Of these, 96% were pain-free or had greater than 75% pain relief 1 year after surgery (89% at 10 years). MVD for pediatric-onset trigeminal neuralgia has a lower success rate than MVD for adults. At the time of discharge, 73% of patients had complete pain relief, with an additional 18% having greater than 75% diminution of pain.40 At last follow-up (mean 105 months), 57% of patients had either complete pain relief or greater than 75% relief of pain. The lower therapeutic response is likely due to the fact that the pathophysiology of the disease is different in pediatric patients. Venous compression was noted in 86% of the cases and was the sole offending vessel in 48%; this is markedly greater than for the adult patients. Additionally, venous compression often results in recurrence of pain because of revascularization. In 393 cases of trigeminal neuralgia caused by veins, 31% developed recurrence (most within 1 year of initial operation) after initial improvement of pain. The morbidity and mortality with MVD for trigeminal neuralgia in elderly patients are similar to morbidity and mortality in younger patients in properly selected patients.37 We do not exclude patients from MVD by age alone. Each patient’s fitness for surgery is evaluated by a medical team, and we typically offer a percutaneous procedure to patients with contraindications to any surgery (e.g., severe lung [emphysema] and heart [congestive
528
OTOLOGIC SURGERY
Chapter 44 • Operations for Vascular Compressive Syndromes
529
FIGURE 44-5. Three-point head fixation device is applied to the patient’s head, and the patient is placed in a lateral park bench position with affected side up. An axillary roll is placed to prevent brachial plexus injury, and all other pressure points are padded appropriately. The head is elevated and distracted, then rotated 10 degrees away from the surgeon. The head is locked into place, and the ipsilateral shoulder is distracted caudally and rotated medially, giving the surgeon maximal working area. The patient is taped securely at the hips and chest to allow rotation of the table during surgery. FIGURE 44-6. Junction of the transverse and sigmoid sinuses must be exposed at superior margin of the craniectomy. Edge of the sigmoid sinus must be identified along the lateral margin of the craniectomy. To minimize cerebellar retraction, craniectomy is extended caudally and laterally for procedures involving lower cranial nerves.
FIGURE 44-7. Intraoperative view of superior cerebellar artery (SCA) and a vein causing compression of the trigeminal nerve in a patient with trigeminal neuralgia.
FIGURE 44-8. Intraoperative view of CN VII compressed by posterior inferior cerebellar artery (PICA) in a patient with hemifacial spasm. VA, vertebral artery.
FIGURE 44-9. Intraoperative view of left lateral medulla and fascicles of CN IX and X compressed by left posterior inferior cerebellar artery (PICA) and left vertebral artery (VA).
heart failure] disease). Other series33-36 comparing MVD in elderly patients and young patients have reached a similar conclusion that, with proper patient selection, MVD can be performed in elderly patients without causing higher postoperative morbidity and mortality. Although the “less invasive” procedures, with the exception of stereotactic radiosurgery, offer a comparably high rate of immediate pain relief, none can compare with the long-term success of MVD. In elderly patients, generalized atrophy and wide cisterns allow ease of access to the cerebellar pontine angle. We believe our findings support the notion that MVD should also be considered as the first-line procedure for trigeminal neuralgia in the elderly as it is in younger patients, and agree that age, by itself, should not be a barrier to MVD.37
Complications In a review of our series by McLaughlin and colleagues,31 hearing loss was noted in 31 of 3196 patients after MVD for trigeminal neuralgia (0.97%).31 Before 1990, the incidence of hearing loss was 1.33%. Since 1990, that percentage has decreased to 0.59%.31 This decline in hearing loss is attributed to the use of brainstem auditory evoked response monitoring. The incidence of other complications is quite rare, with death occurring in 0.14% of cases.39
HEMIFACIAL SPASM
inferiorly to involve the lower portion of the face and the platysma. Approximately 15% of patients have frontalis involvement. Atypical hemifacial spasm differs from the more common form because contractions start in the buccal muscles and progress rostrally. The tonus phenomenon, or sustained contracture, is common in atypical and typical hemifacial spasm resulting in eye closure and ipsilateral abduction of the corner of the mouth. Response to MVD is similar between these two groups. In contrast to patients with trigeminal neuralgia, patients with hemifacial spasm rarely receive temporary benefit from medications such as baclofen, carbamazepine, phenytoin, clonazepam, and other antianxiety medi cations.43,44 Botulinum toxin has been reported to be successful in 80% to 100% of treated patients.45-47 These benefits are transitory, with recurrence following in 12 to 16 weeks.45,46 Obvious facial weakness occurs in nearly 75% of patients treated with repeated injections for at least 3 years.48 A previous analysis showed that the mean life expectancy for patients with hemifacial spasm is 35 years from symptom onset.28 In such patients, botulinum toxin may be less cost-effective and potentially debilitating if given over such a protracted period. Botulinum toxin should not be used in the lower face. Surgery must not be performed within 9 months of botulinum toxin injection because intraoperative electromyography monitoring becomes useless.
Patient Selection
Operative Technique for Microvascular Decompression in Hemifacial Spasm
Hemifacial spasm is more than simply a cosmetic problem. Although uncommon (prevalence of 7/100,000),41 it has significant psychosocial consequences, and may interfere with the patient’s ability to read, drive, or work. In children, hemifacial spasm may retard reading ability, causing significant and profound educational difficulties.42 In contrast to some disorders that may mimic hemifacial spasm, this disorder is characterized by paroxysmal contractions of the facial muscles on one side. The contraction typically begins around the orbicularis oculi muscles and progresses
MVD for hemifacial spasm is similar in several respects to MVD for trigeminal neuralgia. There are several important differences, however. As previously mentioned, the patient’s position is crucial for maximizing surgical success. For hemifacial spasm, the patient’s head is rotated away 10 degrees (as in trigeminal neuralgia), and the vertex is lowered 15 degrees toward the floor.31 This maneuver rotates the vestibulocochlear complex cephalad, while exposing the proximal portion of CN VII. Failure to tilt the vertex down may result in poor
530
OTOLOGIC SURGERY
v isualization of the facial nerve and failure to achieve adequate decompression. The patient is positioned and fixed in the standard lateral position. A second important difference is that the patient must not be given muscle relaxants or paralytics because these interfere with intraoperative monitoring of electromyography potentials recorded from muscles innervated by CN VII. In patients with hemifacial spasm, stimulation of the temporal or zygomatic branch of the facial nerve electrically produces electromyography potentials in the mentalis muscle. This muscle response, specific for hemifacial spasm, has a latency of 10 ms.49 These abnormal electromyography recordings, known as lateral spread, disappear after decompression of the offending vessel.50 Failure to see disappearance of the lateral spread necessitates further inspection of the nerve by the surgical team.51 The skin incision and soft tissue dissection are the same for hemifacial spasm and trigeminal neuralgia. The bone work differs slightly in that more of the craniectomy extends inferiorly and laterally. The dura is opened in a curvilinear or T-shaped fashion and affixed with sutures. As cerebrospinal fluid drains, the operating microscope is employed, and the brain retractor is placed on the inferolateral portion of the cerebellum, elevating the cerebellum and the tonsils. Next, the cisterna magna is opened, sharply exposing CN IX, X, and XI, and, more superiorly, CN VII and VIII.31 Most commonly, a loop of the posterior inferior cerebellar artery is the offending vessel in typical hemifacial spasm (Fig. 44-8). The surgeon typically finds this loop anterior and caudal to the root entry zone. The posterior inferior cerebellar artery is found in 68% of patients, the anterior inferior cerebellar artery is found in 35%, and the vertebral artery is found in 24% (in some cases, multiple offending vessels were found).52 In atypical hemifacial spasm, the surgeon should look for the offending vessel rostrally or lying between CN VII and VIII. Failure to recognize one or all offending vessels is most common when attempting MVD of CN VII.
Results for Microvascular Decompression in Hemifacial Spasm Many clinicians report high success rates with MVD for hemifacial spasm. Iwakuma and coworkers53 reported a 97% success rate after treating 74 patients. Auger and associates54 reported complete relief of spasm in 81% of their patients after nearly 4 years. In an analysis of 16 series, Loeser and Chen55 reported an 84% rate of complete spasm relief and an additional 4% relief after a second operation. In a review of the authors’ series, 86% had excellent relief of spasm, with an additional 5% claiming greater than 75% improvement in spasms.52 These results remained stable at 10-year follow-up. Of 12 patients without any improvement, 11 underwent repeat MVD, and 10 achieved excellent long-term results.52
We recommend that patients without any symptomatic improvement in the initial postoperative period undergo repeat exploration. For patients with even slight improvement in spasm, however, close follow-up is recommended because these patients tend to experience continued improvement. The most common offending vessel in a series of 648 consecutive MVDs was the posterior inferior cerebellar artery, found in 68% of the cases. The anterior inferior cerebellar artery was compressing CN VII in 35% of the cases.52 The pathology for pediatric hemifacial spasm is quite different; venous compression (alone or with another artery) was noted in 75% of the pediatric cases.42 Because venous compression is known to recur after MVD, the rate of excellent outcomes is 67% (mean follow-up 125 months) in the pediatric population.42
Complications In a series of 782 operations for hemifacial spasm, 3.2% of patients experienced transient facial weakness.52 In McLaughlin and colleagues’ report31 of 1069 operations by the senior author, the incidence of ipsilateral hearing loss was 2.99%. After 1990, when the use of intraoperative brainstem evoked potentials became routine, the incidence of hearing loss decreased to 1.59%.31 Cerebrospinal fluid leaks were noted in 2.4% of patients, and 1.2% had wound infections. Other complications were rare (Table 44-1).41
DISABLING POSITIONAL VERTIGO AND TINNITUS Patient Selection Patients with compression of CN VIII may present with myriad symptoms, which include tinnitus, hearing loss, Meniere’s disease, paroxysmal vertigo, and persistent dysequilibrium. Patients with compression of only the cochlear portion or the vestibular portion of CN VIII present with isolated auditory or vestibular dysfunction. Patients with disabling positional vertigo have a characteristic pattern of symptoms, the most common of which is persistent vertigo augmented by activity, but eased by bed rest. This is distinct from patients with Meniere’s disease, characterized by aural fullness, violent attacks of vertigo, tinnitus, and hearing loss lasting for several hours with the patients asymptomatic between attacks.56 In contrast to patients with benign paroxysmal positional vertigo, patients with disabling positional vertigo often complain of associated nausea and occasional vomiting, or they may say that they walk as if they “had a few drinks too many.” Patients often stagger, are unable to make quick turns, and tend to fall to the affected side. Over time, 20% of affected patients develop associated cochlear compression resulting in a slowly progressive loss of hearing in the midfrequency to high-frequency
Chapter 44 • Operations for Vascular Compressive Syndromes TABLE 44-1 Complications after Consecutive
Microvascular Decompressions for Hemifacial Spasm (782 Operations)
Complication Permanent facial weakness Transient facial weakness Complete ipsilateral hearing loss CSF leak Wound infection Pseudomeningocele Bacterial meningitis Cerebellar hematoma Infarct Operative death Other
Percentage 3.3 3.2 2.7 2.4 1.2 0.5 0.5 0.5 0.3 0.1 <1
CSF, cerebrospinal fluid. From Auger RG, Whisnant JP: Hemifacial spasm in Rochester and Olmstead County, Minnesota, 1960 to 1984. Arch Neurol 47:1233-1234, 1990.
range. Occasionally, patients may develop symptoms from adjacent cranial nerves, such as hemifacial spasm, ear pain (geniculate neuralgia), or trigeminal neuralgia (V2 distribution). As previously mentioned, patients with CN VIII compression may present with isolated auditory symptoms. Typically, these patients have hearing loss in the highfrequency range that does not fluctuate.56 Differences in pure tone hearing threshold and tinnitus are also common signs of cochlear nerve compression (see Fig. 44-4). Patients with CN VIII compression differ from patients with other vestibular syndromes in that their symptoms are often unresponsive to common vestibular suppressants, such as meclizine (Antivert) or dimenhydrinate (Dramamine). Some patients report intermittent success with medical therapy involving benzodiazepines such as diazepam and clonazepam.
Operative Technique for Microvascular Decompression in Disabling Positional Vertigo and Tinnitus The position of the patient is similar to the position for hemifacial spasm, with the head rotated away 10 degrees and the vertex lowered 15 degrees toward the floor.31 The bony opening is shaped like an isosceles triangle, with the apex pointed at the junction of the sigmoid and transverse sinuses.31 The dural opening is similar to that of a hemifacial spasm approach. The intradural approach is different, however, in that the retractor blade and cottonoid are placed supralaterally over the cerebellum, elevating it off CN VIII. The root entry zone must be dissected free from the flocculus of the cerebellum more medially than in hemifacial spasm. This dissection allows for greater exposure of the veins along the brainstem, which may be the offending vessels.57-59
531
Patients with vertigo or dysequilibrium or both are likely to have compression of the superior vestibular nerve at or just distal to the root entry zone. Associated nausea is typically associated with compression of the inferior vestibular nerve at the brainstem. In patients with tinnitus, the surgeon must inspect the cochlear nerve from the root entry zone to the internal auditory meatus. Patients with tinnitus and associated deep ear pain with or without vertigo often have compression of the nervus intermedius and the cochlear nerve. Typically, the offending vessel is running between CN VII and VIII. Intrafascicular compression by arteries is often a difficult problem to treat, and may require division of the nervus intermedius.59
Results of Microvascular Decompression for Disabling Positional Vertigo and Tinnitus The group with CN VIII dysfunction comprised 281 patients with disabling positional vertigo, tinnitus, hearing loss, or dysequilibrium. Patients with predominantly vestibular symptoms were noted to have proximal compression at the root entry zone. Cochlear dysfunction was associated with more peripheral compression. Of the patients with vestibular pathology, 79% were markedly improved or symptom-free after surgery, whereas only 40% of the patients with pure cochlear disturbance were significantly better.56 Improvement in tinnitus is delayed and may take 2 years. The results in disabling tinnitus are time-related. At 2 years, 90% do well; this decreases to 80% after 4 years, and then precipitously declines to 10%.
Complications of Microvascular Disabling Positional Vertigo and Tinnitus Hearing was decreased postoperatively in 11 (4.7%) of 235 patients operated on for disabling positional vertigo or tinnitus or both. In two patients, the tinnitus continued on a worsening course. Additionally, there was one patient with transient vocal cord paresis, one patient with trochlear nerve paresis, and two patients with transient mild facial paresis.56
GLOSSOPHARYNGEAL NEURALGIA Patient Selection Glossopharyngeal neuralgia is characterized by sharp, lancinating, intermittent pain involving the posterior tongue, pharynx, and deep ear structures. When other etiologies are eliminated, these patients are excellent candidates for MVD.60 Glossopharyngeal neuralgia is due to vascular compression of CN IX and X. In patients with associated brainstem compression, associated symptoms may include sleep apnea, syncope, ataxia, autonomic dysreflexia, and hypertension (discussed subsequently).
532
OTOLOGIC SURGERY
Some patients find relief with carbamazepine, phenytoin, or baclofen. The efficacy of pharmacotherapy has not been well defined.
NEUROGENIC HYPERTENSION
Operative Technique for Microvascular Disease in Glossopharyngeal Neuralgia
Elevated arterial pressure not caused by well-described pathologic findings such as renal artery stenosis, pheochromocytoma, or renal disease has been termed essential hypertension.1 In 1979, Jannetta and Gendell61 first introduced the concept of neurogenic hypertension. They reported on 16 consecutive hypertensive patients with vascular compression of the left rostral ventrolateral medulla (RVLM). During the past 3 decades, there has been an abundance of scientific evidence in the forms of animal experiments,62-65,73 postmortem anatomic dissection,66,67 operative observations,1,2,61 and radiographic findings68-70 showing that pulsatile compression of the left RVLM elicits sympathetically mediated hypertension. Hypertensive candidates for MVD are refractory to oral antihypertensive medication, have pressure lability interfering with activities of daily living, have associated autonomic dysreflexia, or have intractable side effects from the medication. All patients must be screened for carcinoid, renal artery stenosis, pheochromocytoma, and renal diseases. Additionally, a minimum of three blood pressure measurements confirming hypertension must be documented while adhering to pharmacologic management. Although many patients had demonstration of brainstem compression by ectatic vessels on preoperative MRI, current MRI techniques have been shown to be insufficiently sensitive as a screening tool for neurogenic hypertension.71 Postoperatively, continuous blood pressure measurements are obtained for 24 hours, followed by blood pressure recordings every 6 to 8 hours for 48 hours. Daily measurements are required for 2 weeks.
MVD of CN IX and X requires a low lateral retromastoid craniectomy. When the patient is positioned, prepared, and draped in the usual way, the skin incision is made from the level of the iniomeatal line to the level of the inferior portion of the mastoid tip 0.5 cm behind the hairline. The craniectomy is extended inferiorly to the floor of the posterior fossa, and laterally exposing the sigmoid sinus as it approaches the jugular bulb. The dural opening is a curvilinear or T-shaped incision. The surgeon places the retractor and rubber dam over the inferolateral portion of the cerebellum, elevating the flocculus off CN IX and X. If bridging veins are encountered, they may be coagulated and cut. The cisterna magna is sharply opened, allowing cerebrospinal fluid to drain and the cerebellum to relax, minimizing the need for extensive retraction. The vascular compression is at and on the brainstem on CN IX and on the upper fascicles of CN X. It may be located rostral, anterior, or posterior to CN IX and upper CN X, and may run between them. Before the MVD, arachnoid trabeculae must be meticulously dissected off the cranial nerves and cerebellum, allowing for maximal exposure. If the offending vessel is venous, it may be moved off the nerve and coagulated. If arterial compression is identified, it must be moved away from the nerves and held in place with shredded Teflon felt. Closure is as described previously.
Results of Microvascular Decompression in Glossopharyngeal Neuralgia The prognosis for patients with glossopharyngeal neuralgia is quite good after MVD. Of the patients operated on by the senior author, 79% had at least 95% relief, and another 10% had at least 50% relief, with the remaining 10% considered failures.60 Based on intraoperative observations, the posterior inferior cerebellar artery was the most common cause of neural compression (39%).60
Complications of Microvascular Disease in Glossopharyngeal Neuralgia There was one postoperative death after MVD and one death after nerve section by another surgeon, both from sequelae of intraoperative hypertension. Three patients had permanent CN IX and X palsy (8%), and four others had transient palsies (10%).60
Patient Selection
Operative Technique for Neurogenic Hypertension The operation for neurogenic hypertension is similar to the operation for glossopharyngeal neuralgia with respect to positioning, craniectomy, durotomy, and approach. The surgeon should expect to find the vertebral artery, the posterior inferior cerebellar artery, or the basilar artery causing RVLM compression.1,62,69 Mobilization of the larger vessels, especially if dolichoectatic or calcified, may be quite difficult. Multiple pieces of shredded Teflon felt are usually required to mobilize the vessel away from the point of maximal compression. Great care must be taken to avoid injury to all perforators in this region because inadvertent tearing of small arteries may cause irreversible neurologic deficits.
Results After observing an association between hypertension and left RVLM compression, Jannetta and coworkers72 reported such compression in 51 of 53 hypertensive
Chapter 44 • Operations for Vascular Compressive Syndromes
patients undergoing MVD for either trigeminal neuralgia or hemifacial spasm. Several years later, two articles were published within a few months of each other from different groups describing outcomes of patients with medically intractable hypertension treated with MVD of the RVLM.1,2 The results were similar. Levy and associates1 showed improvement in 72% of the patients treated, with 54% achieving normotensive pressures. Additionally, 54% of the patients with severe medically intractable hypertension were regulating their pressure with fewer medications. Similarly, Geiger and colleagues2 showed that 50% of the patients in their study were normotensive after a 1-year follow-up. The posterior inferior cere bellar artery was compressing the brainstem in 92% of the patients (Fig. 44-9).1 These studies by Geiger and Levy and their colleagues show the potential for operative intervention before irreversible end organ damage typical of poorly controlled hypertensive patients. Current randomized, multicenter, prospective trials are investigating the efficacy of MVD for neurogenic hypertension.
Complications In our more recent study, complications were noted in two patients (17%). These included ipsilateral deafness in one patient after re-exploration for intractable autonomic dysreflexia. A second patient had transient congestive heart failure and paraparesis (of unknown etiology).
SUMMARY The cranial nerves all are susceptible to the forces that cause cranial rhizopathies. These forces, such as atherosclerosis, elongation of vessels, and cerebellar atrophy, place the aging population at significant risk for developing any one of the syndromes discussed in this chapter. An understanding of the cerebellopontine angle anatomy with respect to the cranial nerves enables trained microsurgeons to offer patients a nondestructive cure for their disease with long-term success. Favorable outcomes primarily depend on patient selection, preoperative and intraoperative monitoring, and surgical experience. In experienced hands, the morbidity of MVD is quite low, and the cure rate ranges from 70% to 90% for most cranial rhizopathies.
REFERENCES 1. Levy E I, Clyde B, McLaughlin M R , Jannetta PJ: M icrovascular decompression of the left lateral medulla oblongata for severe refractory neurogenic hypertension. Neurosurgery 43:1-6, 1998. 2. Geiger H, Naraghi R , Schobel H P, et al: Decrease of blood pressure by ventrolateral medullary decompression in essential hypertension. Lancet 352:446-449, 1998.
533
3. Jannetta PJ, Rand RW: Transtentorial retrogasserian rhizotomy in trigeminal neuralgia by microneurosurgical technique. Bull Los Angeles Neurol Soc 31:93-99, 1966. 4. Møller M B, Møller A R : Audiometric abnormalities in hemifacial spasm. Audiology 24:396-405, 1985. 5. Møller A R , Jannetta PJ: Comparison between intracranially recorded potentials from the human auditory nerve and scalp recorded auditory brainstem responses (ABR). Scand Audiol 11:33-40, 1982. 6. Møller A R , Jannetta PJ: Compound action potentials recorded intracranially from the auditory nerve in man. Exp Neurol 74:862-874, 1981. 7. Møller A R , Jannetta PJ, Sekhar L N: Contributions from the auditory nerve to the brainstem auditory evoked potentials (BAEPs): Results of intracranial recording in man. Electroencephalogr Clin Neurophysiol 71:198-211, 1988. 8. Møller A R , Jannetta PJ: Auditory evoked potentials recorded from the cochlear nucleus and its vicinity in man. J Neurosurg 59:1013-1018, 1983. 9. Sweet WH, Poletti C E : Problems with retrogasserian glycerol in the treatment of trigeminal neuralgia. Appl Neurophysiol 48:252-257, 1985. 10. Mullan S, Lichtor T: Percutaneous microcompression of the trigeminal ganglion for trigeminal neuralgia. J Neurosurg 59:1007-1012, 1983. 11. Møller M B, Møller A R : Brainstem auditory evoked potentials in patients with cerebellopontine angle tumors. Ann Otol Rhinol Laryngol 92:645-650, 1983. 12. Laha R K, Jannetta PJ: Glossopharyngeal neuralgia. J Neurosurg 47:316-320, 1977. 13. Broggi G, Franzini A, Lasio G, et al: Long-term results of percutaneous retrogasserian thermorhizotomy for “essential” trigeminal neuralgia: Considerations in 1000 consecutive patients. Neurosurgery 26:783-786, 1990. 14. Taha J M, Tew J M J, Buncher C R : A prospective 15-year follow-up of 154 consecutive patients with trigeminal neuralgia treated by percutaneous stereotactic radiofrequency thermal rhizotomy. J Neurosurg 83:989-993, 1995. 15. Taha J M, Tew J M J: Treatment of trigeminal neuralgia by percutaneous radiofrequency rhizotomy. Neurosurg Clin North Am 8:31-39, 1997. 16. Hakanson S : Trigeminal neuralgia treated by the injection of glycerol into the trigeminal cistern. Neurosurgery 9:638-648, 1981. 17. Zakrzewska J M : Surgery at the level of the gasserian ganglion. In Zakrzewska J M (ed): Trigeminal Neuralgia. London, Saunders, 1995, pp 125–156. 18. Lunsford L D, Bennett M H : Percutaneous retrogasserian glycerol rhizotomy for tic douloureux, I: Technique and results in 112 patients. Neurosurgery 14:424-430, 1984. 19. Lunsford L D, Apfelbaum R I : Choice of surgical therapeutic modalities for treatment of trigeminal neuralgia: Microvascular decompression, percutaneous retrogasserian thermal, or glycerol rhizotomy. Clin Neurosurg 32:319-333, 1985. 20. Burchiel K J: Percutaneous retrogasserian glycerol rhizolysis in the management of trigeminal neuralgia. J Neurosurg 69:361-366, 1988.
534
OTOLOGIC SURGERY
21. Slettebo H, Hirschberg H, Lindegaard K F: Long-term results after percutaneous retrogasserian glycerol rhizotomy in patients with trigeminal neuralgia. Acta Neurochir (Wien) 122:231-235, 1993. 22. Lichtor T, Mullan J F: A 10-year follow-up review of percutaneous microcompression of the trigeminal ganglion. J Neurosurg 72:49-54, 1990. 23. Brown J A, Gouda JJ: Percutaneous balloon compression of the trigeminal nerve. Neurosurg Clin North Am 8:5362, 1997. 24. Maesawa S, Salame C, Flickinger JC, et al: Clinical outcomes after stereotactic radiosurgery for idiopathic t rigeminal neuralgia. J Neurosurg 94:14-20, 2001. 25. Regis J, Metellus P, Hayashi M, et al: Prospective controlled trial of gamma knife surgery for essential trigeminal neuralgia. J Neurosurg 104:913-924, 2006. 26. Sheehan J, Pan HC, Stroila M, Steiner L : Gamma knife surgery for trigeminal neuralgia: Outcomes and prognostic factors. J Neurosurg 102:434-441, 2005. 27. Hasegawa T, Kondziolka D, Spiro R , et al: Repeat radiosurgery for refractory trigeminal neuralgia. Neurosurgery 50:494-500, 2002. 28. Vital statistics of the United States, 1987, Vol II, part A—mortality. Hyattsville, MD, U.S. Department of Health and Human Services, 1990. 29. Casey K : Role of patient history and physical examination in the diagnosis of trigeminal neuralgia. Neurosurg Focus 18:E1, 2005. 30. Taylor JC, Brauer S, Espir M L : Long-term treatment of trigeminal neuralgia with carbamazepine. Postgrad Med J 28:16-18, 1981. 31. McLaughlin M R , Jannetta PJ, Clyde B L , et al: Microvascular decompression of cranial nerves. J Neurosurg 90:1-8, 1999. 32. Jannetta PJ, McLaughlin M R, Casey K F: Technique of microvascular decompression. Neurosurg Focus 18:E5, 2005. 33. Ashkan K, Marsh H : Microvascular decompression for trigeminal neuralgia in the elderly: A review of the safety and efficacy. Neurosurgery 55:840-848, 2004. 34. Javadpour M, Eldridge PR , Varma TR , et al: Microvascular decompression for trigeminal neuralgia in patients over 70 years of age. Neurology 60:520, 2003. 35. Ogungbo B I, Kelly P, Kane PJ, Nath FP: Microvascular decompression for trigeminal neuralgia: Report of outcome in patients over 65 years of age. Br J Neurosurg 14:23-27, 2000. 36. Ryu H, Yamamoto S, Sugiyama K, Nozue M : Neurovascular compression syndrome of the eighth cranial nerve: What are the most reliable diagnostic signs? Acta Neurochir (Wien) 140:1279-1286, 1998. 37. Sekula R F Jr, Marchan E M, Fletcher L H, et al: Microvascular decompression for trigeminal neuralgia in the elderly. J Neurosurg Apr; 108(4):688–691, 2008. 38. Barker FG, Jannetta PJ, Babu R P, et al: Long-term outcome after operation for trigeminal neuralgia in patients with posterior fossa tumors. J Neurosurg 84:818-825, 1996. 39. Barker FG, Jannetta PJ, Bissonette DJ, et al: The longterm outcome of microvascular decompression for trigeminal neuralgia. N Engl J Med 334:1077-1083, 1996. 40. Resnick D K, Levy E I, Jannetta PJ: Microvascular decompression for pediatric onset trigeminal neuralgia. Neurosurgery 43:804-807, 1998.
41. Auger RG, Whisnant J P: Hemifacial spasm in Rochester and Olmstead County, Minnesota, 1960 to 1984. Arch Neurol 47:1233-1234, 1990. 42. Levy E I, Resnick D K, Jannetta PJ, et al: Pediatric hemifacial spasm: The efficacy of microvascular decompression. Pediatr Neurosurg 27:238-241, 1997. 43. Alexander G E, Moses H III: Carbamazepine for hemifacial spasm. Neurology 32:286-287, 1982. 44. Sandyk R , Gillman M A : Clonazepine in hemifacial spasm. Int J Neurosci 33:261-264, 1987. 45. Dutton JJ, Buckley EG: Long-term results and complications of botulinum A toxin in the treatment of blepharospasm. Ophthalmology 95:1529-1534, 1988. 46. Taylor J D N, Kraft S P, Kazdan M S, et al: Treatment of blepharospasm and hemifacial spasm with botulinum A toxin: A Canadian multicentre study. Can J Ophthalmol 26:133-138, 1991. 47. Yoshimura D M, Aminoff M J, Tami TA, et al: Treatment of hemifacial spasm with botulinum toxin. Muscle Nerve 15:1045-1049, 1992. 48. Park YC, Lim J K, Lee D K, et al: Botulinum A toxin treatment of hemifacial spasm and blepharospasm. J Korean Med Sci 8:334-340, 1993. 49. Møller A R , Jannetta PJ: Microvascular decompression in hemifacial spasm: Intraoperative electrophysiological observations. Neurosurgery 16:612-618, 1985. 50. Montero J, Junyent J, Calopa M, et al: Electrophysiological study of ephaptic axono-axonal responses in hemifacial spasm. Muscle Nerve 35:184-188, 2007. 51. Kong DS, Park K, Shin BG, et al: Prognostic value of the lateral spread response for intraoperative electromyography monitoring of the facial musculature during microvascular decompression for hemifacial spasm. J Neurosurg 106:384-387, 2007. 52. Barker FG, Jannetta PJ, Bissonette DJ, et al: Microvascular decompression for hemifacial spasm. J Neurosurg 82:201-210, 1995. 53. Iwakuma T, Matsumoto A, Nakamura N: Hemifacial spasm: Comparison of three different operative procedures in 110 patients. J Neurosurg 57:753-756, 1982. 54. Auger RG, Piepgras DG, Laws ER Jr: Hemifacial spasm: Results of microvascular decompression of the facial nerve in 54 patients. Mayo Clin Proc 61:640-644, 1986. 55. Loeser J D, Chen J: Hemifacial spasm: Treatment by microsurgical facial nerve decompression. Neurosurgery 13:141-146, 1983. 56. Møller MB, Møller AR, Jannetta PJ, et al: Microvascular decompression of the eighth nerve in patients with disabling positional vertigo: Selection criteria and operative results in 207 patients. Acta Neurochir (Wien) 125:75-82, 1993. 57. Jannetta PJ: Trigeminal neuralgia [Letter]. Neurosurgery 18:677, 1986. 58. Jannetta PJ: Neurovascular cross-compression in patients with hyperactive dysfunction symptoms of the eighth cranial nerve. Surg Forum 26:467-469, 1975. 59. Jannetta PJ: Neurovascular cross compression of the eighth nerve in patients with vertigo and tinnitus. In Samii M, Jannetta PJ (eds): The Cranial Nerves. Heidelberg, Springer-Verlag, 1981. 60. Resnick D K, Jannetta PJ, Bissonnette D, et al: Microvascular decompression for glossopharyngeal neuralgia. Neurosurgery 36:64-68, 1995.
Chapter 44 • Operations for Vascular Compressive Syndromes 61. Jannetta PJ, Gendell H M : Clinical observations on etiology of essential hypertension. Surg Forum 30:431-432, 1979. 62. Granata A R , Ernsberger P, Reis DJ: Hypotension and bradycardia elicited by histamine into the C1 area of the rostral ventrolateral medulla. Eur J Pharmacol 136:157162, 1987. 63. Morrison S H, Milner TA, Reis DJ: Reticulospinal vasomotor neurons of the rat rostral ventrolateral medulla: Relationship of sympathetic nerve activity and the C1 adrenergic cell group. J Neurosci 8:1286-1301, 1988. 64. Segal R , Gendell H M, Canfield D, et al: Cardiovascular response to pulsatile pressure applied to ventrolateral medulla. Surg Forum 30:433-435, 1979. 65. Segal R, Jannetta PJ, Wolfson S K J, et al: Implanted pulsatile balloon device for simulation of neurovascular compression syndromes in animals. J Neurosurg 57:646-650, 1982. 66. Naraghi R , Gaab M R , Walter G F: Neurovascular compression as a cause of essential hypertension: A microanatomical study. Adv Neurosurg 17:182-186, 1989. 67. Naraghi R , Gaab M R , Walter G F, Kleineberg B : Arterial hypertension and neurovascular compression at the ventrolateral medulla: A comparative microanatomical and pathological study. J Neurosurg 77:103-112, 1992.
535
68. Kleineberg B, Becker H, Gaab M R : Neurovascular compression and essential hypertension: An angiographic study. Neuroradiology 33:2-8, 1991. 69. Kleineberg B, Becker H, Gaab M R , Naraghi R : Essential hypertension associated with neurovascular compression. Neurosurgery 30:834-841, 1992. 70. Naraghi R , Geiger H, Crnac J, et al: Posterior fossa neuro vascular anomalies in essential hypertension. Lancet 344:1466-1470, 1994. 71. Colon G P, Quint DJ, Dickenson L D, et al: Magnetic resonance imaging of ventrolateral medullary compression in essential hypertension. J Neurosurg 88:226-231, 1998. 72. Jannetta PJ, Segal R , Wolfson S K J: Neurogenic hypertension: Etiology and surgical treatment, I: Observations in 53 patients. Ann Surg 201:391-398, 1985. 73. Jannetta PJ, Segal R , Wolfson S K, et al: Neurogenic hypertension: Etiology and surgical treatment, II: Obser vations in an experimental nonhuman primate model. Ann Surg 202:253-261, 1985.
45
Drainage Procedures for Petrous Apex Lesions Derald E. Brackmann, Neil A. Giddings, and Eric P. Wilkinson Videos corresponding to this chapter are available online at www.expertconsult.com.
With the help of cranial computed tomography (CT) and magnetic resonance imaging (MRI), petrous apex lesions can be correctly diagnosed preoperatively. Consequently, cystic lesions requiring drainage should be approached with procedures designed for drainage, not total en bloc removal. Although the transmastoid infralabyrinthine procedure has been the usual approach for these lesions, transcanal infracochlear drainage offers a more dependent drainage site. The eventual role of the various approaches for petrous apex drainage depends on the long-term follow-up of these patients. Advances in radiologic imaging during the past 2 decades have made it possible to differentiate reliably lesions of the petrous apex preoperatively. The development of CT scanning was the first major step in imaging the temporal bone since the development of polytomography. CT scanning gives the surgeon the ability to visualize the size of the lesion and its relationship to vital structures, including the internal auditory canal (IAC), cochlea, vestibular labyrinth, carotid artery, and jugular bulb. It also helps characterize the border of the lesion as expansile or invasive, which may differentiate between benign lesions and malignant neoplasms. MRI of the temporal bone added the capability of characterizing the substance of the lesion, rather than its effect on bony interfaces, allowing the surgeon to distinguish between mucus, fat, cholesterol granuloma, cholesteatoma, and neoplasm. The combination of CT, with its superior bone imaging algorithms, and MRI, with its enhanced tissue imaging capabilities, allows the surgeon to differentiate accurately and reliably among benign cystic lesions, normal anatomic variants, and neoplastic lesions of the petrous apex. Patients with petrous apex lesions present with various symptoms and physical findings. The most widely recognized finding is Gradenigo’s syndrome, consisting of retro-orbital pain, otorrhea, and ipsilateral sixth cranial nerve paresis. The signs and symptoms of noninflammatory or neoplastic lesions may be more subtle. Hearing loss and vestibular abnormalities are frequently associated with lesions of the petrous apex as they enlarge and compress the IAC. The facial nerve is relatively resistant
to paresis from slowly expansile lesions, but may be involved early with neoplastic lesions of the petrous apex or non-neoplastic lesions compressing the IAC. Pain may be present with benign or cystic lesions, but is more common in neoplastic lesions. Its distribution depends on the region involved. The mastoid cavity is innervated by CN IX and may radiate pain into the neck. The middle fossa and superior petrosal regions are innervated by CN V and may cause retro-orbital or facial pain. Lesions extending into the posterior fossa may cause pain along the routes of distribution of CN IX and X and the first three cervical nerves.1 Although 80% of adult mastoid bones are aerated, only 30% of petrous bones have air cells extending to the apex, and 7% may have asymmetric pneumatization of the petrous apex.2,3 The increasing use of MRI of the head and neck makes it imperative that the clinician can differentiate pathology from normal variant in the temporal bone. Table 45- 1 summarizes common lesions of the petrous apex and their associated radiologic findings. Asymmetric pneumatization is clearly seen on CT scanning, but supplementary MRI may be needed to rule out pathologic lesions in a symptomatic ear. Normal bony architecture can be seen on CT scanning, with hyperintensity on MRI T1-weighted scans and hypointensity on T2weighted scans because of the large fat content in marrow. Retained mucus in the air cells also appears with normal bony architecture on CT scan, but is hypointense on T1weighted MR images and hyperintense on T2-weighted MR images. Cholesteatoma is usually associated with chronic otitis media, but may arise from congenital rest cells in the petrous apex. Because of its high water content, cholesteatoma is isointense with cerebrospinal fluid on CT, and is hypointense on T1-weighted MR images and hyperintense on T2-weighted MR images. Cholesterol granuloma is isointense with brain on CT scanning and appears as a classic image on MRI with hyperintensity on T1-weighted and T2-weighted scans. Radiologic descriptions of other, less common lesions are also summarized in Table 45-1. Differentiating between chordoma, chondroma, and chondrosarcoma of 537
538
OTOLOGIC SURGERY
TABLE 45-1 Radiologic Appearance of Common Petrous Apex Lesions MAGNETIC RESONANCE IMAGING Lesion
Computed Tomography
T1 Weighting
T2 Weighting
Enhancement
Retained mucus Mucocele
Normal bony architecture, nonenhancing Hypodense, expansile smooth border, nonenhancing, normal bony architecture Normal bony architecture, nonenhancing
Hypointense Hypointense
Hyperintense Hyperintense
No No
Hyperintense
Hypointense
No
Loss of normal air cells, nonenhancing, isointense with CSF Expansile smooth border, occasional rim enhancement, isointense with brain Destructive, indistinct border Aggressive bone destruction, calcification
Hypointense
Hyperintense
No
Hyperintense
Hyperintense
No
Isointense Isointense—75% Hypointense—25% Hypointense to isointense
Hyperintense
Yes
Hyperintense
Yes
Asymmetric pneumatization Cholesteatoma Cholesterol granuloma Metastatic lesion
Chondroma
Aggressive bone destruction, calcification
CSF, cerebrospinal fluid.
the temporal bone remains difficult, even with the scanning techniques currently available. The area of origin and the age of the patient must be considered when the pathology of destructive lesions of the petrous apex is determined.4-6
PATIENT SELECTION Most surgical approaches to the petrous apex were developed in the preantibiotic era for drainage of petrous apex abscesses and cure of Gradenigo’s syndrome. With the arrival of modern antibiotics, infectious processes of the petrous apex have greatly declined in frequency, but these same approaches may be equally effective in draining cystic lesions of the petrous apex. Air cell tracts extend above, below, and anterior to the otic capsule, allowing the potential of safe passage to the petrous apex. Approaches that follow superior air cell tracts include middle fossa,7 through the superior semicircular canal,8 the attic, and the root of the zygomatic arch.9 Approaches below the inner ear include the infralabyrinthine and the infracochlear.10-13 Anterior approaches have been described by Ramadier,14 Eagleton,15 and Lempert,16 who used the triangle between the anterior border of the cochlea, the carotid artery, and the middle fossa dura. All these approaches are used for drainage of inflammatory disease processes that are not responsive to antibiotic therapy or simpler operations for chronic ear disease (Fig. 45-1). Infralabyrinthine, infracochlear, and transsphenoidal approaches are most commonly chosen for drainage of cystic lesions of the petrous apex in an ear with serviceable hearing. These lesions are frequently detected at an asymptomatic stage with current imaging techniques. Because the natural history of small benign cystic lesions is not well documented, surgical drainage should be reserved for patients with larger lesions or with
s ymptoms including pain, visual changes, diplopia, hearing loss, vertigo, or facial nerve weakness. For patients without serviceable hearing, these lesions should be drained through a translabyrinthine approach. Because other vital structures may be affected by enlargement of the cyst, delaying surgery in symptomatic patients provides no advantage. Cholesterol granuloma is the most common cystic lesion of the petrous apex, occurring 30 times less frequently than acoustic neuroma.17 It may develop in any aerated portion of the temporal bone, but most commonly occurs in the mastoid air cells distant to a lesion that prevents normal aeration. Cholesterol granuloma of the petrous apex probably develops when a pathologic process or trauma obstructs the air cell tracts to a wellpneumatized petrous apex. The treatment for cholesterol granuloma of the temporal bone is drainage and re-establishment of adequate aeration to the involved area. The cyst wall is composed of a fibrous connective tissue. It is free of keratinizing squamous epithelium that characterizes cholesteatoma, and complete removal of the cyst is unnecessary. Solid tumors of the temporal bone and cholesteatoma are removed when first identified, rather than after further symptoms develop because these symptoms frequently reveal further involvement of other vital structures. Drainage procedures are inadequate treatment for these lesions, and all reasonable efforts should be made to remove them entirely. Total removal may require the sacrifice of cranial nerves and major vascular structures.
PREOPERATIVE EVALUATION AND PATIENT COUNSELING Preoperative evaluation of patients is based on their symptoms. Patients presenting with hearing loss are evaluated initially with audiometric testing, including air, bone, and
Chapter 45 • Drainage Procedures for Petrous Apex Lesions
539
FIGURE 45-1. Surgical approaches for drainage of petrous apex. FIGURE 45-2. A, Exposure of cyst in infralabyrinthine cell tract. An island of bone protects retracted sigmoid sinus (Bill’s island). B, Silicone tube placed into interior of cavity drains into inferior mastoid cavity.
540
OTOLOGIC SURGERY
speech reception thresholds and speech discrimination scores. Electronystagmography is performed in patients who complain of imbalance or vertigo. In patients with otherwise normal results on physical examination, asymmetric hearing is next evaluated with auditory brainstem response testing. If these results are abnormal, MRI is indicated. In patients with cranial nerve involvement other than CN VIII with asymmetric hearing, auditory brainstem response testing is not performed, and the physician proceeds directly to MRI. Patients who have normal hearing but have other cranial nerve deficits that may be referable to the petrous apex may be screened with a high-resolution, thin-section CT scan of the temporal bone or an MRI scan with gadolinium. If an abnormality is found, all patients undergo air, bone, and speech reception thresholds and speech discrimination audiometric testing before surgery to document hearing levels before a procedure that jeopardizes hearing. Preoperatively, patients are counseled to expect resolution of pain, if present, and the possibility of improvement in cranial nerve function if it is decreased preoperatively. Cranial nerves that have been affected for shorter periods seem to have a better prognosis and fewer long-standing deficits than nerves that have been affected longer. Patients are reminded that this is a drainage procedure with the goal of decompressing the lesion and providing an aerated cavity, if possible. The goal is not the removal of the lesion, and close follow-up may be necessary. Recurrence of the lesion secondary to inadequate drainage is usually heralded by the return of preoperative symptoms. Follow-up MRI frequently reveals a cholesterol granuloma cyst that remains full of fluid, but the T1-weighted image is hypointense compared with the preoperative hyperintense image on T1-weighted views. A return of hyperintensity on the T1-weighted image suggests inadequate drainage in a symptomatic lesion.18
SURGICAL TECHNIQUES Infralabyrinthine Drainage of the Petrous Apex In preparation for infralabyrinthine drainage of the petrous apex, the patient is prohibited from eating and drinking for at least 8 hours preoperatively. Unless an infectious process is suspected, no preoperative antibiotics are used. The surgical ear is prepared similarly to any other chronic otitis media case operated on through a postauricular approach. Hair is shaved to one fingerbreadth above the auricle and two fingerbreadths behind the postauricular crease. A facial nerve monitor is typically used, and monitoring electrodes are placed in the standard positions. Surgical preparation is an antibacterial scrub followed by painting with antibacterial solution. Sterile Mastisol is applied around the auricle and allowed to dry.
An adhesive aperture drape is placed over the ear, and sterile sheets cover the patient. Routine chronic otitis media instruments and drill are the only equipment required. 1. The patient is placed in a supine position with the involved ear facing up. The surgeon sits at the side of the patient with the patient’s head turned away. 2. A postauricular incision is made 1 cm behind the postauricular crease down to the temporalis fascia superiorly, and through the periosteum below the temporal line. A second incision is made through the temporalis muscle and fascia at the temporal line, beginning above the external auditory canal (EAC) and ending posteriorly at the postauricular incision, allowing the periosteum to be raised and the ear reflected anteriorly. 3. A simple mastoidectomy is performed, removing all air cells from Trautmann’s triangle (bordered by the middle fossa dural plate superiorly, the semicircular canals anteriorly, and the sigmoid sinus posteriorly) (Fig. 45-2A). 4. The facial nerve should be identified in its vertical portion, but need not be exposed. 5. Mastoid air cells are removed from the mastoid tip, and the sigmoid sinus is followed until the jugular bulb is identified. The superior aspect of the jugular bulb forms the inferiormost portion of the opening into the petrous apex. 6. The posterior half of the horizontal and the inferior portion of the posterior bony semicircular canal are skeletonized, and care is taken not to expose the membranous portions of the canals. 7. When the semicircular canals and the jugular bulb are clearly defined, the infralabyrinthine air cell tract is followed toward the petrous apex with a diamond burr or curette until the cystic lesion is encountered and opened. 8. When the lesion is entered, it is evacuated with suction and copious irrigation. All fluid and loose debris are removed. Removal of tissue lining the cavity is neither necessary nor possible with this approach. 9. The largest silicone catheter that fits into the newly created opening is placed and retained by friction to prevent stenosis of the drainage site (Fig. 45-2B). 10. The periosteum is reapproximated with absorbable suture, and the subcutaneous tissue and skin are closed in layers, followed by the placement of a mastoid pressure dressing. The mastoid pressure dressing is formed by placing Telfa over the postauricular incision. One half a cotton ball is placed in the concha, and two 4 × 4 gauze pads are folded in half and placed in the postauricular crease. A soft, absorbent bolster is placed over the auricle and the mastoid region. A 4 inch Kling bandage is wrapped anteriorly to posteriorly and secured by cinching down with a tracheotomy tie 1 inch posterior to the lateral orbital rim.
Chapter 45 • Drainage Procedures for Petrous Apex Lesions
Most patients are admitted to the hospital for 1 night after this surgery. The dressing is removed the following day, and the incision is cleaned, if necessary. If skin sutures were used, they are removed 1 week later. The patient is allowed to get the postauricular incision wet at this time and should be instructed to clean the incision once per day for the next 2 weeks with hydrogen peroxide to reduce crusting on the incision.
Pitfalls Infralabyrinthine drainage has been the most commonly employed procedure in the United States for drainage of these cystic lesions. It is not without risk, however, to the facial nerve, jugular bulb, and otic capsule. Surgeons attempting this procedure should be intimately familiar with the neural and major vascular anatomy of the petrous bone. When a large cystic lesion expands along the infralabyrinthine cell tracts, this approach may be the simplest. With a small lesion, following this cell tract to the area of pathology may be difficult. Probably the greatest limitation to this approach is its extreme difficulty when used in patients with a high jugular bulb. The drainage opening may be narrowed significantly, even if it is possible to decompress the jugular bulb and retract it inferiorly during the surgical procedure. Consequently, this approach should not be chosen when the jugular bulb is high or narrows the surgical approach to the apex. Injury to the jugular bulb can be temporarily controlled with external packing and light pressure. At this time, the surgeon should arm himself or herself with adequate suction and large sheets of absorbable knitted fabric (Surgicel). The Surgicel is removed, and the damaged jugular bulb is inspected as well as possible. For small lacerations, a small sheet of Surgicel is placed over the bulb and packed in place with absorbable gelatin sponge (Gelfoam). A large defect in the jugular bulb requires that a large piece of Surgicel be placed through the defect into the lumen of the vein. Small pieces should not be placed near the defect or into the lumen because of the possibility of embolization to the lung. Gelfoam is packed over the Surgicel, usually controlling the bleeding with light pressure. Injury to the posterior or horizontal semicircular canal requires early recognition of the surgical misadventure. Little or no suction should be used to inspect the fenestration. A small piece of temporalis fascia is placed over the defect, and bone wax is used to secure the fascia and provide a watertight seal. If recognized early, hearing may be preserved, although vertigo is present for several weeks in the postoperative period.
Transcanal Infracochlear Approach to the Petrous Apex In 1984, Farrior11 described a transcanal approach to small glomus jugulare tumors of the hypotympanum. The infracochlear approach to the petrous apex12 is a
541
combination of this technique and the subcochlear approach described by Ghorayeb and Jahrsdoerfer.13 Facial nerve monitoring is used. 1. A postauricular incision is made, and the auricle is reflected anteriorly. 2. The membranous EAC is completely transected laterally (Fig. 45-3A). 3. A tympanomeatal flap is elevated from 2 to 10 o’clock, leaving the tympanic membrane attached at the umbo and the superior canal wall. 4. The EAC is enlarged anteriorly and inferiorly to expose the hypotympanum. The chorda tympani is followed inferoposteriorly to laterally to define the extent of posterior dissection possible without injury to the facial nerve (Fig. 45-3B). 5. Air cells are removed below the cochlea in the hypotympanum to expose the course of the carotid artery and the jugular bulb. The round window provides the superior line of dissection, and Jacobson’s nerve leads to the “crutch” of the carotid and jugular bulb (Fig. 45-3C). 6. Removal of air cells continues medially. If the plane of dissection remains below the round window, the IAC structures are not at risk (Fig. 45-3D). 7. The cholesterol granuloma cyst is entered and drained. The newly created “window” is enlarged as far anteriorly as possible to the carotid artery, as far inferiorly as the jugular bulb, and superiorly to the inferior aspect of the basal turn of the cochlea (Fig. 45-3E). 8. A silicone catheter of appropriate size is introduced to stent the opening (Fig. 45-3F). 9. The soft tissue of the EAC is returned to its normal position, Gelfoam is packed within the membranous EAC, and bone pâté from the mastoid cortex is placed between the canal wall and the newly enlarged bony canal (Fig. 45-3G). 10. The postauricular incision is closed, and a mastoid dressing is applied (Fig. 45-3H). Postoperative dressings, care, and length of hospitalization are similar to the infralabyrinthine approach except in the care of the EAC. The patient is seen 1 week postoperatively to check the postauricular incision, and to ensure that excessive drainage is not occurring from the EAC. The packing should be moist without evidence of active drainage. If excessive drainage or evidence of infection is present, the patient is prescribed antibiotic ear drops (Cortisporin Otic suspension) three times a day. If the packing is dry, the patient begins using the ear drops 2 weeks after surgery, 2 drops three times a day, until the packing is removed 1 month postoperatively. At this point, the tympanic membrane should be intact and the canal skin healing well. Small areas of exposed bony EAC epithelialize within 1 to 2 months. This procedure is similar to the infralabyrinthine approach, in that both involve an area of the temporal
542
OTOLOGIC SURGERY
FIGURE 45-3. A-H, Transcanal infracochlear approach to petrous apex. EAC, external auditory canal.
Chapter 45 • Drainage Procedures for Petrous Apex Lesions
FIGURE 45-3, cont’d.
543
544
OTOLOGIC SURGERY
TABLE 45-2 Common Drainage Approaches to the Petrous Apex Procedure
Structures at Risk
Advantages
Disadvantages
Infralabyrinthine
Jugular bulb, bony labyrinth, facial nerve
Direct approach that is more familiar to most otolaryngologists
Difficult with high jugular bulb Drainage into mastoid cavity far from eustachian tube
Transcanal infracochlear
Jugular bulb, carotid artery, cochlea
Transsphenoidal
Carotid artery, optic nerve, cavernous sinus, maxillary nerve, pituitary gland
Anatomy may be challenging to surgeons not familiar with hypotympanum and major vessels of temporal bone Sphenoid anatomy highly variable Can be used only for “giant” cysts in contact with sphenoid High rate of failure
Translabyrinthine or subtotal petrosectomy Middle fossa
Labyrinth, jugular bulb, internal auditory canal Temporal lobe, cochlea, labyrinth, internal auditory canal, greater superficial petrosal nerve
Direct drainage near mouth of eustachian tube Can be revised with transtympanic procedure Direct approach to large cysts that are in contact with posterior wall of sphenoid sinus Opening into cyst may be directly observed in clinic with endoscope Cyst and wall can be removed if desired Drains cyst directly into bony eustachian tube
bone to which most otolaryngologists have little exposure. The surgery need not be difficult, but surgeons must spend time in the temporal bone laboratory familiarizing themselves with the relative positions of the basal turn of the cochlea, the jugular bulb, and the carotid artery. After the soft tissue of the canal is reflected superiorly, a large amount of the bone lateral to the annulus can be removed with a cutting burr. When the hypotympanum is entered, the surgeon should switch to successively smaller diamond burrs until the cyst is entered. Injury to the jugular bulb is treated as was described for the infralabyrinthine approach. Injury to the basal turn of the cochlea is more serious than opening into a semicircular canal. It should be approached with minimal suction, and the placement of temporalis fascia should be secured by bone wax. Despite early recognition, the prognosis for hearing is much poorer when the cochlea is violated than when the semicircular canals are violated. This approach may cause injury to the infratemporal portion of the carotid artery. The carotid arterial wall may be thinner in the temporal bone than in the neck. Every effort should be made to leave a thin wall of bone over the carotid artery. If the artery is violated, immediate control can usually be achieved with packing and pressure to the middle ear and EAC. This injury is potentially life-threatening and requires the expertise of a vascular surgeon. Distal control may be achieved with an intraarterial catheter threaded past the point of injury, and proximal control is achieved in the neck. When proximal and distal control are achieved, the injury can be directly
Profound postoperative sensorineural hearing loss Middle fossa craniotomy poorly tolerated in elderly Drainage is not dependent Temporal lobe may obstruct drainage path
repaired. Occlusion of the carotid artery runs the risk of cerebral infarct.
RESULTS Drainage procedures are expected to relieve pain, if it was present preoperatively, and may allow the recovery of some cranial nerve dysfunction. Gherini and associates17 reported that hearing was preserved in 83% of patients who had useful preoperative hearing, but no improvement occurred in patients with preoperative hearing impairment. Vertigo, if present preoperatively, usually resolves with adequate decompression.19 Several patients displayed recovery of cranial nerve function other than the cochlear nerve, but this return of function is not universal, even after adequate decompression.17,19 Stenting with silicone catheters is mandatory in cases employing transcanal infracochlear drainage because patients typically require revision surgery if stenting is not performed.20
COMPLICATIONS AND MANAGEMENT All procedures that approach the petrous apex may cause injury to major vessels and cranial nerves. Each approach has its own set of risks. Table 45-2 compares the common procedures used for drainage of petrous apex lesions. The experience of the surgeon, the position of the lesion, and the needs of the patient all need to be addressed.
Chapter 45 • Drainage Procedures for Petrous Apex Lesions
The infralabyrinthine approach is the most familiar to otolaryngologists and head and neck surgeons. It is a direct extension of the simple mastoidectomy that may be accomplished with minimal morbidity in patients with a large cyst that has expanded between the labyrinth and the jugular bulb. Smaller cysts that are more medially based become more challenging, and a high jugular bulb may preclude this approach entirely. Care must be taken not to disturb the endolymphatic sac or damage the endolymphatic duct during this dissection. Damage to these areas may lead to endolymphatic hydrops. This technique drains the cyst farther from the eustachian tube than any other otologic technique and, consequently, may be more prone to failure. Because of the debris that is never completely removed from these cysts, the mastoid itself may become poorly aerated, leading to the potential for long-term failure. The transcanal infracochlear approach has the advantage of a direct approach to the petrous apex and the maintenance of normal ear anatomy. It requires familiarity with the surgical anatomy of the hypotympanum, but the surgical landmarks are easily identified. The round window is easily seen, and careful removal of bone with a diamond burr reveals the location of the jugular bulb and carotid artery. If major vessel damage occurs, it is easily temporized with direct pressure until the situation can be controlled with direct vessel repair or permanent packing. It also drains the cyst close to the opening of the eustachian tube into the middle ear space, which makes obstruction less likely. Complete transection of the soft tissue EAC may lead to mild stenosis in the postoperative period. If this occurs, it is easily managed and corrected with the placement of a foam ear insert used by audiologists for routine audiometric testing. The insert maintains gentle pressure on the area of stenosis while allowing passage of sound down the central lumen.
ALTERNATIVE TECHNIQUES Several other techniques are available for exposure of the petrous apex.
Transsphenoidal Approach to Petrous Apex Lesions Transsphenoidal drainage of petrous apex cysts may be the most simple and straightforward approach to large cysts that have a large surface area against the posterior wall of the sphenoid sinus. This technique is not useful when the cyst does not impinge on the sphenoid sinus. The risk of damage to the cochlea and labyrinth is decreased with this technique, but the risk to the carotid artery increases, and an added risk of damage to the optic nerve exists (a reported
545
c omplication of endoscopic sinus surgery). This technique also has a high failure rate compared with otologic drainage procedures for cholesterol granuloma of the petrous apex. Thedinger and colleagues19 reported that 80% of patients who had a transsphenoidal initial approach for drainage eventually required one or more procedures to drain the lesion effectively. In their series, the only patient who required revision surgery after an otologic procedure had previously undergone a middle fossa attempt that did not provide dependent drainage. 1. An external ethmoidectomy, sphenoidotomy, intranasal sphenoethmoidectomy, intranasal sphenoidotomy, transseptal sphenoidotomy, or transpalatal approach may be used to expose the posterior and lateral walls of the sphenoid sinus.21-23 2. Indentations in the lateral and superior walls may reveal the location of vital nervous and vascular structures, including the pituitary gland, optic nerve, maxillary nerve, carotid artery, and cavernous sinus. 3. A cruciate incision is made in the posterior sphenoid sinus mucosa, and flaps are elevated from the bony posterior wall. 4. Based on preoperative CT scanning, a diamond burr or sharp chisel is used to remove a small portion of the posterior bony wall of the sphenoid sinus. Pituitary rongeurs are used to enlarge the bony orifice as much as possible without violating vital structures, exposing the cystic lesion. 5. The cyst wall is opened, and the contents are drained. The cyst is irrigated, and all debris is removed. The mucosal flaps previously raised are laid into the cavity, and no synthetic drain is used. 6. A large sphenoidotomy is performed for postoperative inspection and cleaning. 7. If an intranasal approach was used, the sphenoid sinus should remain unpacked, but the nasal cavity may be lightly packed. If an external approach was chosen, the wound is closed in layers. With the increased interest in endoscopic sinus surgery, the transsphenoidal approach to the petrous apex is gaining in popularity. This approach does not work in smaller lesions that do not directly abut the posterior wall of the sphenoid sinus. Many endoscopists have gained a great deal of expertise with the surface anatomy of the sphenoid and other paranasal sinuses. This procedure should not be confused with routine sinusotomy or removal of infected mucosa from the sinus walls. This approach to draining the petrous apex requires bone removal from the posterior wall of the sphenoid sinus, placing the carotid artery and optic nerve at risk. Skilled endoscopists perform this procedure with a low complication rate, but endoscopists with less experience need to note that the sphenoid sinus is the most variable in
546
OTOLOGIC SURGERY
form of any bilateral cavity or organ in the human body. The sphenoid sinus may vary in length from 4 to 44 mm, in width from 2.5 to 34 mm, and in height from 5 to 33 mm.23
Subtotal Petrosectomy Fisch and Mattox24 described a subtotal petrosectomy in combination with removal of the otic capsule to gain full exposure of the petrous apex. This technique offers the advantage of direct exposure from the carotid artery to the sigmoid sinus in an anteroposterior direction and exposure from the hypotympanum to the middle fossa dura in an inferosuperior direction. The contents of the IAC are left undisturbed, and the facial nerve does not require transposition, making possible complete removal of benign lesions of the petrous apex, including cystic structures with their lining matrix. This increased exposure results in total sensorineural hearing loss with the removal of the otic capsule. 1. A postauricular incision is made down to the temporalis fascia superiorly and through the periosteum posteriorly and inferiorly. A second incision is made along the temporal line through the temporalis muscle, the auricle is reflected anteriorly, and the cartilaginous canal is transected. EAC epithelium is everted, and the canal is sutured closed with a layer of periosteum sutured medially to form a two-layer closure (Fig. 45-4A). 2. A canal wall down tympanomastoidectomy is performed, and the posterior canal wall is lowered to the level of the facial nerve (Fig. 45-4B). 3. The mastoid tip cells are removed to the level of the digastric muscle (Fig. 45-4C). 4. The remaining EAC skin, tympanic membrane, and annulus are removed after section of the incudostapedial joint and tensor tympani muscle (Fig. 45-4D). 5. Retrolabyrinthine, supralabyrinthine, supratubal, infralabyrinthine, and retrofacial air cells are removed under direct vision. The bony labyrinth and cochlea are also removed without transposition of the facial nerve (Fig. 45-4E). 6. The lesion and its matrix or capsule are removed under direct vision (Fig. 45-4F). 7. The remaining mucosa is gently curetted from the middle ear space. The introitus from the middle ear to the eustachian tube is drilled with a diamond burr, and the remaining mucosa is cauterized. The tube is obliterated with bone wax, muscle, and fibrin glue (Fig. 45-4G). 8. The cavity is obliterated with abdominal fat, and a temporalis muscle flap is rotated inferiorly, then sutured lateral to the fat graft (Fig. 45-4H and I). 9. The incision is closed in layers, and a mastoid dressing is applied.
Middle Fossa Drainage of the Petrous Apex Lesions of the superior petrous apex air cells may not be amenable to drainage via the transcanal infracochlear technique. These lesions may be approached through a middle fossa craniotomy with drainage of the cyst. The steps in a middle fossa approach to the petrous apex are described elsewhere. The IAC is identified as necessary, but its bony covering is left. When the cyst is identified at the petrous apex, it is entered and drained. A portion of the cyst wall around the bony opening is removed to allow for continued drainage. An opening is made through the tegmen mastoideum into the mastoid, near the antrum. A silicone catheter of the ventriculostomy type is cut to size and is placed with one end in the cyst and the other end in the mastoid (Fig. 45-5). Bone wax may be used as necessary to secure the catheter into position.25 The dura is allowed to relax, and the surgical wound is closed in the standard fashion.
CONCLUSION The combination of high-resolution CT and MRI makes it possible to predict reliably the character of petrous apex lesions preoperatively. Solid tumors and cholesteatoma may require destructive procedures, with cranial nerve and major vessel sacrifice for complete removal.
FIGURE 45-4. A-I, Subtotal petrosectomy approach to petrous apex. EAC, external auditory canal.
Chapter 45 • Drainage Procedures for Petrous Apex Lesions
FIGURE 45-4, cont’d.
547
548
OTOLOGIC SURGERY
FIGURE 45-4, cont’d.
Chapter 45 • Drainage Procedures for Petrous Apex Lesions Petrous apex cyst opened
Opening into mastoid through tegmen mastoideum
Ae V
FIGURE 45-5. Middle fossa drainage of petrous apex.
Cystic lesions may be approached conservatively. Providing long-term adequate drainage frequently relieves all symptoms and provides long-term control. We prefer the transcanal infracochlear approach to the petrous apex for cystic lesions because of its direct approach, dependent drainage, and ease of revision if necessary.
REFERENCES 1. Hollinshead WH : The ear. In Anatomy for Surgeons, vol 1. Philadelphia, Harper & Row, 1982, pp 159-221. 2. Myerson MC, Rubin J, Gilbert JG: Anatomic studies of the petrous portion of the temporal bone. Arch Otolaryngol Head Neck Surg 20:195-210, 1934. 3. Roland PS, Meyerhoff WL , Judge LO, Mickey B E : Asymmetric pneumatization of the petrous apex. Otolaryngol Head Neck Surg 103:80-88, 1990. 4. Lipper M H, Cail WS : Chordoma of the petrous bone. South Med J 84:629-631, 1991. 5. Sekhar L N, Pomeranz S, Janecka I P, et al: Temporal bone neoplasms: A report on 20 surgically treated cases. J Neurosurg 76:578-587, 1992. 6. Bourgouin PM, Tampieri D, Robitaille Y, et al: Lowgrade myxoid chondrosarcoma of the base of the skull: CT, MR, and histopathology. J Comput Assist Tomogr 16:268-273, 1992. 7. Hendershot E L , Wood JW: The middle fossa approach in the treatment of petrositis. Arch Otolaryngol Head Neck Surg 98:426-427, 1973.
549
8. Frenckner P: Some remarks on the treatment of apicitis (petrositis) with or without Gradenigo’s syndrome. Acta Otolaryngol (Stockh) 17:97-120, 1932. 9. Mawson S R : Complications of otitis media. In Ludman H (ed): Mawson’s Diseases of the Ear. Baltimore, Williams & Wilkins, 1963, pp 347-352. 10. Dearmin R M : A logical approach to the tip cells of the petrous pyramid. Arch Otolaryngol Head Neck Surg 26:314-320, 1937. 11. Farrior J B : Anterior hypotympanic approach for glomus tumor of the infratemporal fossa. Laryngoscope 94:10161020, 1984. 12. Giddings N A, Brackmann D E, Kwartler J A : Transcanal infracochlear approach to the petrous apex. Otolaryngol Head Neck Surg 104:29-36, 1991. 13. Ghorayeb BY, Jahrsdoerfer R A : Subcochlear approach for cholesterol granulomas of the inferior petrous apex. Otolaryngol Head Neck Surg 103:60-65, 1990. 14. Ramadier J: Exploration de la pointe du rocher par la voie du canal carotidien. Ann Otolaryngol 4:422-444, 1933. 15. Eagleton WP: Unlocking the petrous apex for localized bulbar (pontile) meningitis secondary to suppuration of the petrous apex. Arch Otolaryngol Head Neck Surg 13:386-422, 1931. 16. Lempert J: Complete apicectomy (mastoidotympanoapicectemy). Arch Otolaryngol Head Neck Surg 25:144-177, 1937. 17. Gherini SG, Brackmann D E, Low WW, Solti-Bohman LG: Cholesterol granuloma of the petrous apex. Laryngoscope 95:659-664, 1985. 18. Jackler R K, Parker D A : Radiographic differential diag nosis of petrous apex lesions. Am J Otol 13:561-574, 1992. 19. Thedinger B A, Nadol J B, Montgomery WW, et al: Radiographic diagnosis, surgical treatment, and longterm follow-up of cholesterol granulomas of the petrous apex. Laryngoscope 99:896-907, 1989. 20. Brackmann D E, Toh E H : Surgical management of petrous apex cholesterol granulomas. Otol Neurotol 23:529-533, 2002. 21. Davis A E, Robin PE : Transpalatal approach to the petrous apex. J Laryngol Otol 103:94-96, 1989. 22. Congdon E : The distribution and mode of origin of septa and walls of the sphenoid sinus. Anat Rec 18:97, 1920. 23. Van Alyea O E : Sphenoid sinus: Anatomic study, with consideration of the clinical significance of the structural characteristics of the sphenoid sinus. Arch Otolaryngol Head Neck Surg 34:225, 1941. 24. Fisch U, Mattox D: Microsurgery of the Skull Base. New York, Thieme, 1988. 25. Arriaga M A : Petrous apex effusion: A clinical disorder. Laryngoscope 116:1349-1356, 2006.
46
Surgery for Glomus Tumors and Other Lesions of the Jugular Foramen Moisés A. Arriaga and Derald E. Brackmann Videos corresponding to this chapter are available online at www.expertconsult.com.
The therapy for glomus tumors of the temporal bone is controversial. Because the clinical characteristics and growth rates of these tumors vary, the full gamut of management has been recommended from observation alone through radiation therapy and surgical management. Although there are isolated case reports of prolonged survival without treatment, these lesions can be quite deadly. The studies of Rosenwasser,1 Bickerstff and Howell,2 Steinberg and Holz,3 Brown,4 and Spector and colleagues5 have documented mortality rates of 5% to 13% for glomus jugulare tumors. This chapter outlines the diagnostic and preoperative evaluation, surgical techniques, and results and complications in the management of glomus tumors of the temporal bone. The application of these surgical techniques for other lesions of the jugular foramen is also reviewed.
PATIENT SELECTION Temporal bone glomus tumors are neoplasms of the normal paraganglioma in the temporal bone that principally occur in the adventitia of the dome of the jugular bulb, but are also found in the submucosa of the cochlear promontory within the tympanic plexus. Numerous classification schemes have been proposed for these lesions, primarily by their origin (tympanic plexus versus jugular bulb) and the anatomic extent of lesion. The clinical surgical classification proposed by De La Cruz is particularly useful in planning the clinical management of patients with glomus tumors. The extent of the tumor is described by the involvement of structures of the temporal bone and skull base. A series of operations that correspond to the extent of the tumor is used (Table 46-1). Other classification schemes include the Fisch classification (Table 46-2) and the Glasscock-Jackson classification (Table 46-3).
Tympanic Tumor Tympanic tumors arise from the glomus body of the promontory along Jacobson’s nerve. The tumor is confined entirely to the mesotympanum, and all of its borders can be seen with routine otoscopy. This small tumor could
not arise from the jugular bulb, or it would extend beyond the inferior margins of the tympanic annulus. Although no additional studies are necessary to define the extent of the tumor, any vascular tumor of the middle ear must be differentiated from an aberrant carotid artery or a dehiscent jugular bulb. An aberrant carotid artery lies more anteriorly and is paler than the glomus tumor. The jugular bulb lies more posteriorly and is darker blue. If there is any question about the existence of either of these lesions, cranial computed tomography (CT) must be done to exclude them.
Tympanomastoid Tumor Similar to tympanic tumors, tympanomastoid tumors arise from the glomus body on the promontory. It extends beyond the tympanic annulus, however, inferiorly or posteriorly. Because there is no way clinically to delineate the tumor’s extent, any patient with a tumor that extends beyond the tympanic annulus must have a thorough radiographic evaluation. The key feature of this tumor category is that studies show the lesion does not involve the jugular bulb itself. Tympanomastoid tumors may extend into the mastoid and into the retrofacial air cells.
Jugular Bulb Tumor Jugular bulb tumors arise from the glomus body of the dome of the jugular bulb. Lesions may extend into the middle ear and into the bulb itself. This lesion is limited to involvement of the middle ear, mastoid, and jugular bulb. By definition, it does not extend onto the carotid artery or medially into the skull base or intracranially.
Carotid Artery Glomus Tumor Carotid artery glomus tumors arise from the jugular bulb, but extend beyond the confines of the jugular bulb and vein and contact the carotid artery. Small tumors of this category may involve only the carotid artery at the skull base, whereas larger tumors may extend far medially, and may involve the horizontal portion of the internal carotid artery (ICA) and the petrous apex. 551
552
OTOLOGIC SURGERY
TABLE 46-1 Modified De La Cruz Glomus Tumor
Classification with Associated Surgical Approach
Classification
Surgical Approach
Tympanic Tympanomastoid Jugular bulb
Transcanal Mastoid/extended facial recess Mastoid/neck (possible limited facial nerve rerouting) Infratemporal fossa ± subtemporal Infratemporal fossa/intracranial Transcondylar Cervical
Carotid artery Transdural Craniocervical Vagale
Surgery for Glomus Tumors, Chapter 49 Brackman DE, Arriaga MA in Otologic Surgery, eds Brackmann DE, Shelton C, Arriaga MA WB Saunders, Philadelphia. p579-593
TABLE 46-2 Fisch Classification of Glomus Tumors Type A Type B
Type C1, C2, C3
Type D1 Type D2, D3
Tumors limited to middle ear cleft Tumors limited to tympanomastoid area with no infralabyrinthine compartment involvement Tumors involving infralabyrinthine compartment of temporal bone and extending into petrous apex Tumors with intracranial extension <2 cm in diameter Tumors with intracranial extension >2 cm in diameter
Fisch U, Infratemporal Fossa Approach for Glomus Tumors of the Temporal Bone. nn Otol Rhinol LAryngol 91–474-479, 1982.
Transdural Tumor Transdural tumors arise from the jugular bulb and extend not only to the ICA, but also through the jugular foramen intracranially.
Glomus Vagale Tumor Glomus vagale tumors arise from the glomus body along the vagus nerve at the base of the skull. Because glomus vagale tumors do not begin within the temporal bone, they are often larger than glomus tumors of the temporal bone itself because they later produce symptoms of pulsatile tinnitus and hearing loss. Clinically, these lesions produce vocal cord paralysis before the onset of hearing loss or tinnitus or the appearance of a vascular middle ear mass. In contrast, glomus tumors of the temporal bone produce otologic symptoms before the onset of vocal cord paralysis.6
Jugular Foramen Schwannoma Although schwannomas are the second most frequently occurring tumor of the jugular foramen, they still are extremely rare. They arise from the Schwann cells of CN IX, X, and XI. The most common cranial nerve to
TABLE 46-3 Glasscock-Jackson Glomus Tumor
Classification
Type
Physical Findings
Glomus Tympanicum I II III IV
Small mass limited to promontory Tumor completely filling middle ear space Tumor filling middle ear and extending into mastoid Tumor filling middle ear, extending into mastoid or through tympanic membrane to fill external auditory canal; may also extend anterior to internal carotid artery
Glomus Jugulare I II III IV
Small tumor involving jugulare bulb, middle ear, and mastoid Tumor extending under internal auditory canal; may have intracranial extension Tumor extending into petrous apex; may have intracranial extension Tumor extending beyond petrous apex into clivus or infratemporal fossa; may have intracranial extension
Lateral Temporal Approach to the Skull Base, Chapter 8, Jackson G, Johnson GD, Poe DS in Surgery of the Skull Base, ed Jackson CG. New York, Churchill Livingston, 1991 p 141-196.
be affected is the vagus nerve.7,8 The clinical distinction between glomus vagale tumors and vagale schwannomas is that glomus vagale tumors develop vocal cord paralysis early in their course.9 Presentation of a jugular foramen schwannoma depends on the growth pattern. Kaye and colleagues10 categorized three different patterns: A, B, and C. Type A tumors occur primarily in the posterior fossa. Type A tumors with intracranial extension may manifest with CN IX, X, or XI palsies. Similar to most cerebellopontine angle tumors, hearing loss or imbalance or both may be the only significant presenting symptoms. Type B tumors remain confined to the skull base with extension often into the clivus. Type C tumors begin in the jugular foramen and extend inferiorly into the neck. Type B and C tumors manifest more frequently with lower cranial neuropathies than type A tumors.10 Radiologic assessment is performed with magnetic resonance imaging (MRI) and CT. MRI of the jugular foramen schwannoma shows a smooth contoured mass isodense on T1-weighted images, high signal intensity is usually noted on T2-weighted images, and significant enhancement occurs on T1-weighted gadolinium scans. CT is helpful in differentiating schwannoma from glomus jugulare tumors. Schwannomas tend to produce smooth erosion of the jugular foramen. Glomus tumors tend to show an irregular bone margin at the jugular foramen. The third most common lesions in this area are meningiomas of the jugular foramen. These lesions may be difficult to distinguish from schwannomas preoperatively. An enhancing dural tail on MRI is helpful. They also tend to infiltrate the bone around the jugular foramen, rather than smoothly enlarge it.
Chapter 46 • Surgery for Glomus Tumors and Other Lesions of the Jugular Foramen
PREOPERATIVE EVALUATION Thorough preoperative evaluation permits surgical planning for complete and safe tumor removal. Advances in imaging technology now permit accurate preoperative assessment of tumor involvement within the temporal bone. The following tests are routinely used in the evaluation of patients with glomus tumors of the temporal bone.
Routine Hearing Tests Air, bone, and speech audiometry are performed to assess the degree of conductive and sensorineural hearing impairments.
Cranial Computed Tomography The mainstay of assessment of glomus tumors of the temporal bone is thin-section (1.5 mm thick) cranial CT using the bone algorithm. Tumors confined to the middle ear and mastoids are distinguished from tumors that involve the jugular bulb. Extensive lesions that extend onto or medial to the ICA and lesions that extend transdurally are also defined by this technique. In many cases, cranial CT is the only examination necessary for planning treatment. Because the jugular bulb is the major consideration in preoperative planning, cranial CT has supplanted retrograde jugular venography in assessing involvement of the jugular bulb.
Magnetic Resonance Imaging Because bone involvement by tumor is not clearly shown on MRI, this technique provides only adjunctive information regarding the extent of tumor involvement. If the diagnosis is in question, MRI combined with CT provides exquisite preoperative guidance in the differential diagnosis of petrous apex lesions.11 MRI can indicate occlusion of the jugular bulb and vein because the normal flow signals are altered. In intradural tumors, MRI can delineate more clearly the tumor-brain interface and the relationship of the lesion to the intradural structures. MRI must be interpreted cautiously because T1 images of glomus tumors may overestimate the degree of tumor involvement. Marrow-containing bone of the petrous apex is hyperintense and indistinguishable from enhancing tumor in the petrous apex. Magnetic resonance angiography and MR venography offer another diagnostic tool in evaluating glomus tumors. In their present form, however, MR angiography techniques cannot provide adequate imaging resolution to define feeding vessels to the tumor. CT angiography is another promising technology in evaluating vascular temporal bone lesions12; however, the role of these techniques in the preoperative assessment of glomus tumors has not been fully clarified.13,14
553
Arteriography Four-vessel angiography is necessary when cranial CT shows a glomus tumor involving the jugular bulb, carotid artery, or intradural structures. The principal indication for angiography is to assess the involvement of the ICA by tumor. In addition, this technique ensures preoperative identification of additional glomus lesions, if present. Although estimates vary, paragangliomas are multicentric in approximately 10% of nonfamilial cases and 33% of familial cases.
Brain Perfusion and Flow Studies For tumors that abut the ICA, it is necessary to assess the adequacy of the cerebral cross-perfusion from the contralateral ICA. Cross-compression angiography, stump pressure measurements, and clinical evaluation during test occlusion of the involved carotid are basic guides to the risk of stroke in the event that the affected carotid artery must be sacrificed. Xenon blood flow and radioisotope studies offer much more precise quantification of the risk of stroke and the possible need for surgical replacement of the ICA.15 In certain cases with extensive invasion of the carotid artery and acceptable results on the perfusion studies of the contralateral artery, the involved carotid artery may be permanently occluded with a detachable balloon. We generally do not recommend carotid sacrifice. Despite excellent advances in diagnostic flow studies, an appreciable risk (about 5%) of stroke exists with carotid artery sacrifice even in cases with favorable functional and perfusion characteristics or test occlusion studies. If possible, repair or graft replacement of the carotid is recommended if the carotid is injured during tumor removal.
Embolization Large glomus tumors may result in significant intraoperative blood loss. We have found that preoperative embolization of feeding vessels can significantly reduce such loss.16 The embolization is usually performed with polyvinyl alcohol (Ivalon) and is performed at the time of angiography 1 or 2 days before surgery. A longer interval between embolization and surgery may result in collateral blood flow to the tumor, which may paradoxically increase tumor perfusion and intraoperative blood loss.
Biopsy Biopsy of vascular middle ear masses is not recommended. The clinical and radiographic appearance of glomus tumors is characteristic enough to permit definitive management without a tissue diagnosis. Efforts at obtaining a tissue diagnosis before completion of the radiologic evaluation can result in injury to an aberrant carotid artery or a high jugular bulb, and significant bleeding from the tumor itself.
554
OTOLOGIC SURGERY
PATIENT COUNSELING Patients with tympanic and tympanomastoid glomus tumors are advised of the routine risks of exploratory tympanotomy and tympanomastoidectomy surgery. As discussed later in the results section of this chapter, however, the prognosis for these patients is excellent if total tumor removal is accomplished. Patients with jugular bulb, carotid artery, or transdural tumors should be aware of the additional risk inherent in complete removal of their tumors. Specifically, facial nerve transposition is generally necessary and carries the attendant risk of facial paresis. These patients also should be aware of the risk of lower cranial nerve injury and possible vascular complications. Any patient with intradural tumor extension must be aware of the risks of craniotomy, including postoperative hemorrhage, cerebrospinal fluid leakage, meningitis, and stroke.
SURGICAL APPROACHES The following surgical techniques are used for removal of glomus tumors of the temporal bone. The specific approaches described correspond to the tumor classification system described earlier.
Transcanal Approach The transcanal approach is used for small glomus tympanicum tumors that are limited to the mesotympanum. Because the entire circumference of the tumor is visible within the middle ear, preoperative imaging studies are unnecessary. If there is any doubt of the possibility of an aberrant ICA, however, preoperative CT scanning should be obtained to exclude this possibility. The tympanomeatal flap is modified with the inferior incision extending more anteriorly so that the inferior aspect of the tympanic membrane can be elevated. The tumor is identified on the promontory (Fig. 46-1). The inferior tympanic branch of the ascending pharyngeal artery supplies the glomus tympanicum tumor. The vessel may be controlled with bipolar cautery or a small piece of absorbable knitted fabric (Surgicel) to occlude the bony canaliculus from which the vessel arises. It is crucial to avoid unipolar cautery on the tympanic promontory because this may severely injure the cochlea. The tumor is removed with cup forceps. The brisk bleeding from the distal end of the artery anterior to the stapes (near the cochleariform process) is difficult to control directly; however, it usually clots readily. Small pledgets of Surgicel assist in hemostasis; the tympanomeatal flap is replaced, and the ear canal is packed. No
FIGURE 46-1. Transcanal exposure of glomus tympanicum tumor limited to promontory.
Chapter 46 • Surgery for Glomus Tumors and Other Lesions of the Jugular Foramen
specific dressing is necessary, and the patient is ready for hospital discharge the following day. Accurate preoperative assessment is crucial for success with the transcanal approach for glomus tympanicum tumors. If the entire circumference of the lesion is not visible through the meatus, the hypotympanotomy or mastoid–extended facial recess approach should be considered. Tumors with limited hypotympanic extension without posterior involvement on CT scan may be removed by a modified transcanal (hypotympanic) approach.17 After a postauricular incision is performed with transection of the ear canal, a superiorly based tympanomeatal flap is elevated to permit access to the inferior aspect of the tympanic ring. This bone is progressively drilled until the inferior limit of the tumor is identified. The drilling involved with this exposure is usually much less than the drilling required for the transcanal infracochlear drainage procedures for the petrous apex described in Chapter 45.
Mastoid–Extended Facial Recess Approach The mastoid–extended facial recess approach is used for tympanomastoid glomus tumors. The tumor may extensively involve the middle ear and mastoid, but has arisen from the glomus tympanicum body and does not involve the jugular bulb. Preoperative CT evaluation is crucial before this approach is undertaken. Because the limits of the tumor are not visible through the tympanic membrane, CT provides an assessment of the extent of tumor involvement. Patient preparation and draping are performed as for a routine tympanomastoidectomy. A wide shave is performed, and the incision is made 1.5 cm posterior to the postauricular sulcus. After a complete mastoidectomy, the facial recess is opened. The extended facial recess exposure is performed by further bone removal inferiorly accomplished by severing the chorda tympani nerve and following the fibrous annulus of the tympanic membrane as a landmark. Such an approach allows complete exposure of the middle ear and the hypotympanum. After the tumor is exposed, bipolar cautery is helpful in shrinking the tumor and in controlling the blood supply. Surgicel packs are used to tamponade further the main arterial supply in the hypotympanum, and the tumor is removed with cup forceps. The tumor can be stripped from the ossicles if necessary. Extension of glomus tympanicum tumors into the retrofacial air cells is managed by direct exposure. A cutting burr is used to remove the air cells inferior to the labyrinth and beneath the facial nerve, leaving the facial nerve suspended with a thin layer of bone to allow the surgeon access to the entire hypotympanum (Fig. 46-2). A small curette is used to remove bits of the tumor from crevices in the hypotympanum. The dome of the jugular bulb can be inspected to ensure that it is free of the tumor. Extensive tumor involvement may necessitate removal of the ossicles and tympanic membrane. In such circumstances, a tympanoplasty and ossicular
555
r econstruction may be accomplished in the routine manner. Similarly, if the tumor has produced extensive destruction of the posterior canal wall, such cases may be managed with a canal wall down technique combined with tympanoplasty and mastoid obliteration after complete tumor removal. At the conclusion of the procedure, a mastoid dressing is applied, and the patient is usually ready for discharge by the first postoperative morning.
Mastoid-Neck Approach The mastoid-neck approach is used for small glomus jugulare tumors. These tumors involve the jugular bulb, but do not extend onto the ICA or into the neck or posterior fossa. The preoperative evaluation of these patients may include angiography because involvement of the jugular bulb raises the question of possible carotid artery involvement. Continuous intraoperative facial nerve monitoring is used. Additionally, electromyography electrodes in the sternocleidomastoid muscle are useful for monitoring CN XI, electrodes in the lateral pharyngeal wall can monitor CN IX, and electrodes in the vocalis muscle can monitor CN X. Specially designed endotracheal tubes with
1. 2. 3.
1. Extended facial recess 2. Tumor in hypotympanum 3. Tumor in extension beneath facial n.
FIGURE 46-2. Facial recess has been opened widely and extended inferiorly to expose tumor in hypotympanum. Tumor extension beneath facial nerve is exposed by removing retrofacial air cells and skeletonizing facial nerve.
556
OTOLOGIC SURGERY Digastric muscle and parotid gland retracted
Facial n. in fallopian canal
FIGURE 46-3. Completed procedure using mastoid-neck approach. Jugular bulb is resected along with tumor after proximal and distal vessels are controlled.
attached electromyography electrodes are particularly useful for monitoring lower cranial nerve function. The procedure is performed by initially completing the same exposure as that described in the mastoid–extended facial recess approach (Fig. 46-3), then amputating the mastoid tip. The periosteum of the digastric groove is followed anteriorly until it turns abruptly laterally at the stylomastoid foramen. Drilling laterally along the digastric ridge anteriorly and posteriorly frees the entire mastoid tip. The incision is carried into the neck along the anterior border of the sternocleidomastoid muscle. This muscle is freed from the mastoid tip and retracted posteriorly. The mastoid tip can be removed by grasping it with a Kocher clamp and cutting with a curved Mayo scissors along the bone. The posterior belly of the digastric muscle is identified and freed from the digastric groove and retracted anteriorly to allow exposure of the major neurovascular structures of the neck. When the internal jugular vein is identified and dissected free from surrounding tissues, it is occluded with multiple 2-0 silk sutures. The jugular vein is followed over the transverse process of the first cervical vertebra into the base of the skull. In this fashion, CN XI is identified (usually lateral to the vein) and preserved. In limited tumors that do not extend into the neck or skull base, it is usually possible to preserve CN IX, X, and XI. The neck exposure is necessary to permit ligation of the jugular vein. Exposure of the sigmoid sinus and jugular bulb is completed with diamond burrs. Although limited tumors
do not involve the medial wall of the jugular foramen, the tumor arises from the dome of the jugular bulb, and the bulb must be resected in continuity with the tumor. The proximal sigmoid sinus is controlled with extraluminal packing of Surgicel. Preservation of the bone over the midportion of the sigmoid sinus permits extraluminal packing of this portion of the sigmoid sinus. Because the jugular vein has been tied in the neck, the sigmoid sinus may be opened just distal to the proximal packing. Bleeding occurs at this point from the patent inferior petrosal sinus and condylar vein. Surgicel is advanced into the jugular bulb to control this bleeding. This packing must not be placed too firmly because a weakness of CN IX, X, and XI may result. The tumor is now ready for resection in continuity with the dome of the jugular bulb. Bipolar cautery helps in hemostasis and shrinkage of the tumor bulk. When the tumor and dome of the jugular bulb are excised, hemostasis is completed with Surgicel packing (see Fig. 46-3). Complete tumor resection is accomplished, and if the tympanic membrane or ossicles are involved, reconstruction may be performed as described earlier. It is usually possible to preserve CN IX, X, and XI with these limited tumors, unless there is preoperative involvement of these structures. Postoperatively, these patients are cared for in the intensive care unit because the possibility of an acute lower cranial neuropathy or major postoperative hemorrhage exists. Usually, minimal morbidity is associated with this approach, and patients are stable and may
Chapter 46 • Surgery for Glomus Tumors and Other Lesions of the Jugular Foramen
557
Tumor
Facial n. transposed from bony canal and elevated against external auditory canal
FIGURE 46-4. Mastoid-neck approach with limited facial nerve rerouting. Displacement of facial nerve exposes larger tumors of jugular bulb.
be discharged from the hospital within a few days. The major pitfall associated with this approach is related to inaccurate preoperative assessment of tumor extent. The mastoid-neck approach is too limited if the tumor extensively involves the carotid artery.
Mastoid-Neck with Limited Facial Nerve Rerouting A useful modification of the mastoid-neck approach adds a limited facial nerve rerouting to the procedure. Additional exposure in the mastoid-neck approach may be achieved by totally decompressing the facial nerve from the second genu throughout the entire vertical segment. The periosteum of the facial nerve at the stylomastoid foramen is preserved, but the fibrous attachments to the nerve in its vertical portion are sharply transected. The mobilized nerve can be transposed laterally along with the periosteum at the stylomastoid foramen and the attached posterior belly of the digastric muscle. A suture through the stylomastoid periosteum can be used to hold the nerve laterally and prevent traction on this structure. Such transposition of the facial nerve permits further bone removal in the area of the vertical facial canal and retrofacial air cells along the infralabyrinthine air cell tract (Fig. 46-4). This modification of the mastoid-neck approach with limited facial nerve rerouting is ideal for neuromas of the jugular foramen because they are not as intimately involved with the carotid artery as glomus tumors are. The surgeon should be cautious about applying this approach for glomus tumors that involve the carotid artery. The annulus and posterior canal wall still limit the surgeon’s view of the vertical portion of the carotid artery if extensive manipulation is necessary in this area.
FIGURE 46-5. Overlying facial nerve prevents exposure of internal carotid artery for tumor removal.
Infratemporal Fossa Approach The development of the infratemporal fossa approach by Fisch18 has been a significant advance in the total removal of large glomus jugulare tumors. Previous approaches that did not remove the external auditory canal or reroute the facial nerve did not allow adequate exposure of the tumor or ICA (Fig. 46-5). There are eight distinct steps in the
558
OTOLOGIC SURGERY Tumor
ICA
Jugular vein
FIGURE 46-6. Temporal bone dissection completed. Facial nerve is skeletonized, and tympanic ring is removed. Internal carotid artery (ICA) and jugular bulb with tumor are exposed.
infratemporal fossa exposure: (1) patient preparation, (2) management of the ear canal and tympanic ring, (3) mastoidectomy, (4) initial preparation of the jugular vein and neck exposure, (5) transposition of the facial nerve, (6) completion of the neck exposure and identification of the lower cranial nerves and skull base carotid artery, (7) tumor removal including intracranial extension, and (8) wound closure. Continuous facial nerve monitoring and electromyography monitoring of the lower cranial nerves are employed as previously described. A wide shave is performed, and a large postauricular C-shaped incision is made. The incision is carried anteriorly, and the ear canal is transected slightly medially to the bone-cartilage junction of the ear canal. Cartilage is removed from the ear canal to permit fashioning of the ear canal skin as a cuff that can be everted. The skin of the meatus is closed with 4-0 nylon sutures. The periosteum of the postauricular area is elevated as a flap and sutured behind the opening in the meatus to reinforce the closure further. Next, a mastoidectomy is completed, and the facial recess is opened to allow separation of the incudostapedial joint. The posterior wall of the ear canal can be removed with rongeurs and cutting burrs. The remaining skin of the ear canal and the tympanic membrane, malleus, and incus are removed. The facial nerve is decompressed from the geniculate to the stylomastoid foramen. An eggshell-thin layer of bone is left over the nerve itself. By use of cutting and diamond burrs, the bone of the tympanic ring is progressively removed, the level of the jugular bulb is identified, and the bone over the temporomandibular joint and vertical segment of the petrous carotid artery is removed anteriorly (Fig. 46-6).
Attention is then focused on the neck. The incision is continued vertically along the anterior border of the sternocleidomastoid muscle, the mastoid tip is removed as previously described, the jugular vein is identified, and ligatures are placed around the vein, but are not yet tied at this point. The carotid artery is identified and marked with a ligature. The posterior belly of the digastric muscle is transected. The facial nerve is transposed. The transposition technique originally described by Fisch has been modified because of temporary and sometimes permanent residual facial weakness.19 Rather than exposing the facial nerve in the parotid, the surgeon transposes the nerve with periosteum of the stylomastoid foramen and elevates the entire tail of the parotid.20 After decompressing the nerve, the remaining eggshell-thin bone over the facial nerve is removed with a blunt instrument. The multiple fibrous connections along the descending portion of the nerve are sharply transected. The tympanic portion of the nerve does not have such adhesions, and this section elevates readily. The posterior belly of the digastric muscle is moved anteriorly because the fascia of this muscle contributes to the stylomastoid foramen periosteum. The nerve can be transposed anteriorly along with the tail of the parotid. A large suture is placed through the periosteum of the stylomastoid foramen and attached to the soft tissues in the area of the root of the zygoma (Fig. 46-7), elevating the facial nerve and preventing it from being stretched when retractors are placed. The use of continuous facial nerve monitoring during this maneuver has significantly improved postoperative facial nerve function.21 After the facial nerve is decompressed and elevated, a large Perkins retractor is placed beneath the angle of the
Chapter 46 • Surgery for Glomus Tumors and Other Lesions of the Jugular Foramen Periosteum of stylomastoid foramen tacked superiorly
559
Facial nerve, rerouted anteriorly
Tumor
FIGURE 46-7. Facial nerve transposed anteriorly. Large suture tacks periosteum of stylomastoid foramen and facial nerve superiorly to avoid tension on transposed nerve.
mandible, and the entire mandible is retracted forward. This exposure avoids the need to resect the mandibular condyle, even in large tumors that extend into the infratemporal fossa extensively. The remaining bone over the distal sigmoid sinus, jugular bulb, and vertical portion of the petrous carotid artery can be removed with diamond burrs. Attention is again focused in the neck, where the ICA is followed through the skull base into its intratemporal course. The lower cranial nerves (CN IX, X, and XI) are followed into the jugular foramen as well. CN XII is also identified and followed to its foramen. The jugular vein is doubly ligated and transected between ligatures, and the external carotid artery is ligated. If the tumor extends intradurally, the proximal sigmoid sinus is doubly ligated with silk sutures passed through openings in the dura with an aneurysm suture passer (Fig. 46-8). If the tumor is not intradural, the sigmoid can be packed with Surgicel without violating the dura. The jugular vein is elevated, and the tumor in the area of the jugular bulb is freed inferiorly to superiorly following the jugular vein into the jugular bulb. The tumor is freed from the carotid artery anteriorly, and bleeding from the caroticotympanic vessels is controlled with bipolar cautery. If the tumor is adherent to the ICA, it is best to leave a portion of it on the artery and remove the bulk of the tumor. Lower cranial nerve preservation is enhanced if the medial wall of the jugular bulb is left in situ protecting the pathway of the lower cranial nerves through the pars nervosa in the anteromedial portion of the jugular bulb. Tumor hemostasis is continued with
bipolar cautery and Surgicel packing of the inferior petrosal sinus; the tumor can be removed in continuity with the dome of the jugular bulb. If the last bit of the tumor from the carotid artery is not removed until the conclusion of the procedure, a small entry into the carotid artery that may occur at the location of the caroticotympanic artery can be repaired directly (Fig. 46-9). Small intracranial tumor extensions are usually removed at the time of removal of the jugular bulb because this is the usual location of dural penetration. If there is extensive intracranial extension, however, a decision must be made about whether to attempt total removal of the tumor. The decision is largely based on the amount of blood lost to this point. If blood loss has been limited to less than 2000 mL, as is almost always the case, removal of the intracranial extension of the tumor may proceed. If the amount of blood loss has been greater, however, problems with bleeding may occur despite the replenishment of the known clotting factors with fresh frozen plasma and platelet packs. In such cases, a twostage procedure with removal of the intracranial portion of the tumor at a later date is planned. The removal of the intracranial portion of the tumor is often easier than the removal of tumor within the temporal bone. By the time the surgeon is ready for the removal of the intracranial extension, the blood supply has often been controlled. The blood supply to the intracranial portion of the tumor is often discrete and can be controlled with bipolar cautery as with other cerebellopontine angle tumors (Fig. 46-10). If tumor has
560
OTOLOGIC SURGERY
Facial n. transposed Glomus jugulare tumor Lateral semicircular canal
FIGURE 46-8. Jugular vein and sigmoid sinus are doubly ligated. If tumor does not extend intradurally, sigmoid is occluded with extraluminal packing. Tumor dissected from internal carotid artery
Tumor delivered posteriorly
FIGURE 46-9. Final piece of tumor removed from internal carotid artery. Artery may be repaired with vascular suture if injury occurs with total tumor removal.
been left along the ICA, it is now removed. Closure is accomplished by closing the eustachian tube with Surgicel, muscle, and strips of abdominal fat (Fig. 46-11). If cerebrospinal fluid has been encountered, continuous lumbar drainage is used for approximately 5 days until the wound is sealed; however, the use of cranioplasty techniques with hydroxyapatite cement or titanium mesh is effective at controlling cerebrospinal fluid fistulas.22
A drain is left in the neck wound and removed on the first postoperative morning. A pressure dressing is placed for 4 postoperative days. One technique with encouraging results in cases with large blood loss has been the use of a Cell Saver (Hemonics, Braintree, MA) (which is commonly used in cardiovascular surgery). By use of a special irrigating suction device with a heparinized reservoir,
Chapter 46 • Surgery for Glomus Tumors and Other Lesions of the Jugular Foramen
IX
PICA
Tumor
561
XII
X J.B. XI
A Tumor removed
PICA Pons
Cerebellum
B FIGURE 46-10. A and B, Removal of intracranial extension of tumor. Intracranial feeding vessels are controlled by bipolar cautery. JB, jugular bulb; PICA, posterior inferior cerebellar artery.
intraoperative blood loss can be salvaged and prepared for replacement to the patient during the same procedure. Also, we routinely counsel patients on preoperative autologous donation of blood to minimize the need for banked blood replacement. With these techniques, it is unusual to require banked blood transfusion even in large tumors.
Fallopian Bridge Technique A new strategy for minimizing postoperative facial nerve dysfunction and enhancing preservation of the middle ear in jugular foramen surgery is the fallopian bridge technique (Fig 46-12). In this technique, the facial nerve is left in situ, but the bone surrounding the nerve is almost
562
OTOLOGIC SURGERY Parotid gland
Digastric muscle
Facial nerve
Strips of fat
FIGURE 46-11. Eustachian tube is closed with Surgicel and muscle. For intracranial tumors, dura is sutured to the extent possible. Strips of abdominal fat obliterate middle ear and mastoid.
completely removed.23 This technique can be used with the mastoid-neck approach as an alternative to partial mobilization. It is even applicable with the full infratemporal fossa technique. Although the technique is relatively new, it promises to decrease the number of cases requiring facial nerve translocation, with beneficial outcomes for postoperative facial nerve function and preservation of the conductive hearing mechanism.
Transcondylar Approach In certain recurrent glomus jugulare tumors, the intracranial tumor may extend significantly toward the foramen magnum. In these situations, a wider exposure of the cranial cervical junction is necessary. Although the transsigmoid, translabyrinthine exposure combined with the infratemporal fossa approach offers a wide view, the cranial cervical junction itself is not fully exposed with this procedure. In this case, the transcondylar exposure is helpful with the infratemporal fossa approach. The steps involved in the transcondylar approach24 are initially extension of the muscular incisions of the posterior superior aspect of the neck to identify the vertebral artery posterior and inferior to the mastoid tip on the
transverse process of C1. When this landmark is clearly identified, wide exposure of the occipital condyle and jugular tubercle permits full exposure of the hypoglossal nerve and an unlimited view of the cranial cervical junction (Fig. 46-13). If more than half of the occipital condyle is resected, careful consideration should be given to a simultaneous cervical stabilization procedure. When transcondylar exposure is needed in addition to standard otologic exposure, we find the use of head holding pins in the lateral position helpful for access to the posterior superior skull base.
Complete Carotid Mobilization In very rare cases of recurring glomus jugulare tumors after previous surgery and radiation, the tumor may completely encase the petrous carotid artery with significant extension into the far anterior infratemporal fossa. In these cases, we have found the combination of standard otologic (posterior) infratemporal fossa approach with the preauricular infratemporal fossa approach quite useful. Specifically, an orbital zygotomy in continuity with the glenoid fossa is performed after a temporal craniotomy. In this fashion, the glenoid can be reconstructed
Chapter 46 • Surgery for Glomus Tumors and Other Lesions of the Jugular Foramen
563
Skeletonized facial nerve
FIGURE 46-12.
Jugular bulb
•
•
•
Facial nerve
•
Common carotid
•
Jugular vein
•
Sigmoid sinus
Extent of bone removal
Dural incision
•
Sigmoid sinus
• •
Occipital condyle to be drilled
• •
C1
• A
Jugular bulb
Vertebral artery
B
FIGURE 46-13. FIGURE 46-12. Fallopian bridge technique. In this modification, facial nerve is left in situ with only an island of bone to permit tumor dissection lateral and medial to facial nerve. FIGURE 46-13. A and B, Transcondylar exposure. Once the vertebral artery is identified, the occipital condyle can be progressively removed to provide wide exposure of the craniocervical junction.
564
OTOLOGIC SURGERY Venous Sinus Drainage Anatomy
A
B
I.P.S. 7.
3. SS.I.
6. M F D
L S C
2. 5. Ce.
3.
4. 1. 4.
Thinned bone over MFD
1. 2. 3. 4. 5. 6. 7.
Sigmoid sinus Jugular bulb Superior petrosal sinus Emissary v. Posterior fossa dura (bone removed) Facial recess Inferior petrosal sinus
FIGURE 46-14. Surgical view (A) and schematic view (B) of venous sinuses. LSC, lateral semicircular canal; Ce, cerebellum; IPS, inferior petrosal sinus; MFD, middle fossa dura; SS, sigmoid sinus.
at the conclusion of the procedure. The carotid canal can be followed directly and fully from the base of the skull all the way to the cavernous sinus. In this way, the carotid artery can be mobilized anteriorly and inferiorly if necessary for complete tumor resection. This approach facilitates any carotid reconstruction if necessary. In cases with extensive carotid involvement, carotid injury is a real hazard. One center advocates preoperative stenting as a safety measure.25
RESULTS Glomus Tympanicum Tumors O’Leary and colleagues26 reviewed the results of glomus tympanicum surgery at the House Ear Clinic. Between 1957 and 1990, 73 glomus tympanicum tumors (tympanic tumor or tympanomastoid tumor) were managed at the House Ear Clinic. In 80% of these tumors, a mastoid– facial recess approach was required, and in 20%, a transcanal removal was possible. Although significant intraoperative blood loss occurred (average >500 mL), the morbidity was minimal. Hearing levels remained stable: the mean speech reception threshold increased
1 dB postoperatively. Of the five complications, three were residual tympanic membrane perforations requiring a secondary tympanoplasty. One patient developed a cholesteatoma postoperatively. One patient developed a facial nerve weakness requiring re-exploration and facial nerve decompression with an ultimately good outcome (grade II/VI on the House-Brackmann scale). This series showed that the critical feature of glomus tympanicum management is total tumor resection. The three cases in which an incomplete resection was performed resulted in tumor recurrence. Overall, the recurrence rate in this series was less than 5%.
Glomus Jugulare Tumors Green and associates27 reviewed the House Ear Clinic experience with glomus jugulare tumors (jugular bulb tumors, carotid artery involvement tumors, and intradural tumors) between 1980 and 1991. During this interval, 52 patients were surgically treated for glomus jugulare tumors who had undergone no prior radiotherapy or surgery. The techniques involved were the infratemporal fossa approach in 83%, the mastoid-neck approach in 7%, and the mastoid-neck with limited facial nerve
Chapter 46 • Surgery for Glomus Tumors and Other Lesions of the Jugular Foramen
565
Gentle pressure
A
C.
C. S. G.
S.S.
B
C. Cottonoid S. Surgicel
C
C. Cottonoid G. And/or Gelfoam
FIGURE 46-15. A and B, Management of bleeding from sigmoid sinus and control strategies, including gentle pressure, and Cottonoid over Surgicel or absorbable gelatin sponge (Gelfoam) CV, condylar vein; IPS, inferior petrosal Sinus; JB, jugular bulb; SS, superior petrosal sinus.
mobilization in 10%. Complete surgical removal was possible in 85% of the patients. Eight patients required transection of the facial nerve for complete tumor removal. These patients underwent segmental grafting or greater auricular nerve reconstruction and achieved a grade III/VI facial nerve recovery. The mean intraoperative blood loss was 1500 mL. Long-term facial function was good: 95% of patients who had facial nerve rerouting had grade I/VI or II/VI facial function at 1 year follow-up or greater. Nearly 20% of patients required a vocal cord augmentation procedure. No patient required a tracheotomy in the immediate postoperative period. Four patients required prolonged nasogastric tube feeding, and two ultimately required gastrostomy tempo rarily. No patients required long-term gastrostomy. Of all patients, 85% were able to resume the same activity level as before surgery.
COMPLICATIONS One of the most common issues in glomus surgery or any complex temporal bone surgery is appropriate management of bleeding from the venous sinuses. Figures 46-14 to 46-17 illustrate normal drainage patterns from
the venous sinuses, and systematic strategies for managing bleeding or injury from these sinuses.
Glomus Tympanicum Tumors Glomus tympanicum surgery risks the same morbidity as that involved in tympanomastoid surgery. The management of associated facial nerve, tympanic membrane, sigmoid sinus, and healing complications is identical to management of these complications in tympanomastoid surgery for chronic otitis media.
Glomus Jugulare Tumors As illustrated in the results section, the categories of morbidity in glomus jugulare surgery include facial nerve injury, lower cranial nerve dysfunction, carotid artery injury, bleeding problems, and intracranial complications. Direct tumor infiltration of the facial nerve necessitates transection of the involved segment and replacement of that segment with a nerve graft using standard techniques. The advent of continuous intraoperative facial nerve monitoring has significantly improved postoperative facial nerve function in infratemporal fossa surgery.21 In our experience, the need for
566
OTOLOGIC SURGERY
J.B. 1. S.S. 2.
A
J.B.
1. Cottonoid 2. Surgicel and wax
SPS
B
1.
1. Cottonoid 2. Bone wax
2.
SPS
FIGURE 46-16. A and B, Bleeding from jugular bulb controlled with bone wax and Surgicel (A), and bleeding from superior petrosal sinus controlled with bone wax and pressure from a Cottonoid (B). JB, jugular bulb; SPS, superior petrosal sinus; SS, sigmoid sinus.
tracheotomy or gastrostomy has been infrequent (<4%). We recommend early vocal cord augmentation for vagal paralysis.28 The possibility of carotid artery injury must be anticipated. Careful preoperative imaging, including CT scans and angiography, is helpful in defining the anatomic relationships of the tumor to the ICA. Preoperative flow studies (i.e., balloon occlusion or xenon or technetium blood flow studies) are helpful in predicting how well a patient would tolerate total occlusion of the ICA. The indications for replacement versus permanent occlusion have been summarized elsewhere.29 The possibility of significant intraoperative blood loss must be anticipated. Autologous blood donation has been helpful in avoiding transfusions of banked blood. In the weeks preceding surgery, the patient donates his or her own blood for possible autologous transfusion intraoperatively. Similarly, the Cell Saver, which recycles the patient’s intraoperative blood loss, may reduce the requirement for banked blood transfusions. Finally, the surgeon must be cognizant of the extent of blood replacement and be certain that the appropriate ratio of platelets and fresh frozen plasma is replaced in addition to the red blood cell products themselves. Intracranial glomus tumor extension should be managed cooperatively with a neurosurgeon. The techniques for tumor removal are similar to those used for other posterior fossa neoplasms. Specifically, the possibility of
intracranial hemorrhage, wound infection, and cerebrospinal fluid leakage may require neurosurgical expertise. In our experience, patients with cerebrospinal fluid fistulas respond to conservative management with pressure dressing and lumbar drainage.
RADIATION The role of radiation therapy in the management of glomus jugulare tumors is still controversial, but increasingly accepted. Histologic studies have shown that the effects of irradiation seem to be on the blood vessels and fibrous elements of the tumor, rather than on the tumor cells themselves.30 With external-beam strategies, these tumors often begin growing after a 10 to 15 year period of control. In addition, the potential for malignant transformation of radiating benign tumors must not be underestimated. Fatal malignant transformations of benign jugular foramen tumors after radiation therapy have been reported. More recent studies have indicated excellent shortterm and intermediate-term control with stereotactic radiosurgery techniques that limit the dose to surrounding tissue, but can effectively shrink paragangliomas in anatomically eloquent areas such as the cavernous sinus.31 The Mayo Clinic series reported greater than 92% control in a “long-term series” of 30 patients with median
Chapter 46 • Surgery for Glomus Tumors and Other Lesions of the Jugular Foramen
567
Excessive bleeding near (IPS and/or condylar vein) C.V. Avoid injury to CN’s IX, X, XI from excessive packing in J.B.
J.B.
3 1
First ligate jugular v.
S.S.
IPS
A
2
Packing of oxidized cellulose extraluminally
SPS C.V.
S.S.
B
FIGURE 46-17. A-C, Technique for packing jugular bulb and sigmoid sinus in the case of massive uncontrollable bleeding by ligating the jugular vein and occluding the sigmoid sinus before placing intraluminal packing materials. Intraluminal packing can dislodge and produce a pulmonary embolus if the jugular vein is not occluded first. SS, sigmoid sinus.
13 year follow-up.32 Nonetheless, the longer term results (>15 to 20 years) are unknown, especially for stereotactic radiosurgery; the data can be considered promising, but not yet definitive. Although we previously recommended radiation therapy only in elderly patients with symptomatic glomus tumors or in patients who otherwise could not withstand a surgical removal, radiation (primarily stereotactic radiosurgery) can be considered as having a role in the modern management of these tumors and should be discussed with patients. Although we generally recommend surgery as definitive therapy for patients who are medically stable enough to undergo an operative procedure because this is the only therapy with the potential for actually curing (eliminating) the tumor, radiation may be part of a strategy to limit cranial neuropathies. Although the patient’s preferences should be discussed before surgery, it may be reasonable to avoid transection and sacrifice of cranial nerves and employ a near-total tumor resection if removing the last bit of tumor would result in cranial nerve morbidity. Similar to glomus tumors, jugular foramen schwannomas show short-term effective control and excellent cranial nerve outcomes with stereotactic radiation therapy.33
SUMMARY Modern imaging studies accurately delineate the extent of glomus tumors of the temporal bone. Microsurgical techniques allow total removal of even the largest tumors with acceptable morbidity and virtually no mortality. Although we favor surgery as the principal treatment of glomus tumors except in elderly or infirm patients, radiation therapy is increasingly used as an alternative treatment when the risk of cranial neuropathy with surgical management is high. The role of radiation for glomus tumors is currently a subject of intense investigation.
REFERENCES 1. Rosenwasser H : Carotid body-like tumor of the middle ear and mastoid bone. Arch Otolaryngol 41:64-67, 1945. 2. Bickerstff E R , Howell J S : The neurological importance of tumors of the glomus jugulare. Brain 76:576-693, 1953. 3. Steinberg N, Holz WG: Glomus jugulare tumors. Arch Otolaryngol Head Neck Surg 82:387-394, 1965. 4. Brown J S : Glomus jugulare tumors revisited: A ten-year statistical follow-up of 231 cases. Laryngoscope 95:284285, 1985.
568
OTOLOGIC SURGERY
5. Spector GJ, Fierstein J, Ogura J H : A comparison of therapeutic modalities of glomus tumors in the temporal bone. Laryngoscope 86:690-969, 1976. 6. Leonetti J P, Brackmann D E : Glomus vagale tumors: The significance of early vocal cord paralysis. Otolaryngol Head Neck Surg 100:533-537, 1989. 7. Horn K L , Hankinson H : Tumors of the jugular foramen. In Brackmann D E, Jackler R K (eds): Neurotology. Philadelphia, Mosby, 1994, pp 1059-1068. 8. Horn K L , House WF, Hitselberger WE : Schwannomas of the jugular foramen. Laryngoscope 95:761-765, 1985. 9. Leonetti J P, Brackmann D E : Glomus vagale tumors: The significance of early vocal cord paralysis. Otolaryngol Head Neck Surg 100:533-537, 1989. 10. Kaye A H, Hahn J F, Kinney J E : Jugular foramen schwannomas. J Neurosurg 60:1045-1053, 1984. 11. Arriaga M A, Brackmann D E : Differential diagnosis of primary petrous apex lesions. Am J Otol 12:470-474, 1991. 12. Krishnan A, Mattox D E, Fountain A J, Hudgins PA : CT arteriography and venography in pulsatile tinnitus: Preliminary report. AJNR. Am J Neuroradiol 27:1635-1639, 2006. 13. Arriaga M A, Lo WWM, Brackmann D E : Imaging case study of the month: Magnetic resonance angiography of synchronized bilateral carotid body paragangliomas and bilateral vagal paragangliomas. Ann Otol Rhinolaryngol 101:955-957, 1992. 14. Rogers G P, Brackmann D E, Lo WWM : Magnetic resonance angiography, a technique for evaluation of skull base lesions. J Otol 14:56-62, 1993. 15. Janecka I P, Sekhar L N, Horton J A : General blood flow evaluation. In Cummings CW, Fredrickson J, Harker L, et al (eds): Otolaryngology Head and Neck Surgery Update II. St. Louis, Mosby-Year Book, 1990, pp. 54-63. 16. Murphy TP, Brackmann D E : Effects of preoperative embolization on glomus jugulare tumors. Laryngoscope 99:1244-1247, 1989. 17. Farrior J B : Glomus tumors—postauricular hypotympanotomy. Arch Otolaryngol Head Neck Surg 86:367-373, 1967. 18. Fisch U: Infratemporal fossa approach for glomus tumors of the temporal bone. Ann Otol Rhinol Laryngol 91:474479, 1982.
19. Fisch U, Faga P, Valvanis A : The infratemporal fossa approach for the lateral skull base. Otolaryngol Clin North Am 17:513-552, 1984. 20. Brackmann D E : The facial nerve in the infratemporal approach. Otolaryngol Head Neck Surg 97:15-17, 1987. 21. Leonetti J P, Brackmann D E, Prass RC : Improved preservation of facial function in the infratemporal fossa approach to the skull base. Otolaryngol Head Neck Surg 101:74-78, 1989. 22. Arriaga M A, Chen D A, Burke E L : Hydroxyapatite cement cranioplasty in translabyrinthine acoustic neuroma surgery—update. Otol Neurotol 28:538-540, 2007. 23. Pensak M L , Jackler R K : Removal of jugular foramen t umors. Otolaryngol Head Neck Surg 117:586-591, 1997. 24. Fukushima T: Manual of Skull Base Dissection. Pittsburgh, AF, Neurovideo Inc, 1996. 25. Sanna M, Khrais T, Menozi R , et al: Surgical removal of jugular paraganglioma after stenting of the intratemporal internal carotid artery: A preliminary report. Laryngoscope 116:742-746, 2006. 26. O’Leary M J, Shelton C, Giddings N, et al: Glomus tympanicum tumors: A clinical perspective. Laryngoscope 101:74-78, 1989. 27. Green JD, Nguyen CD, Arriaga MA, et al: Technical modifications in the surgical management of glomus jugulare tumors. Laryngoscope v 104, p 917-24, 1994. 28. Netterville J D: Primary thyroplasty in glomus jugulare surgery. Presented at American Neurotology Society Fall Meeting, Washington DC, September 13, 1992. 29. deVries EJ: A new method to predict safe resection of the internal carotid artery. Laryngoscope 100:85-89, 1990. 30. Brackmann D E, House WF, Terry R , et al: Glomus jugulare tumors: Effects of irradiation. Trans Am Acad Ophthalmol Otolaryngol 76:1423-1431, 1972. 31. Gerosa M, Visca A, Rizzo P, et al: Glomus jugulare t umors: The option of gamma knife radiosurgery. Neurosurgery 59:561-569, 2006. 32. Krych A J, Foote R L , Brown PD, et al: Long-term results of irradiation for paraganglioma. Int J Radiat Oncol Biol Phys 65:1063-1066, 2006. 33. Martin JJ, Kondziolka D, Flickinger JC, et al: Cranial nerve preservation and outcomes after stereotactic radiosurgery for jugular foramen schwannomas. Neurosurgery 61:76-81, 2007.
47
Rehabilitation of Lower Cranial Nerve Deficits after Neurotologic Skull Base Surgery James L. Netterville and Christopher A. Sullivan
Before the advent of modern skull base surgery, the treatment of cranial base lesions was associated with significant perioperative complications and long-term morbidity. Rapid advances in medical imaging have made possible early detection of lesions and preservation of vital structures at the time of surgery. The development of advanced microsurgical techniques has made the removal of most skull base tumors not only possible, but also feasible. Today, the major source of short-term and long-term morbidity results from the loss of cranial nerves at the time of resection. Innovative approaches to cranial nerve rehabilitation have allowed many of these patients productive and enjoyable lives after cranial base surgery. The lower cranial nerves function in concert to facilitate speech, swallowing, and airway protection (Fig. 47-1). Interruption of the complex interactions of these nerves results in articulation deficits, inanition, and aspiration. With speech and swallowing therapy, most patients are able to compensate for the loss of a single lower cranial nerve. The loss of multiple nerves, particularly in an elderly patient, may result in permanent inability to swallow despite intensive therapy. Paradoxically, preoperative loss of cranial nerve function from tumor compression allows most patients to compensate slowly over time, and may predict better postoperative speech and swallowing rehabilitation. Patient age and preoperative cranial nerve function are important factors in the decision to proceed with complete surgical resection, partial resection, or radiation therapy.
PATIENT EVALUATION When lateral skull base surgery is elected, CN V3 to XII and the sympathetic trunk are at risk. A careful head and neck examination including cranial nerve evaluation and flexible fiberoptic examination of the hypopharynx and larynx should be performed at the bedside on the 1st postoperative day to confirm known deficits and to determine the extent of any additional cranial nerve deficits. The state of the airway, vocal fold function and cord
position, pooling of secretions, and the ability to clear secretions by coughing should be noted. A modified barium swallow is the “gold standard” for evaluating aspiration, and should be performed as soon as the patient is awake and alert enough to cooperate with the study. The level of oropharyngeal dysphagia may be identified and addressed by the speech and language pathologist at the time of this study. Flexible videoendoscopic evaluation has been described to evaluate swallowing at the bedside, but is not the study of choice because it cannot identify aspiration during the pharyngeal phase of swallowing, which is the most significant component of swallowing dysfunction in lateral skull base surgery.1 After thorough evaluation has been completed, efforts are directed at rehabilitation. The remainder of this chapter describes the pertinent regional cranial nerve anatomy and function,2,3 associated postoperative deficits, and methods for rehabilitation.
LOWER CRANIAL NERVE DEFICITS AND REHABILITATION Trigeminal Nerve: Mandibular Division (Cranial Nerve V3) The mandibular division of the trigeminal nerve (CN V3) carries a branchial motor and a general sensory component. The sensory pathway travels via five orocutaneous nerves: auriculotemporal, meningeal, buccal, lingual, and inferior alveolar. Loss of the first two nerves leaves little functional deficit, and compensation readily occurs; however, severe burns to the cutaneous distribution of these nerves may occur with the use of curling irons and other heated hair grooming devices. Patients are instructed to take care in daily hairdressing. Loss of the buccal, lingual, and inferior alveolar nerves affects the sensory feedback loop of the initial aspects of swallowing. The food bolus is not detected in the nonsensate area, affecting the preparatory phase of oral swallowing. Grafting of these nerves is recommended when possible; however, because these nerves are usually resected intracranially as the nerve exits the 569
570
OTOLOGIC SURGERY
FIGURE 47-1. Function of upper aerodigestive tract is coordinated by complex interaction of lower cranial nerves in speech and swallowing.
dura, it is impossible to identify the appropriate proximal fibers for nerve grafting. Swallowing therapy is the mainstay of rehabilitation, and allows compensation in most patients. The motor division of CN V3 provides function to six muscles: tensor veli palatini, tensor tympani, medial pterygoid, lateral pterygoid, masseter, and temporalis. The medial pterygoid nerve is the first branch off CV V3 after it exits the foramen ovale. Before it reaches the medial pterygoid muscle, it gives a branch to the tensor veli palatini and to the tensor tympani, both of which pass through the otic ganglion without synapsing. Isolated tensor veli palatini paralysis is compensated for by the action of the levator palatini muscle with minimal palatal dysfunction. Aural attenuation is decreased with loss of tensor tympani function, but is adequately compensated for by the action of the stapedius muscle as long as CN VII has been preserved. Therapy for loss of tensor tympani or stapedius function is usually unnecessary. Just past the otic ganglion, CN V3 gives off branches to the lateral pterygoid and the masseter muscle. Two or three deep temporal nerve branches arise in the same region and pass up into the temporalis muscle. As the inferior alveolar nerve enters its canal in the mandible, the final motor branches of CN V3 depart from it and go to the mylohyoid muscle and anterior belly of the digastric muscle. If the contralateral muscles of mastication are intact, patients usually compensate with little trismus or chewing dysfunction. Often, early trismus associated with exposure of the temporomandibular joint during surgery in this region may be overcome by propping the
FIGURE 47-2. Significant atrophy of temporalis has occurred 1 year after infratemporal fossa dissection with sacrifice of CN V3 and mobilization of temporalis muscle. (From Netterville JL, Civantos FJ: Rehabilitation of cranial nerve deficits after neurotologic skull base surgery. Laryngoscope 103(Suppl 60):45-54, 1993.)
mouth open with a stack of tongue depressors for several minutes two or three times daily. The number of tongue depressors is gradually increased until adequate opening is achieved. Shifting of the mandibular arch is usually corrected over time with this therapy and increasing use of the mandible. Atrophy of the temporalis muscle is an inevitable sequela of CN V3 sacrifice. For this reason, attempts at masseter or temporalis transfer for facial nerve rehabilitation are doomed to failure and should not be attempted. By 1 year, noticeable temporalis wasting occurs, and results in a significant cosmetic defect. Most patients desire placement of a silicone or methylmethacrylate implant beneath the residual temporalis muscle (Fig. 47-2). Care must be taken to contour the implant where it abuts the lateral orbital rim, or else a residual crease results in this region.
Abducens Nerve (Cranial Nerve VI) During lateral transtemporal approaches to the cranial base, the abducens nerve is often encountered in its course along the clivus and medial aspect of the petrous apex. Its sole function is to supply somatic motor innervation to the lateral rectus muscle. After exiting the brainstem, the abducens nerve lies along the clivus and enters Dorello’s canal inferomedial to the root of the trigeminal nerve. It passes beneath (rarely above) the superior sphenopetrous ligament in a sulcus on the petrous apex. It courses through the cavernous sinus lateral to the carotid into the superior orbital fissure. Injury may occur anywhere along
Chapter 47 • Rehabilitation of Lower Cranial Nerve Deficits after Neurotologic Skull Base Surgery
this course and result in lateral rectus muscle dysfunction with limitation of lateral gaze with associated diplopia. CN VI is exquisitely sensitive to pressure injury in the cavernous sinus, and great care must be taken not to pack the cavernous sinus vigorously to control bleeding because this may result in permanent lateral rectus palsy despite anatomic integrity of the nerve. If the nerve is known to be intact, treatment is conservative; prism glasses allow compensation until function returns, usually within 3 to 4 months. Botulinum toxin injection into the ipsilateral medial rectus has been used successfully to relieve diplopia by weakening the antagonist action of the medial rectus muscle.4 When it seems that permanent palsy has occurred, superior and inferior rectus muscle transposition may be performed to improve lateral gaze function.4
Glossopharyngeal Nerve (Cranial Nerve IX) The glossopharyngeal nerve has branchial motor, visceral motor, visceral sensory, general sensory, and special sensory components. Its branchial motor component is limited to the stylopharyngeus muscle, which elevates the pharynx during swallowing and speech. Isolated loss of motor fibers to this muscle results in little swallowing dysfunction; however, when combined with the loss of general sensory fibers of the glossopharyngeal plexus and vagal motor fibers, severe swallowing deficits result. The visceral motor component of the glossopharyngeal nerve exerts parasympathetic control over parotid salivary secretion. As CN IX exits the skull base through the pars nervosa of the jugular foramen, it forms superior and inferior ganglia that contain nerve cell bodies that mediate general, visceral, and special sensory function. Parasympathetic fibers leave the inferior ganglion and pass into the middle ear as the tympanic nerve. In the middle ear, these fibers form the tympanic plexus; branches from the tympanic plexus form the lesser petrosal nerve, which passes back up through the floor of the middle cranial fossa, travels intracranially, and descends through the foramen ovale in the greater wing of the sphenoid bone where it synapses at the otic ganglion. Postganglionic fibers travel with the auriculotemporal nerve to reach the parotid gland. Injury to these fibers may occur at many levels and lead to decreased parotid salivary flow; rarely, this leads to chronic parotitis. The general sensory component provides afferent feedback from the base of tongue and lateral pharyngeal wall, and the external ear and inner aspect of the tympanic membrane. The sensory fibers to the base of tongue and pharynx run in a plexus on the undersurface of the stylopharyngeus muscle and pierce the middle constrictor to reach the mucosa of the base of tongue. Disruption of this glossopharyngeal plexus causes significant delay in the oropharyngeal phase of swallowing on the ipsilateral side. In glomus tumor surgery, the glossopharyngeal nerve is taken at the pars nervosa
571
in the jugular foramen precluding nerve grafting. Isolated, unilateral CN IX deficits are well compensated for with swallowing therapy that focuses on maintaining passage of the food bolus adjacent to the contralateral sensate side of the pharynx using a chin-tuck and headturn maneuver toward the side of the deficit. When this deficit is combined with the loss of CN X and XII at the skull base, dysfunction becomes severe. Rehabilitation of this combined loss involves correction of glottal incompetence with thyroplasty and intensive swallowing therapy. The final function of CN IX is visceral sensory from the carotid body and carotid sinus by way of the nerve of Hering. Disruption of CN IX at the skull base causes loss of the carotid sinus reflex on the ipsilateral side and was originally described for the treatment of carotid sinus syndrome.5 Unilateral loss does not interfere with the control of blood pressure and pulse, presumably because of the presence of an intact reflex on the contralateral side. When there is bilateral alteration of this system, acute elevation of blood pressure to greater than 220 mm Hg may be seen within a 60 second period. Preoperatively, one should consider the potential for a bilateral deficit in any patient who has had surgery on the contralateral neck where CN IX fibers may have been disrupted, such as in carotid endarterectomy. Appropriate intraoperative management of this scenario includes administration of a pure α-blocker, such as phenoxybenzamine hydrochloride. Sodium nitroprusside may be added for further control. The patient is weaned to clonidine hydrochloride in the postoperative period. Permanent maintenance of blood pressure may be required. Consultation with an intensivist in the preoperative period is strongly recommended if a bilateral loss is anticipated.
Vagus Nerve (Cranial Nerve X) Of the lower cranial nerves, the vagus is most intimately involved in control of the airway and swallowing function. Isolated loss of vagal function yields a far greater deficit than the loss of any other single lower cranial nerve. The vagus nerve carries general sensory, visceral sensory, visceral motor, and branchial motor fibers. The general sensory component provides afferent signals from the external auditory canal, tympanic membrane, supraglottic larynx, and lateral pharyngeal wall. Two ganglia are formed as it exits the skull base through the pars nervosa of the jugular foramen: the superior (jugular) ganglion lies in the jugular foramen, and the inferior (nodose) ganglion is located 1 to 2 cm outside the foramen. The sensory fibers from the supraglottic larynx form the superior laryngeal nerve, which passes deep to the external and internal carotid arteries to join the vagus at the level of the inferior ganglion. Isolated loss of this nerve can result in swallowing difficulties; it is grafted on rare occasions in selected patients. Most often, the vagus is resected so proximally at the skull base that grafting is technically
572
OTOLOGIC SURGERY
impossible. With time and swallowing therapy, compensation for the sensory loss occurs. The visceral sensory and branchial motor components of CN X provide afferent sensory and parasympathetic function to the pharynx, larynx, trachea, esophagus, thoracic viscera, and abdominal viscera distally to the level of the splenic flexure of the colon. Unilateral loss of vagal function may cause decreased gastroesophageal motility, loss of lower esophageal sphincter tone, and delayed gastric emptying owing to inadequate pyloric sphincter function. Subsequent regurgitation during the early postoperative period is common, and may limit adequate nutrition, cause transient increases in intracranial pressure and potential for cerebrospinal fluid leak, and lead to life-threatening aspiration pneumonia. Pain medication and anticholinergic drying agents add further to gastric stasis. Treatment consists of administration of a motility agent, such as metoclopramide hydrochloride, and a temporary decrease in feeding rate. Over time, these symptoms gradually improve. If feeding intolerance continues, a jejunal feeding tube may be necessary. Rarely, usually in cases of bilateral vagal injury, total loss of lower esophageal sphincter tone can occur and may require Nissen fundoplication. The branchial motor component of CN X provides motor function to the palate, pharynx, and larynx, with the exception of the stylopharyngeus muscle (CN IX) and the tensor veli palatini muscle (CN V3). These fibers depart the vagus in three distinct bundles: the pharyngeal branch, the external branch of the superior laryngeal nerve, and the recurrent laryngeal nerve. The pharyngeal branch departs the vagus as the inferior (nodose) ganglion passes over the internal carotid, deep to the external carotid artery, and enters the pharynx at the upper border of the middle constrictor. Damage to this branch results in unilateral palatal and pharyngeal paralysis, which causes a loss of lateral wall motion on the affected side and an ineffective sphincter for closing the nasopharyngeal port; this leads to nasal regurgitation and hypernasal speech. In the early postoperative period, unilateral palatal paralysis causes no significant morbidity relative to more immediate concerns regarding airway and swallowing function. The pharyngeal dysfunction responds well to swallowing therapy. Subsequent velopharyngeal insufficiency (VPI) resulting in nasal regurgitation and hypernasal speech is embarrassing and quite bothersome to most patients. Nonsurgical approaches to VPI have included speech and swallowing therapy, palatal lift prostheses, and palatal obturators. Various surgical procedures have been described for the treatment of VPI secondary to cleft palate, including pharyngeal augmentation and pharyngoplasty. The most popular of the pharyngoplasty techniques is the superiorly based pharyngeal flap as described by Jackson.6 Vagal injury at or above the jugular foramen causes paralysis of not only the palate, but also the pharyngeal constrictors, resulting in loss of lateral wall motion on the affected side. Superiorly based flaps and
pharyngeal augmentation are used to address VPI in cleft palate patients and are designed to leave an open lateral port. Lateral pharyngoplasty techniques7 are adynamic, and do not recreate lateral pharyngeal wall movement. A simple alternative technique that addresses the palatal and pharyngeal component of VPI in skull base patients is to close the lateral port by adhering the palate on the affected side to the posterior pharyngeal wall. The chief advantage of this procedure in patients who may already have multiple cranial nerve deficits and abnormal swallowing is that it does not alter pharyngeal anatomy with long mucosal flaps, and does not carry the risk of pharyngeal stenosis. Unilateral palatal adhesion successfully eliminates hypernasality and nasal regurgitation, and has become the procedure of choice for the correction of velopharyngeal incompetence in neurotologic skull base patients.8,9 Palatal adhesion is usually performed several months after resection when swallowing function has stabilized, and when it is certain that an injured but intact vagus nerve has not recovered. Preoperative assessment by nasopharyngoscopy with either a rigid or a flexible scope shows unilateral closure of the nasopharynx. Under general anesthesia, the palate is exposed with a Dingman mouth gag. The paralyzed hemipalate and posterior pharyngeal wall are injected with epinephrine. Care is taken to inject just below any residual adenoid tissue. A transpalatal incision is performed in the area of the palatal crease that forms with normal palatal elevation, and the posterior pharyngeal wall is viewed through this incision (Fig. 47-3). A similar incision is created into the posterior pharyngeal wall down to the prevertebral fascia. The pharyngeal mucosa is elevated for just a few millimeters all around to create an edge to which the nasopharyngeal mucosa of the palate is sewn. Multiple deep mattress sutures are used to suture the nasopharyngeal surface of the palate to the posterior pharyngeal wall (Fig. 47-4). The oral surface of the palate is closed on itself, creating a unilateral palatal adhesion (Figs. 47-5 to 47-7). Postoperatively, the patient is immediately allowed to take a liquid diet and is discharged from the hospital as soon as pain control is adequate. The only complication that has occurred from this procedure is a wound dehiscence that granulates over several months, producing a secondary adhesion with minimal VPI. Sleep apnea has not been seen as a complication of this procedure. Improvement of nasal regurgitation and reduction in hypernasal speech occur in all patients.8,9 The effect of paralysis of the superior, middle, and inferior constrictor muscles causes more morbidity than the palatal dysfunction. As the food bolus passes into the oropharynx, the paralyzed side dilates laterally and forms a pseudopocket that collects the bolus. The normal contraction on the contralateral side pushes the bolus into this region, rather than down into the hypopharynx. This delay interrupts the normal timing of the swallowing event so that when the larynx reopens, the food bolus,
Chapter 47 • Rehabilitation of Lower Cranial Nerve Deficits after Neurotologic Skull Base Surgery
573
FIGURE 47-3. Transpalatal incision is made in area of palatal crease that forms with normal palatal elevation, and posterior pharyngeal wall is viewed.
FIGURE 47-4. Incisions are made in palate and posterior pharyngeal wall. Multiple mattress stitches are placed to complete adhesion. FIGURE 47-5. Superior endoscopic view of left-sided palatal adhesion.
574
OTOLOGIC SURGERY
Chapter 47 • Rehabilitation of Lower Cranial Nerve Deficits after Neurotologic Skull Base Surgery
575
FIGURE 47-6. Sagittal view of completed palatal adhesion. FIGURE 47-7. A and B, Endoscopic view from above of right-sided palatal adhesion during quiet respiration (A), and showing complete closure during production of pressure consonant (B). FIGURE 47-8. Alar window is positioned as shown, taking care to preserve inferior alar strut. FIGURE 47-9. Depth gauge is used to medialize true vocal fold based on auditory and visual feedback.
which should have been in the esophagus, is still partially in the hypopharynx, where it is then aspirated. Treatment is centered around an intensive swallowing therapy program in which a head positioning technique is used to obliterate physically the paralyzed side and force the food bolus onto the normal side. Avoidance of a tracheotomy, if possible, further hastens recovery of swallowing function because the tracheotomy would interfere with laryngeal elevation. Loss of cricopharyngeal relaxation combined with an indwelling tracheotomy tube (particularly a cuffed tube) further hinders the passage of the food bolus through the esophageal inlet and encourages aspiration. The second motor branch of CN X is the superior laryngeal nerve. It provides motor control to the ipsilateral cricothyroid muscle. Loss of this nerve results in decreased vocal range, which is usually a problem only for professional voice patients. Because the vagus is usually resected or injured at the skull base, patients have superior and recurrent laryngeal nerve paralyses, which are rehabilitated at the same time with silicone elastomer (Silastic) medialization thyroplasty and arytenoid adduction. The third motor branch of the vagus is the recurrent laryngeal nerve, which provides innervation to the intrinsic laryngeal musculature and the cricopharyngeus muscle. True vocal fold paralysis and lack of cricopharyngeal relaxation result from resection or neural injury at the time of surgery. The resultant glottal incompetence leads to a weak, breathy voice and an inefficient cough. This deficit combined with lack of cricopharyngeal relaxation results in aspiration and poor pulmonary hygiene. Tracheotomy has been universally performed to protect the airway until early swallowing rehabilitation can be accomplished. From an airway perspective, most patients with high vagal injury tolerate a unilateral vocal fold paralysis, making tracheotomy generally unnecessary; for the reasons discussed earlier, tracheotomy prolongs swallowing rehabilitation and adds morbidity to the postoperative course. Silastic medialization with arytenoid adduction has been the procedure of choice for addressing vocal fold paralysis in these patients.10 Although cricopharyngeal myotomy is not routinely necessary in these patients, it has been shown to decrease aspiration and promote early swallowing.11 The primary goal for vocal fold medialization in skull base patients is (1) to provide glottal competence, (2) to provide an efficient mechanism for coughing, and (3) to give support and volume to the voice. Generally, most patients with a unilateral vocal fold paralysis undergo a
absorbable gelatin sponge (Gelfoam) injection postoperatively regardless of whether the nerve was taken. This injection lasts approximately 6 to 8 weeks, and greatly facilitates speech and swallowing rehabilitation while the nerve regains function or until Silastic medialization and arytenoid adduction may be performed. It is prudent to wait 2 to 3 months before proceeding with thyroplasty because denervation vocal fold atrophy occurs over several weeks, which alters the size of the Silastic implant and the degree of arytenoid adduction necessary to achieve good voice and airway results. Several weeks postoperatively, the patient should return for airway evaluation, videostroboscopy, and airflow measurements. A Silastic medialization under local anesthesia is performed. The technique is outlined in the following paragraphs and has been described elsewhere.10,12-14 Under local anesthesia, a midline horizontal incision or an extension of the cervical portion of the skull base incision is made overlying the midpoint of the thyroid cartilage. The sternohyoid muscle is divided at its medial attachments to the hyoid bone superiorly. A perichondrial flap is created from the midline back to the posterior edge of the thyroid cartilage, elevating the remaining attachments of the sternohyoid muscle and the thyrohyoid muscle with the flap, and exposing the inferior edge of the thyroid ala. A rectangular window is outlined so that the anterior extent lies 5 mm back from the anterior commissure in women and 7 mm back in men. The window is placed as low as possible, leaving an inferior 3 mm thyroid cartilage strut that is wide enough not to fracture when the implant is placed (Fig. 47-8). The final outline of the window is usually 6 mm in height and 13 mm in length. The location of the implant relative to the anterior commissure is based on the angle of the thyroid ala. With an increased thyroid cartilage angle as seen in women, the implant must be brought closer to the anterior commissure, and must have an increased slope to compensate for the wider angle. If the implant is placed too far anteriorly, overmedialization of the anterior commissure occurs, producing a strained quality to the voice. Using 4× loupes, a high-speed drill with a 2 mm cutting burr is employed to drill away the outline of the window. The cartilage may be partially ossified, particularly inferiorly and posteriorly. The window is usually 3 to 4 mm thick anteriorly and 5 to 7 mm thick posteriorly. Next, the long and short phonosurgery intralaryngeal elevators (Xomed, Jacksonville, MS) are used to elevate the inner perichondrium in all directions except anteriorly. Medialization is attempted, but it is rarely possible
576
OTOLOGIC SURGERY
Chapter 47 • Rehabilitation of Lower Cranial Nerve Deficits after Neurotologic Skull Base Surgery
577
FIGURE 47-10. Preformed Silastic block is trimmed to fit auditory and visual feedback. FIGURE 47-11. Final position of vocal cord after placement of Silastic implant. FIGURE 47-12. Exposure is achieved by rotating thyroid cartilage upward with a hook. A Kerrison rongeur is used to remove posterior thyroid ala. Muscular process is palpated and moved on its plane of abduction and adduction.
FIGURE 47-13. Arytenoid stitch is placed to mimic vector of force that rotates vocal process downward and inward.
to achieve the required medialization with the inner perichondrium intact. The superior, posterior, and inferior margins of the perichondrium are incised discretely without injuring the lateral fascia of the thyroarytenoid muscle just deep to this. Branches of the superior laryngeal artery and vein can usually be seen lying just under this fascia. Although some surgeons fear that dividing the perichondrium leads to implant extrusion, this has not occurred in several hundred medialization procedures using this technique.13 The depth gauge is used to medialize the cord, and voice quality and cord position are assessed (Fig. 47-9). Based on these visual and auditory data, an implant is carved to appropriate dimensions from a preformed Silastic block (Fig. 47-10). Its inferior border is thicker than the superior border, and the posterior border is thicker than the anterior border. This creates an even medialization of the vocal fold with the point of maximum medialization at the lower edge of the window. Good alignment of the true vocal cord in the midline is usually obtained with little medialization of the false cord. All edges of the Silastic block are beveled to allow ease of insertion through the window. The average size of the block that is used when performing this under general anesthesia is 2 to 3 mm thick anteriorly and 5 to 7 mm at the point of maximal medialization, with an overall length of 13 to 18 mm. The block is inserted in the window by placing the lower flange behind the lower window strut while compressing the upper flange until it expands under the cartilage. Some 4-0 polypropylene (Prolene) sutures are used to stabilize the implant to the inferior strut of thyroid cartilage (Fig. 47-11). All patients are given 10 mg of dexamethasone at the time of the procedure and at least three doses every 6 hours postoperatively. Patients are treated with glycopyrrolate, 0.2 mg every 6 hours, for 3 days postoperatively to decrease salivary secretions. High vagal injury with loss of the superior laryngeal nerve is usually best addressed with medialization and arytenoid adduction. Arytenoid adduction has been described elsewhere,15,16 and is performed at the time of Silastic medialization. When the cartilage window has been created, the voice is assessed using the depth gauge as described earlier, and the decision to proceed with arytenoid adduction is made. Exposure of the arytenoid cartilage is achieved by first placing a hook beneath the posterior border of the thyroid cartilage and rotating the larynx forward. The perichondrium at the posterior border is divided, and
the piriform sinus mucosa is elevated away from the inner surface of the thyroid cartilage. A 5 mm Kerrison rongeur is used to remove the posterior border of the thyroid cartilage, exposing the paraglottic space lateral to the muscular process of the arytenoid. Patients are then asked to purse their lips and blow out to confirm the location of the piriform mucosa so that dissection may be carried out anteriorly to expose the posterior cricoarytenoid muscle. The muscular process is palpated and moved in its plane of abduction and adduction while observing the monitor (Fig. 47-12). A double-armed 4-0 Prolene suture is passed through the lateral edge of the muscular process in a figure-eight fashion to secure the arytenoid (Fig. 47-13). The goal of this stitch is to mimic the vector of force that rotates the vocal process of the arytenoid down (inferior) and in (medial) during phonation. To accomplish this, one end of the suture is brought through the paraglottic space and out through the cartilage just anterior to the window. The other end of the suture is passed from the window below the lower cartilage strut and then through the cricothyroid membrane soft tissue in the midline. Gentle traction is applied to the arytenoid, and its motion is observed on the monitor. The implant is carved and placed. Both arytenoid sutures pass deep to the implant. Final adjustments are made to the arytenoid stitch, and it is tied down. The perichondrium and strap muscles are laid back into position, and the wound is closed with absorbable suture.
Spinal Accessory Nerve (Cranial Nerve XI) The spinal accessory nerve has only a branchial motor component. Its lower motor neuron cell bodies arise in the spinal cord, and its motor fibers ascend into the cranium and then descend through the pars nervosa of the jugular foramen, passing superficial to (70%), posterior to (27%), or through (3%) the jugular vein to innervate the sternocleidomastoid and trapezius muscles.17 Loss of sternocleidomastoid function results in weakness when turning the head away from the operated side, although this is rarely noticed by most patients. Loss of trapezius function results in downward and lateral rotation of the scapula with a shoulder droop. This causes severe shoulder disability secondary to weakness and pain. Whether the nerve may be grafted successfully or not, a shoulder exercise program that emphasizes strengthening of the levator, scalene, and rhomboid muscles should be administered by a qualified physical therapist. This program
578
OTOLOGIC SURGERY
should continue indefinitely to maintain the strength and support of the shoulder girdle.
intake. Early management consists of dietary modification with bland foods and oral carbamazepine (100 to 200 mg twice daily). Slowly, over time, the symptoms improve.
Hypoglossal Nerve (Cranial Nerve XII) The hypoglossal nerve provides somatic motor control to all the intrinsic and extrinsic muscles of the tongue except the palatoglossus. The nerve exits the skull base through the hypoglossal canal medial to the jugular foramen. As it passes laterally, it shares fibers with the vagus nerve near the inferior (nodose) ganglion. Separation of these fibers leads to vocal fold paresis or paralysis and should be avoided.18 In time, most patients can compensate for unilateral tongue paralysis. The deficit is characterized by difficulty in positioning the bolus during the oral phase of swallowing. Often the bolus becomes lodged beneath the tongue on the paralyzed side. With swallowing therapy and lingual exercise, patients learn to position the bolus on the nonparalyzed side.
Cervical Sympathetic Chain The sympathetic innervation to the head and neck structures originates in the upper thoracic segment of the spinal column. The preganglionic fibers ascend in the sympathetic chain to exit through one of its four ganglia (superior, middle, vertebral, and stellate). The superior cervical ganglion lies on the surface of the longus capitus muscle at the level of the second or third cervical vertebra, deep to the internal carotid artery at the most distal or superior point of the cervical sympathetic chain. From this ganglion, postganglionic fibers typically reach the first three or four cervical rootlets. They also communicate with CN IX, X, XI, and XII as part of the pharyngeal plexus. Finally, they form the sympathetic plexus that ascends along the internal carotid artery.2,19 Damage to the superior cervical ganglion or the internal carotid plexus results in Horner’s syndrome, consisting of ptosis, miosis, and anhidrosis. For most patients, there is little functional deficit. Rarely, visual fields may be partially obstructed because of the ptosis. More commonly, it is a cosmetic nuisance. Elevation of the eyelid to its normal open position can be accomplished by either levator shortening or resection of Müller’s muscle.20 The other significant sequela from injury to the cervical sympathetics at the skull base is loss of sympathetic innervation to the parotid gland. This complication is frequently seen with resection of high vagal paragangliomas when the sympathetic chain is either resected or damaged. In this setting, a prolonged course of cramping pain is associated with the first bite of each meal and has been named “firstbite syndrome.”18 Patients usually describe this pain as a spasm over the parotid region that begins with the first bite and subsides within the next several bites. The intensity of the pain is increased with strong sialogogues, such as vinegar or lemon. In the early postoperative period, the pain can be so severe as to limit oral
SUMMARY Damage to any single lower cranial nerve results in sufficient morbidity to warrant therapy; however, most patients compensate for isolated loss of function. Multiple cranial nerve deficits (CN IX, IX, and XII) as seen in glomus tumor surgery employ the full efforts of the rehabilitative team. After vocal cord medialization, arytenoid adduction, palatal adhesion, and speech and swallowing therapy, most younger (<55 years old) patients resume adequate oral intake; however, the time it takes to return to a reasonable, enjoyable diet often extends to 1 year postoperatively. Some patients never attain the goal of enjoyable intake and continue to struggle to maintain adequate nutrition. This is generally the rule in elderly patients, and careful consideration should be given to the slow growth potential of certain tumors, the resulting postoperative cranial nerve deficits, and the patient’s age before proceeding with aggressive surgical management of lateral skull base tumors.
REFERENCES 1. Bastian RW: Videoendoscopic evaluation of patients with dysphagia: An adjunct to the modified barium swallow. Otolaryngol Head Neck Surg 104:339, 1991. 2. Hollinshead WH : Anatomy for Surgeons: The Head and Neck, 3rd ed. Philadelphia, Lippincott, 1982. 3. Wilson-Pouwels L , Akesson E J, Stewart PA : Cranial Nerves: Anatomy and Clinical Comments. Philadelphia, Decker, 1988. 4. Rosenbaum A L , Kushyner B : Vertical rectus muscle transposition and botulinum toxin after abducens nerve palsy. Arch Ophthalmol 107:820, 1989. 5. Ray B S, Stewart H J: Role of the glossopharyngeal nerve in the carotid sinus reflex in man: Relief of carotid sinus syndrome by intracranial section of the glossopharyngeal nerve. Surgery 23:411, 1948. 6. Jackson I : Discussion: A review of 236 cleft palate patients treated with dynamic muscle sphincter. Plast Reconstr Surg 71:187, 1983. 7. Orticochea M : A review of 236 cleft palate patients treated with dynamic muscle sphincter. Plast Reconstr Surg 71:180, 1983. 8. Netterville J L , Vrabec JT: Unilateral palatal adhesion for paralysis after high vagal injury. Arch Otolaryngol Head Neck Surg 120:218, 1994. 9. Netterville J L , Civantos FJ: Rehabilitation of cranial nerve deficits after neurotologic skull base surgery. Laryngoscope 103:45, 1993. 10. Netterville J L , Jackson CG, Civantos FJ: Thyroplasty in the functional rehabilitation of neurotologic skull base surgery patients. Am J Otol 14:460, 1993.
Chapter 47 • Rehabilitation of Lower Cranial Nerve Deficits after Neurotologic Skull Base Surgery 11. Montgomery WW, Hillman R E, Varvares M A : Combined thyroplasty type I and inferior constrictor myotomy. Ann Otol Rhinol Laryngol 103:858, 1994. 12. Isshiki N: Recent advances in phonosurgery. Folia Phoniatr Logop 32:119, 1984. 13. Netterville J L , Stone R E, Lukens L S, et al: Silastic medialization and arytenoid adduction: The Vanderbilt experience—a review of 116 phonosurgical procedures. Ann Otol Rhinol Laryngol 102:413, 1993. 14. Wanamaker J R, Netterville J L, Ossoff R H: Phonosurgery: Silastic medialization for unilateral vocal fold paralysis. Operative Tech Otolaryngol Head Neck Surg 4:207, 1993. 15. Isshiki N, Tanabe M, Sawada M : Arytenoid adduction for unilateral vocal fold paralysis. Arch Otolaryngol 104:555, 1978. 16. Miller FR , Bryant G L , Netterville J L : Arytenoid adduction in vocal fold paralysis. Operative Tech Otolaryngol Head Neck Surg 10:36, 1999.
579
17. Parsons FG, Keith A : Seventh report of the Committee of Collective Investigation of the Anatomical Society of Great Britain and Ireland, for the year 1896-97: Question III: The position of the spinal accessory nerve. Whether it passes outward between the jugular vein and internal carotid artery, or between the jugular vein and the atlas? Whether it perforates the sterno-mastoid or not: If so does the whole nerve perforate or only part? Which division of the sterno-mastoid does it perforate? J Anat Physiol 32:177, 1897. 18. Netterville J L , Reilly K M, Robertson D, et al: Carotid body tumors: A review of 30 patients with 46 tumors. Laryngoscope 105:115, 1995. 19. Collins S L : Cervical sympathetic nerves in the surgery of the neck. Otolaryngol Head Neck Surg 105:544, 1992. 20. Dortzbach R K : Superior tarsal muscle resection to correct blepharoptosis. Ophthalmology 86:1883, 1979.
48
Middle Fossa Approach Clough Shelton, Derald E. Brackmann, and William F. House Videos corresponding to this chapter are available online at www.expertconsult.com.
The middle fossa approach for vestibular nerve section was reported in 1904; however, hammer and chisel were used at that time, which put the facial nerve at risk.1 The middle fossa approach did not have widespread application until refined by the senior author (W.F.H.) in 1961.2 The approach was used initially for decompression of the internal auditory canal (IAC) in cases of extensive otosclerosis. That therapy was later abandoned, but it became evident that this approach was suitable for removal of acoustic tumors. Initially, the middle fossa approach was used for tumors of all sizes. Further experience showed that it was most suitable for small tumors,3-5 however, and that preservation of hearing and facial nerve function was possible in a significant proportion of operated patients.6 The middle fossa approach was used infrequently until the development of gadolinium-enhanced magnetic resonance imaging (MRI). With this development, a larger number of acoustic tumors are diagnosed when they are small and before hearing has been significantly affected, making an attempt at hearing preservation desirable. The middle fossa approach provides complete exposure of the contents of the IAC, allowing removal of laterally placed tumors without the need for blind dissection.7 This exposure ensures total removal and is well suited for the removal of very small acoustic tumors.8 The facial nerve can be located in its bony canal, allowing positive identification in a location not involved by tumor. The middle fossa approach is technically difficult because of the lack of robust landmarks and the limited exposure. Bleeding in the posterior fossa can be difficult to control because of the limited access. Because of its location in the superior aspect of the IAC, the facial nerve is subjected to more manipulation in this approach than in other approaches.9,10 In the past, facial nerve results in middle fossa cases have not been as good as results from the translabyrinthine approach for similar-sized tumors.11 The routine use of the facial nerve monitor has helped improve these results, however, and now the results are comparable between the two approaches.12 Several authors use an extended middle fossa approach for large tumors.13-15 The tentorium is divided
to give wider access to the posterior fossa. Some authors also perform a labyrinthectomy to enlarge the exposure when hearing preservation is not attempted.16-18
INDICATIONS The primary indications for the middle fossa approach are a small acoustic tumor, in the IAC or with moderate extension into the cerebellopontine angle, and good preoperative hearing. In contrast, the retrosigmoid approach is used for patients with good hearing and small tumors located mostly in the cerebellopontine angle, without extension into the lateral aspect of the IAC. For hearing conservation surgery, we use the arbitrary audiometric criteria of speech reception threshold of better than 50 dB and speech discrimination score of better than 50%, although these indications must be individualized to the needs of the patient.19 Some patients have a subjective assessment about the usefulness of the preoperative hearing, such as the ability to use the telephone with the involved ear or the ability to localize a sound source. These subjective valuations should also be taken into consideration when determining hearing preservation candidacy. Some authors advocate attempting hearing preservation in the removal of small acoustic tumors if any measurable preoperative hearing exists.20 Patients older than 65 years do not tolerate the middle fossa approach as well as younger patients because of the fragility of the dura and retraction of the temporal lobe.
PREOPERATIVE EVALUATION Several preoperative factors may predict postoperative hearing preservation. The most obvious is tumor size. Intuitively, the smaller the tumor, the easier it is to remove, and the more likely that hearing will be saved. This trend has been substantiated by several authors.11,21,22 Some authors have also found that the better the preoperative hearing, the more likely it will be preserved,23,24 whereas 581
582
OTOLOGIC SURGERY
others have failed to identify such a relationship.11,22,25 Also, an intact preoperative stapedial reflex has been associated with successful postoperative hearing preservation.23 Several authors have reported a relationship between preoperative auditory brainstem response (ABR), and audiometry and hearing preservation.17,22 In one report, hearing was preserved in 78% of patients with an interaural wave V latency difference of 0.4 ms or less.11 For latency differences of 0.5 to 2 ms, the hearing preservation rate decreased to 58%. In patients with no response on the ABR, postoperative measurable hearing remained in only 50%. Patients with a more normal preoperative ABR result apparently have a greater success rate for postoperative hearing preservation. This result may reflect less tumor involvement of the cochlear nerve. Others do not find preoperative ABR to be predictive of hearing outcome.26,27 Tumors arising from the superior vestibular nerve have a higher rate of hearing preservation than tumors arising from the inferior vestibular nerve. Acoustic tumors developing in the inferior portion of the IAC may involve the cochlear nerve earlier and more severely.27,28 In a series of middle fossa acoustic tumor removals, 68% of patients whose tumors were found intraoperatively to arise from the superior vestibular nerve had measurable hearing preservation, whereas only 43% of patients whose tumors originated from the inferior vestibular nerve had measurable postoperative hearing.11 This difference was statistically significant. Preoperative electronystagmography may predict tumor origin and hearing preservation. The caloric response reflects superior vestibular nerve function. In the presence of a small acoustic neuroma, a normal caloric response indicates an inferior vestibular nerve tumor, whereas a decreased response suggests a tumor arising from the superior vestibular nerve. Of a group of 54 patients who had preoperative electronystagmography, hearing was preserved in 64% with hypoactive caloric responses, whereas postoperative hearing remained in only 45% of patients with normal caloric responses.11 The association of normal caloric tests with nonpreservation of hearing has also been reported by others,10,17 although some series have not found this correlation.29 Vestibular evoked myogenic potentials may prove useful in determining the nerve of tumor origin. Early experience indicates that a combination of normal electronystagmography and a reduced vestibular evoked myogenic potentials predicts the inferior vestibular nerve as the nerve of origin. For intracanalicular tumors, the radiographic appearance may predict success at hearing preservation. Small tumors that enlarge the IAC have a poorer prognosis for hearing preservation (Jackler RK, personal communication, 1990). In our experience, these small tumors that expand the canal are very adherent to the cochlear nerve, which adversely affects the hearing outcome. Fast spin
echo MRI provides ultra-high-resolution images of IAC anatomy.30 With this imaging technique, it is possible to determine the nerve of origin for small tumors. Tumors that are impacted into the lateral end of the IAC, especially tumors that impinge on the cochlear nerve canal, have a lower rate of hearing preservation.31
PATIENT COUNSELING After a thorough discussion of the relevant anatomy and the necessity to treat acoustic tumors, the options regarding surgical approaches to remove acoustic tumors are described to the patient. For patients with small tumors and good preoperative hearing, the issues of hearing preservation are discussed. We tell such patients that there is approximately a 50% chance of saving hearing. It is also important for patients to understand that hearing rarely improves after tumor removal.32 If the preoperative electronystagmography and ABR data are available and are favorable (see earlier), patients are informed that the prognosis for hearing preservation is above average. The patient is told that there is approximately a 90% chance that normal or near-normal facial nerve function will be obtained in the long-term, but that there is a 20% to 30% chance of having temporary facial paresis in the early postoperative period. Although the facial nerve results for either the middle fossa or retrosigmoid approach are excellent, our best and most consistent facial nerve results for small tumors occur with the translabyrinthine approach. Patients with preoperative tinnitus are counseled that the problem will likely get better, but probably will not disappear. Patients with no preoperative tinnitus have approximately a 25% chance of developing it postoperatively.33 Other important possible but rare complications are discussed, including cerebrospinal fluid leak, meningitis, serious brain complications, death, and blood transfusion options. The patient can donate 1 U of autologous blood before surgery, although transfusion is rarely needed. Recuperation can take weeks to months, and most patients return to work within 4 to 6 weeks. The patient should expect to be dizzy postoperatively, and the rapidity of the central compensation greatly influences the time course of the recuperation.
SURGERY Preoperative Preparation Intraoperative furosemide and mannitol are given to allow easier temporal lobe retraction. A single dose of corticosteroid such as dexamethasone is routinely used intravenously at the beginning of surgery. This single dose of steroid does not seem to affect wound healing adversely. Long-acting muscle relaxants are avoided during surgery
Chapter 48 • Middle Fossa Approach
so as not to interfere with facial nerve monitoring. Preoperative antibiotics are administered. (Chapter 1 of this text details surgical site preparation and draping, along with the instruments used, including the House-Urban middle fossa retractor.) Intraoperative ABR is routinely used. One of us (D.E.B.) also uses direct CN VIII recordings.34
Surgical Anatomy The surgical anatomy of the temporal bone from the middle fossa approach is compact and complex (Fig. 48-1). Landmarks are not as apparent as with other approaches through the temporal bone, so laboratory dissection is useful for the surgeon to become familiar with the anatomy from above. Anteriorly, the limit of the dissection is the middle meningeal artery, which is lateral to the greater superficial petrosal nerve. The arcuate eminence marks the position of the superior semicircular canal and may be readily apparent in some patients, but obscure in others. Kartush and coworkers35 cautioned that the relationship between the arcuate eminence and the superior semicircular canal may be variable in some patients, but the superior canal tends to be perpendicular to the petrous ridge. Medially, the superior petrosal sinus runs along the petrous ridge.
583
Surgical tolerances are very tight in the area of the lateral IAC. The labyrinthine portion of the facial nerve lies immediately posterior to the basal turn of the cochlea. Bill’s bar separates the facial and superior vestibular nerves. Slightly posterior and lateral to this area is the vestibule and ampullated end of the superior semicircular canal. Identification of the geniculate ganglion can be accomplished by tracing the greater superficial petrosal nerve posteriorly to it. If the tegmen is unroofed, the geniculate is found to be slightly anterior to the head of the malleus. The IAC lies approximately on the same axis as the external auditory canal; this relationship is useful in orienting the surgical field.12 The more medial one progresses along the IAC, the more space exists around it,36 allowing for safe dissection in this area. Several methods can be used to locate the IAC, and are reviewed in detail elsewhere.12,35 The technique of Garcia-Ibanez and Garcia-Ibanez37 provides a reliable and safe method to locate the IAC. It involves drilling on the bisection of the angle formed by the superior semicircular canal and greater superficial petrosal nerve. The IAC can be initially located in the “safe” medial area of the temporal bone and followed laterally.
Surgical Technique
FIGURE 48-1. Surgical anatomy of temporal bone as viewed from middle fossa approach.
The patient is placed in the supine position with the head turned to the side. The surgeon is seated at the head of the table, and the anesthesiologist is at the foot. An incision is made in the pretragal area and extended superiorly in a gently curving fashion (Fig. 48-2). The midportion of the incision curves posteriorly, which keeps it in the hair of most patients with male pattern baldness. An inferiorly based U-shaped flap is fashioned of the temporalis muscle and fascia, and is reflected inferiorly. A craniotomy opening is made in the squamous portion of the temporal bone (Fig. 48-3). It is located approximately two thirds anterior and one third posterior to the external auditory canal and is approximately 5 cm by 5 cm. Anterior placement of the craniotomy is important for adequate exposure, especially for a left ear. This bone flap is based on the root of the zygoma as close to the floor of the middle fossa as possible. The craniotomy flap can be fashioned using a high-speed drill with a footplate attachment that protects the underlying dura. Dura is initially exposed in two corners of the bone flap. These exposures are arranged diagonally, which allows separation of the dura from the bone flap and introduction of the footplate drill. During creation of this flap, care is taken to avoid laceration of the underlying dura. It is important to keep the footplate drill perpendicular to the bone to avoid tearing the dura. The extradural position of the footplate should be periodically verified during the creation of the craniotomy. It may be necessary to remove additional craniotomy window bone
584
OTOLOGIC SURGERY FIGURE 48-2.
EAC Craniotomy window
Incision
FIGURE 48-3. FIGURE 48-2. Incision begins in pretragal area and extends 7 to 8 cm superiorly in gently curving fashion. FIGURE 48-3. Two thirds of craniotomy window is located anterior to external auditory canal (EAC).
toward the floor of the middle fossa with a cutting burr. Alternatively, the whole craniotomy flap can be outlined with a high-speed drill, using standard cutting and then diamond burrs. The bone flap is set aside for later replacement. The dura is elevated from the floor of the middle fossa. The initial landmark is the middle meningeal artery, which marks the anterior extent of the dissection. Frequently, venous bleeding is encountered from this area and can be controlled with absorbable knitted fabric (Surgicel). Powdered absorbable gelatin sponge (Gelfoam) and thrombin, mixed into a slurry, can also be used for hemostasis. Dissection of the dura proceeds in a posterior-to-anterior manner. In approximately 5% of cases, the geniculate ganglion of the facial nerve is dehiscent, but injury can be avoided with dural elevation from posterior to anterior. The petrous ridge is identified, and care is taken not to injure the superior petrosal sinus, which is elevated from its groove at the time the true ridge is identified. The arcuate eminence and greater superficial petrosal nerve are identified. These are the major landmarks
to the subsequent intratemporal dissection. When the dura has been elevated, typically with a suction-irrigator and a blunt dural elevator, the House-Urban retractor is placed over the medial ridge of the superior petrosal sinus (true ridge) to support the temporal lobe. The tip of the retractor is placed beneath the true petrous ridge. To maintain a secure position, the teeth of the retaining retractor should be locked against the bony margins of the craniotomy window, and the tip of the retractor should be placed beneath the true petrous ridge (Fig. 48-4). One of us (C.S.) uses a Layla retractor system, which offers a lower profile than the House-Urban. Two retractor blades are employed to support the widely elevated temporal lobe.38 The tip of the anterior blade is placed into the groove of the fifth cranial nerve, and the tip of the posterior blade is placed medial to the petrous ridge near the arcuate eminence. The greater superficial petrosal nerve is located medial to the middle meningeal artery (Fig. 48-5). Using a large diamond drill and continuous suction-irrigation, the superior semicircular canal is identified. As the superior semicircular canal is skeletonized and followed
Chapter 48 • Middle Fossa Approach
585
FIGURE 48-4. Temporal lobe is supported by HouseUrban retractor.
Geniculate ganglion Cochlea Greater superficial petrosal n. Carotid artery
Incus
Bill’s bar Inferior vestibular n. Arcuate eminence Cochlear n. Facial n. Superior vestibular n.
FIGURE 48-5. Greater superficial petrosal nerve is identified medial to middle meningeal artery.
a nteriorly, the geniculate ganglion can be identified. Bone is removed at the medial aspect of the petrous ridge at the bisection of the angle formed by the greater superficial petrosal nerve and the superior semicircular canal. The IAC is identified in this medial location and traced laterally. The dura of the posterior fossa is widely exposed
(2 cm), and the circumference of the porus acusticus is exposed for approximately 270 degrees (Fig. 48-6). The posterior fossa dura can be opened with a microblade (No. 59; Beaver Company) to release the cerebrospinal fluid. This relaxes the temporal lobe and facilitates the removal of the retractors to allow an unimpeded view of
586
OTOLOGIC SURGERY
Facial n.
Bill’s bar Superior vestibular n. Arcuate eminence Posterior fossa dura
FIGURE 48-6. Geniculate ganglion is found by following superficial petrosal nerve posteriorly. Bill’s bar separates facial nerve from superior vestibular nerve at lateral end of internal auditory canal. Internal auditory canal is skeletonized through the entire length. Bone is removed around porus acusticus, uncovering dura of posterior fossa.
the lateral end of the IAC. As the dissection proceeds laterally, it must narrow to about 90 degrees because of the encroachment of the cochlea and superior semicircular canal. At the lateral end of the IAC, Bill’s bar and the labyrinthine facial nerve are exposed. The dura of the IAC is divided along the posterior aspect (Fig. 48-7). The facial nerve is clearly identified in the anterior portion of the IAC. The superior vestibular nerve is divided at its lateral end, and the tumor is separated from the facial nerve using high magnification (Fig. 48-8). The tumor is separated initially at Bill’s bar, but all dissection along the facial nerve is from medial to lateral. The arachnoid is divided, and the edge of the facial nerve is identified. The facial-vestibular anastomosis is sharply cut. Following the border of the facial nerve can facilitate tracing its course. Intracapsular debulking is performed, if needed, using microscissors and cup forceps. Tumor removal is accomplished from a medial-to-lateral direction to prevent traction on the cochlear nerve and blood supply as it enters the modiolus. If uninvolved, the inferior vestibular nerve can be left in an effort to preserve the labyrinthine artery. Persistent unsteadiness from a partial vestibulopathy can occur in patients with a retained inferior vestibular nerve. To avoid persistent unsteadiness, we recommend cutting the inferior vestibular nerve medial to Scarpa’s ganglion, but not dissecting it at the fundus of the IAC. After irrigation of the tumor bed (Fig. 48-9) and establishment of hemostasis, papaverine-soaked Gelfoam is placed on the cochlear nerve to prevent vasospasm. The first author (C.S.) also places methylprednisolone (SoluMedrol)–soaked Gelfoam on the facial nerve to decrease postoperative edema. Abdominal fat is used to close the
defect in the IAC. The retractor is removed, and the temporal lobe is allowed to re-expand. The craniotomy flap is replaced. The wound is closed with absorbable subcutaneous sutures. A mastoid-type pressure dressing is maintained for 4 days postoperatively. The patient is observed in the intensive care unit overnight, and typically stays in the hospital for 3 to 4 days. Once the patient leaves the intensive care unit, ambulation is encouraged. We believe that early ambulation is important for rapid vestibular compensation. Although not severe, postoperative pain after the middle fossa approach is more intense than the postoperative pain from the other approaches. This pain results from muscle spasm from division of the temporalis muscle. Some degree of temporary trismus may also result. Routine postoperative analgesics are usually sufficient to control this pain.
RESULTS Course of Healing Postoperative dizziness varies with the amount of remaining vestibular function in the ear preoperatively. The more function remaining, the greater the postoperative dizziness, and the longer the time for central compensation to occur. Patients tend to have the most severe dizziness the first 1 to 2 days postoperatively, and by the end of the first week they are left with unsteadiness. By this time, they generally can ambulate without assistance. The middle fossa and abdominal incisions are usually well healed, and patients are able to get them wet 1 week after surgery. Other postoperative instructions
Chapter 48 • Middle Fossa Approach
587
Facial n.
Incision along superior vestibular n.
FIGURE 48-7. Dura of internal auditory canal is incised along posterior aspect.
FIGURE 48-8. Transverse crest Bill’s bar Facial n.
Cut ends of superior and inferior vestibular nerves
Tumor
FIGURE 48-9. Superior vestibular n. Inferior vestibular n.
FIGURE 48-8. Intracanalicular acoustic tumor is dissected from facial and cochlear nerves. FIGURE 48-9. Divided vestibular nerves are visible after tumor removal.
include avoidance of vigorous activity and heavy lifting for 6 weeks after surgery. When patients no longer feel dizzy, they may begin driving.
Success Rate In a series of 106 patients over 25 years with tumors removed through the middle fossa approach, hearing was preserved in 59% and preserved near the preoperative
level in 35% of cases.11 Hearing preservation rates have improved owing to technical improvements (wider bony exposure of medial IAC, medial-to-lateral tumor dissection, use of high magnification, and use of topical papaverine)39 and increased experience because of more frequent diagnosis of smaller tumors owing to improved imaging. A more recent series of 333 middle fossa acoustic tumor removals performed over a 7 year period yielded a hearing preservation rate of 80% with hearing preserved near
588
OTOLOGIC SURGERY
the preoperative level (within 15 dB pure tone average and 15% speech discrimination score) in 50%.29 At 1 year, 95% of patients had normal or near-normal facial nerve results (House-Brackmann grade I or II).27
Complications Postoperative complications are uncommon in reported middle fossa series. Postoperative cerebrospinal fluid leakage occurs in 2% to 7% of patients, and meningitis occurs in 2% to 5%.11,27 Not all of the postoperative meningitis cases were culture proven. There was a single patient with an epidural hematoma reported in one series.39 Postoperative seizures have been reported in only two patients from one series.22 Electroencephalographic studies in both of these patients were consistent with an ipsilateral temporal lobe source for the epileptic activity, which was believed to result from temporal lobe retraction. These complications are best avoided by limited temporal lobe retraction of short duration (1 to 1.5 hours).
Patient Selection Pitfalls For the middle fossa approach, the most important limitation in patient selection is tumor size. Because of the limited access to the posterior fossa, this approach is best suited for intracanalicular tumors and tumors with moderate cerebellopontine angle extension. With larger tumors, difficulty in controlling bleeding in the posterior fossa may be encountered.
SUMMARY The middle fossa approach is well suited for the removal of small acoustic tumors, and may preserve hearing. The most appropriate candidates have tumors with moderate extension into the cerebellopontine angle and good preoperative hearing. Measurable postoperative hearing can be preserved in 80% of patients, and normal or nearnormal facial function occurs in 95%. Serious postoperative complications are rare with this approach. With the advent of gadolinium-enhanced MRI, it is now possible to diagnose reliably acoustic tumors when they are small, and before hearing has been significantly affected. The middle fossa approach provides excellent access for the removal of these small tumors.
REFERENCES 1. Parry R H : A case of tinnitus and vertigo treated by division of the auditory nerve. J Laryngol Otol 19:402, 1991. 2. House WF: Surgical exposure of the internal auditory canal and its contents through the middle cranial fossa. Laryngoscope 71:1363, 1961.
3. House WF: Middle cranial fossa approach to the petrous pyramid: Report of 50 cases. Arch Otolaryngol Head Neck Surg 78:460, 1963. 4. House WF, Gardner G, Hughes R L : Middle cranial fossa approach to acoustic tumor surgery. Arch Otolaryngol Head Neck Surg 88:631, 1968. 5. Kurze T, Doyle J B Jr: Extradural intracranial (middle fossa) approach to the internal auditory canal. J Neurosurg 19:1033, 1962. 6. House F, Hitselberger WE : The middle fossa approach for removal of small acoustic tumors. Acta Otolaryngol (Stockh) 67:413, 1969. 7. Wade PJ, House WF: Hearing preservation in patients with acoustic neuromas via the middle fossa approach. Otolaryngol Head Neck Surg 92:184, 1984. 8. Shelton C, Hitselberger WE: The treatment of small acoustic tumors: Now or later? Laryngoscope 101:925, 1991. 9. Brackmann D E : Middle cranial fossa approach. In House WF, Luetje C M (eds): Acoustic Tumors Management. Baltimore, University Park Press, 1979, vol. 2, p 15. 10. Glasscock M E, Poe DS, Johnson G D: Hearing preservation in surgery of cerebellopontine angle tumors. In Fisch U, Valavanis A, Yasargil MG (eds): Neurological Surgery of the Ear and Skullbase. Amsterdam, Kugler & Ghedini, 1989, p 207. 11. Shelton C, Brackmann D E, House WF, et al: Middle fossa acoustic tumor surgery: Results in 106 cases. Laryngoscope 99:405, 1989. 12. Arriaga M A, Luxford WM, Berliner K I : Facial nerve function following middle fossa and translabyrinthine acoustic tumor surgery: A comparison. Am J Otol 15:620624, 1994. 13. Dautheribes M, Migueis A, Vital J M, et al: Anatomical basis of the extended subtemporal approach to the cerebellopontine angle: Its value and limitations. Surg Radiol Anat 11:187, 1989. 14. Rosomoff H L : The subtemporal transtentorial approach to the cerebellopontine angle. Laryngoscope 81:1448, 1971. 15. Wigand M E, Haid T, Berg M : The enlarged middle cranial fossa approach for surgery of the temporal bone and of the cerebellopontine angle. Arch Otorhinolaryngol 246:299, 1989. 16. Bochenek Z, Kukwa A : An extended approach through the middle cranial fossa to the internal auditory meatus and the cerebellopontine angle. Acta Otolaryngol (Stockh) 80:410, 1975. 17. Kanzaki J, Ogawa K, Shiobara R , et al: Hearing preservation in acoustic neuroma surgery and postoperative findings. Acta Otolaryngol (Stockh) 107:474, 1989. 18. Shiobara R, Ohira T, Kanzaki J, et al: A modified extended middle cranial fossa approach for acoustic nerve tumors: Results of 125 operations. J Neurosurg 68:358, 1988. 19. Shelton C, Brackmann D E, House WF, et al: Acoustic tumor surgery: Prognostic factors in hearing conservation. Arch Otolaryngol Head Neck Surg 115:1213, 1989. 20. Gantz B J, Parnes L S, Harker L A, et al: Middle cranial fossa acoustic neuroma excision: Results and complications. Ann Otol Rhinol Laryngol 95:454, 1986. 21. Frerebeau PH, Benezech J, Uziel A, et al: Hearing preservation after acoustic neurinoma operation. Neurosurgery 21:197, 1987.
Chapter 48 • Middle Fossa Approach 22. Glasscock M III, McKennan K X, Levine SC : Acoustic neuroma surgery: The results of hearing conservation surgery. Laryngoscope 97:785, 1987. 23. Josey A F, Glasscock M E, Jackson CG: Preservation of hearing in acoustic tumor surgery: Audiologic indicators. Ann Otol Rhinol Laryngol 97:626, 1988. 24. Nadol J B Jr, Levine R , Ojemann RG, et al: Preservation of hearing in surgical removal of acoustic neuromas of the internal auditory canal and cerebellar pontine angle. Laryngoscope 97:1287, 1987. 25. Mangham C A, Skalabrin TA : Indications for hearing preservation in acoustic tumor surgery. Kona, Hawaii, Presented at the Annual Meeting of the American Neurotology Society, May 4, 1991. 26. Kemink J L , LaRouere M J, Kileny R P, et al: Hearing preservation following suboccipital removal of acoustic neuromas. Laryngoscope 100:597, 1990. 27. Slattery WH, Brackmann D E, Hitselberger WE : Middle fossa approach for hearing preservation with acoustic neuromas. Am J Otol 18:596-601, 1997. 28. Glasscock M E, Woods C I, Jackson CG, et al: Management of bilateral acoustic tumors. Laryngoscope 99:475, 1989. 29. Brackmann D E, Owens R M, Friedman R A, et al: Prognostic factors for hearing preservation in vestibular schwannoma surgery. Am J Otol 21:417-424, 2000. 30. Shelton C, Harnsberger H R , Allen R , King B : Fast spin echo magnetic resonance imaging: Clinical application in screening for acoustic neuroma. Otolaryngol Head Neck Surg 114:71-76, 1996. 31. Dubrulle F, Ernst o, Vincent C, et al: Cochlear fossa enhancement at MR evaluation of vestibular schwannoma: Correlation with success at hearing-preservation surgery. Radiology 215:458-462, 2000.
589
32. Shelton C, House WF: Hearing improvement after acoustic tumor removal. Otolaryngol Head Neck Surg 103:963-965, 1990. 33. Berliner K I, Shelton C, Hitselberger WE, Luxford WM : Acoustic tumors: Effect of surgical removal on tinnitus. Am J Otol 13:13, 1992. 34. Roberson J, Senne A, Brackmann D, et al: Direct cochlear nerve action potentials as an aid to hearing preservation in middle fossa acoustic neuroma resection. Am J Otol 17:653-657, 1996. 35. Kartush J M, Kemink J L , Graham M D: The arcuate eminence: Topographic orientation in middle cranial fossa surgery. Ann Otol Rhinol Laryngol 94:25, 1985. 36. Parisier SC : The middle cranial fossa approach to the internal auditory canal: An anatomical study stressing critical distances between surgical landmarks. Laryngoscope 87(Suppl 4):1, 1977. 37. Garcia-Ibanez E, Garcia-Ibanez J L : Middle fossa vestibular neurectomy: A report of 373 cases. Otolaryngol Head Neck Surg 88:486, 1980. 38. Chen D, Arriaga M, Fukushima T: Technical refinements in retraction for middle fossa surgery. Am J Otol 19:208-211, 1998. 39. Brackmann D E, House J R , Hitselberger WE : Technical modifications to the middle fossa craniotomy approach in removal of acoustic neuromas. Am J Otol 15:614-619, 1994.
49
Translabyrinthine Approach John W. House, James Lin, Rick A. Friedman, and Robert D. Cullen Videos corresponding to this chapter are available online at www.expertconsult.com.
House1,2 first began removing acoustic tumors through the translabyrinthine approach in 1960. Physicians at the House Ear Clinic have been using this approach for most larger acoustic tumor removals. Since 1983, we have performed more than 2700 tumors using this approach. The translabyrinthine procedure allows excellent access to the cerebellopontine angle (CPA) and exposure of the entire facial nerve from the brainstem to the stylomastoid foramen. The approach is extradural through most of the surgery. The primary disadvantage is sacrifice of hearing. Regardless of the size of the tumor, the translabyrinthine approach is ideal for acoustic neuromas when the hearing is poor. It is also ideal for facial nerve lesions, such as neuromas; trauma owing to operative injury; or head trauma. The approach has many advantages. It is the most direct route to the structures of the CPA (Figs. 49-1 and 49-2).3 The lateral end of the internal auditory canal (IAC) can be dissected to ensure complete tumor removal from this area, and to allow consistent anatomic identification of the facial nerve.4 The approach for facial nerve lesions offers exposure of the mastoid, tympanic, and labyrinthine portions of the facial nerve. Identification of the facial nerve in the mastoid is facilitated after removal of the semicircular canals (Fig. 49-3). Because the labyrinth has been removed, the labyrinthine segment of the nerve is readily followed into the IAC. The IAC and CPA can be exposed widely if the lesion involves the facial nerve in the posterior fossa. The facial nerve is readily accessible from the brainstem to the stylomastoid foramen and beyond into the parotid gland. In addition to vestibular schwannoma removal, the translabyrinthine craniotomy approach is used for other tumors (e.g., meningiomas, cholesteatomas involving the petrous bone and posterior fossa, cholesterol granulomas, glomus tumors, and adenomas), for decompression of the facial nerve, and for repair of the facial nerve by either direct end-to-end anastomoses or nerve grafting (Fig. 49-4). For tumors involving the area anterior to the internal auditory nerve at the clivus, the standard translabyrinthine approach is modified to allow anterior exposure (Fig. 49-5). The facial nerve is removed from the fallopian canal in the tympanic and mastoid segments and is
reflected anteriorly. The cochlea is removed to provide excellent exposure anterior to the IAC (see Chapter 52). A major advantage of the translabyrinthine approach is that the patient is in the supine position with the head turned away from the surgeon (Figs. 49-6 and 49-7). This position eliminates some of the possible complications of the classic suboccipital approach to the CPA in which the sitting position is used, such as risks of air embolism and injury to the cerebellum from retraction. Quadriplegia has been reported in association with the sitting position.5 The translabyrinthine approach poses diminished danger of air embolism and requires less retraction of the cerebellum. Much of the surgery is extradural, lessening the chances of injury to the brain. The extradural dissection also greatly decreases the seeding of bone dust into the subarachnoid space.
PATIENT SELECTION Tumor size and residual hearing are the principal factors influencing the choice of the translabyrinthine approach. We use the translabyrinthine approach in most cases if the tumor is larger than 2.5 cm in maximal dimension, and in all cases of nonserviceable hearing in the involved ear. When tumors are confined to the IAC in an ear with serviceable hearing, we use the middle cranial fossa approach in an attempt to save hearing. For patients with good hearing and tumors primarily based in the CPA with minimal extension into the IAC, the retrosigmoid approach is chosen. Our definition of serviceable hearing is a pure tone average threshold better than 50 dB, a speech discrimination score of greater than 50%, or both. This definition is referred to as the 50/50 rule. Exceptions exist if the hearing in the contralateral ear is poor, or if bilateral tumors are present. In such cases, we may attempt tumor removal through the middle fossa approach, even if the tumor extends 1 cm into the CPA. We may use the retrosigmoid approach in an attempt to save hearing if the tumor is smaller than 1.5 to 2 cm and does not extend into the lateral half of the IAC. 591
592
OTOLOGIC SURGERY
FIGURE 49-1. MR image of large acoustic neuroma, illustrating direct route to cerebellopontine angle through mastoid and labyrinth. FIGURE 49-2. Postoperative CT scan of patient shown in Figure 49-1. Note extent of removal of bone from mastoid and labyrinth. FIGURE 49-3. Facial nerve identified in mastoid portion after removal of semicircular canals. Note island of bone over sigmoid sinus. FIGURE 49-4. Facial nerve graft through translabyrinthine approach. Greater auricular nerve graft has been placed from internal auditory canal (IAC) to proximal mastoid facial nerve.
Chapter 49 • Translabyrinthine Approach
SURGICAL PROCEDURE The procedure is performed with general endotracheal anesthesia with inhalation agents. Muscle relaxants are used only for the induction because they may interfere with facial nerve monitoring. An orogastric tube and Foley catheter are placed. Antibiotics with good cerebrospinal fluid (CSF) penetration are given intravenously before skin incision. Cefuroxime is our antibiotic of choice in non–penicillin-allergic patients, whereas vancomycin is used in patients with penicillin allergy. These intravenous antibiotics are maintained for 24 hours perioperatively. A generous amount of hair is shaved from the postauricular and temporal areas. The skin is cleaned with povidone-iodine (Betadine), and a sheet of Ioban is placed to cover the entire area. Because facial nerve monitoring is routine during all of our translabyrinthine procedures, needle electrodes are inserted into the orbicularis oris muscles before the drape is applied. The lower abdomen is also prepared and draped with Ioban to allow for the harvesting of fat; the navel is prepared in the field for orientation. Lidocaine 1% (Xylocaine) with epinephrine 1:100,000 is injected into the postauricular region. The epinephrine assists with homeostasis. The incision is performed about 2 to 4 cm posterior to the postauricular sulcus. Generally, the larger the tumor, the more posterior an incision is required, to allow for extended decompression of the dura posterior to the sigmoid sinus. The incision is curved anteriorly to allow anterior retraction of the pinna. The posterior curve of the incision allows exposure of the area posterior to the sigmoid sinus. This exposure is important to allow access to the CPA. A scalp flap just superficial to the temporalis fascia and mastoid periosteum is developed anterior and posterior to the skin incision. The incision through the next layer is brought down to the bone and is typically staggered either anterior or posterior to the skin incision to decrease risk of CSF wound leak after closure. The Lempert elevator is used to elevate the periosteum off the bone of the mastoid. Soft tissue must be removed from the posterior edge of the external auditory canal to an area far posterior to the sigmoid sinus. Care must be taken not to tear the skin of the external auditory canal; otherwise, CSF otorrhea may develop postoperatively. Self-retaining retractors are placed to expose the mastoid and cranium for extensive bony dissection. At this point, muscle may be harvested for packing of the eustachian tube and epitympanic space. Temporalis fascia may also be taken for support of the dural closure. A complete mastoidectomy is performed with a highspeed drill with various sizes of cutting and diamond burrs. Removing bone 2 cm posterior to the sigmoid sinus is crucial for adequate exposure of the dura of the posterior cranial fossa. A small island of bone (Bill’s island— named for William F. House, who first suggested it) is left over the otherwise exposed sigmoid sinus (Fig. 49-8).
593
This bony cover protects the sigmoid sinus from the shaft of the burr as the drilling proceeds medially to remove the labyrinth. The dissection continues with the removal of all bone covering the posterior fossa dura medial to the sigmoid sinus and down to the labyrinth. It is important to remove all bone over the sinal dural angle and a small amount of bone over the middle fossa dura adjacent to the angle. In larger tumors or contracted mastoids, we recommend removal of 2 to 3 cm of bone from temporal squama. We prefer to perform all of the lateral bone work before beginning the labyrinthectomy to obtain better exposure of the deep structures. After the complete mastoidectomy is performed, and the bone is removed from posterior fossa dura, sigmoid sinus, and some of the middle fossa dura, the deepest point of dissection shifts to the sinal-dural angle. The labyrinthe ctomy begins with the removal of the lateral semicircular canal and extends posterior to the posterior canal. The bone removal is continued inferior and anterior toward the ampullated end of this canal (Fig. 49-9). The ampulla of the posterior canal is the landmark for the inferior border of the IAC. The inferior extent of bony removal is the jugular bulb. The posterior semicircular canal is opened inferiorly to the vestibule and superiorly to the common crus. The facial nerve is identified in its descending portion in the mastoid and skeletonized to just proximal to the stylomastoid foramen. We prefer to identify the facial nerve after the posterior semicircular canal has been removed to use the side of the diamond burr rather than the end of it; this helps reduce the possibility of injury to the nerve. The mastoid segment of the facial nerve serves as the anterior limit of dissection, and it is important to remove as much of its overlying bone as possible to maximize exposure for tumor dissection. With this portion of the facial nerve identified, the remainder of the bone of the inferior IAC is removed to the vestibule. After opening the vestibule widely, the removal of the superior portion (nonampullated end) of the posterior canal is carried to the common crus, which is composed of the nonampullated ends of the posterior and superior semicircular canal. The common crus is opened to the vestibule. The superior canal is now opened and removed to its ampullated end in the vestibule. This portion of the superior canal identifies the area where the superior vestibular nerve exits the lateral end of the IAC and is in close proximity to the labyrinthine segment of the facial nerve. Similarly, the singular nerve exits the IAC at the posterior semicircular canal ampulla, and the inferior vestibular nerve exits the canal at the saccule and the spherical recess. Identification of these structures delineates the superior and inferior extent of the IAC. As the bone posterior to the IAC is removed, the vestibular aqueduct and the beginning of the endolymphatic sac are removed. An eggshell thickness of bone is left over the dura of the IAC to avoid injury to the underlying structures until all of the bony dissection is completed (Fig. 49-10).
594
OTOLOGIC SURGERY
Chapter 49 • Translabyrinthine Approach
595
FIGURE 49-5. Meningioma involving internal auditory canal and extending anterior to clivus can be removed through translabyrinthine/ transcochlear approach.
FIGURE 49-6. Patient placed in supine position with head turned away from surgeon. Anesthesiologist is at foot of table. FIGURE 49-7. Surgeon is seated with operating microscope. Nurse is across table. Facial nerve monitor is at lower left.
The IAC is not entered at this time because bone must be removed medially to the porus acusticus. The IAC runs deep from the vestibule and away from the surgeon. A great deal of bone must be removed to expose the contents of the IAC and the CPA properly. Bone is removed around the canal superiorly and inferiorly to expose at least 270 degrees in circumference. The inferior limit of bone removal is the cochlear aqueduct and the jugular bulb. The cochlear aqueduct enters the posterior fossa directly inferior to the midportion of the IAC, superior to the jugular bulb (Fig. 49-11).6 It identifies the location of the neural compartment of the jugular foramen anterior to the jugular bulb. By not removing bone from anterior and deep to the cochlear aqueduct, injury to CN IX, X, and XI is avoided. Bone is removed from the inferior portion of the IAC and particularly the inferior lip, affording access to the inferior poll of the tumor in the CPA. The bone is removed from the superolateral IAC last because of its close proximity to the facial nerve. Bone should be removed from the superior lip of the IAC. This dissection is tedious because the facial nerve often underlies the dura along the anterosuperior aspect of the IAC. The surgeon must be careful not to allow the burr to drop into the canal and possibly injure the nerve. As with the inferior lip, all of the bone must be removed from the superior lip to allow access to the superior pole of the tumor. With the superior lip, the facial nerve may be close to the surface, making this part of the removal laborious. The facial nerve is identified as it exits the lateral end of the IAC at the vertical crest of bone (Bill’s bar) with a sharp 3 mm hook. The facial nerve may be identified further in its proximal labyrinthine segment by additional bone removal. The hook is passed carefully along the inside of the superior distal IAC until Bill’s bar is palpated (Fig. 49-12). It is not unusual for the facial nerve monitor to sound a warning as the hook passes along the nerve at the proximal portion of the fallopian canal. All of the dissection so far has been extradural, and morbidity should be minimal. When the facial nerve has been positively identified, the dura of the posterior fossa over the midportion of the IAC is opened with sharp scissors. The length of the incision depends on the size of the tumor. For smaller tumors and nerve sections, the incision is made close to the IAC. For larger tumors, it is started closer to the sigmoid sinus. The incision extends to the IAC, and then curves superiorly and inferiorly around the porus acusticus. The surgeon must take care to avoid blood vessels on the surface of the tumor and adjacent to the dura. Posteriorly, the petrosal vein lies
close to the dura. The IAC is opened over the inferior vestibular nerve and reflected superiorly to avoid injury to the facial nerve. Cottonoids are placed posteriorly between the tumor and the cerebellum. It is important to develop this plane accurately because doing so separates the major vessels of the CPA from the tumor. With larger tumors, the size of the tumor is reduced by the use of the House-Urban (Fig. 49-13) or ultrasonic dissector. The surface of the tumor is carefully inspected first to identify nerves. Occasionally, the facial nerve is deflected posterior. The tumor capsule is incised, and the dissector is inserted to begin the intracapsular removal of the bulk of the tumor. Excessive manipulation of the tumor must be avoided to prevent traction of the facial nerve. The capsule of the tumor is collapsed toward its center, greatly facilitating its dissection from the CPA. The tumor is followed medially to the brainstem. The plane between the tumor and the brainstem is developed with sharp and blunt dissection. Cottonoids are placed between the brainstem and the tumor to protect the underlying structures. At this point, attempts are made to identify the facial nerve superiorly. It is usually anterior to the tumor, but may be draped over the top of it. CN IX is identified inferiorly. In large tumors, CN IX, X, and XI may be stretched over the inferior surface of the tumor. Manipulation of the tumor may cause a change in the pulse rate or blood pressure. During this phase of the tumor removal, use of a fenestrated neurotologic suction tip helps avoid injury to surrounding structures.7 The vestibular nerve and tumor are separated from the facial nerve in the IAC and carefully dissected medially to the porus acusticus and into the CPA (Fig. 49-14). Some tumors involve the lateral end of the IAC, complicating identification of the facial nerve in the canal. In these cases, bone is removed from the proximal fallopian canal to allow positive identification of the facial nerve where it is not involved with tumor. This maneuver greatly reduces injury to the facial nerve. It may be necessary to identify the facial nerve at the brainstem and to begin to separate the tumor from the facial nerve medially to laterally. Careful, patient dissection results in complete separation of the facial nerve from the tumor as the tumor is dissected out of the posterior fossa. Continuous intraoperative facial nerve monitoring has greatly facilitated this process. It has been shown that use of scissors to free the nerve carefully from the tumor causes less trauma than the use of blunt dissection to establish this plane. As the tumor is dissected free, bleeding is controlled with bipolar cautery. Only the vessels that enter the tumor
596
OTOLOGIC SURGERY
Chapter 49 • Translabyrinthine Approach
597
FIGURE 49-8. Mastoidectomy with exposure of sigmoid sinus. Small bony island is left over sinus to protect it from shaft of burr. FIGURE 49-9. Completed labyrinthectomy with ampullated ends of posterior and superior semicircular canals. FIGURE 49-10. All bone has been removed, exposing dura of posterior fossa and internal auditory canal (IAC). FIGURE 49-11. Bone has been removed from internal auditory canal and posterior fossa. Cochlear aqueduct can be seen inferior to internal auditory canal and superior to jugular bulb.
capsule are coagulated. The other vessels are freed from the tumor capsule. A small blood vessel may accompany CN VIII. As the nerve is cut, control of bleeding with bipolar cautery may be necessary. Careful control of bleeding produces minimal blood loss. With an average blood loss of about 250 mL, our patients rarely require transfusions. We offer our patients the opportunity to withdraw 1 to 2 U of their own blood 1 month before the scheduled surgery. During the drilling, the wound is periodically irrigated with a solution containing bacitracin to reduce the chance of infection. After the drilling has been completed, the bacitracin irrigant is attached to the suction-irrigator and used during the dissection of the tumor. This solution cannot be used during drilling because it tends to produce foam. We have reduced the rate of meningitis in our patients by using the bacitracin irrigation and administering perioperative antibiotics. Closure involves the use of abdominal fat and, if the opening is large, a partial closure of the dura. Fat is obtained from the lower abdomen through a small transverse incision. The wound is closed in layers with poly glactin 910 (Vicryl) sutures, and a Penrose drain is inserted. Steri-Strips are used to reinforce the skin closure. Obliteration of “dead space” with deep sutures is important in reducing the risk of an abdominal wall hematoma. The fat is cut into strips and soaked in the bacitracin irrigant. The dura is closed with 4-0 silk along the posterior fossa incision. This closure may be reinforced with temporalis fascia or DuraGen dural graft matrix. The strips of fat are inserted through the dural opening and the IAC. These strips are tightly packed into the defect expanding on both sides of the dura to prevent leakage. Muscle is packed into the attic. The incus may be removed to pack the eustachian tube and the middle ear with muscle. The mastoid is also filled with fat. Since 2003, we have begun performing a cranioplasty with titanium mesh overlying the fat. The wound is closed in layers with 0 chromic and 3-0 Vicryl. SteriStrips are applied to the scalp incision. A head dressing and abdominal pressure dressings are applied.
Management of the Contracted Mastoid There are few absolute contraindications to the use of the translabyrinthine approach in acoustic tumor removal. One absolute contraindication is active infection in the affected ear. In addition, there has been controversy regarding the use of this approach in anatomically
c onstricted mastoids. A low-lying tegmen, an anterior sigmoid sinus, and a high jugular bulb are individually or collectively considered contraindications to this approach. Using several crucial maneuvers, we have never altered our approach to the CPA based on anatomic variations.8 Wide removal of the bone of the temporal squama and presigmoid and postsigmoid posterior fossa dura overcomes the limitations imposed by a low-lying tegmen and anterior-placed sigmoid sinus. For a high jugular bulb, we recommend skeletonization along its anteromedial, medial, and posterior surfaces without compression. This skeletonization, combined with the ability to retract the widely exposed temporal lobe dura superiorly, provides an improved line of sight in the deep field of the CPA.
POSTOPERATIVE CARE The abdominal drain is removed the following day. We no longer use a nasogastric tube and remove the orogastric tube at the time of extubation. In addition, the urinary catheter inserted at the beginning of the procedure is removed on postoperative day 1 or 2. The head dressing remains for 3 days, and the Steri-Strips remain for 1 week. The patient stays in the intensive care unit usually for 1 day after surgery and stays in the hospital for 3 to 5 days. Progressive ambulation begins the day after surgery. The patient sits on the side of the bed, stands by the bed, and is encouraged to sit in a chair. We believe that early ambulation helps reduce complications and speeds recovery. Vital signs and temperature are monitored frequently during the first 2 to 3 days after surgery. In addition, the nurses are instructed to observe the patient for possible CSF leak and neurologic changes. If the patient has drainage from the nose, the nurses test it for glucose and alert the physician.
COMPLICATIONS We performed a review of complications and facial nerve outcomes on 512 consecutive patients undergoing translabyrinthine craniotomy for unilateral, sporadic acoustic neuromas. In this section, we quantify many complication rates based on this review.
598
OTOLOGIC SURGERY
Figure 49-12. Internal auditory canal (IAC) is skeletonized, and bone is removed to expose 180 to 240 degrees of canal. Facial nerve is identified as it exits IAC at Bill’s bar.
Figure 49-13. House-Urban dissector is used for intracapsular removal of tumor bulk. Figure 49-14. Vestibular nerve and tumor are separated from facial nerve in internal auditory canal and dissected to porus acusticus and cerebellopontine angle.
Facial Nerve Facial nerve weakness or paralysis is not a complication, but a risk that cannot be entirely eliminated. The primary concern of most patients undergoing tumor removal is the ultimate facial nerve result. Continuous intraoperative facial nerve monitoring and meticulous dissection of the tumor from the facial nerve usually yield good
facial nerve results. Eighty percent of our tumor patients are House-Brackmann grade I or II. Only about 5% of patients have a grade VI outcome. If the facial nerve is intimately involved with the tumor, or if the tumor is a facial neuroma, preservation of the anatomic continuity of the facial nerve may be impossible. We believe that repairing the nerve during the initial surgery is important. With the
Chapter 49 • Translabyrinthine Approach
translabyrinthine approach, the nerve is identified in the labyrinthine segment, and exposed in the tympanic and mastoid portions, and either the nerve is rerouted or an interposed graft is sutured to the proximal and distal portions of the nerve. We currently use NeuraGen tubules to aid in our nerve anastomoses. Suturing the nerve in the CPA is difficult; if this is possible, usually only one suture can be placed. The result allows for normal facial tone and good emotional and voluntary motion. Mass action or synkinesis is always present. Grade III is the best result that can be expected. When a primary anastomosis or nerve graft is impossible, or if facial nerve function does not return after 1 year, we perform a facial-hypoglossal (CN VII-XII) anastomosis. It gives good resting tone, fair voluntary motion with synkinesis, and usually a grade IV recovery. If the nerve has been paralyzed for several years, and the patient has poor tone, we combine the CN VII-XII anastomosis with a temporalis muscle transposition to the orbicularis oris. This procedure gives an immediate cosmetic improvement to the face at rest and gradual return of voluntary motion over 6 to 12 months. Our review of patients from 2000 to 2004 revealed anatomic preservation of the facial nerve in 97.5%. Twelve of the 512 patients had a severed facial nerve; the average tumor size in those patients was 3.7 cm. Six of these patients underwent primary anastomosis of the nerve, whereas three patients underwent greater auricular nerve grafting at the time of tumor removal. Of the six patients who underwent primary anastomosis, five obtained a House-Brackmann grade of III or IV, whereas the remaining one patient retained a grade of VI. Only one of the patients who underwent cable-grafting obtained a House-Brackmann grade of III, whereas one remained completely paralyzed, and the third was lost to followup. Of the remaining three patients who did not undergo immediate facial nerve reconstruction, two underwent hypoglossal-facial anastomosis in the early postoperative period, and one elected for no reconstruction. One patient who underwent hypoglossal-facial anastomosis achieved a grade IV, and the other patient was lost to follow-up.
Bleeding The most dramatic and potentially fatal complication is an early postoperative hematoma in the CPA. This complication is manifested by signs of increased central nervous system pressure, such as loss of consciousness and nonreactive pupils. It is managed by immediate opening of the wound and removal of the fat while the patient is in the intensive care unit. An advantage of the translabyrinthine approach is that the angle may be rapidly decompressed for this uncommon complication. The patient is taken back to surgery, and the bleeder is identified and controlled. In the 512 patients reviewed from 2000-2004, there were four subdural hematomas (0.8%) and three CPA clots (0.6%).
599
Bleeding is part of the procedure. The most dramatic bleeding occurs if the sigmoid sinus is entered. Opening of the sinus produces profuse bleeding; because the bleeding is venous, it is easy to stop with light pressure over the sinus. Bleeding is controlled with extraluminal packing with absorbable knitted fabric (Surgicel), thrombin-soaked absorbable gelatin sponge (Gelfoam), or microfibrillar collagen (Avitene). Great care is taken to prevent the packing from entering the lumen of the sinus. If this occurs, the packing then enters the pulmonary circulation, resulting in a pulmonary embolism. If the lumen of the jugular bulb is opened, the jugular vein is ligated in the neck, and the bulb is packed to control bleeding. This complication is extremely rare. Arterial bleeding is seldom a problem. Great care is taken to identify and avoid the anterior cerebellar artery because thrombosis or injury to this artery can be fatal. This injury is extremely rare with modern microsurgical techniques.
Cerebrospinal Fluid Leak Before 1974, we used temporalis muscle to close the dura, and our incidence of CSF leak was 20%.2 Since we began using abdominal fat instead, this incidence has been reduced to less than 7% in translabyrinthine acoustic tumor removals.9 Since we began using titanium mesh for cranioplasty in mid-2003, our incidence of CSF rhinorrhea has been reduced to 3.3%. Most leaks can be stopped with a pressure head dressing and bed rest with the patient’s head elevated. Typically, we leave the dressing in place for 3 days. If the CSF leak continues with the dressing in place, or if it recurs when the dressing is removed, a lumbar spinal drain is inserted, and the patient is placed at bed rest for 3 to 4 days. This conservative management controls CSF rhinorrhea in two thirds of patients. In the remaining one third of cases with persistent leak through the nose, the patient is taken back to surgery, and the wound is reopened. The ear canal is oversewn, and the epithelium medial to the overclosure is removed completely followed by wide canalplasty and packing of the eustachian tube. In cases of wound leak, the wound may be oversewn in addition to pressure dressing. If this fails, the wound is reopened and reclosed in layers occasionally with additional fat placement. A wound exploration typically requires harvesting of additional fat.
Meningitis The incidence of meningitis has been declining with the use of bacitracin irrigation and reduced surgical time. The review of 512 patients operated on from 2000-2004 revealed postoperative meningitis in 3 patients (0.6%). A spinal tap is performed when the patient has a fever, elevated white blood cell count, and a stiff neck. If the CSF white blood cell count is greater than 100 wbc’s/ml,
600
OTOLOGIC SURGERY
we use high doses of intravenous antibiotics. In addition to the cell count, the spinal fluid is cultured for anaerobic and aerobic bacteria, and total protein and glucose levels are measured. In typical patients with meningitis, the CSF protein level is elevated, and the glucose level is reduced. Antibiotics are altered if the cultures so indicate. The usual response to treatment is reduction in fever, decrease in white blood cell count, and reduction in headache. The spinal tap is repeated on the 4th or 5th treatment day. Treatment is continued until the CSF white blood cell count is less than 100 wbc’s/ml, and the CSF is composed mostly of monocytes.
RESULTS Greater than 5000 acoustic neuromas have been removed at the House Ear Clinic. With experience and refinements of technique, the results have progressively improved. A review from our database of acoustic tumor cases provides data from 1302 patients who underwent a translabyrinthine acoustic tumor removal between 1982 and 1993. The mean age of patients was 50 years; 46% were men, and 54% were women. Tumor size ranged from 0.5 to 6.5 cm, with a mean size of 2.4 cm. Operating time averaged 3.3 hours. Three (0.2%) deaths occurred in this series. Previously acquired data on long-term (6 month) facial nerve function as determined by the House-Brackmann scale were available on 889 cases, with a mean follow-up time of 2.1 years. Of these, 58.2% had a grade I function; 12.6%, grade II; 13.2%, grade III; 7.8%, grade IV; 3.3%, grade V; and 5.1%, grade VI. For cases undergoing surgery since the advent of facial nerve monitoring in 1988 and with at least 1 year follow-up, 59% of the 312 patients had grade I facial nerve function; 15.4%, grade II; 9.3%, grade III; 7.7%, grade IV; 4.2%, grade V; and 4.5%, grade VI. In our review of 512 patients who underwent translabyrinthine removal of unilateral, sporadic acoustic neuromas between 2000-2004, the mean age was 49 years; 49% were men, and 51% were women. The average tumor size was 2.4 cm with a range of 0.5 to 5 cm. Of these 512 patients, 392 (77%) were examined at 1 year or returned a validated facial nerve questionnaire. Of these patients, 67.6% had grade I facial nerve function at 1 year; 13.3%, grade II; 8.2%, grade III; 7.1%, grade IV; 0.8%, grade V; and 3.1%, grade VI. This study also detected a decrease in facial outcome with larger acoustic neuromas, particularly greater than 3.5 cm in size. For these patients with “giant” vestibular schwannomas, a House-Brackmann grade of I or II at 1 year postoperatively could be achieved in only about 50%.
Postoperative Follow-up In our experience, vestibular schwannomas rarely recur after translabyrinthine removal. Our recurrence rate for unilateral tumors removed through the translabyrinthine
approach treated between 1961 and 1995 was 0.3%.10 The average interval to recurrence was 10 years. Based on these findings, we have recommended a single gadolinium-enhanced magnetic resonance imaging (MRI) study 5 years postoperatively.
SUMMARY The translabyrinthine approach to tumors involving the temporal bone and CPA offers excellent and safe exposure. We have used this approach for more than 40 years to remove greater than 3000 tumors. Our experience has yielded many refinements of the original procedure, and we continue to seek improvements to reduce the morbidity. Mortality now approaches 0%. We continue to believe that the translabyrinthine approach is the approach of choice for most large acoustic neuromas and all tumors when preoperative hearing is poor or unlikely to be preserved.
EDITORIAL COMMENT Neurotologic approaches to cranial base tumors are a team endeavor in which the neurotologist and the neurosurgeon must be fully familiar with the other members’ techniques. A frequent comment by former clinical fellows after completing training at the House Ear Clinic is that differences exist in the removal of acoustic tumors, especially larger tumors, between neurosurgeons familiar with the translabyrinthine approach and surgeons who are adjusting to this anterior exposure. The editors have invited William E. Hitselberger and Marc Schwartz to provide the following neurosurgical perspective on tumor removal through the translabyrinthine approach. With a personal experience of more than 6000 acoustic tumors removed with this technique, Dr. Hitselberger has insights that are particularly relevant for neurotologic teams in the early phases of collaboration.
NEUROSURGICAL TECHNIQUES IN ACOUSTIC TUMOR SURGERY The strategy for the removal of an acoustic neuroma depends on whether the tumor is large or small. The techniques used in each of these situations, although similar, vary enough in important details that differences are described and emphasized here. The slight variation in technique is necessary because of the variation in the difficulty in preserving neurologic structures in each of these situations. Although variations exist in the technique used, the size of an acoustic neuroma is not a limiting factor in the choice of the translabyrinthine approach. In 30 years and more than 3000 acoustic neuromas,
Chapter 49 • Translabyrinthine Approach
I Units have never found a tumor that was too large to take out through this approach.
Removal of Small Acoustic Tumor In the removal of a small or a large acoustic neuroma through the translabyrinthine approach, the importance of exposure cannot be overemphasized. Adequate exposure of the tumor is the sine qua non of the procedure. Bone removal should include the bone over the middle fossa dura, the sigmoid sinus should be skeletonized so that it can be easily compressed, and bone removal should extend down to the jugular bulb. The posterior fossa dura should be cleared and easily retractable. The IAC should be skeletonized for at least 180 degrees from the posterior presentation. The labyrinthine portion of the facial nerve should be identified and uncovered. During the drill-out of the IAC, the rotation of the drill is important because the facial nerve is near the surface of the canal and exposed to a greater degree when the tumor is small than when it is larger. In a larger tumor, the tumor usually acts as a buffer between the facial nerve and the drill. If the drill is rotating into the IAC, the bit can catch on the bony edge, and the fast-moving burr may injure the nerve. The dura should be opened over the end of the IAC, and the facial nerve should be positively identified visually and with the facial nerve stimulator. The nerve at the end of the canal is anterior and superior to the superior vestibular nerve. The plane between these nerves can be readily developed if the bony dissection of the IAC has been completed at its lateral extent. After division of the facial-vestibular anastomosis, the plane between the superior vestibular nerve and the facial nerve leads the surgeon into the plane between the facial nerve and the tumor. This latter plane is carried medially down to the porus acusticus. At this point, the nerve may be bound down in adhesions and tumor. High magnification (×40) may be necessary to keep the facial nerve plane delineated from the adhesions and tumor. Facial nerve dissection may be facilitated if the nerve is identified at the brainstem. The facial nerve arises anterior and medial to the cochleovestibular nerve. The anterior inferior cerebellar artery usually crosses between CN VII and VIII near the brainstem. The facial nerve is generally quite distinct and has a whiter color than CN VIII because of a heavier concentration of myelin. Developing the facial nerve plane from medial to lateral leads to the medial extent of the tumor. Sometimes, the tumor “shells out” from the bed of CN VIII. More important, the continuing plane of the facial nerve can be developed back to the porus. The facial nerve plane is developed from the medial brainstem side and the lateral IAC side. Usually, when the facial nerve has been cleared from the surface, the tumor can be easily delivered. The importance of facial nerve dissection lies in the
601
reservation of the facial nerve and in the delineation and p ultimate removal of the tumor. Blood vessels lying on the surface of the tumor should be dissected free without coagulation, if possible. This method preserves the blood supply to the adjacent brainstem and the facial nerve. A very fine bayonet forceps and scissors can be used to free the vessels from the surface of the tumor without coagulation. After the tumor has been removed, the tumor bed is carefully evaluated for any small bleeders. Bipolar coagulation set at a low level and oxidized cellulose are generally sufficient to treat such bleeding.
Removal of Large Acoustic Neuroma The bony removal for a large acoustic neuroma, although similar to that for a small neuroma, varies enough that important variations should be mentioned. Wide, adequate exposure is the key, and is accomplished by thorough removal of bone over the middle fossa, posterior fossa, and IAC. Additionally, bone should be removed for at least 1 cm posterior to the sigmoid sinus over the subocciput. Even with a contracted mastoid, adequate exposure can be obtained for removal of any-sized neuroma if the bone removal has been exploited to the maximum. Another key for removal of a large acoustic neuroma from the cerebellopontine angle is extradural retraction. If the bone removal is inadequate, this retraction against the residual bone is impossible. Bony overhangs at the external genu of the facial nerve and the overhang of the posterior external auditory canal should also be removed. The facial nerve is identified at the end of the IAC, as in a smaller tumor. This dissection can be more difficult if the end of the canal is distorted by an impacted tumor, which is occasionally invasive into the otic capsule. After identification of the facial nerve at the end of the IAC, the dura is opened over the posterior fossa anterior to the sigmoid sinus. The tumor may bulge into the dural opening. A rapid decompression of the interior of the tumor can be carried out with the House-Urban rotary dissector, starting from the posterosuperior compartment of the tumor adjacent to the tentorium. At this point, who am not concerned with moderate venous bleeding; rather, the goal is rapid debulking of the tumor. Coagulating small bleeding vessels is unnecessary; these stop bleeding when the tumor is removed. After the tumor has been decompressed, the cisterna lateralis is emptied of CSF. This step adds further to the available space and allows greater room for retraction. The position of the facial nerve should be ascertained with the facial nerve monitor before radical resection of the tumor is undertaken. Usually, the facial nerve is anterior to the tumor, but it may be superiorly placed. Rarely, it can even be in a posterior position, which is an especially dangerous position for the facial nerve because the surgeon must operate past the facial nerve to remove the tumor, subjecting it to increased risk.
602
OTOLOGIC SURGERY
After decompression of the tumor, the facial nerve dissection can be started. High magnification (×40) and sharp dissection are best suited for this process. The nerve is usually stretched and attenuated by the tumor. The facial nerve monitor is invaluable in delineating the nerve when it has been thinned out over the surface of the tumor. The facial nerve dissection proceeds from the medial and lateral ends of the tumor. Vascular radicles are removed from the surface of the tumor, avoiding coagulation whenever possible. A large branch of the petrosal vein is usually positioned on the posterior medial surface of the tumor and should be sought after, identified, and dealt with, either with preservation, if easily accomplished, or with bipolar coagulation. After tumor removal, all bleeding points in the tumor bed should be controlled with bipolar coagulation. The smallest bipolar coagulating tips possible should be used to avoid excessive heat and facial nerve injury.
REFERENCES 1. House WF: Acoustic neuroma [Monograph]. Arch Otolaryngol Head Neck Surg 80:598-757, 1964. 2. House WF: Translabyrinthine approach. In House WF, Luetje C M (eds): Acoustic Tumors Management. Baltimore, University Park Press, 1979, vol. 2, pp 43-87.
3. Brackmann D E : Translabyrinthine removal of acoustic neurinomas. In Brackmann D E (ed): Neurological Surgery of the Ear and Skull Base. New York, Raven Press, 1982, pp 235-241. 4. House WF, Leutje C M (eds): Acoustic Tumors Management. ������������������������������������������������������������������������ vol. 1. �������������������������������������������������������������� Baltimore, University Park Press, 1979. 5. Hitselberger WE, House WF: A warning regarding the sitting position for acoustic tumor surgery [Editorial]. Arch Otolaryngol Head Neck Surg 106:69, 1980. 6. Brackmann D E, Green D: Translabyrinthine approach for acoustic tumor removal. Otolaryngol Clin North Am 25:311-329, 1992. 7. Brackmann D E : Fenestrated suction for neuro-otologic surgery. Trans Am Acad Ophthalmol Otolaryngol 84:975, 1977. 8. Friedman R A, Brackmann D E, van Loveren H R , Hitselberger WE : Management of the contracted mastoid in the translabyrinthine removal of acoustic neuroma. Arch Otolaryngol Head Neck Surg 123:342-344, 1997. 9. Rodgers G K, Luxford WM : Factors affecting the development of cerebrospinal fluid leak and meningitis after translabyrinthine acoustic tumor surgery. Laryngoscope 103:959-962, 1993. 10. Shelton C : Unilateral acoustic tumors: How often do they recur after translabyrinthine removal? Laryngoscope 105:958-966, 1995. 11. Brackmann D E, Cullen R D, Fisher L M : Facial nerve function after translabyrinthine vestibular schwannoma surgery. Otolaryngol Head Neck Surg 136:773-777, 2007.
50
Retrosigmoid Approach to Tumors of the Cerebellopontine Angle Robert K. Jackler and David W. Sim Videos corresponding to this chapter are available online at www.expertconsult.com.
The retrosigmoid approach is a versatile type of craniotomy that creates a panoramic view of the posterior fossa from the tentorium cerebelli to the foramen magnum. Indications for the retrosigmoid approach include (1) resection of extra-axial lesions, such as schwannoma, meningioma, and epidermoid; (2) cranial nerve neurectomy (e.g., CN V, VIII, and IX); (3) vascular decompression of cranial nerves (e.g., CN V, VII, and IX); (4) vascular disorders of the vertebrobasilar system; and (5) parenchymal lesions of the brainstem and cerebellum. The primary advantages of the retrosigmoid ap proach are the potential for hearing preservation and an unhindered exposure of the inferior portion of the cerebellopontine angle (CPA). The principal disadvantages are a substantially higher incidence of persistent postoperative headache and a higher incidence of cerebrospinal fluid (CSF) leaks compared with transtemporal approaches. Although the retrosigmoid approach is technically capable of addressing most lesions involving the CPA, it is best used selectively to gain optimal benefit from its advantages, while avoiding its occasional disadvantages. This chapter concentrates on the use of the retrosigmoid approach for tumors of the CPA, with an emphasis on acoustic neuroma resection.
SURGICAL ANATOMY Historically, the earliest approach to the posterior fossa was undertaken through the suboccipital convexity. Krause1 first employed this technique during the latter portion of the 19th century. Until the 1970s, the technique in widespread use was the so-called suboccipital approach. In this procedure, a large bone window is removed, and the anterior limit of the craniectomy is the first mastoid air cell encountered. Curtailment of the anterior opening at the first contact with pneumatization was predicated on the assumption that the mastoid was bacterially contaminated, and that opening its air cell tracts created an increased risk of meningitis. Because of its more posterior angle of view, the suboccipital approach
required a greater degree of cerebellar retraction, and sometimes necessitated a partial cerebellar resection. In recent years, as a result of increased experience with CPA surgery, the classic suboccipital approach has been modified to become the retrosigmoid approach, which is now the preferred method for exposing the CPA behind the sigmoid sinus. In this technique, bone is removed anteriorly up to the level of the posterior border of the sigmoid sinus and superiorly to the inferior margin of the transverse sinus (Fig. 50-1). Although mastoid air cells are frequently transected during this maneuver, experience has not shown an increased incidence of postoperative infection. The slightly higher risk of CSF leak associated with this more anterior exposure is more than offset by its more favorable angle of view into the CPA and the markedly reduced need for cerebellar retraction with this approach. The anatomic exposure of the posterior fossa provided by the retrosigmoid approach is bounded superiorly by the tentorium cerebelli and inferiorly by the jugular foramen and foramen magnum (Fig. 50-2).2-5 Access to the central nervous system includes the lateral cerebellar hemisphere and the lateral surface of the pons and upper medulla. CN V through XI are visible at their root entry zones and over their cisternal courses. Although the theoretical anterior limit of exposure is the clivus and the apical portion of the petrous pyramid, in practice, access to these ventral structures is usually limited by CN VII and VIII superiorly and CN IX through XI inferiorly, which bridge across the CPA, restricting ventral access to narrow intervals. Exposure of the prepontine cistern is largely obstructed by the lateral aspect of the pons, which does not tolerate medial retraction well. Anatomic variations may affect the CPA exposure provided by the retrosigmoid approach. A posteriorly placed sigmoid sinus course results in the anterior edge of the craniectomy being placed more posteriorly. This placement creates a deeper field of action and a less favorable angle of view with the consequent need for more cerebellar retraction. This disadvantageous exposure may be compromised further by a low transverse sinus course, particularly if the patient also has a short neck and a prominent shoulder. This problem of 603
604
OTOLOGIC SURGERY
Chapter 50 • Retrosigmoid Approach to Tumors of the Cerebellopontine Angle
605
FIGURE 50-1. Incision used in retrosigmoid approach is located approximately 6 cm behind postauricular sulcus. After craniectomy and retraction of cerebellum, tumor becomes visible within cerebellopontine angle. Anterior edge of craniotomy is placed immediately behind sigmoid sinus and just inferior to lower margin of transverse sinus.
FIGURE 50-2. Operative view of cerebellopontine angle as seen through retrosigmoid approach. Superiorly is trigeminal nerve (CN V) and petrosal vein. Inferiorly are lower cranial nerves (IX, X, and XI). In midsection of the exposure, a medium-sized acoustic neuroma is seen in relation to facial and audiovestibular nerves.
restricted exposure may be overcome by combining the retrosigmoid approach with an anterosigmoid, retrolabyrinthine decompression to allow anterior retraction of the sigmoid sinus.6 A highly placed jugular bulb restricts access to the internal auditory canal (IAC), and can make the dissection of the inferior bony trough between the canal and the bulb difficult. Occasionally, the bulb may extend superiorly to overlap the IAC, partially obscuring access to the medial aspect of the canal.7
PREOPERATIVE EVALUATION AND PATIENT COUNSELING The minimal preoperative evaluation for a patient with a CPA tumor comprises a clinical history, a physical examination, pure tone and speech audiometry, and an imaging study (preferably, gadolinium-enhanced magnetic resonance imaging [MRI]). In nonacoustic tumors, computed tomography (CT) scanning for evaluation of the osseous characteristics of the cranial base and angiography to address vascular anatomy and possibly to perform embolization are occasionally indicated. Neither vestibular diagnostic testing nor auditory evoked responses are routinely obtained in patients already diagnosed with an acoustic neuroma.8 Numerous factors affect the selection of posterior fossa craniotomy for tumors of the CPA.7,9,10 As advocates of selective management of these lesions according to the unique attributes of each tumor and the potential surgical options, we involve the patient in the discussion of the relative advantages and disadvantages of each technique. In most cases, an obvious choice can be made, whereas in others, patient preference is important. Our customary preoperative counseling includes the anticipated and potential risks to hearing, balance, and facial motor function. Less common complications that are discussed include CSF leak, meningitis, cerebrovascular accident, and death.11 Although blood transfusion is seldom required, we encourage the patient to donate 1 U of autologous blood.
PATIENT SELECTION Common Indications in Neurotology Hearing Preservation The primary aim of acoustic neuroma management is removing the threat of progressive tumor growth, while avoiding injury to the central nervous system. Preservation
of cranial nerve function (facial movement, facial sensation, and hearing), which has become the primary focus of acoustic neuroma surgery in recent years, is a secondary goal. Patients with acoustic tumors can be classified into three groups in terms of potential for hearing preservation. Patients for whom hearing preservation is highly improbable generally undergo translabyrinthine removal. Criteria that place a patient into this group include poor hearing (<30% speech discrimination, >70 dB speech reception threshold), large CPA component (>3 cm), and deep penetration of the IAC. Conversely, patients with good hearing (>70% speech discrimination, <30 dB speech reception threshold), small CPA component (<1 cm), and shallow IAC involvement are considered excellent candidates for a hearing conservation approach.7 It is difficult to codify a set of rules concerning selection of a hearing conservation approach for the numerous patients who lie between these parameters. Each surgical team must rely on its own criteria, based on experience, together with the patient’s wishes in coming to a selection of surgical approach. Neurotologists would always favor undertaking a hearing conservation approach, even when the chances of success were remote, were there not potential adverse consequences from the endeavor. The lower morbidity of the translabyrinthine approach, especially in terms of persistent headache and CSF leak, leads the clinician away from the retrosigmoid hearing conservation approach when the chances of success are limited. The concept of useful hearing is context dependent. In a patient with a normal contralateral ear, imperfect residual hearing in the tumor ear is often of little practical benefit. When hearing in the contralateral ear is impaired or threatened, such as in cases of bilateral acoustic neuromas associated with neurofibromatosis type 2, a conservative approach to hearing conservation is prudent, occasionally even at the expense of complete tumor excision.12 Hearing preservation is seldom achieved when tumors with a CPA component exceeding 2 cm in diameter are removed.13,14 This rule should not be applied in nonacoustic CPA tumors (e.g., meningiomas), however, because hearing preservation is frequently achieved even with large tumors.15 The retrosigmoid approach exposes a variable amount of the IAC without violating the inner ear while the canal is being drilled open. Two factors should be considered in the decision of whether hearing conservation via the retrosigmoid approach is feasible: the depth to which the tumor penetrates the IAC, and the degree of
606
OTOLOGIC SURGERY
IAC exposable in that patient. The relationship between the inner ear and the lateralmost extension of the tumor into the IAC may be predicated by preoperative gadolinium-enhanced MRI.16
Use of Retrosigmoid Approach in Combined Therapy of Acoustic Neuroma Numerous studies have shown that functional outcome after conventional microsurgery is substantially poorer in patients with acoustic neuroma larger than approximately 3 cm. In these patients, the incidence of persistent facial dysfunction is high. There is also an increased risk of persistent balance dysfunction because of infarction of the middle cerebellar peduncle.17 In an effort to improve functional outcome, some centers have begun approaching larger tumors with subtotal resection leaving a rind of tumor on the pons and along the course of the facial nerve. When the patient has serviceable hearing, the retrosigmoid approach is typically used. To reduce the risk of recurrence, it is essential to remove the IAC component. Such surgical remnants resume growth in approximately one third of cases.18 If the remnant grows on serial imaging, it may be treated with stereotactic radiation with a greater than 90% probability of halting its growth.
Acoustic Neuroma in a Patient with Chronic Otitis Media Although patients with acoustic neuromas rarely have concomitant chronic middle ear infection, in patients who do the translabyrinthine approach for acoustic neuroma resection is contraindicated. The retrosigmoid approach may also open into potentially contaminated mastoid air cells lying behind the sigmoid sinus, and into air cells that may surround the IAC. To avoid potential intracranial infection, chronic middle ear infection should be controlled with tympanoplasty, antibiotics, or both, before tumor surgery whenever possible.7
Tumors Extending into Inferior Portion of Cerebellopontine Angle The retrosigmoid approach provides the best access to the lower portion of the CPA and can be extended to expose the foramen magnum when required. Transtem poral approaches to the CPA are limited in their inferior exposure by the sigmoid sinus and the jugular bulb. Acoustic neuromas seldom extend into the inferior reaches of the CPA. Even when they do, the capsular peel is readily mobilized superiorly after tumor debulking. Meningiomas and other extra-axial tumors usually do not mobilize easily, however, and are often entwined with the lower cranial nerves (CN IX through XII) and vital vascular structures (e.g., posterior inferior cerebellar and vertebral arteries). In such cases, the retrosigmoid approach is chosen for its superior ability to expose this region.
In patients with neurofibromatosis type 2, concurrent schwannomas on the lower cranial nerves are a common finding at the time of acoustic neuroma surgery. The retrosigmoid approach permits a thorough inspection of the jugular foramen contents and the dural lining of the posterior cranial base for possible early meningioma formation. Small, asymptomatic schwannomas on the lower cranial nerves are typically left alone, whereas early meningiomas are excised.
Tumors with Limited Extension into Meckel’s Cave The retrosigmoid approach is also useful in approaching extra-axial posterior fossa tumors that possess minor extensions into Meckel’s cave (cavum trigeminale). Most such tumors are trigeminal schwannomas and petroclival meningiomas. Added exposure is obtained by removing the apical petrous bone between the IAC and the tentorium. This maneuver provides access to approximately 1 to 2 cm of the posterior aspect of Meckel’s cave for tumor removal.19-21
Revision Surgery The retrosigmoid approach is favored in cases of recurrence after a previous translabyrinthine removal of an acoustic neuroma to avoid the dural scar from the prior procedure and to allow identification of the facial nerve as it emerges from the fat graft located in the surgical defect.
Relative Contraindications Deep Extension into Internal Auditory Canal Generally, tumors extending into the lateral one third of the IAC are not resectable by the retrosigmoid approach without destroying hearing. In such cases, the translabyrinthine approach ensures complete resection and reduces operative morbidity.7
Extension into Cranial Base Tumor penetration into the posterolateral cranial base is a relative contraindication to the retrosigmoid approach. CPA tumors that invade the temporal bone (other than the medial two thirds of the IAC), the jugular foramen, or the hypoglossal canal are generally best addressed via a lateral, transbasal craniotomy.
Large Tumors Although large CPA tumors (>3 cm) may be approached through either the retrosigmoid or the translabyrinthine technique, when we intend radical removal we prefer to use the latter because it provides ample exposure and minimizes the need for cerebellar retraction. In addition, when the pons and the cerebellar peduncle are
Chapter 50 • Retrosigmoid Approach to Tumors of the Cerebellopontine Angle
s ubstantially displaced medially and posteriorly, the translabyrinthine approach, by virtue of its more anterior placement, provides a more favorable angle of view posteriorly toward the brainstem interface. In cases of facial nerve disruption, which is more common in large tumors, the translabyrinthine approach affords more reconstructive options through mastoid meatal rerouting.7
PATIENT PREPARATION AND POSITIONING At most centers, the operation is performed by a multi disciplinary team consisting of a neurotologist, neurosurgeon, neuroanesthesiologist, neurophysiologist, and specialized operating room nurses. The operation is done with the patient under general anesthesia. A shortduration muscle relaxant is used to facilitate endotracheal intubation. Thereafter, anesthesia is maintained with inhalational agents alone, avoiding the use of muscle relaxants, which would prevent effective intraoperative cranial nerve electrophysiologic monitoring. In addition to the routine neuroanesthesia monitoring equipment, antithrombotic stockings and a urinary catheter are used. The retrosigmoid approach may be carried out in one of three surgical positions: supine, lateral supine (park bench position), and sitting.10 Supine is the favored position because it affords excellent exposure and carries the lowest risk of complication (discussed later). The patient is secured in the optimal operating position by means of a head holder attached to the bed frame (e.g., Mayfield). This apparatus facilitates exposure of the suboccipital region while the patient is in the supine position. Optimal surgical field exposure is obtained by a combination of head rotation, neck flexion, and ipsilateral shoulder elevation. Excessive neck torsion should be avoided to prevent cervical injury and to reduce the risk of cerebellar swelling secondary to compromised flow through the vertebral venous system. The cranial nerve– monitoring electromyographic electrodes are placed into the muscles supplied by CN V, VII, and XI. When intraoperative auditory brainstem monitoring is indicated, scalp electrodes are placed, and an earphone is inserted into the ipsilateral external auditory canal.22 We favor using an operating room table with enhanced lateral rotation capability (up to 30 degrees), which permits optimal visualization of the lateral end of the IAC at a comfortable working angle. When the surgeon works at relatively extreme rotations, the patient must be securely supported on the operating table by a lumbar support and placed on the contralateral side to the operative exposure, with chest and thigh safety straps. The bed is reversed, with the patient’s head on the foot section to allow the surgeon to sit during the microsurgical portion of the procedure. A perioperative prophylactic antibiotic with good CSF penetration (e.g., ceftizoxime, 2 g) is administered intravenously. Mannitol (1 g/kg) is administered
607
intravenously when the scalp incision is made so that its effectiveness in reducing brain swelling coincides with dural entry. We do not routinely give corticosteroids except in patients with larger tumors (>3 cm) or in patients with peritumoral brain edema, when dexamethasone (10 mg) is administered intravenously. To reduce the risk of CSF fistulization, an indwelling lumbar CSF drain is used when extensive peri-IAC pneumatization is encountered.
SURGICAL SITE PREPARATION We ask the patient to wash his or her hair thoroughly either the morning of surgery or on the evening before with an antiseptic shampoo. In the operating room, after induction of anesthesia, the hair over the suboccipital area is removed with electric clippers. It is also acceptable to use a racing stripe technique. The upper neck is included in the operative site, allowing potential access to the great auricular nerve in case a graft is required for facial nerve reconstruction. The scalp is washed with povidone-iodine soap, and the clipped area is shaved. To improve attachment of adhesive drapes, the surgical site is defatted with alcohol and dried. Sterile drapes are placed 1 cm from the hair edge around the prepared scalp and held in place with surgical adhesive (e.g., Mastisol). The field is prepared with povidone-iodine solution and dried. Sterile towels are placed around the operative field and held into position by an adhesive plastic sheet.
SPECIAL INSTRUMENTS Various instruments are used for the retrosigmoid approach to the CPA, including craniectomy instruments, retractors, high-speed surgical drill with a selection of cutting and diamond burrs, suction and suction-irrigation tips of the fenestrated and nonfenestrated types, bipolar cautery, microdissection instruments, and a binocular operating microscope. We perform the craniectomy with an Acra-Cut disposable cranial perforator burr-hole maker in a Hudson brace, a system that allows rapid bone removal, while minimizing the chance of dural or venous sinus injury. The craniectomy is completed with rongeurs. To retract the thick suboccipital musculature, a deep-bladed Weitlaner-type retractor is used. For brain retraction, several sizes of malleable blades are used that may be held in position in several ways. We prefer to use the Apfelbaum base, which combines a Weitlaner-type retractor with a movable arm to affix the retractor. Other options for basing the brain retractors during retrosigmoid craniotomy include a C-clamp placed on the head holder frame or a table-based system (e.g., Greenberg). Either an electric or an air-powered drill is suitable to use for this approach. When the exposure is narrow, an angled handpiece is advantageous because it is less
608
OTOLOGIC SURGERY
obstructing to the surgeon’s point of view. An operating microscope with an inclinable optical pathway is desirable to accommodate the various exposure angles required during the procedure while maintaining a comfortable operating position. Insulated bipolar cautery forceps are essential for obtaining hemostasis during CPA tumor surgery. Large tips for handling substantial vessels and slender, fine tips for use when the coagulation must be confined to a narrow region are needed. We have found that a self-irrigating system (e.g., Malis bipolar irrigating system) is valuable because it discourages tissue adhesion to the forceps tips. We use a microsurgical instrument set that includes sharp and blunt dissectors in various shapes and sizes, needles, and small bone curettes (e.g., Rhoton microneurosurgical instruments). A set of sharp scissors of different sizes and angles is also important. Many special tools are available to facilitate rapid intracapsular debulking of the tumor. We prefer to use a Cavitron ultrasonic surgical aspirator (CUSA), which allows debulking without traction or torsion, minimizes hemorrhage, and respects tumor capsular planes, avoiding inadvertent neural or vascular injury. Other options include the surgical laser and a rotatory surgical aspirator (House-Urban). Operating room electric circuitry and neuroanesthesia electric monitoring equipment should be grounded and electronically quiet to minimize 60 Hz noise production, which interferes with the cranial nerve electrophysio logic monitoring setup. The specialized equipment for intraoperative cranial nerve monitoring used in our institution has been described elsewhere in detail.22
SURGICAL TECHNIQUE Acoustic neuroma excision by the retrosigmoid approach to the CPA can be subdivided into seven stages: (1) craniectomy, (2) exposure of the CPA, (3) exposure of the IAC, (4) tumor resection, (5) hemostasis, (6) IAC closure, and (7) craniotomy closure.
Craniectomy A curvilinear paramedian incision 3 cm behind the postauricular sulcus is made down to bone. The cervical muscles are detached anteriorly and posteriorly, exposing the mastoid and suboccipital areas. Emissary venous bleeding is controlled with bone wax. The mastoid tip is exposed, and the posterior belly of digastric muscle is elevated from its groove. Dissection directly on the bone preserves the occipital nerves and vessels. A posterior fossa craniotomy window of approximately 3 × 3 cm is made in the retrosigmoid approach. It is bounded anteriorly by the sigmoid sinus and superiorly by the transverse sinus. The craniectomy begins with two or three closely approximated burr holes. The burr holes are joined up with rongeurs, creating a craniotomy window. The bone fragments are collected and stored in sterile
antibiotic-saline solution for replacement in the cranial defect at the end of the procedure. Development of the craniectomy anteriorly usually opens the mastoid air cell system to a variable degree. When the bony craniectomy is complete, the opened mastoid air cells are sealed with bone wax. Wax is also used to control bleeding from diploic bone at the craniotomy margins. Many styles of dural opening are described in the literature. We use a posteriorly based dural flap to enter the posterior fossa. The posterior fossa dura is opened 2 to 3 mm from its junction with the sigmoid and transverse sinus dura and at a similar distance from the inferior bony margin. The dural flap is reflected posteriorly. Small, relaxing incisions are made superiorly and inferiorly in the marginal dura to create small anterior and superior dural flaps, which are retracted with stay sutures, completing the dural opening.
Exposure of Cerebellopontine Angle When the dural flap has been reflected posteriorly, it and the craniotomy margins are covered with moist Telfa strips. To drain CSF from the cisterna magna, the cerebellum is gently retracted superiorly with a polytef (Teflon)-coated malleable retractor. The arachnoid of the cistern is lanced with a bayoneted suction tip, which decompresses the posterior fossa, relaxes the cerebellum, and allows it to fall away medially. Premature medially directed cerebellar retraction, before draining the cisterna magna, risks inducing massive cerebellar swelling. After this maneuver, the retractor is withdrawn and repositioned anteriorly to develop posteromedial cerebellar retraction. Retraction in this manner, accompanied by division of arachnoid bands and bridging veins, opens the CPA. The degree of CPA exposure required varies with the size and location of the tumor being addressed. Superiorly, the petrosal veins (or Dandy’s veins), which lie just below the tentorium cerebelli and run parallel to the course of the trigeminal nerve, may hinder exposure or appear to be in jeopardy of tearing with retraction. When necessary, these may be coagulated and divided. After their division, the superior pole of the cerebellar hemisphere falls posteromedially away from the tentorium, providing access to the superior aspect of the CPA. The cerebellar flocculus often overlies the brainstem root entry zones of CN VII and VIII, and must be gently mobilized from the cerebellar peduncle and lateral pontine surfaces. A tuft of choroid plexus, emanating from the lateral recess of the fourth ventricle, is also frequently encountered in this area. Mobilizing these structures from the root entry zone need not be performed during hearing conservation procedures when the proximal portion of the nerves are not involved with tumor because this maneuver places the internal auditory artery at risk. The cranial nerve electrophysiologic monitoring circuitry is tested by stimulating CN XI, which is usually readily accessible at the inferior pole of the exposure. Particular attention is
Chapter 50 • Retrosigmoid Approach to Tumors of the Cerebellopontine Angle
paid to the location of the anterior inferior cerebellar artery (AICA) and its branches. Inferiorly, the posterior inferior cerebellar artery may be seen in relation to the lower cranial nerves, and superiorly the superior cerebellar artery may be identified coursing through the region of the tentorial notch. In acoustic neuroma surgery, we prefer to begin the drill excavation of the IAC at an early stage, before extensive opening of the arachnoid planes above and below the CPA component. This method helps reduce bone debris contamination of the subarachnoid space.
Exposure of Internal Auditory Canal Exposure of the IAC and its contents involves removal of the bone surrounding the posterior, superior, and inferior aspects (Figs. 50-3 to 50-6). Optimal canal visualization may be obtained through a combination of rotation of the operating table away from the side of the surgeon and microscope positioning. These maneuvers bring the posterior petrous face into view centered over the region of the IAC. To locate the canal, the opening of the meatus is gently probed with a blunt, right angle hook. Before IAC opening begins, the operative field is set up to contain as much bone debris as possible and to prevent its dissemination into the subarachnoid space. Absorbable gelatin sponge (Gelfoam) pledgets are placed into the superior and inferior portions of the CPA. A rectangular-shaped rubber dam is fashioned from a surgical glove, placed over the occluding pledgets, and held in place with the cerebellar retractor. An H-shaped dural incision, centered on the long axis of the IAC, is outlined on the posterior petrous face by use of a bipolar cautery. When the dura has been incised with the tip of a No. 11 blade, superior and inferior dural flaps are elevated with a small Lempert mastoid elevator. The surgeon should exercise caution when incising inferiorly because the jugular bulb is occasionally dehiscent on the posterior petrous face. Similarly, the incision should not be carried too far laterally because laceration of the sigmoid sinus can occur. Care is taken to identify and preserve the endolymphatic sac and duct, which are located posterolaterally. The dura usually can be elevated off the endolymphatic sac. The entry point of the vestibular aqueduct into bone is a useful anatomic landmark. When the bony dissection of the IAC does not extend lateral to the operculum of the aqueduct, the labyrinth is unlikely to be breached. The posterior IAC wall is rapidly removed by the drilling of a trough over the posterior petrous face. Drilling from medial to lateral in the line of the IAC reduces the risk of the burr slipping into the CPA. The canal should be opened only as much as required to expose the lateralmost aspect of the tumor. Excessive bony opening does not enhance exposure further, but may increase the risk of CSF leak through the opening of additional petrous air cells. Initially, the bone is removed with a cutting burr until the IAC dura is identified through a thin bony
609
plate. To expose the dura of the posterior aspect of the IAC, the dural cuff of the meatus is first elevated from the thin residual plate. Then the remaining bony shell over the posterior aspect of the IAC is drilled away. To reduce the risk of traumatizing the IAC dural lining or its neural structures, removal of the last eggshell of bone is accomplished with diamond burrs. Diamond burrs are more controllable by virtue of their reduced tendency to run, and are less likely to cause injury if they come into contact with soft tissue structures. Bony troughs, 3 to 4 mm in diameter, are developed above and below the canal. These troughs are important for three reasons: (1) to provide working space for the insertion of angled instruments needed to establish a plane of dissection between the tumor and the facial and cochlear nerves, (2) to permit visualization of the facial nerve when it is acutely angled superiorly or inferiorly as a result of tumor displacement, and (3) to enhance exposure of the anterior aspect of the CPA. In preparation for the drilling of bony troughs around the canal, the IAC dura is elevated from the upper and lower canal walls with a blunt dissector. The troughs, which should be widest at the level of the porus, are excavated with a cutting burr. As the troughs are developed, a thin shell of bone is left over the dura of the superior and inferior walls of the IAC. When the troughs are fully developed, the remaining bony shells are progressively thinned with the side of the diamond burr until the dura is exposed. Copious irrigation is used to prevent thermal injury to neural structures. Often, the IAC dura can be gently retracted with a fenestrated suction, permitting completely atraumatic removal of the remaining bony eggshell fragments, which can be elevated from the exposed IAC dura. This technique exposes 180 to 270 degrees of the IAC circumference. Caution must be exercised in development of the superior and inferior troughs because of the proximity of the facial nerve and the jugular bulb. The width of the inferior trough varies with the location of the jugular bulb. When the jugular bulb is unusually high, creating an inferior bony trough at the level of the meatus may be impossible, although exposure of the fundus is typically unhindered. Compensating for the limited inferior access associated with a high jugular bulb is usually possible through creation of an unusually wide and deep superior trough. Additional exposure of the IAC from above may also be gained through retraction of the tentorium. In hearing conservation attempts, the lateral extent of the IAC opening should be restricted to approximately the medial two thirds of the IAC because opening of the lateral one third to expose the fundus may result in a breach of the vestibule or crus commune, militating against hearing conservation. The decision as to how far laterally the IAC is opened depends on the lateral intracanalicular extent of the tumor, which may be accurately predicted from the preoperative gadolinium-enhanced MRI.16,23 Alternatively, the lateral opening can be limited
610
OTOLOGIC SURGERY
Chapter 50 • Retrosigmoid Approach to Tumors of the Cerebellopontine Angle
611
FIGURE 50-3. Step 1 of exposure of internal auditory canal (IAC) during retrosigmoid approach. After localization of porus acusticus through palpation with a ball hook, an H-shaped dural incision is created over the long axis of IAC. Dural flaps are reflected anteriorly and posteriorly to expose posterior aspect of petrous pyramid. Dura usually can be dissected from posterior surface of endolymphatic sac. To maximize the possibility of hearing preservation, care must be exercised to avoid avulsion of sac from its aqueduct. Before drilling, Gelfoam pledgets are positioned above and below 7 to 8 neurovascular bundle in posterior fossa in an effort to minimize the spread of bone dust onto arachnoidal surfaces. FIGURE 50-4. Step 2 of exposure of internal auditory canal (IAC) during retrosigmoid approach. A cutting burr is used to excavate rapidly bone overlying IAC. When canal dura is encountered, a diamond burr is used. Drilling is carried out from medial to lateral (from porus to fundus), in part to minimize the possibility that drill could accidentally run into posterior fossa.
FIGURE 50-5. Step 3 of exposure of internal auditory canal (IAC) during retrosigmoid approach. A diamond burr is used to create deep troughs around IAC that should extend well deep to canal plane to provide adequate room for microdissection with the angled instruments needed to remove facial nerve from tumor safely. Particular care should be exercised while superior trough is developed because facial nerve may lie immediately beneath dura in this location.
FIGURE 50-6. Step 4 of exposure of internal auditory canal (IAC) during retrosigmoid approach. In preparation for tumor removal, dura of IAC is incised. After an incision along the length of the canal is created with upbiting scissors, two small relaxing incisions are created at porus and fundus to develop dural flaps. These are reflected anteriorly and posteriorly to expose canal contents.
on the premise that an indirect inspection and clearance of the tumor from the lateral IAC can be satisfactorily achieved. This method has the attendant risk of leaving residual tumor in the lateral IAC, however. To avoid this problem, some surgeons have advocated blind curettage using special right angle curettes followed by inspection of the fundus with a small mirror or endoscope to validate the extent of tumor resection.24,25 With these methods, distinguishing residual tumor from the transected vestibular nerves and traumatized dura is sometimes difficult. Dissection of tumor from the fundus without direct visualization risks leaving wellvascularized residual tumor with the potential for clinically significant recurrence.26,27 We advocate exposure of the IAC laterally to a point beyond the tumor interface, where the naked CN VII and residual CN VIII may be visualized. This process sometimes may require opening the canal to the fundus, with resultant entry into labyrinthine structures and sacrifice of residual hearing. It has been proposed that enhanced visualization of the fundus can be achieved by skeletonization of the posterior and superior semicircular canals.28 Similarly, it has been shown that partial resection of the posterior semicircular canal may be helpful in augmenting fundal exposure.29 After completion of the IAC exposure, the rubber dam and Gelfoam pledgets are removed. The dura of the IAC is opened along the long axis of the canal with sharp, upturned, right angle microscissors working from a medial-to-lateral direction. This incision is placed slightly eccentrically and is biased to the superior side to avoid the creation of a long flap over the facial nerve course. The dural flaps are reflected superiorly and inferiorly, exposing the IAC contents.
Acoustic Neuroma Resection Attention is now turned to planning the actual resection of the tumor; the size of the tumor largely dictates the actual sequence and pattern of removal (Figs. 50-7 to 50-9). We prefer to dissect the IAC initially because this step helps ascertain the probable course of the facial
nerve outside of the porus into the CPA, and allows early identification of the facial nerve. A test run of the neural monitoring system can be performed in which positive identification of the nerve by its anatomic relationships is possible. In many cases, ascertaining whether the tumor has arisen from the superior or inferior vestibular nerve is possible. When only one of these nerves is visible on the posterior surface of the tumor, it may be assumed that the other was the nerve of origin. Dissection is begun laterally by identification of the plane between the facial nerve and the tumor. A fine-tipped dissector is insinuated between the superior dural leaf of the canal and the tumor while gentle downward pressure and a rotating motion are applied. Gradually, this process brings into view the interface between the facial nerve and the lateral end of the tumor. When this plane has become established, a sharp, right angle instrument is used to dissect the tumor from the posterior surface of the facial and cochlear nerves. All tissue superficial to this plane, including the tumor and both vestibular nerves, is transected either with curved microscissors or through an upward motion with the sharp edge of the dissector. When the intracanalicular tumor component is bulky, it may require debulking to a variable degree to permit microdissection of the capsular peel from the facial and cochlear nerves. This initial dissection of the intracanalicular portion should proceed only to the lip of the porus acusticus or just beyond it, to avoid dissection of the typically most adherent section at this stage. It is important to avoid inducing neurapraxic injury, which might impair later electric identification of the facial nerve medially at its brainstem exit. After removal of the intracanalicular portion of the tumor, the CPA component is addressed. The posterior capsule is swept with the neural monitoring probe to ensure that the facial nerve is not on this surface (a rarity in acoustic neuroma). A rectangular incision is made in the posterior capsule with the point of a No. 11 blade, and the peel is resected with scissors. Intracapsular debulking may be done with cupped forceps, sharp dissection with scissors, an ultrasonic aspirator (e.g., Cavitron), a rotatory aspiration device (e.g., House-Urban), or the
612
OTOLOGIC SURGERY
Chapter 50 • Retrosigmoid Approach to Tumors of the Cerebellopontine Angle
613
FIGURE 50-7. Step 1 of removal of acoustic neuroma via retrosigmoid approach. After intracanalicular portion of tumor is debulked, lateralmost extension of tumor is reflected medially, and plane between tumor capsule and facial nerve is developed. This maneuver ensures complete removal of tumor from fundus. It also affords an early opportunity to confirm function of cranial nerve monitoring system by stimulation of distal facial nerve under direct vision in a region where it is characteristically not especially adherent to tumor surface.
FIGURE 50-8. Step 2 of removal of acoustic neuroma via retrosigmoid approach. Main portion of tumor in cerebellopontine angle is rapidly debulked. To facilitate rapid and safe tumor removal, we use a Cavitron ultrasonic aspirator. Tumor capsule is first liberated from cerebellum and middle cerebellar peduncle. Pontine surface, including root entry zones of CN VII and VIII, can be exposed. In larger tumors, lesion also must be microdissected from trigeminal and lower cranial nerves (IX and X). FIGURE 50-9. Step 3 of removal of acoustic neuroma via retrosigmoid approach. Characteristically, facial nerve is most adherent to tumor capsule between brainstem surface and anterior lip of porus acusticus. Liberation of nerve from tumor surface in this location often requires particularly delicate microdissection techniques.
FIGURE 50-10. Closure of internal auditory canal defect at completion of retrosigmoid craniotomy. After waxing of cut bony walls to seal any transected air cells, muscle graft harvested from nuchal area is mortised into bony defect. Graft is retained in position by sutures, which are anchored in dural flaps previously developed from posterior petrous surface.
surgical laser. We favor using the CUSA because it efficiently removes the tumor core while respecting its capsule, avoiding potential injury of adherent nerves and vessels. Tumor resection proceeds with alternate intracapsular debulking, followed by microdissection of the thin capsule from the brain surface and cranial nerves, and, ultimately, resection of the liberated capsular segment. The most crucial aspect of CPA tumor removal is identification and preservation of the cranial nerves and blood vessels that lie draped on the capsular surface. In larger tumors, the medial tumor brain dissection plane begins posteriorly along the middle cerebellar peduncle. When this arachnoid plane has become established, it is gradually developed onto the lateral surface of the pons. Attention is turned inferiorly to the probable root entry zone of the seventh and eighth cranial nerve complexes. As an aid to facial nerve identification, electric stimulation is periodically performed along the meniscus of dissection. The course and appearance of the nerve vary depending on its displacement by the tumor. It may be thinned and fanned to a variable degree, making it difficult to delineate from surrounding thickened arachnoid tissue without the use of the microneural stimulator. The brainstem entry of CN VIII is usually encountered lateral to and immediately above the seventh cranial nerve entry zone. A small branch of AICA typically passes between the two nerves and may be a useful guide in orienting the surgeon. In hearing conservation approaches, the vestibular fibers must be separated from the cochlear fibers and divided proximally to establish a tumor dissection plane. When no effort is being made at hearing preservation, CN VIII may simply be transected, a maneuver that simplifies identification of the proximal seventh cranial nerve. When the proximal plane over these two nerves is established, an arachnoid plane can be developed between them and the tumor capsule. While the tumor capsule is dissected, large and small arteries, potential AICA branches, are meticulously preserved. Vessels directly entering the tumor capsule can generally be safely coagulated and divided at the capsular surface without adverse consequences.
While the tumor neural plane is dissected, use of the microneural stimulator (e.g., Xomed Treace-Yingling) with a curved, pliable wire allows blind stimulation of the yet undissected anterior capsule. By localizing the facial nerve course before dissecting the tumor nerve interface, the surgeon may rapidly resect uninvolved capsule and direct meticulous efforts along the actual course of the nerve. Although the course of the facial nerve varies, it characteristically lies anterior to the tumor, occasionally with an anterosuperior or anteroinferior bias. In small tumors, the entire dissection may be accomplished from a medial-to-lateral direction. In larger tumors, medialto-lateral dissection becomes difficult when the nerve is anteriorly angulated toward the porus acusticus. When this occurs, we return to the lateral tumor nerve interface at the end of the IAC and work medially. Alternatively, the anterior tumor capsule with attached nerve may be lifted and rotated to bring the facial nerve course into the surgeon’s view. This action is quite traumatic to the facial nerve, however, and risks disruption of its attenuated fibers. We prefer to dissect the tumor from the facial nerve in situ without mobilizing it from its bed, where it lies supported by an arachnoidal mesh. When the facial nerve is splayed and tightly adherent to the tumor capsule, removing the last remnant of capsule may be impossible without disruption of the nerve.30 In such cases, we prefer to perform a near-total removal, leaving a thin velum of capsule, only 1 to 2 mm thick, attached to the nerve. We believe that this minuscule residual capsule, hanging free in the CPA, is unlikely to generate a recurrent tumor.26 By contrast, tumor left in the distal IAC or in contact with brainstem possesses a vascular supply, and the possibility of regrowth is greater. Several modifications in the strategy of tumor removal are used during hearing conservation approaches. The direction of dissection should be from medial to lateral, whenever possible, to reduce the risk of traumatic avulsion of the delicate cochlear nerve fibers from their entry into modiolus. Throughout the cochlear nerve dissection, changes in auditory brainstem responses relative to the previously recorded baseline waveforms obtained at the start of the procedure are reported. Continuity of
614
OTOLOGIC SURGERY
the cochlear nerve is maintained if possible; however, tumor adherence to it may necessitate its resection. Even when the cochlear nerve is well preserved during dissection, hearing is often lost because of interruption of the cochlear blood supply. This may occur either in the CPA, where the labyrinthine artery branches from a loop of AICA, or in the IAC, where it courses between the inferior vestibular and cochlear nerves. After tumor resection, anatomic and electrical continuity of the cochlear and facial nerves are checked. The facial nerve stimulation threshold voltage at the root entry zone and the intraoperative auditory brainstem response waveform pattern and latencies are recorded. We believe that electrophysiologic monitoring of the auditory nerve is not clearly beneficial, other than in the prognostic sense, in the maintenance of hearing. Monitoring of the facial nerve is indispensable, however, if an optimal outcome is to be obtained.
sutures. At closure of the dura, bacitracin-saline solution is instilled into the subarachnoid space. The margins of the craniectomy are reinspected for opened air cells and smeared with bone wax as indicated. When the transected air cells are large, rather than merely impacting the wax into the exposed cavities, a thin sheet of wax is applied. A Gelfoam pad cut in the shape of the craniectomy defect is placed over the dura, and the previously preserved bone chips are replaced. In our experience, the bone regenerates over several months into a strong, bony plate that restores the cranial contour. After removal of the remaining retractors, the soft tissues of the neck are closed in a series of layers with interrupted 2-0 Surgilon sutures, closing any potential dead space, after the wound is irrigated with antibiotic solution. The skin is sutured with interrupted 4-0 nylon sutures. Electromyography electrodes, ground pads, and the external auditory canal earphone are removed after completion of wound closure and dressing.
Hemostasis After the tumor resection is completed, the wound is irrigated with bacitracin-saline solution, the blood clot is removed, and all bleeding points are identified and controlled with bipolar cautery or by application of thrombin-soaked Gelfoam. As a means of detecting subtle or intermittent bleeding, the anesthesiologist gives the patient a Valsalva maneuver for 20 seconds. Because postoperative hemorrhage into the CPA is a potentially devastating complication, hemostatic efforts should be diligent.
DRESSING After closure of the wound, it is cleaned, dried, and covered with a Telfa strip, to which a sterile adhesive, Op-Site, or similar dressing is applied. To discourage subcutaneous accumulation of CSF, a mastoid-type padded pressure bandage is applied. The dressing is removed 48 hours after surgery, and the wound is inspected and left open to the air. The skin sutures are removed 7 to 10 days after surgery.
Internal Auditory Canal Closure At the time of IAC closure (Fig. 50-10), the bony troughs developed for the IAC exposure are inspected for opened air cells by palpation with a ball hook. Inspection of the cut bony edge may also be carried out through use of a 90 degree angled rigid endoscope. Bone wax is applied to a small Cottonoid and smeared over the exposed bony trough surfaces to seal overtly and covertly opened air cells to prevent CSF leakage. A small fat graft is harvested from the abdomen and is used to seal the IAC. We prefer fat to muscle as an IAC sealant because, with fat-suppressed MRI, fat creates less obscuration of the tumor bed on follow-up imaging. A 7-0 monofilament nylon suture may be placed through the dural flaps of the posterior petrous face, although this is not always needed. The fat is positioned in the IAC, and the suture is tied. Auditory and facial nerve monitoring are maintained until the fat is secured in place so that any possible neural irritation induced by its placement can be identified.
Craniotomy Closure After removal of all the Cottonoids and Telfa strips, the dural flap is sutured back into place with multiple, closely positioned, interrupted 4-0 braided nylon (Surgilon)
POSTOPERATIVE CARE The anesthesiologist awakens the patient, ideally with a smooth extubation that avoids straining and coughing. Antiemetics are given prophylactically to prevent vomiting, which could cause aspiration and associated pneumonitis during recovery from anesthesia. Postoperative monitoring is carried out initially in the postanesthesia care unit and then in the neurosurgical intensive care unit for 24 hours after surgery. After the initial 24-hour period, patients spend an average of 5 to 6 days on a hospital unit staffed by nurses experienced in postcraniotomy care. In addition to the monitoring of temperature, cardiorespiratory status, consciousness level, and fluid balance, the nursing staff and patients are instructed to identify and report any CSF wound leakage or rhinorrhea. If a postoperative facial palsy is present, its grade is recorded according to the House-Brackmann scale, and preventive eye care is instituted. When eye closure is incomplete, artificial tears are applied hourly, or more often as needed, while the patient is awake. During sleep, a plastic eye shield is placed to prevent drying and development of corneal abrasion. Special attention needs to be directed toward patients with dysfunction of the facial
Chapter 50 • Retrosigmoid Approach to Tumors of the Cerebellopontine Angle
and trigeminal nerves. When the cornea is dry, exposed, and insensitive, early gold weight placement is performed even when facial nerve recovery is expected. Moderate to severe headache for several days is typical and may require narcotic analgesia for a variable period. Global headache that is delayed in onset by several days may signify the evolution of aseptic or bacterial meningitis and is discussed later. We try to wean patients from narcotics quickly and try to get their headaches under control with nonsteroidal anti-inflammatory preparations. In the few cases in which corticosteroids have been used, they are tapered over a 7 to 10 days. Vertigo can be controlled with parenterally administered antivertiginous agents, if the condition is severe, and with oral agents, if it is mild. We generally avoid vestibular suppressants in the postoperative period because they may retard vestibular compensation. Diet and increasingly independent mobilization are encouraged under the guidance of a dietitian and physical therapist. We usually restrict the fluid intake for 3 days to a total of 1.5 L/24 hour period. Most patients are able to start a light diet 24 to 48 hours after surgery. Constipation and straining are avoided by the administration of stool softeners to prevent aggravation of headache and possible development of CSF leakage. Most patients begin mobility around 48 hours after surgery, although they have been encouraged to exercise their legs actively while they are recumbent in bed to reduce the risk of deep venous thrombosis. Antiembolism stockings are used until the patient is mobile. Mobilization usually takes the form of initially sitting at the bedside chair, followed by accompanied walks to the bathroom and then farther afield to the hospital corridors, and then onto a trial of practice on the stairs. Walking aids are provided by the physical therapist as required by the patient, depending on his or her progress. We usually discourage hair washing until 1 week after surgery to prevent the wound from getting wet and macerated. Patients may use a dry shampoo if desired. Most patients are usually ready for discharge 5 to 7 days after surgery. Even when all other functions have recovered fully, easy fatigability often persists for 1 to 3 months postoperatively. The convalescent period required before returning to full-time employment and all the previous activities of daily living varies, but is usually 2 to 3 months.
RESULTS Historically, the primary issue in acoustic neuroma surgery was the survival of the patient. With the evolution of microsurgical techniques, mortality from acoustic neuroma surgery has become very low—less than 2% in most recent series. Contemporary emphasis includes tumor control and, particularly, functional preservation. Before the data from our own experience and the data published
615
in the literature are addressed, however, it is important to appreciate that limited international standardization exists in the criteria used for reporting results on degree of resection,31 facial nerve function,32 and hearing preservation.33 In our opinion, the goal of acoustic neuroma resection should be tumor control and not complete resection in every case. Nevertheless, we perform a complete removal in most cases. Incomplete removal can be considered in two categories: subtotal and near-total excision. Subtotal removal, in which a substantial bulk of tumor remains, used to be reserved for elderly or infirm patients with a short anticipated life span in whom shortening of the operative procedure is thought to be in the patient’s interest. In recent years, it is increasingly used in large tumors, backed up by stereotactic radiation if the remnant grows. As previously discussed, near-total excision, in which a thin peel of capsule is left on the most adherent portion of the facial nerve, is occasionally used. Although few data are published on the recurrence risk for this group of patients, we have observed many individuals with serial gadolinium-enhanced MRI and have found a 3% risk of recurrence. The decision to undertake a near-total resection depends on the patient’s age (i.e., less desirable in a younger individual) and preference as to whether the slightly higher risk of recurrence is justified by the improved facial nerve outcome. There are numerous articles in the literature on the subject of facial nerve preservation in acoustic neuroma surgery citing varying degrees of success. Because these results are difficult to compare and draw conclusions from, we confine our commentary to our own series. In our experience, facial nerve outcome from the retrosigmoid approach is similar to that from the other methods of removing acoustic neuromas for tumors of similar size.34 In our acoustic neuroma patients, anatomic continuity of the facial nerve was maintained in 99.2% of cases. Anatomic continuity does not imply functional integrity. The probability of a grade I or II facial function at 1 year after surgery in the context of tumor size was 100% for tumors less than 1 cm; 90% for tumors 1 to 3 cm; and 82% for tumors greater than 3 cm. With regard to hearing preservation, most published series address residual “measurable” hearing in contrast to the much more relevant concept of “useful” hearing.35 For a patient with a unilateral acoustic neuroma, it could be argued that unless the conserved hearing maintains an interaural difference of less than 30 dB hearing loss with good speech discrimination (>50%), it would be likely to be beneficial. Preservation of useful hearing has been reported to be achieved in 25% to 58% of hearing conservation candidates.13 Very little information is available on the long-term follow-up of patients with preserved hearing. In two published series, significant late decline occurred in 22% to 56% of ears with successful hearing conservation.36,37
616
OTOLOGIC SURGERY
Factors relevant to success in hearing conservation approaches to acoustic neuroma include tumor size in the CPA, the depth to which the tumor penetrates the IAC, pure tone hearing level, and auditory brainstem response results. A full discussion of these criteria is beyond the scope of this chapter. It is not yet well established whether intraoperative auditory monitoring materially improves hearing conservation results. In one study using auditory brainstem response monitoring, it was found to be of marginal benefit overall, with the possible exception of tumors less than 1 cm in diameter.38 Several retrospective studies have compared hearing preservation rates after the retrosigmoid and middle fossa approaches.39-41 In each study, the middle fossa approach yielded significantly better hearing results. Although that choice has not yet been universally accepted, the trend among centers undertaking a large volume of acoustic neuroma surgical procedures is to use the middle fossa approach as the preferred means of attempting hearing conservation. In our institution, the upper limit on the use of the extended middle fossa approach is a tumor in the 15 mm range of extracurricular diameter. The retrosigmoid approach is reserved for acoustic neuromas when three conditions are met: (1) excellent hearing, (2) a cisternal component 15 to 20 mm, and (3) no tumor involvement of the distal one third of the IAC. The retrosigmoid approach is still used for selected nonacoustic tumors of the CPA (e.g., meningiomas and epidermoids). Incomplete resection also has a potential role in hearing preservation, particularly in patients with neurofibromatosis type 2 or in patients with a tumor in an only hearing ear.12,42
COMPLICATIONS The common complications of the retrosigmoid approach to the CPA are persistent headache and CSF leakage.11,43,44 Less common complications include (aseptic or bacterial) meningitis, hydrocephalus, cerebellar dysfunction, vascular compromise (thrombosis and hemorrhage), and problems associated with patient malpositioning during surgery. Medical complications, such as pulmonary thromboembolism and pneumonia, may also occur, but are not specific to surgery of this region. Although the potential complications of acoustic neuroma surgery are similar among the various operative approaches, their relative incidence varies considerably. In the retrosigmoid approach, persistent headache and CSF leakage occur more frequently than with the other technique used in approaching CPA tumors.
Vascular Complications Hemorrhage Vascular complications may be extra-axial or intra-axial. The main extra-axial problem is bleeding into the CPA. CPA hematomas may cause brainstem compression
and acute obstructive hydrocephalus. The incidence of acute CPA hematomas has been reported to be 0.5% to 2%; however, with modern hemostatic techniques, the incidence is probably considerably less frequent.11 This diagnosis should be suspected when a patient does not promptly awaken after surgery or has a delayed deterioration in the level of consciousness. The diagnosis may be made by noncontrast CT scan, in which fresh blood appears as a hyperdense mass in the CPA and extrinsic pontine compression is noted. If serious neurologic sequelae and death are to be avoided, prompt surgical evacuation of the hemorrhage is essential. Intra-axial pontine hemorrhage may occur, particularly after removal of very large tumors that have greatly deflected the brainstem. Although major parenchymal hemorrhage is rare, minor amounts of intrinsic pontine bleeding are often evident radiographically after extirpation of giant tumors. Presumably, these bleeds result form the sudden re-expansion of the deeply compressed parenchyma. Supratentorial intra-axial hemorrhages have been reported after retrosigmoid approaches performed with the patient in the sitting position. These hemorrhages were associated with hypertension and may have resulted from subcortical venous tearing resulting from mechanical stress induced by the sitting position.45-47 Extradural hematoma formation, a concern in the middle fossa approach, is uncommon after the retrosigmoid approach.
Anterior Inferior Cerebellar Artery Syndrome Brainstem infarction may occur after damage to the AICA, the vascular supply to the pons and cerebellar peduncle. Mechanisms of injury include disruption, cauterization, and arteriospasm with thrombosis. A full-fledged AICA syndrome is extremely serious and is often fatal because it results in the loss of respiratory center control.48 Partial interruption of flow in the AICA system, avulsion of one or more of its branches, or obstruction of a nondominant AICA may result in an incomplete AICA syndrome.17 More recently, we have recognized several patients operated on for acoustic neuromas greater than 3 cm in diameter in whom gadolinium-enhanced MRI detected an infarction in the region of the middle cerebellar peduncle. These patients had unilaterally impaired cerebellar function and required prolonged physical therapy rehabilitation.49
Nonvascular Complications Complications from Patient Positioning As with any craniotomy, air embolism through breach of the major venous sinuses is a potential hazard. This risk is minimal, however, when a supine or lateral patient position is used.50 Air embolism is the main complication of the sitting position and has been reported in 30% of cases.51 When the sitting position is used, intraoperative monitoring with precordial Doppler ultrasonography alerts the anesthesiologist to venous air entry.
Chapter 50 • Retrosigmoid Approach to Tumors of the Cerebellopontine Angle
The initial maneuvers to perform when air embolism has been detected are to flood the field with fluid and lower the head of the bed. Quadriplegia (in four cases) and paraplegia have also been reported after acoustic neuroma resection in the sitting position. The degree of cervical flexion in the absence of protective spinal reflexes during anesthesia was thought to have caused spinal cord compression and infarction.52,53 In the supine and lateral supine positions, the unconscious patient must be handled carefully, especially when the head holder is positioned. Excessive head rotation risks cervical injury and may obstruct vertebral venous drainage and contribute to cerebellar swelling. Excessive downward displacement of the shoulder risks traction injury on the brachial plexus. As with any prolonged surgical procedure, adequate padding under pressure points is important to avoid pressure ulceration. Despite the best of precautions, patients frequently complain of discomfort over the ischium or other bony prominences for a few weeks postoperatively.
Cerebrospinal Fluid Leakage CSF leakage is the most common postoperative complication, occurring in approximately 15% of patients who undergo retrosigmoid approaches for acoustic neuroma.54-56 The patient must be counseled to recognize and report CSF leakage so that steps can be taken to control it rapidly to prevent infectious meningitis. CSF leak occurs either directly through the wound or indirectly through the ear and auditory tube to the nasopharynx, where it manifests as a watery rhinorrhea or salty postnasal discharge. CSF escape into the ear may occur through opened and unsealed mastoid air cells in the region of the craniectomy or through air cells opened and unsealed in the bony IAC dissection.57 CSF drainage often stops spontaneously with simple fluid restriction and avoidance of straining. The use of acetazolamide, a carbonic anhydrase–inhibiting diuretic, may also be beneficial. Alternatively, the early use of a lumbar CSF drain for 48 to 72 hours may halt the drainage. Some authors have advocated (1) wound reexploration with rewaxing of the bone to close covert open air cells, (2) replacement of the muscle graft plug to close CSF leakage, and (3) continued lumbar drainage.50 We prefer to address persistent, intractable CSF otorhinorrhea transtemporally. When useful residual hearing is present, a canal wall up mastoidectomy is performed, perilabyrinthine cells are copiously waxed, the fossa incudis is occluded with a fascia graft, and fat is used to obliterate the cavity. When the operated ear is deaf, a canal wall down mastoidectomy is performed. The external auditory canal is sutured closed, and the auditory tube is sealed under direct vision with bone wax and muscle. The mastoid air cells are also waxed, and the cavity is obliterated with fat. Lumbar CSF drainage is maintained for approximately 72 hours after surgery.
617
Aseptic and Bacterial Meningitis Entry of blood and bone dust into the subarachnoid space can result in aseptic meningitis. Care is taken during the drilling of the posterior petrous face during the IAC exposure to prevent contamination of the subarachnoid space with bone dust. Gelfoam is placed in the CPA superior and inferior to the tumor and CN VII-VIII complex, and a rubber dam is placed over the cerebellum. After completion of the bone work, the wound is thoroughly irrigated, and the bone debris is removed. Similarly, throughout the tumor dissection and at its completion, a combination of suction and irrigation is used to prevent the buildup of blood and clots because blood and bone debris produce an irritative or chemical aseptic meningitis.11 To reduce the risk of bacterial meningitis, intravenous prophylactic antibiotics are administered at the start of surgery, and bacitracin is added to the irrigant solution used to flush the CPA at the end of the procedure. This complication should be suspected if the patient develops headache, fever, and malaise in the first postoperative week. Nuchal rigidity, usually considered a sign of meningeal irritation, is not a useful sign after retrosigmoid craniotomy because the neck muscles may be in spasm owing to direct surgical trauma. Bacterial meningitis may also occur in the late postoperative period, particularly when a CSF leak is present. The clinician should maintain a high degree of suspicion about bacterial meningitis and, when in doubt, should obtain a sample of CSF via lumbar puncture for analysis. In patients in whom the clinical picture is suggestive, intravenous antibiotics should be instituted pending results of culture and sensitivity testing.
Hydrocephalus Hydrocephalus can occur as a result of blood and bone debris contamination of the posterior fossa subarachnoid space. Particulate and proteinaceous debris becomes ingested by arachnoid granulations, impairing their absorptive capabilities, which results in raised intracranial pressure. Cerebellar retraction with subsequent swelling at release may also result in the development of hydrocephalus.11
Cerebellar Dysfunction Prolonged cerebellar retraction may result in edema and swelling and possibly contusion, with resultant dysmetria and impaired balance in the postoperative period.
Persistent Headache Headache is encountered more frequently after the retrosigmoid approach than after other types of posterior fossa craniotomy.58 In our experience, nearly all retrosigmoid patients have substantial headache during the first
618
OTOLOGIC SURGERY
postoperative month. By 3 months after surgery, approximately one third continue to complain of this symptom. By 1 year, approximately 15% of patients continue to have chronic moderate to severe headaches compared with very few headaches for patients who underwent the translabyrinthine procedure. Some individuals are unable to return to work or resume other life activities because of this symptom. The highest incidence of persistent headache in our series has been in patients with small tumors who underwent the retrosigmoid approach in an effort to preserve hearing. Although the headache may have myriad presentations, it is most commonly either frontal or referred to the area of surgery and is often triggered by cough. Numerous potential underlying causes exist for chronic headache after retrosigmoid craniotomy, including aseptic meningitis, coupling of the suboccipital dura to the nuchal musculature, occipital neuralgia, and exacerbation of an underlying headache tendency, such as migraine. Although numerous mechanisms are possible, we believe that most headaches are a result of chronic arachnoiditis incited by contamination with bone dust and blood at the time of surgery. One apparent risk factor for the development of chronic headaches is retrosigmoid craniectomy, in which the calvarial bony defect is left unreconstructed.59,60 Replacement of the retrosigmoidal bone with a flap, bone chips, or even alloplastic material diminishes the incidence of persistent headache.61,62
Residual or Recurrent Tumor We prefer to revise recurrent tumors after retrosigmoid craniotomy using a translabyrinthine approach. This method avoids the previously scarred dural areas and tends to present more favorable arachnoid dissection planes during the early portion of the procedure.63
Acknowledgment Figures 51-1 to 1-10 were adapted from artwork produced by the authors from Jackler RK: Atlas of Skull Base Surgery and Neurotology, 2nd ed. New York, Thieme. 2009. The original drawings in this chapter were produced by Christine Gralapp, MA.
REFERENCES 1. Krause F: Zur Freilegung der hinteren Felsenbeinflache und des Kleinhirns. Beitr Klin Chir 37:728-764, 1903. 2. Rhoton A L Jr, Tedeschi H : Microsurgical anatomy of acoustic neuroma. Otolaryngol Clin North Am 25:257294, 1992. 3. Lang J: Clinical Anatomy of the Posterior Cranial Fossa and its Foramina. New York, Thieme, 1991. 4. Lang J Jr, Samii A : Retrosigmoidal approach to the posterior cranial fossa: An anatomical study. Acta Neurochir (Wien) 111:147-153, 1991.
5. Camins M B, Oppenheim J S : Anatomy and surgical techniques in the suboccipital transmeatal approach to acoustic neuromas. Clin Neurosurg 38:567-588, 1992. 6. Silverstein H, Morrell H, Smouha E, Jones R : Combined retrolabretrosigmoid vestibular neurectomy: An evolution in approach. Am J Otol 10:166-169, 1989. 7. Jackler R K, Pitts L H : Selection of surgical approach to acoustic neuroma. Otolaryngol Clin North Am 25:361387, 1992. 8. Selesnick S H, Jackler R K : Clinical manifestations and audiologic diagnosis of acoustic neuromas. Otolaryngol Clin North Am 25:521-551, 1992. 9. Cohen N L , Hammerschlag P, Berg H, Ransohoff J: Acoustic neuroma surgery: An eclectic approach with an emphasis on hearing preservation. Ann Otol Rhinol Laryngol 95:21-27, 1986. 10. Cohen N L : Retrosigmoid approach for acoustic tumor removal. Otolaryngol Clin North Am 25:295-310, 1992. 11. Wiet R J, Teixido M, Liang JG: Complications in acoustic neuroma surgery. Otolaryngol Clin North Am 25:389412, 1992. 12. Glasscock M E III, Hart M J, Vrabec JT: Management of bilateral acoustic neuroma. Otolaryngol Clin North Am 5:449-469, 1992. 13. Shelton C : Hearing preservation in acoustic tumor surgery. Otolaryngol Clin North Am 25:609-621, 1992. 14. Yates PD, Jackler R K, Satar B, et al: Is it worthwhile to attempt hearing preservation in larger acoustic neuromas? Otol Neurotol 24:460-464, 2003. 15. Nassif PS, Shelton C, Arriaga M M : Hearing preservation following surgical removal of meningiomas affecting the temporal bone. Laryngoscope 102:1357-1362, 1992. 16. Blevins N, Jackler R K : Exposure of the lateral extremity of the internal auditory canal via the retrosigmoid approach: A radioanatomic study. Otolaryngol Head Neck Surg 11:81-90, 1994. 17. Hegarty J L , Jackler R K, Rigby PL , et al: Distal AICA syndrome following acoustic neuroma surgery. Otol Neurotol 23:560-571, 2002. 18. Bloch D, Oghalai J S, Jackler R K, Pitts L H : The role of less-than-complete resection of acoustic neuroma. Otolaryngol Head Neck Surg 130:104-112, 2004. 19. Cheung SW, Jackler R K, Pitts L P, Gutin PH : Interconnecting the posterior and middle fossa for tumors which traverse Meckel’s cave. Am J Otol 16:200-208, 1995. 20. Seoane E, Rhoton A L Jr: Suprameatal extension of the retrosigmoid approach: Microsurgical anatomy. Neurosurgery 44:553-560, 1999. 21. Samii M, Tatagiba M, Carvalho G A : Retrosigmoid intradural suprameatal approach to Meckel’s cave and the middle fossa: Surgical technique and outcome. J Neurosurg 92:235-241, 2000. 22. Yingling C D, Gardi J N: Intraoperative monitoring of facial and cochlear nerves during acoustic neuroma surgery. Otolaryngol Clin North Am 25:413-448, 1992. 23. MacDonald C B, Hirsch B E, Kamerer D B, Sekhar L : Acoustic neuroma surgery: Predictive criteria for hearing preservation. Otolaryngol Head Neck Surg 104:128, 1991. 24. Goksu N, Bayazit Y, Kemaloglu Y: Endoscopy of the posterior fossa and dissection of acoustic neuroma. J Neurosurg 91:776-780, 1999.
Chapter 50 • Retrosigmoid Approach to Tumors of the Cerebellopontine Angle 25. Wackym PA, King WA, Poe DS, et al: Adjunctive use of endoscopy during acoustic neuroma surgery. Laryngoscope 109:1193-1201, 1999. 26. Lye R H, Pace-Balzan A, Ramsden RT, et al: The fate of tumour rests following removal of acoustic neuromas: An MRI Gd-DTPA study. Br J Neurosurg 6:195-201, 1992. 27. Thedinger B S, Whittaker C K, Luetje C M : Recurrent acoustic tumor after a suboccipital removal. Neurosurgery 29:681-687, 1991. 28. Mazzoni A, Calabrese V, Danesi G: A modified retrosigmoid approach for direct exposure of the fundus of the internal auditory canal for hearing preservation in acoustic neuroma surgery. Am J Otol 21:98-109, 2000. 29. Arriaga M, Gorum M : Enhanced retrosigmoid exposure with posterior semicircular canal resection. Otolaryngol Head Neck Surg 115:46-48, 1996. 30. Kemink J L , Langman AW, Niparko J K, Graham M D: Operative management of acoustic neuromas: The priority of neurologic function over complete resection. Otolaryngol Head Neck Surg 104:96-99, 1991. 31. Moffat D A : Synopsis on near-total, subtotal, or partial removal: Acoustic neuroma. In Tos M, Thomsen J (eds): Proceedings of the First International Conference on Acoustic Neuroma, Copenhagen, August 25-29, 1991. New York, Kugler Publications, 1992, pp 983-984. 32. Baer S, Tos M, Thomsen J, Hughes G: Synopsis on grading of facial nerve function after acoustic neuroma treatment: Acoustic neuroma. In Tos M, Thomsen J (eds): Proceedings of the First International Conference on Acoustic Neuroma, Copenhagen, August 25-29, 1991. New York, Kugler Publications, 1992, pp 993-995. 33. Sanna M, Gamoletti J, Tos M, Thomsen J: Synopsis on hearing preservation following acoustic neuroma surgery: Acoustic neuroma. In Tos M, Thomsen J (eds): Proceedings of the First International Conference on Acoustic Neuroma, Copenhagen, August 25-29, 1991. New York, Kugler Publications, 1992, pp 985-987. 34. Lalwani A, Butt FY, Jackler R K, et al: Facial nerve outcome after acoustic neuroma surgery: A study from the era of cranial nerve monitoring. Otolaryngol Head Neck Surg 111:561-570, 1994. 35. Hinton A E, Ramsden RT, Lye R H, Dutton J E : Criteria for hearing preservation in acoustic schwannoma surgery: The concept of useful hearing. J Laryngol Otol 106:500503, 1992. 36. Shelton C, Hitselberger WE, House WF, Brackmann D E : Long-term results of hearing after acoustic tumor removal: Acoustic neuroma. In Tos M, Thomsen J (eds): Proceedings of the First International Conference on Acoustic Neuroma, Copenhagen, August 25-29, 1991. New York, Kugler Publications, 1992, pp 661-664. 37. McKenna M J, Halpin C, Ojemann RG, et al: Long-term hearing results in patients after surgical removal of acoustic tumors with hearing preservation. Am J Otol 13:134136, 1992. 38. Slavit D H, Harner SG, Harper C M Jr, Beatty CW: Auditory monitoring during acoustic neuroma removal. Arch Otolaryngol Head Neck Surg 17:1153-1157, 1991. 39. Staecker H, Nadol J B, Ojeman R , et al: Hearing preservation in acoustic neuroma surgery: Middle fossa versus retrosigmoid approach. Am J Otol 21:399-404, 2000.
619
40. Irving R M, Jackler R K, Pitts L H : Hearing preservation in patients undergoing vestibular schwannoma surgery: Comparison of middle fossa and retrosigmoid approaches. J Neurosurg 88:840-845, 1998. 41. Arriaga M A, Chen D A, Fukushima T: Individualizing hearing preservation in acoustic neuroma surgery. Laryngoscope 107:1043-1047, 1997. 42. Wigand M E, Haid T, Goertzen W, Wolf S : Preservation of hearing in bilateral acoustic neurinomas by deliberate partial resection. Acta Otolaryngol (Stockh) 112:237-241, 1992. 43. Mangham C A : Complications of translabyrinthine versus suboccipital approach for acoustic tumor surgery. Otolaryngol Head Neck Surg 99:396-400, 1988. 44. Ebersold M J, Harner SG, Beatty CW, et al: Current results of the retrosigmoid approach to acoustic neurinoma. J Neurosurg 76:901-909, 1991. 45. Haines J H, Maroon JC, Janetta PJ: Supratentorial intracerebral hemorrhage following posterior fossa surgery. J Neurosurg 49:881, 1978. 46. Harders A, Gilbach J, Weigel K : Supratentorial spaceoccupying lesions following infratentorial surgery: Early diagnosis and treatment. Acta Neurochir (Wien) 74:57, 1985. 47. Seiler RW, Zurbrugg H R : Supratentorial intracerebral hemorrhage after posterior fossa operation. Neurosurgery 18:472, 1986. 48. Atkinson J: The anterior cerebellar artery: Its variations, pontine distribution, and significance in the surgery of cerebellopontine angle tumours. J Neurol Neurosurg Psychiatry 12:137-151, 1949. 49. Hegarty JL, Jackler RK, Rigby PL, Pitts LP, Cheung SC: Distal AICA syndrome following acoustic neuroma surgery. Otol Neurotol 23:560-571, 2002. 50. Harner SG, Beatty CW, Ebersold M J: Retrosigmoid removal of acoustic neuroma: Experience 1978-1988. Otolaryngol Head Neck Surg 103:40-45, 1990. 51. Duke D A, Lynch JJ, Harner SG, et al: Venous air embolism in sitting and supine patients undergoing vestibular schwannoma resection. Neurosurgery 42:1282-1286, 1998. 52. Hitselberger WE, House WF: A warning regarding the sitting position for acoustic tumor surgery. Arch Otolaryngol Head Neck Surg 106:69, 1980. 53. Samii M, Turel K E, Penker G: Management of seventh and eighth nerve involvement by cerebellopontine angle tumors. Clin Neurosurg 32:242, 1985. 54. Becker S S, Jackler R K, Pitts L P. CSF Leak after acoustic neuroma surgery: A comparison of the translabyrinthine, middle fossa, and retrosigmoid approaches. Otol Neurotol 24:107-112, 2003. 55. Baird C J, Hdeib A, Suk I, et al: Reduction of cerebrospinal fluid rhinorrhea after vestibular schwannoma surgery by reconstruction of the drilled porus acusticus with hydroxyapatite bone cement. J Neurosurg 107:347-351, 2007. 56. Falcioni M, Romano G, Aggarwal N, Sanna M : Cerebrospinal fluid leak after retrosigmoid excision of vestibular schwannomas. Otol Neurotol 29:384-386, 2008. 57. Smith PG, Leonetti J P, Grubb R L : Management of cerebrospinal fluid otorhinorrhea complicating the retrosigmoid approach to the cerebellopontine angle. Am J Otol 11:178-180, 1990.
620
OTOLOGIC SURGERY
58. Schessel D A, Nedzelski J M, Rowed Feghali JG: Headache and local discomfort following surgery of the cerebellopontine angle: Acoustic neuroma. In Tos M, Thomsen J (eds): Proceedings of the First International Conference on Acoustic Neuroma, Copenhagen, August 25-29, 1991. New York, ������������������������������������������������������������ Kugler Publications, 1992, pp 899-904. 59. Koperer H, Deinsberger W, Jodicke A, Boker D K : Postoperative headache after the lateral suboccipital approach: Craniotomy versus craniectomy. Minim Invasive Neurosurg 42:175-178, 1999. 60. Feghali JG, Elowitz E H: Split calvarial graft cranioplasty for the prevention of headache after retrosigmoid resection of acoustic neuromas. Laryngoscope 108:1450-1452, 1998.
61. Silverman D A, Hughes G B, Kinney S E, Lee J H : Technical modifications of suboccipital craniectomy for prevention of postoperative headache. Skull Base 14:77-84, 2004. 62. Schaller B, Baumann A : Headache after removal of vestibular schwannoma via the retrosigmoid approach: A long-term follow-up-study. Otolaryngol Head Neck Surg 128:387-395, 2003. 63. Beatty CW, Ebersold M J, Harner SG: Residual and recurrent acoustic neuromas. Laryngoscope 97:11681171, 1987.
51
Transotic Approach Ugo Fisch and Joseph M. Chen
The transotic approach to the cerebellopontine angle (CPA) was first introduced in 1979 by one of us (U.F.) in response to the limitations of the translabyrinthine technique. The objective of this approach is to obtain a direct lateral exposure and the widest possible access to the CPA through the medial wall of the temporal bone, from the superior petrosal sinus to the jugular bulb, and from the internal carotid artery to the sigmoid sinus. The tympanic and mastoid portions of the fallopian canal are left in situ. This transtemporal access is achieved at the expense of bony exenteration, rather than cerebellar retraction. Despite well-documented technical details,1 there is a general misconception equating the transotic approach with the transcochlear approach of House and Hitselberger.2 Significant differences exist between the two approaches in the extent of exposure, the management of the facial nerve, and the obliteration of the surgical cavity. As a natural extension of subtotal petrosectomy, which forms the basis of lateral and posterior skull base surgery at the University of Zurich,1 the trans otic approach was initially designed for acoustic neuromas and has since expanded to include other pathology. Several modifications were also made over the years to optimize its use.3-5
INDICATIONS Acoustic Neuroma Although the transotic approach, similar to the translabyrinthine approach, can be used for tumors of all sizes, it is ideal for tumors 2.5 cm or less in their mediolateral extent, in patients with no serviceable hearing. In this clinical setting, the transotic approach offers the best possible exposure for tumor extirpation and preservation of facial nerve with minimal morbidity. At the University of Zurich, tumors larger than 2.5 cm that cause significant brainstem compression are managed by the neurosurgery department as a matter of departmental policy. Small intracanalicular tumors in patients with good hearing (using the 50/50 rule of at least
50 dB hearing loss and 50% discrimination score) are managed through a middle cranial fossa (transtemporalsupralabyrinthine) approach (see Chapter 35).
Other Lesions Other lesions involving the CPA or the temporal bone with invasion of the internal auditory canal (IAC) or the otic capsule could also be approached via the transotic technique. These lesions include the following:
• Epithelial cysts (congenital cholesteatoma) • Arachnoid cysts • Hemangiomas • Giant cholesterol and mucosal cysts • Jugular foramen schwannomas • Temporal paragangliomas (glomus tumors) These lesions can be quite extensive and may require a combined infratemporal fossa type A or B approach for added exposure.
PREOPERATIVE EVALUATION The evaluation for retrocochlear lesions, such as acoustic neuromas, is standard at the University of Zurich and includes routine audiometry, auditory brainstem response, electronystagmography, and magnetic resonance imaging (MRI) with gadolinium enhancement. High-resolution computed tomography (CT) is still performed for bony assessment of lesions within or invading the temporal bone. Facial nerve status is recorded clinically using the Fisch grading system6 and quantified by electroneuronography before surgery. In addition to a candid discussion of surgical and postoperative complications, patients are informed of the advantages of the transotic approach specifically with regard to the preservation of the facial nerve and the complete obliteration of the surgical cavity, with blind sac closure of the external auditory canal to minimize cerebrospinal fluid (CSF) leak. The option 621
622
OTOLOGIC SURGERY
of conservative management by close monitoring with serial MRI to follow tumor growth is presented to all patients and is recommended for patients with nonprogressive long-standing symptoms (especially elderly patients), patients with small tumors and normal hearing, patients with significant medical illnesses, or patients who refuse to undergo surgery. These patients are told that rapid tumor growth ultimately requires surgical attention, and that facial nerve function and hearing preservation may be compromised as a result of the delay in surgery.
Skin Incision
SURGICAL TECHNIQUE
A mastoid periosteal flap is developed while the postauricular skin flap is elevated. The external auditory canal is transected, and its skin is elevated, everted externally, and closed as a blind sac. A second layer of closure using the mastoid periosteal flap ensures a complete seal (Fig. 51-2).
Preoperative Preparation The patient is premedicated with clonidine, metoclopramide, and midazolam before surgery. A perioperative antibiotic, ceftriaxone (Rocephin), 2 g intravenously for 24 hours, is given at the time of surgery until the removal of intravenous infusion, usually by the 3rd day after surgery.
Surgical Site Preparation, Positioning, and Draping The surgical site is prepared in the operating room after induction of anesthesia. Hair over the temporal area is shaved 9 cm above and 5 cm behind the pinna. The skin is washed with povidone-iodine. The abdomen and the contralateral leg are also shaved and prepared for fat harvesting and the possible need of a sural nerve graft. The positioning and draping for this procedure are similar to positioning and draping described in Chapter 1, with some minor differences. The patient is secured in supine position on the Fisch operating table (see Chapter 35), with the head turned away from the surgeon. A large plastic bag is incorporated into the draping to catch excess irrigation and blood.
Intraoperative Monitoring and Concerns Intraoperative facial nerve monitoring using the Xomed nerve integrity monitor (NIM-II) and percutaneous electromyography needles is standard with this approach. Intracranial pressure is controlled by deep anesthesia induced intravenously before introduction of inhalational anesthetics. Partial pressure of carbon dioxide is maintained at 30 to 40 mm Hg. Pharmacologic manipulation with dexamethasone (Decadron), 4 mg every 8 hours perioperatively and 4 days postoperatively, and mannitol, 0.5 mg/kg intravenously intraoperatively, is also standard. Furosemide is added when necessary. Lumbar CSF drainage is not routinely performed. Hypotensive anesthesia with nitroglycerin or clonidine (Catapres), or both, is used in most cases to maintain a systolic blood pressure of 80 to 100 mm Hg.
A postauricular incision is placed along the hairline to keep it behind the operative cavity (Fig. 51-1). The incision is made from the mastoid tip to the temporal region for the surgical approach; its superior extension is made at the time of wound closure for the exposure of the temporalis muscle flap.
Blind Sac Closure of External Auditory Canal
Subtotal Petrosectomy: Exposure of Jugular Bulb and Petrous Carotid A complete mastoidectomy is performed, and the remaining external auditory canal skin, tympanic membrane, and ossicles are removed in a stepwise fashion. The tympanic bone is progressively thinned out, and a complete exenteration of the pneumatic spaces (retrofacial, retrolabyrinthine, supralabyrinthine, hypotympanic, infralabyrinthine, and pericarotid) is carried out. Figure 51-3 shows the surgical cavity at the completion of this step. The middle fossa dura, sigmoid sinus, and jugular bulb are blue-lined; the fallopian canal and the vertical portion of the petrous carotid artery are skeletonized. The mastoid tip is removed to reduce the depth of the surgical cavity.
Obliteration of Eustachian Tube The mucosa of the membranous eustachian tube orifice is coagulated, and the bony canal is obliterated with bone wax at the isthmus. An additional muscle plug is used before closure.
Exenteration of Otic Capsule With the completion of subtotal petrosectomy, the surgical cavity is divided into two compartments by the fallopian canal (Fig. 51-4). Because the enlarged IAC lies mostly deep within the anterior compartment, the advantage of the transotic approach to access this region fully is clear. To begin this step, the semicircular canals are removed, and the vestibule is opened as in the translabyrinthine approach. The posterior aspect of the IAC is exposed from the fundus to the porus, leaving a thin layer of bone over the meatal dura. The posterior fossa dura of the posterior compartment is exposed caudal to the
623
Chapter 51 • Transotic Approach BLIND SAC CLOSURE OF THE EXTERNAL AUDITORY CANAL AFTER EVERSION OF THE CANAL SKIN
Vertical internal carotid a.
FIGURE 51-1. Skin incision. FIGURE 51-2. A and B, Blind sac closure of external auditory canal. FIGURE 51-3. Subtotal petrosectomy. EAC, external auditory canal. FIGURE 51-4. Position of internal auditory canal in relation to anterior and posterior compartments of operative cavity.
624
OTOLOGIC SURGERY
superior petrosal sinus and anterior to the sigmoid sinus. Retrofacial cells are subsequently removed to gain access over the inferior aspect of the IAC. Attention is now focused on the anterior compartment. The cochlea is drilled away to expose the enlarged IAC, which lies predominantly within this compartment, and as the dissection is carried forward, the dura anterior to the porus is also exposed to the level of the vertical segment of the ICA. Bony reduction between the jugular bulb and the inferior aspect of the IAC requires working beneath and over the fallopian canal, which is left in its anatomic position across the surgical field. Sufficient bone is left surrounding the canal to prevent accidental fracture. The cochlear aqueduct is identified between the jugular bulb and the IAC. The arachnoid of the aqueduct is opened to allow the outflow of CSF, decompressing the lateral cistern before the posterior fossa dura is opened. The tensor tympani muscle and bone medial to it are removed to gain more anterior access; likewise, bone medial to the vertical carotid artery is removed as much as possible.
Unroofing of Labyrinthine Portion of Facial Nerve Since 1988, the unroofing of the labyrinthine segment of the facial nerve from the meatal foramen to the geniculate ganglion has been incorporated as a standard step in the transotic approach. The meatal foramen can be found approximately 2 mm anterior and superior to the meatal fundus. This exposure provides additional room for the portion of the facial nerve most likely to sustain traction injury and edema subsequent to tumor manipulation. Also, the labyrinthine portion of the facial nerve serves as an important landmark while further access over the porus is gained along the superior petrosal sinus. The completed exposure is shown in Figure 51-5. The posterior fossa dura surrounding the porus is circumferentially exposed from the carotid artery to the sigmoid sinus, and from the jugular bulb to the level of the superior petrosal sinus.
Tumor Removal A few instruments are required for tumor removal. Bayonet and angled bipolar forceps, cup forceps, microraspatories, and a long suction with finger control are the most essential instruments. The intrameatal portion of the tumor is approached first and is separated from the facial nerve until the level of the porus. Figure 51-6 illustrates the advantage of the additional space obtained with the transotic exenteration, whereby the intrameatal portion of the tumor can be easily displaced and mobilized during its removal. The posterior fossa dura is incised between the sinodural angle and the posterior edge of the porus. The incision is extended superiorly and inferiorly along the
porus (Fig. 51-7). It is important to elevate the dura with a hook before making an incision to prevent the inadvertent injury of vessels over the cerebellum. The dural edges must be cauterized before extending the incision to facilitate hemostasis. One must also be acutely aware of the variations of the course of the anterior inferior cerebellar artery (AICA) and its branches. The superior and inferior dural flaps are retracted with 4-0 polyglactin 910 (Vicryl) sutures, which are clipped to the wound edges (Fig. 51-8). The full extent of the tumor usually can be shown. The posterior pole of the tumor abuts against the cerebellum and the petrosal vein, and the AICA courses anteroinferior to the tumor. Intracapsular reduction of the tumor can now begin and is continued until tumor margins can be seen without tension being placed on the facial nerve. During this step, the meatal dura at the superior pole of the porus is not detached to render some stability to the tumor. It is crucial to handle the tumor meticulously; manipulations should be carried out with suction over a Cottonoid, and the displaced facial nerve should always be in view to avoid undue traction (Fig. 51-9). Bleeding is diminished by coagulation of all visible vessels over the tumor capsule. The main blood supply to the tumor generally runs along CN VIII, and some may come from branches of the AICA. These vessels should be coagulated, cut on the tumor, and gently pulled away. With sufficient reduction, separation of the facial nerve can now be attempted. The advantage of the transotic approach is now easily appreciated because the displaced facial nerve can be followed in its entirety. The dural attachments of the tumor at the porus are cut, and the nerve can be gently grasped with bipolar forceps and teased away from the tumor (Fig. 51-10). Likewise, the AICA can be separated from the nerve by using the tips of the forceps or by pulling on the coagulated branches. At the inferior pole of the tumor, the origin of CN VIII and the course of the AICA looping around it are identified. In many instances, the root exit zone of the facial nerve, always anterior to CN VIII, is identified only after CN VIII is cut. The anterior access of the transotic approach offers an unparalleled view to an area that is usually partially hidden from the surgeon during suboccipital or translabyrinthine surgery.7,8 In this exact region, the facial nerve is most tenuous and frequently appears as a thin, transparent band. Any manipulation not under direct vision can easily rupture the nerve. When it is completely detached from all vital structures, the tumor can now be removed. The CPA and all its structures are exposed in Figure 51-11. The facial nerve is stimulated electrically to obtain a threshold response. Despite a normal response intraoperatively, the patient may still show an immediate or delayed facial paralysis because of impaired vascular supply and the inevitable trauma to the nerve during dissection. If stimulation fails to produce a response or facial contraction, and if the anatomic integrity of the nerve is precarious,
Chapter 51 • Transotic Approach
625
FIGURE 51-5. A and B, Exposure of internal auditory canal and posterior fossa dura. FIGURE 51-6. Intrameatal tumor dissection.
it is best to proceed with nerve grafting immediately. Failure to do so while waiting for the improbable return of facial function may delay reinnervation for 2 years. The details of intracranial/intratemporal and hypoglossal/facial crossover grafting techniques are beyond the scope of this chapter and are discussed elsewhere.1,9
Wound Closure A musculofascial graft that is slightly larger than the dural defect is taken from the temporalis muscle. It is placed under the dura and fixed in place with the two 4-0 Vicryl sutures used previously as stay sutures (Fig. 51-12). These sutures are passed through the edges of the graft and secured to the dura. A second temporalis fascial graft
is used to cover the opened IAC. A small muscle graft is also used as a plug to supplement the prior wax obliteration of the eustachian tube. Both grafts are stabilized with fibrin glue. A second layer of closure with abdominal fat grafts is to follow. A large piece of fat is first passed under the fallopian canal and firmly anchored (Fig. 51-13). Several small pieces of fat are used to fill out the surgical cavity and are stabilized with fibrin glue. The posterior half of the temporalis muscle is now transposed and sutured in place with 2-0 Vicryl sutures. Additional fat is placed under the muscle flap to create a slight compressive tension (Fig. 51-14). This type of closure has consistently minimized the incidence of postoperative CSF leaks, which is another advantage of
626
OTOLOGIC SURGERY DURAL INCISIONS
Facial nerve
INITIAL CPA EXPOSURE
V Suction
FIGURE 51-7. Dural incisions. CSF, cerebrospinal fluid; PFD, posterior fossa dura. FIGURE 51-8. Initial cerebellopontine angle (CPA) exposure. AICA, anterior inferior cerebellar artery. FIGURE 51-9. Intracapsular tumor reduction. FIGURE 51-10. Intracranial facial nerve dissection. AICA, anterior inferior cerebellar artery. FIGURE 51-11. View of cerebellopontine angle after tumor removal. AICA, anterior inferior cerebellar artery.
Chapter 51 • Transotic Approach
627
DURAL CLOSURE FASCIA
TEMPORALIS MUSCLE FLAP FAT OBLITERATION OF THE SURGICAL CAVITY
FIGURE 51-12. Dural closure. IAM, internal auditory meatus. FIGURE 51-13. Fat obliteration of surgical cavity. FIGURE 51-14. Temporalis muscle flap. SCM, Sternocleidomastoid. FIGURE 51-15. Cross-sectional views of transotic approach versus translabyrinthine approach. AN, acoustic neuroma; C, carotid; CER, cerebellum; CVN, cochleovestibular nerve; EAM, external auditory meatus; FN, facial nerve; ME, middle ear; SS, sigmoid sinus.
628
OTOLOGIC SURGERY
the transotic approach. A small plastic suction drain is inserted over the muscle flap while the skin incision is closed in two layers with 2-0 polyglycolic acid (Dexon) and 3-0 nylon sutures.
DRESSING AND POSTOPERATIVE CARE The suction drain is removed as soon as a compression dressing is applied. The dressing is left in place for 5 days, and if there is any evidence of CSF leak or subcutaneous CSF accumulation, the compression dressing is reapplied. The patient is transferred to the postanesthesia care unit extubated and fully awake. Routine and neurologic vital signs are closely monitored every 30 to 60 minutes. Adequate analgesics and antiemetics are ordered to keep the patient comfortably at bed rest, usually for 72 hours after surgery. Ambulation and oral intake are started slowly thereafter. Subcutaneous heparin is often given during the early convalescent period. An oral antibiotic, trimethoprim and sulfamethoxazole (Bactrim Forte) or ciprofloxacin (Ciproxin), is prescribed for at least 5 days after intravenous fluid and ceftriaxone therapy are discontinued. Head and abdominal wound sutures are removed on day 12, and leg sutures are removed after 2 weeks. The patient is discharged from the hospital at this time, barring any complication.
TIPS AND PITFALLS The transotic approach is more than a combination of the translabyrinthine and transcochlear approaches. It uses the complete infralabyrinthine compartment of the temporal bone, from the carotid artery to the sigmoid sinus, and from the jugular bulb to the superior petrosal sinus. It provides the largest possible transtemporal access to the CPA, which can be best appreciated by comparing the cross-sectional surgical exposure of the transotic approach with the translabyrinthine approach shown in Figure 51-15. The preservation of the facial nerve in its anatomic position within the fallopian canal does not limit the visibility or illumination. Enough bone must be kept surrounding the canal initially during subtotal petrosectomy and subsequently while the otic capsule is exenterated. Skeletonization of the fallopian canal must be done progressively as the surgical cavity enlarges, and with only diamond burrs. Bone surrounding the proximal tympanic segment of the facial nerve and the superior aspect of the IAC should be left intact to support the facial nerve. If the fallopian canal is inadvertently fractured during dissection, it is likely to remain undisplaced and not to impede surgery. If no significant torsion or traction has occurred, no major adverse effects should result,
rovided that no further manipulation occurs. The fracp tured edges can be supported with fibrin glue, and at the time of closure, abdominal fat adequately renders support from beneath. If the fracture is displaced and unstable, a small, malleable aluminum strip can be used as a retractor and can maintain the fallopian canal in position. If this is impossible, the nerve may have to be fully unroofed and transposed anteriorly as in the infratemporal fossa type A approach. This situation has not occurred in our hands, however. The three fundamental principles for the removal of acoustic neuromas are as follows: 1. Perform intracapsular reduction of the tumor to gain better visibility progressively around the circumference of the tumor 2. Expose tumor from lateral to medial, coagulate all visible vessels over the tumor capsule, and remove the devascularized portion in a piecemeal fashion 3. Separate the facial nerve from the tumor and not vice versa CSF outflow after the opening of the cochlear aqueduct decompresses the lateral cistern before the dural incision is made. It also indicates that the pars nervosa of the jugular foramen and the lower cranial nerves have not yet been reached by the tumor. Even in the presence of a high jugular bulb, a few millimeters of exposure can be obtained between it and the IAC to allow adequate access to the inferior pole of the tumor. Unroofing and compressing the jugular bulb to gain more exposure are unnecessary and dangerous. If access to the CPA is severely limited by the jugular bulb, the facial nerve can be unroofed and transposed anteriorly, as in the infratemporal fossa type A approach. Bony exenteration to expose the posterior fossa dura should be done in a stepwise manner, gaining as much exposure of the posterior fossa dura around the porus as possible. The time spent in the initial bony exenteration may be tedious at first, but it is well rewarded by much expanded access and improved illumination, which facilitate tumor removal dramatically. The dura should not be opened until all bony work has been completed, and hemostasis has been perfectly controlled. While incising the dura, the surgeons should beware of the AICA, which may loop underneath, and make the initial cut in the center of the exposure to avoid this artery, which usually lies in the inferior half of the CPA. Intracapsular reduction of the tumor is performed while it is still attached to the meatal dura at the superior pole of the porus to prevent excessive traction on the facial nerve. It also stabilizes the tumor during reduction. The major blood supply of acoustic tumors runs along CN VIII; one should always check for vessels on the undersurface of the tumor before removal. The most delicate portion of the facial nerve is just proximal to the acoustic porus, where the nerve can be flattened to a thin
629
Chapter 51 • Transotic Approach
COMPLICATIONS AND MANAGEMENT
transparent band. It is most frequently pushed anteriorly and superiorly. The surgeon must keep an eye on this area while working deeper in the CPA. If the facial nerve appears to be significantly traumatized and cannot be stimulated at the end of surgery, one must proceed directly to an intracranial-intratemporal grafting or CN XII-VII cross-innervation procedure, depending on the clinical setting. Spontaneous return of function with conservative treatment is not likely to occur. Posterior fossa dura should not be resected to gain exposure; the dural edges are retracted with stay sutures. One should not forget to obliterate the eustachian tube with bone wax and a muscle plug to prevent CSF rhinorrhea. Fat graft anchored under the fallopian canal supports the musculofascial repair and prevents lateralization of these autogenous tissues.
The obliteration of the surgical cavity and eustachian tube, along with the blind sac closure of the external auditory canal, has significantly reduced the incidence of CSF leak. In the series mentioned earlier, 4% of patients developed a subcutaneous CSF collection without leakage, and the collections usually resolved within 3 to 4 weeks with conservative management, including bed rest and prolonged compressive dressing over the surgical site. Three patients (4%) developed either immediate or delayed CSF leaks and were treated with lumbar drainage and bed rest; only one required surgical revision. One patient had meningitis and responded to antibiotic treatment without sequelae. Most notably is the lack of any other central nervous system complication in this series. One death occurred as a result of postoperative pulmonary embolism and cardiorespiratory failure. Wound infection with necrosis of the abdominal fat graft or temporalis muscle flap was a rare complication and was thought to be the inciting cause in the case of meningitis.
RESULTS Between 1979 and 1990, 147 consecutive transotic app roaches were performed for the removal of unilateral acoustic neuromas by the senior author (U.F.). Tumors in this series were limited in size from 1 to 2.5 cm in mediolateral extension. Tumors may fill the lateral cistern, abutting but without significantly compressing the brainstem. Complete tumor removal was achieved in all cases. Optimal visualization of the facial nerve was obtained, and the anatomic preservation of the nerve was possible in 139 cases (94.6%). In eight cases in which facial nerve integrity could not be preserved, intracranialintratemporal nerve grafting resulted in an average of 66% return of facial function by the Fisch facial nerve grading system (or House-Brackmann grade III or better), during a mean follow-up of 4 years. In this series, 66 patients were available for followup of at least 2 years. Tumors 1 to 1.4 cm were removed with no incidence of permanent facial injury, and 80% of patients with tumors 1.5 to 2.5 cm had normal or near-normal facial function (Table 51-1). Since 1988, the unroofing of the labyrinthine portion of the facial nerve has been a standard step in the transotic approach, which is believed to be a major contributing factor in the diminished incidence of delayed facial palsy in acoustic neuroma surgery.
ALTERNATIVE TECHNIQUES When and how acoustic neuromas should be operated on are issues of ongoing and often emotional debates. If surgery is contemplated, the aim is to try to obtain the safest and best possible exposure that would allow complete tumor extirpation and the preservation of facial nerve. Hearing preservation is of secondary concern if the opposite ear is functional. Translabyrinthine and suboccipital approaches are perhaps the most established and popular techniques, whereas the middle fossa approach has traditionally been reserved for small tumors in patients with serviceable hearing; an extended version of the middle fossa approach has gained popularity in some centers to remove tumors measuring 4.5 cm10,11 despite a seemingly high morbidity.12 The relative efficacy of each of these approaches is difficult to quantify without a randomized multi-institutional study. Our own experience with the translabyrinthine removal of acoustic neuromas before 1979 was unsatisfactory in many respects, and the problems were subsequently rectified with the transotic approach.1
TABLE 51-1 Facial Function Two Years Postoperatively % RECOVERY Tumor Size (cm) 1-1.4 1.5-2.5* Total *No
No.
100
80-99
60-79
40-59
0-39
14 52 66
100 61 70
— 19 15
— 12 9
— 4 3
— 4 3
difference in facial function after removal of tumors of 1.5-1.9 cm versus 2-2.5 cm.
630
OTOLOGIC SURGERY
There are three advantages of the transotic approach over the translabyrinthine approach, as follows: 1. A wider surgical access with a near-circumferential exposure of the IAC and the porus acusticus; this added exposure is particularly important in the presence of a high-riding jugular bulb and an anteriorly positioned sigmoid sinus 2. The direct visualization and access to the anterior CPA where the facial nerve is most tenuous and vulnerable 3. A much reduced rate of CSF leakage as a result of permanent closure of the ear canal and eustachian tube and complete obliteration of the surgical cavity
REFERENCES 1. Fisch U, Mattox D: Microsurgery of the Skull Base. New York, Thieme Publishers, 1988. 2. House WF, Hitselberger WE : The transcochlear approach to the skull base. Arch Otolaryngol Head Neck Surg 102:334-342, 1976. 3. Jenkins H A, Fisch U: The transotic approach to resection of difficult acoustic tumors of the cerebellopontine angle. Am J Otol 2:70-76, 1980.
4. Gantz B J, Fisch U: Modified transotic approach to the cerebellopontine angle. Arch Otolaryngol Head Neck Surg 109:252-256, 1983. 5. Chen J M, Fisch U: The transotic approach in acoustic neuroma surgery. J Otolaryngol 22:331-336, 1993. 6. Burres S, Fisch U: The comparison of facial grading systems. Arch Otolaryngol Head Neck Surg 112:755-758, 1986. 7. Whittaker C K, Leutje C M : Translabyrinthine removal of large acoustic neuromas. Am J Otol 7(Suppl):155-160, 1985. 8. Gardner G, Robertson J H : Transtemporal approaches to the cranial cavity. Am J Otol 7(Suppl):114-120, 1985. 9. Fisch U, Lanser M J: Facial nerve grafting. Otolaryngol Clin North Am 24:691-708, 1991. 10. Wigand M E, Haid T: Extended middle cranial fossa approach for acoustic neuroma surgery. Skull Base Surg 1:183-187, 1991. 11. Kanzaki J, Ogawa K, Yamamoto M, et al: Results of acoustic neuroma surgery by the extended middle cranial fossa approach. Acta Otolaryngol Suppl (Stockh) 487:1721, 1991. 12. Kanzaki J, Ogawa K, Tsuchihashi N, et al: Postoperative complications in acoustic neuroma surgery by the extended middle cranial fossa approach. Acta Otolaryngol Suppl (Stockh) 487:75-79, 1991.
52
Transcochlear Approach to Cerebellopontine Angle Lesions Antonio De la Cruz and Karen B. Teufert Videos corresponding to this chapter are available online at www.expertconsult.com.
The transcochlear approach was developed to treat midline intracranial lesions arising from the clivus, and cerebellopontine angle (CPA) masses arising anterior to the internal auditory canal (IAC), without requiring the use of brain retractors. These lesions may extend around the vertebrobasilar arteries. Because in the 1970s traditional surgical approaches were limited by the cerebellum and the brainstem, these lesions had been considered inoperable by many surgeons. The transcochlear approach does not have these limita tions, and was designed primarily for meningiomas arising from the petroclinoid ridge, intradural clivus lesions, chordomas, congenital petrous apex cholestea tomas, and primary intradural epidermoids anterior to the IAC. The transcochlear approach evolved from the inabil ity to excise the base of implantation and control the blood supply of these near-midline and midline tumors with other surgical approaches. Total removal of these lesions through a suboccipital approach is often impos sible because of the interposition of the cerebellum and the brainstem.1,2 The transpalatal-transclival approach was attempted for these intradural midline lesions, with little success, during the early 1970s.3 The exposure was often inadequate; the field is relatively far from the surgeon; the blood supply is lateral, away from the sur geon’s view; and the risk of intracranial complications caused by oral contamination is increased. The retro labyrinthine approach is limited in its forward exten sion by the posterior semicircular canal. Tumor access with the translabyrinthine approach is limited anteri orly by the facial nerve, which impedes removal of the tumor’s base of implantation, which is anterior to the IAC, around the intrapetrous carotid artery, or ante rior to the brainstem. The development of the extended middle fossa approach and combined transpetrous approach enables complete removal of petroclinoid meningiomas, and is used in patients with useful hear ing.4-6 The primary limitation with this approach is poor access to tumors with inferior or midline extensions.7,8 The endoscopic-assisted transsphenoidal approach has
been gaining acceptance as an alternative approach to access midline intracranial lesions arising from the clivus and petrous apex lesions (Stamm A, personal communication, 2008). The transcochlear approach was developed by House and Hitselberger2,3 in the early 1970s as an anterior extension of the translabyrinthine approach. It involves rerouting of the facial nerve posteriorly and the removal of the cochlea and petrous apex, which exposes the area of the intrapetrous internal carotid artery. This approach affords wide intradural exposure of the anterior CPA; CN V, VII, VIII, IX, X, and XI; both sixth cranial nerves; the clivus; and the basilar and vertebral arteries, without using any brain retractors. The contralateral cranial nerves and the opposite CPA are also visible.9 This wide exposure affords removal of the tumor base and its arterial blood supply from the internal carotid artery, which is particu larly important in the treatment of meningiomas.2 Adding excision and closure of the external auditory canal (EAC), as advocated by Brackmann (personal communication, 1987), increases further the anterior exposure for lesions of the petroclival regions and prepontine cistern. Smaller lesions can be reached without rerouting the facial nerve (transotic approach).
ADVANTAGES OF TRANSCOCHLEAR APPROACH This approach requires no cerebellar or temporal lobe retraction. Exposure and dissection of the petrous apex and clivus allows excellent exposure of the midline and complete removal of the tumor, its base of implantation, and its blood supply. This is of particular importance in meningiomas.10,11 Careful handling and constant monitoring of the facial nerve during rerouting prevent injury to the intra temporal portion of the nerve. Cholesteatomas tend to wrap themselves around it. If the facial nerve is lost dur ing tumor removal, we recommend immediate repair by end-to-end anastomosis or nerve graft interposition. 631
632
OTOLOGIC SURGERY
DISADVANTAGES OF TRANSCOCHLEAR APPROACH The main disadvantages of this approach are sacrifice of residual hearing in the operated ear and risk of temporary facial palsy. This technique is indicated when no service able hearing exists in the involved ear, or when the tumor is too far anterior for the extended middle fossa crani otomy approach or transpetrous approach. With the use of continuous facial nerve monitoring, the incidence of permanent facial nerve paralysis is low.
PATIENT EVALUATION AND PREOPERATIVE COUNSELING Individuals with tumors that require transcochlear surgery may have minimal symptoms, with tumors that can be quite large at the time of diagnosis.1,12 Petrous apex epidermoids manifest with unilateral hearing loss and tin nitus in 80% of the cases. Facial twitch is also common. Imbalance, ataxia, and parietal or vertex headaches may be the only complaints in 20% of patients.13 Patients with meningiomas and intradural epidermoids may be nearly symptom-free until they present with CN V findings and signs of increased intracranial pressure.14,15 There is a high rate of jugular foramen syndrome in patients with meningiomas.1 Seizures, dysarthria, and late signs of dementia from hydrocephalus were common presenting symptoms in the past.13 Hearing and vestibular functions are frequently normal, and acoustic reflex decay or abnor mal auditory brainstem response audiometric results may be the only anomalies.1 Imaging using high-resolution computed tomography (CT) with contrast enhancement, magnetic resonance imaging (MRI), and magnetic resonance angiography (MRA) is essential for diagnosis and surgical planning.16 Petrous apex and intradural epidermoids are expansile, spherical, or oval lesions, with scalloped bone edges on CT. They are isodense to cerebrospinal fluid on CT, with capsular enhancement. On MRI, they are hypoin tense on T1-weighted images and hyperintense on T2weighted images. Meningiomas enhance with contrast on MRI, and manifest with a “dural tail.” Evaluation of blood supply of some tumors may also require MRA. In tumors surrounding or invading the intrapetrous carotid artery, patency of the circle of Willis is accessed, and carotid residual pressure is measured; preopera tive balloon occlusion of the carotid artery and selec tive embolization are performed 1 day before surgery. Radioisotope, xenon, or positron emission tomography studies are used to assess cerebral perfusion during occlusion studies. The natural history of petrous apex epidermoids is that they grow slowly and may produce cranial neu ropathies; they may also become infected. Treatment is difficult when infection occurs, and meningitis, sepsis,
and death may result. Intracranial epidermoids spread through the cisterns and subarachnoid planes to neigh boring regions, including the opposite CPA. Petrous ridge meningiomas grow and are space-occupying lesions that increase intracranial pressure. After surgery, intracranial pressure is reduced, and cranial nerve symptoms tend to improve. Risks and complications in the immediate postoperative period include transient vertigo; com plete hearing loss; temporary CN VII, IX, X, XI, and XII paresis; infection; bleeding; swallowing difficulties; aspiration pneumonia; cerebrovascular accidents; and, rarely, death.
SURGICAL ANATOMY Intracranial structures that may be exposed by the trans cochlear approach include the entire lateral aspect of the pons and upper medulla, CN V through XI, and the mid basilar artery. Posterior fossa exposure is extensive except inferiorly, where it is limited in the area of the jugular foramen and foramen magnum. The degree to which the neural compartment of the jugular foramen is visible depends on the height of the jugular bulb. Modifications to the transcochlear approach permit identification of the anterior aspect of the pons and both sixth cranial nerves, and improved identification of the basilar artery and vertebrobasilar junction.
SURGICAL TECHNIQUE A wide mastoidectomy and labyrinthectomy are per formed, exposing the IAC. When first described by House and Hitselberger in 1976,3 the tympanic ring was not removed, and only the facial recess was opened to permit anterior exposure. Brackmann, after Fisch, modified the approach by removing the entire tympanic ring, malleus, incus and stapes, and blind-sac closing the EAC.17,18 The facial nerve is completely skeletonized, with transection of the greater superficial petrosal and chorda tympani nerves, and is rerouted posteriorly out of the fallopian canal. The cochlea and the fallo pian canal are completely drilled out, and the internal carotid artery is skeletonized. A large triangular win dow is created into the skull base. Its superior bound ary is the superior petrosal sinus; inferiorly, it extends below and medial to the inferior petrosal sinus into the clivus. Anteriorly lies the region of the intrapetrous internal carotid artery, and the apex of the triangle is just beneath Meckel’s cave. When the dura is opened, this window gives excellent direct access to the midline without need of any retraction (Fig. 52-1). After tumor removal, the dura is reapproximated with dural silk, the eustachian tube is packed with absorbable knitted fab ric (Surgicel) and muscle, and abdominal fat is used to
Chapter 52 • Transcochlear Approach to Cerebellopontine Angle Lesions
633
I.C.A.
T. B.a.
EAC
C.N. VII, VIII
FIGURE 52-1. Transcochlear approach. Exposure obtained at skull base at comple tion of approach. Ba, basilar artery; EAC, external auditory canal; ICA, internal carotid artery; SPS, superior petrosal sinus; SS, sigmoid sinus; T, tumor.
S. S. SPS
fill the dura and mastoidectomy defects and to cushion the facial nerve.
Setup General endotracheal anesthesia with direct arterial blood pressure monitoring is used, and a urinary catheter and a nasogastric tube are inserted. Long-acting muscle relaxants are avoided. Intraoperative nerve monitoring is used in all cases. Anesthesia is kept light so that changes in blood pres sure and pulse brought about by tumor manipulation are not masked. Prophylactic third-generation cephalosporin antibiotics and steroids are used routinely before the skin incision is made. Venous antiembolism compression boots are placed on the patient’s legs before the procedure begins. The patient is placed supine on the operating room table, with the head turned to the opposite side, and is maintained in a natural position without fixation. This position avoids air embolization, minimizes surgeon fatigue, and allows stabilization of the surgeon’s hands during the microsurgical procedure.
Incision A postauricular suboccipital incision is made 5 cm behind the postauricular fold, starting 1 cm above the ear, extend ing through the occipital bone and ending at the level of the mastoid tip. This incision can be extended inferiorly into the neck to provide control of the great vessels and of the lower cranial nerves, if necessary. The scalp flap is lifted anteriorly uncovering the temporalis fascia. The periosteum is incised just above the linea temporalis from the zygomatic root anteriorly to a level posterior to the sigmoid sinus. A second periosteal incision perpendicular to the previous one is carried inferiorly in the direction of the mastoid tip. The periosteal flap is elevated forward to the spine of Henle and to the level of the EAC. The skin of the EAC is usually left in place. In some cases
of lesions placed far anteriorly, removal of the tympanic bone and blind closure of the EAC are necessary. In this case, the skin, tympanic membrane, malleus, and incus are removed, and the meatus is closed in three layers.
Mastoidectomy An extended mastoidectomy (Fig. 52-2) is carried out with microsurgical cutting and diamond burrs and con tinuous suction-irrigation. Bone removal is started along two lines: one along the linea temporalis and another tangential to the EAC. The mastoid antrum is opened, and the lateral semicircular canal is identified. The lat eral semicircular canal is the most reliable landmark in the temporal bone, and allows the dissection to proceed toward delineating the vertical fallopian canal and the osseous labyrinth. The opening of the mastoid cavity must be as large as possible and is extended posterior to the sigmoid sinus, exposing 1 to 2 cm of suboccipital dura. The larger the tumor, the further back the posterior fossa dura is exposed, to a maximum of 2 to 3 cm. Mastoid emissary veins are dissected, and bleeding is controlled using bipolar cautery and Surgicel packing. Removal of bone over the sigmoid sinus is performed with diamond burrs, and an island of bone (Bill’s island) may be left over the dome of the sinus initially. This eggshell of bone protects the sinus from being injured by the shaft of the burr. Bone is removed from the sinodural angle along the superior petrosal sinus. The mastoid air cells are exenterated from the sinodural angle, skeletonizing the dura of the posterior and the middle fossae.
Closure of External Auditory Canal When further anterior exposure is required, removal and three-layer closure of the EAC are included. The canal skin is transected at the bony-cartilaginous junction
634
OTOLOGIC SURGERY Facial recess
EAC skin
MFD
PCW .
Fn
B. I.
FIGURE 52-3. Exposure of internal auditory canal, skeletonization of facial nerve from internal auditory canal to stylomastoid foramen, and extended facial recess. Labyrinthectomy has been completed. PFD bone Dura
FIGURE 52-2. Mastoidectomy. There is wide exposure of poste rior and middle fossa dura, with identification of bony labyrinth and s keletonization of sigmoid sinus. Bill’s island (BI) is preserved initial ly, then removed after internal auditory canal skeletonization. EAC, external auditory canal; MFD, middle fossa dura; PCW, posterior canal wall; PFD, posterior fossa dura; FN, facial nerve.
and is undermined laterally. Extra cartilage is removed, and the skin is closed with interrupted nylon sutures in a dimple-like fashion at the external auditory meatus. A flap of mastoid periosteum is developed on a pedicle just posterior to the EAC. This flap is rotated anteriorly and secured as a second layer of closure for the meatus. After removal of all of the canal skin, tympanic mem brane, and malleus, and incus, the bony EAC is drilled out and excised circumferentially. An extended facial recess is created before the posterior bony wall is removed. The eustachian tube is curetted and packed with Surgicel and temporalis muscle. Care is taken to avoid entering the glenoid fossa.
Labyrinthectomy and Skeletonization of Internal Auditory Canal Dissection of the perilabyrinthine cells down to the lateral semicircular canal is completed (Fig. 52-3). The facial nerve is identified in its vertical portion between the nonampullated end of the lateral semicircular canal and the stylomastoid foramen. At this stage, exposing the perineurium of the nerve is unnecessary, but it should be clearly and unmistakably identified in its vertical course.
The lateral semicircular canal is fenestrated superi orly, and the membranous portion is identified and fol lowed anteriorly to its ampullated end and posteriorly to the posterior semicircular canal. All three membranous and bony semicircular canals are removed, and the saccule and utricle in the vestibule are identified and removed. The dissection proceeds along the sinodural angle and the superior petrosal sinus. The dura of the posterior fossa is exposed anteriorly. The cells over the jugular bulb are removed, skeletonizing it, and the cochlear aqueduct is removed. To obtain a better control of the lower cra nial nerves and the vertebrobasilar junction, the jugular bulb is completely exposed and, if needed, compressed. The IAC is skeletonized, beginning inferiorly and then around the porus acusticus. The falciform crest (trans verse) and vertical crest (Bill’s bar) are used as identifying landmarks for the superior and inferior vestibular nerves and the facial nerve. The bone of Bill’s island over the sigmoid sinus is removed. The superior and inferior ves tibular and cochlear nerves are excised.
Facial Nerve Rerouting (Figs. 52-4 and 52-5) After removal of the incus, if the bony EAC was not removed, an extended facial recess opening is created. The facial nerve is completely skeletonized from the IAC to the stylomastoid foramen, including the genicu late ganglion, with diamond burrs. An area comprising 180 degrees of the bony fallopian canal is uncovered. The greater superficial petrosal nerve is cut at its origin from the geniculate ganglion. The facial nerve is reflected posteriorly out of the fallopian canal with care taken to avoid traction or bending on the nerve, especially near the geniculate ganglion, where the nerve is reduced in size and forms an acute angle, and near the mastoid genu, which is the site of several branches to the stapedius muscle.
635
Chapter 52 • Transcochlear Approach to Cerebellopontine Angle Lesions
G.G. F.C. IAC
VII
FD
MFD
M
JB
D PF
Fn. S. S.
D
S. S.
PF
FIGURE 52-4. Bone over facial nerve is removed. At this point, the
SDA
g reater superficial petrosal nerve is sectioned. GG, geniculate ganglion; IAC, internal auditory canal; MFD, middle fossa dura; PDF, posterior fossa dura; SS, sigmoid sinus.
Care is also taken to avoid kinking of the nerve near the stylomastoid foramen when reflecting it posteriorly. The facial nerve is protected at all times and kept wet. The IAC dura is preserved, and CN VII and VIII, with all the protecting dura, are rerouted posteriorly.
FIGURE 52-5. Location of facial nerve after it is completely reflected posteriorly out of bony fallopian canal. FC, fallopian canal; Fn, facial nerve; JB, jugular bulb; MFD, middle fossa dura; PFD, posterior fossa dura; SDA, sinodural angle; SS, sigmoid sinus.
Tumor Removal (Fig. 52-7) With meningiomas, arterial feeder vessels from the inter nal carotid artery are encountered and eliminated during the approach. The diamond burr is used to excise these vessels and the base of implantation at the petroclival area. The dura is opened anterior to the IAC, and the open ing is extended as far forward as necessary for complete
.
Co.
ICA
Fn
The fallopian canal and the ossicles have been removed, and the promontory is now exposed. Starting with the basal coil, the cochlea is completely drilled out. Bone removal is carried forward around the internal carotid artery, and inferiorly the bone removal extends to the inferior petrosal sinus and jugular bulb. Superi orly, the superior petrosal sinus is followed to Meckel’s cave. Medially, bone removal extends to the clivus. At this stage, a large triangular window, covered by dura, has been created into the midline of the skull base. Its boundaries are the superior petrosal sinus superiorly; below and medial to the inferior petrosal sinus into the clivus inferiorly; the region of the internal carotid artery anteriorly; and the lateral clivus medially. The apex of the triangle is just beneath Meckel’s cave.
MFD
Transcochlear Drill-out (Fig. 52-6)
JB
MFD
PFD
. S.S
FIGURE 52-6. Cochlear drill-out. Co, cochlea; Fn, facial nerve; ICA, internal carotid artery; JB, jugular bulb; MFD, middle fossa dura; PFD, posterior fossa dura; SS, sigmoid sinus.
636
OTOLOGIC SURGERY
tumor exposure. The dural opening extends from the superior petrosal sinus superiorly to the inferior petrosal sinus inferiorly. CN V, VII, and VIII are identified. The facial nerve is kept on the posterior surface of the tumor. The junction of the intracranial portion of the facial nerve and the skeletonized intratemporal por tion is now identified. The nerve is protected and kept moist.
Tumor
B.A .
V
ICA
Fn.
VI
S. B.
The tumor pseudocapsule is opened, and the center of the main mass of the tumor is removed with the HouseUrban rotatory dissector or with an ultrasonic aspira tor. As the dissection proceeds forward and medially, the basilar artery and CN VI are identified anterosu periorly. The vertebral arteries appear posteroinferiorly. The tumor is removed from these vessels and their major tributaries under direct vision. In tumors extend ing across the midline, the basilar artery and its major branches can be dissected posteriorly off the tumor capsule. When the lesion is removed in this fashion, the cranial nerves and the IAC in the opposite CPA come into view. No brain retractors are needed to allow for this exposure. For dumbbell-shaped tumors, the tentorium can be opened to excise the part of the tumor that is lying in the middle fossa. Care is taken not to injure the vein of Labbé posteriorly or CN IV at the edge of the tento rium. Figures 52-8 to 52-12 illustrate the transcochlear approach with removal of the tympanic bone and blind closure of the EAC, which is done when further anterior exposure is necessary.
VIII
. S.S
FIGURE 52-7. Tumor removal. Tumor is removed with a HouseUrban rotatory dissector or with an ultrasonic aspirator. BA, basilar artery; BS, brainstem; Fn, facial nerve; ICA, internal carotid artery; SS, sigmoid sinus.
Closure After the tumor has been removed, hemostasis is secured. The dura is reapproximated, and the facial nerve is replaced forward. The eustachian tube orifice is plugged with Surgicel, bone wax, and muscle. Abdominal fat strips are used to fill the dural defect, and the mastoid and skull base defect, and to form a bed for the facial nerve. A titanium mesh cranioplasty is performed. The postauricular incision is closed in three layers, and a compressive dressing is placed securely around the head.
E.
FIGURE 52-8. Increased anterior exposure obtained with removal of bony external auditory canal. E, exposure.
Chapter 52 • Transcochlear Approach to Cerebellopontine Angle Lesions
637
Lumbar drainage may be initiated and continued for 4 to 5 days.
POSTOPERATIVE CARE The patient is observed in the intensive care unit for 24 hours after surgery and remains in the hospital for 4 to 5 days. Steroids are continued for 48 hours. Antibiotics are routinely used after the perioperative period. With early mobilization and ambulation, thromboembolism is avoided, and there is a speedy return of balance.
RESULTS JB
F.C. one
Db
PF
B.I
Dura
FIGURE 52-9. Mastoidectomy. Meatus is closed in three layers; skin, tympanic membrane, malleus, and incus are removed; and bony exter nal auditory canal is drilled out. BI, Bill’s island; FC, fallopian canal; JB, jugular bulb; PFD, posterior fossa dura.
In 1982, De la Cruz1 reviewed the results of 16 patients in whom the transcochlear approach was used. A com bination transcochlear/middle fossa approach was used in the three cases involving dumbbell-shaped tumors in the posterior and middle cranial fossae. Total tumor removal was possible in 13 of the 16 patients. Each of the other three patients had had surgery elsewhere and presented with large recurrent meningiomas and exten sive neurologic deficits preoperatively. During the trans cochlear approach, scraps of tumor were left behind on the vertebral artery in two of these cases. None of these patients had tumor recurrence. Four patients had facial paresis or twitch preoperatively; of the other 12, 4 had permanent facial paralysis because of tumor
ICA
ICA
P.A bone.
Co.
JB
JB Fn.
Fn.
S. S.
A
B
FIGURE 52-10. A and B, Cochlea and adjacent part of petrous apex (PA) are drilled away. Internal carotid artery (ICA) is exposed at anterior limit of dissection. Co, cochlea; Fn, facial nerve; JB, jugular bulb; SS, sigmoid sinus.
638
OTOLOGIC SURGERY
Dorsal incision
ICA
JB
MFD PFD
Fn.
. S.S
FIGURE 52-11. Final dural exposure and outlining of dural incision. Fn, facial nerve; ICA, internal carotid artery; JB, jugular bulb; MFD, middle fossa dura; PFD, posterior fossa dura; SS, sigmoid sinus.
involvement of the nerve, and 7 had temporary paresis with good recovery of facial function. This paresis was attributed primarily to removal of blood supply to the nerve, and these patients were operated on before facial nerve monitoring was available. Two deaths occurred in this series. One patient had bleeding from the ver tebral artery 1 week postoperatively, requiring clipping of the vessel, with subsequent infarction of the brain stem; the other patient was diabetic and died 1 month postoperatively because of gram-negative shock from pyelonephritis. He was one of the patients reported as having a “permanent” facial paralysis. Autopsy revealed no evidence of residual tumor on the facial nerve or elsewhere. In 1989, Yamakawa and colleagues15 published their results with the suboccipital approach, reporting subto tal tumor removal in 17 of 29 patients with intracranial epidermoids and tumor recurrence in 7 patients. One of 14 patients with CPA tumors had postoperative CN VII paralysis, 6 had CN VI palsy, 4 had dysphagia, and 2 had hearing loss. The report by Yasargil and associates19 pub lished in the same year analyzed results in 43 patients; 35 had epidermoid tumors, whereas others had intracranial dermoid tumors. There were no recurrences. Aseptic meningitis and transient cranial nerve palsies were the most common complications.
In 2001, Angeli and coworkers20 reviewed 24 cases operated on between 1985-1995 at the House Ear Clinic using the transcochlear approach or 1 of its modifica tions. In 1 case, a modified transotic approach was used (small melanoma); in 2 other cases, the transcochlear approach was extended inferiorly with an infratemporal and upper neck dissection (1 glomus and 1 meningi oma). Most of the tumors (16 of 24) were meningiomas. The other tumors included 4 cholesteatomas, 2 mela nomas, 1 glomus, and 1 ependymoma. The EAC was closed in 12 patients. Complete removal was achieved in 82% of tumors (average follow-up time 36 months). In 2 cases (8%), it was elected intraoperatively to do a subtotal tumor resection owing to excessive blood loss (intracranial glomus jugulare tumor) in 1 and unresect ability (melanoma invading the brainstem) in 1. Most patients had some degree of facial nerve dysfunction immediately after surgery, and 12 of 20 patients sub sequently improved to House-Brackmann grade III or better. A significant incidence of temporary facial weak ness was expected as a result of posterior facial nerve transposition. 59% of patients had permanent neuro logic sequelae because of either the surgery or their dis ease. The most common neurologic deficit was diplopia (27%). Other complications included dysphagia, facial numbness, unsteadiness, hoarseness, hemiparesis, and dysarthria. In 2004, Gonzalez and associates21 reported 32 patients with 34 anterior inferior cerebellar artery aneu rysms. Surgical approaches included retrosigmoid, trans cochlear, translabyrinthine, and orbitozygomatic. The transcochlear approach was used in 4 patients; 1 patient had small bilateral aneurysms that were approached, with good results, from the same side by taking advantage of the wide corridor obtained with a transcochlear approach. Complications reported were involvement of CN VI, VII, and VIII, and CSF leak. Siwanuwatn and coworkers22 quantitatively assessed the working areas and angles of attack associated with retrosigmoid, combined petrosal, and transcochlear cra niotomies, using silicone-injected cadaveric heads. They reported that the transcochlear approach provided sig nificantly greater working areas at the petroclivus and brainstem than the combined petrosal and retrosigmoid approaches (P < .001). The horizontal and vertical angles of attack achieved using the transcochlear approach were wider than the angles of the combined petrosal and retro sigmoid at the Dorello canal and the origin of the anterior inferior cerebellar artery (P < .001). They concluded that the transcochlear approach provides the widest corridor, improving the working area and angle of attack to both areas. In 2006, Leonetti and colleagues24 reported 29 patients with large meningiomas of the CPA surgically treated through a combined retrosigmoid/transpetrosal/ transcochlear approach. Total tumor removal was achieved in 19 of 29 (67%) of the patients, and the facial
Chapter 52 • Transcochlear Approach to Cerebellopontine Angle Lesions
639
ICA VI T.
JB
BA.
V I
VI
SPS
. S.S
FIGURE 52-12. Tumor removal and structures seen at skull base at completion of approach. Basilar artery (BA) and CN VI are seen. ICA, internal carotid artery; JB, jugular bulb; SPS, superior petrosal sinus; SS, sigmoid sinus; T, tumor.
nerve was anatomically preserved in 26 of 29 (89%) of the cases. Cerebrospinal fluid leakage was seen in 3.5% of the patients, and additional transient cranial nerve deficits were noted in 14% of the cases, but no signifi cant neurologic sequelae occurred. Of the 10 patients with residual tumor, 6 have been stable without growth, 2 were treated with reoperation for regrowth of disease, and 2 were controlled with localized radiotherapy. These investigators concluded that this combined lateral trans temporal approach provided wide exposure to the CPA and optimized the surgical extirpation of the 29 menin giomas presented in their series.
COMPLICATIONS AND THEIR MANAGEMENT Temporary facial nerve paresis is the most common complication. If facial paresis occurs, prompt eye care is essential; adequate lubrication with drops, nighttime ointments, and a moisture shield prevent corneal compli cations. Unless the facial nerve has been severed, surgical intervention for facial reanimation is not indicated. The best approach in individuals with even complete paralysis after transcochlear surgery, if the facial nerve is anatomi cally intact, is eye care that includes soft lenses, spring and gold weights, sometimes canthoplasty, and “watch ful waiting” because most of these patients recover to an
acceptable grade of facial function within the first year after surgery. Other cranial nerve palsies may occur and should be addressed individually. The neurotologist must attain a good working relationship not only with a neurosur geon, but also with an ophthalmologist and a laryngolo gist to help in the management of these cranial nerve deficits. Intracranial bleeding is controlled at the time of surgery. Close observation of the patient in the intensive care unit for the initial 24 hours postoperatively allows early recognition of delayed postoperative intracranial hemorrhage. In these cases, treatment consists of imme diate reopening of the surgical wound and removal of the fat in the intensive care unit while the operating room is being prepared, and then operative evacuation of the hematoma and control of the bleeding site or sites. Meningitis may occur after complete excision of intracranial epidermoids, and the incidence increases when the tumor capsule is left in place.19,24 Meningitis may be fatal if it is infectious and requires early, aggressive antibiotic therapy. More commonly, however, it is a chemical aseptic meningitis, and the patient is treated with dexamethasone. Postoperative pain is not as severe as that seen with the suboccipital approach and is managed adequately with analgesics. With this approach, tumor recurrence is rare when all visible tumor has been removed. Patients
640
OTOLOGIC SURGERY
with recurrences do not present typically and may have vague complaints of unsteadiness or trigeminal symp toms several years after the initial resection.1 Annual follow-up with gadolinium-enhanced and fat-suppression MRI is necessary. In cases of suspected tumor regrowth or recurrence, complete re-evaluation is performed, and removal of the recurrent tumor is advised. Gamma knife stereotactic radiosurgery and stereotactic radiotherapy are also treatment options for residual or recurrent meningiomas.26
SUMMARY Access to midline intradural lesions, intradural petroclival tumors, and CPA tumors arising anterior to the IAC has traditionally been difficult. With the transcochlear approach, the facial nerve is mobilized, the cochlea is removed, and the petrous apex is dissected around the internal carotid artery, allowing direct exposure of these lesions and of midline and contralateral CPA structures, without using retraction. Total removal of the tumor and its base and blood supply is possible with this approach. The transcochlear approach is recommended for these lesions in patients with poor hearing. Its safety and efficacy encourage its use.
REFERENCES 1. De la Cruz A: The transcochlear approach to meningio mas and cholesteatomas of the cerebellopontine angle. In Brackmann D E (ed): Neurological Surgery of the Ear and Skull Base. New York, Raven Press, 1982, pp 353-360. 2. House WF, De la Cruz A: Transcochlear approach to the petrous apex and clivus. Trans Am Acad Ophthalmol Otolaryngol 84:927-931, 1977. 3. House WF, Hitselberger WE: The transcochlear approach to the skull base. Arch Otolaryngol Head Neck Surg 102:334-342, 1976. 4. Hitselberger WE, Horn K L, Hankinson H, et al: The middle fossa transpetrous approach for petroclival meningiomas. Skull Base Surg 3:130-135, 1993. 5. Spetzler RF, Daspit CP, Pappas CT: The combined supratentorial and infratentorial approach for lesions of the petrous and clival regions: Experience with 46 cases. J Neurosurg 76:588-599, 1992. 6. Daspit CP, Spetzler RF, Pappas CT: Combined approach for lesions involving the cerebellopontine angle and skull base: Experience with 20 cases-preliminary report. Otolaryngol Head Neck Surg 105:788-796, 1991. 7. Shiobara R, Ohira T, Kanzaki J, Toya S: A modified extended middle cranial fossa approach for acoustic nerve tumors: Results of 125 operations. J Neurosurg 68: 358-365, 1988. 8. Wigand M E, Haid T, Berg M: The enlarged middle cranial fossa approach for surgery of the temporal bone and the cerebellopontine angle. Arch Otol Rhinol Otolar yngol 246:299, 1989.
9. Jackler R K, Sim DW, Gutin PH, Pitts L H: Systematic approach to intradural tumors ventral to the brain stem. Am J Otol 16:39-51, 1995. 10. Arriaga M, Shelton C, Nassif P, Brackmann D E: Selection of surgical approaches for meningiomas affect ing the temporal bone. Otolaryngol Head Neck Surg 107: 738-744, 1992. 11. Thedinger B A, Glasscock M E III, Cueva R A: Transco chlear transtentorial approach for removal of large cer ebellopontine angle meningiomas. Am J Otol 13:408-415, 1992. 12. Brackmann D E, Anderson RG: Cholesteatomas of the cerebellopontine angle. In Silverstein H, Norrell H (eds): Neurological Surgery of the Ear. Birmingham, Aescula pius, 1979, pp 340-344. 13. De la Cruz A, Doyle KJ: Epidermoids of the cerebello pontine angle. In Jackler R A, Brackmann D E (eds): Neu rotology. York, PA, Spectrum, 1994, pp 823-834. 14. Nager GT: Epidermoids involving the temporal bone: Clinical, radiological, and pathological aspects. Laryngo scope 2(Suppl):1-22, 1975. 15. Yamakawa K, Shitara N, Genka S, et al: Clinical course and surgical prognosis of 33 cases of intracranial epider moid tumors. Neurosurgery 24:568-573, 1989. 16. Mafee MF: MRI and CT in the evaluation of acquired and congenital cholesteatomas of the temporal bone. J Otolaryngol 22:239-248, 1993. 17. Brackmann DE: Translabyrinthine Transcochlear ap proaches. In Sekhas LN, Janecka IP (eds): Surgery of Cranial Base Tumors. New York, Raven Press, 1993, pp 351-365. 18. Friedman RA, Brackmann DE: Transcochlear Approach Operative Techniques in Neurosurgery Vol 2, No 1, pp 39-45, 1999. 19. Yasargil MG, Abernathy C D, Sarioglu AC: Microneu rosurgical treatment of intracranial dermoid and epider moid tumors. Neurosurgery 24:561-567, 1989. 20. Angeli S I, De la Cruz A, Hitselberger WE: The trans cochlear approach revisited. Otol Neurotol 22:690-695, 2001. 21. Gonzalez L F, Alexander M J, McDougall CG, Spetzler R F: Anteroinferior cerebellar artery aneurysms: Surgical approaches and outcomes—a review of 34 cases. Neuro surgery 55:1025-1035, 2004. 22. Siwanuwatn R, Deshmukh P, Figueiredo EG, et al: Quantitative analysis of the working area and angle of at tack for the retrosigmoid, combined petrosal, and trans cochlear approaches to the petroclival region. J Neuro surg 104:137-142, 2006. 23. Leonetti J P, Anderson D E, Marzo S J, et al: Combined transtemporal access for large (>3 cm) meningiomas of the cerebellopontine angle. Otolaryngol Head Neck Surg 134:949-952, 2006. 24. Cantu RC, Ojemann RG: Lucosteroid treatment of kera tin meningitis following removal of a fourth ventricle epidermoid tumor. J Neurol Neurosurg Psychiatry 31:75, 1968. 25. Guidetti B, Gagliardi FM: Epidermoid and dermoid cysts. J Neurosurg 47:12-18, 1977. 26. Mattozo C A, De Salles A A, Klement I A, et al: Stereotactic radiation treatment for recurrent nonbenign meningiomas. J Neurosurg 106:846-854, 2007.
53
Extended Middle Cranial Fossa Approach Rick A. Friedman
Gaining access to the cerebellopontine angle (CPA) and prepontine cisterns has presented a formidable challenge. Approaches designed to expose infraclinoid basilar tip aneurysms, petroclival meningiomas, chondromas, chondrosarcomas, and chordomas involving the petrous apex and clivus must take into consideration the vital neighboring neurovascular structures.1,2 The risks encountered in the region of the prepontine cistern during the management of aneurysms of the posterior circulation were best described by Drake3 in 1961, when he said, “the upper clival region is to be considered no-man’s land.” Several conventional neurosurgical approaches to this region have been described, including the subtemporal and trans-sylvian and a combination half-andhalf approach incorporating both techniques. Despite advances in microsurgical techniques and neuroanesthesia, the petrous bone has previously been an impediment to satisfactory exposure in this anatomically complex region. Modern skull base approaches, including the middle cranial fossa and the middle fossa transpetrous or extended middle fossa (EMF), have been instrumental in removing the petrous barrier, minimizing temporal lobe retraction, and improving the line of sight for the neurosurgeon. The middle fossa approach was first described in the literature in 1904.4 The sentinel work of House in 19615 led to a revitalization and refinement of this approach to the internal auditory canal (IAC), CPA, and prepontine cisterns. By extending the traditional middle fossa dissection anteriorly to the clivus, the EMF approach provides complete exposure of the IAC and the prepontine cisterns and the mid to upper clivus.6-11 Posteriorly, the approach allows access to tumors approaching, but not entering the jugular foramen (Table 53-1).
TECHNIQUE We administer preoperative antibiotics and continue them for 24 hours postoperatively. Intraoperative furosemide and mannitol are given to allow easier temporal
lobe retraction. We administer dexamethasone intravenously during the procedure and continue this for 24 hours postoperatively. Long-acting muscle relaxants are avoided during surgery so as not to interfere with facial nerve monitoring.
SURGICAL ANATOMY The surgical anatomy of the temporal bone from the middle fossa approach is complex (Fig. 53-1).12 Landmarks are not as apparent as with other approaches through the temporal bone. Laboratory dissection is essential so that the surgeon may become familiar with the anatomy from above. Anteriorly, the limit of the dissection is the middle meningeal artery, which is anterior and lateral to the greater superficial petrosal nerve. Excessive anterior retraction can lead to postoperative paresthesias in CN V3. The arcuate eminence roughly marks the position of the superior semicircular canal. The relationship between the arcuate eminence and the superior semicircular canal is inconstant.13 The superior semicircular canal tends to be perpendicular to the petrous ridge. Medially, the superior petrosal sinus runs along the petrous ridge. Surgical tolerances are tight in the area of the lateral IAC. The labyrinthine portion of the facial nerve lies immediately posterior to the basal turn of the cochlea. Bill’s bar separates the facial and superior vestibular nerves. Slightly posterior and lateral to this area is the vestibule and ampullated end of the superior semicircular canal. Identification of the geniculate ganglion can be accomplished by tracing the greater superficial petrosal nerve posteriorly. The ganglion is dehiscent in approximately 16% of patients. The IAC lies roughly on the same axis as the external auditory canal; this relationship is useful in orienting the surgical field. The area around the porus acusticus medially is a “safe zone” compared with the lateral or fundal region, and provides an excellent place to begin IAC dissection. We begin our dissection medially, by 641
642
OTOLOGIC SURGERY
SURGICAL TECHNIQUE
drilling in the meatal plane in the area of the bisection of the angle formed by the superior semicircular canal and the greater superficial petrosal nerve. The IAC can be located initially in this medial area of the temporal bone and followed laterally.
The patient is placed in the supine position with the head turned to the side opposite the lesion (Fig. 53-2). The surgeon is seated at the head of the table, and the anesthesiologist is seated at the foot. An incision is made in the preauricular area and extended superiorly in a gently curving fashion. Care must be taken near the anterior extension of the incision to avoid injury to the temporal branch of the facial nerve (see Fig. 53-2). The temporalis muscle is incised, beginning at the zygomatic root, along the linea temporalis, and the muscle is elevated from the temporal fossa and reflected anteroinferiorly. The temporal squama is exposed. Using cutting and diamond burrs, a temporal craniotomy is performed. The craniotomy measures approximately 5 × 5 cm, and is two thirds anterior and one third posterior to the zygomatic root (Fig. 53-3). The inferior
TABLE 53-1 Indications for Extended Middle Fossa
Approach
Vestibular schwannoma (<2 cm) Petroclival meningioma Chondroma Chondrosarcoma Chondroblastoma Chordoma Trigeminal schwannoma Infraclinoidal basilar tip aneurysms
10.
9.
12. 8.
ELS 5.
2.
11.
7.
6.
14.
4. 1.
16. 13.
3.
V 13. 15. 14.
F.M.
I.P.S. 1. 2. 3. 4. 5. 6. 7. 8.
Greater superficial petrosal n. Middle meningeal a. I.C.A. Cochlear n. Cochlea Geniculate ganglion Vestibule Incus, malleus
9. 10. 11. 12. 13. 14. 15. 16.
EAC Facial n. Superior semicircular canal and arcuate eminence Superior petrosal sinus SAFE ZONES Facial n. (CN VII) Vestibulocochlear n. (CN VIII) Sigmoid sinus
FIGURE 53-1. Anatomic landmarks of middle cranial fossa. EAC, external auditory canal; ICA, internal carotid artery; IPS, inferior petrosal sinus.
Chapter 53 • Extended Middle Cranial Fossa Approach
643
Temporal branch of VII
Skin incision
FIGURE 53-2. Preauricular/temporal scalp incision and relationship to temporal branch of facial nerve.
5 X 5 cm
Craniotomy
FIGURE 53-3. Position of craniotomy.
644
OTOLOGIC SURGERY Superior semicircular canal makes a 45–60 degree angle with IAC
EAC
Mma.
G.g. G.S.P.N. S V N
C
AE V.
VII
FIGURE 53-4. Surface localization
45°–60°
of trajectory of internal auditory canal (IAC). AE, arcuate eminence; C, cochlea; EAC, external auditory canal; Gg, geniculate ganglion; GSPN, greater superficial petrosal nerve; Mma, middle meningeal artery; SVN, superior vestibular nerve; V, vestibule; VII, facial nerve.
limit of the flap should be at the level of the zygoma, approximating the floor of the middle cranial fossa. Care must be taken to avoid laceration of the underlying dura. The bone flap is set aside for later replacement. The middle fossa floor is exposed extradurally with identification of the middle meningeal artery at the foramen spinosum, the greater superficial petrosal nerve at the facial hiatus and along the petrosal ridge, the arcuate eminence, and the mandibular branch of the trigeminal nerve (Fig. 53-4). The middle meningeal artery is sacrificed, and the horizontal segment of the petrous internal carotid artery is exposed. Frequently, venous bleeding is encountered from this area and can be controlled with absorbable knitted fabric (Surgicel). Elevation of the dura proceeds in a posterior-to-anterior fashion. As stated previously, in approximately 16% of cases the geniculate ganglion of the facial nerve is dehiscent, and injury can be avoided with posterior-to-anterior dural elevation. The petrous ridge is identified and care is taken not to lacerate the superior petrosal sinus as it is elevated from its sulcus. The arcuate eminence and greater superficial petrosal nerve are identified (see Fig. 53-4). These are the major landmarks for the subsequent intratemporal dissection. When the dura has been elevated, typically with a suction-irrigator and a blunt dural elevator, the HouseUrban retractor is placed to support the temporal lobe. To maintain a secure position, the teeth of the retaining retractor should be locked against the bone margins of the craniotomy window, and the retractor blade must be placed on the true petrous ridge (Fig. 53-5). Using a large diamond burr and continuous suction irrigation, the superior semicircular canal is identified, but not blue-lined, at the arcuate eminence. The superior semicircular canal makes a 45 to 60 degree angle with the IAC (see Fig. 53-4).
Bone removal over the IAC begins medially at the porus acusticus with a large diamond burr. The area of bone anteromedial to the IAC, and medial to the petrous carotid artery can be saucerized exposing its anterior surface. Next, the bone in the “postmeatal triangle” can be removed exposing the posterior surface of the IAC. Medially, 270 degrees of bone can be removed from its circumference (Fig. 53-6). The circumference of the IAC is less exposed laterally because of the location of the inner ear. The lateral end of the IAC is dissected with clear identification of the labyrinthine segment of the facial nerve, Bill’s bar, and the superior vestibular nerve. The labyrinthine portion of the facial nerve is identified proximal to the geniculate ganglion. Care must be taken to avoid the cochlea, which lies 1 mm anterior to the labyrinthine portion of the facial nerve; this is best accomplished by careful delineation of the anterior limit of the IAC with a blunt hook. The superior vestibular nerve can be followed laterally approximately one half the distance of the labyrinthine facial nerve to avoid entering the ampulla of the superior semicircular canal (Fig. 53-7). The lateral dissection is essential for clear identification of the facial nerve and complete removal of tumor from the fundus of the IAC. The area of bone removed consists of the entire petrous apex from the IAC posteriorly to the internal carotid artery laterally and the petroclival junction anteromedially. Inferiorly, the bone removal extends to the inferior petrosal sinus limiting access to lesions of the lower one third of the clivus (Fig. 53-8). The dura is opened along the inferior edge of the temporal lobe. The subtemporal dura is sectioned perpendicular to the dural opening directly to the segment of the superior petrosal sinus exposed by the bone removed
Chapter 53 • Extended Middle Cranial Fossa Approach
645
Temporalis m. flap
House-Urban retractor
FIGURE 53-5. Temporal lobe retractor in place.
from the anterior petrous dissection. The superior petrosal sinus is sectioned between two titanium clips exposing the junction of the posterior fossa dura and the tentorium cerebelli. The posterior fossa dura is opened to the apex of the anterior petrosectomy. The tentorium cerebelli is sectioned to a point posterior to the entry of the trochlear nerve, to the tentorial edge. Care must be taken during this step to avoid sectioning the tentorium too far posteriorly and inadvertently injuring the vein of Labbé as it enters the transverse sinus. The combination of the extended middle cranial fossa and the translabyrinthine approaches provides wide exposure of the CPA and prepontine region without the need for facial nerve rerouting. Technical pearls for surgical technique are as follows: 1. Avoid frontal branch of facial nerve 2. Place retractor blade in true petrous ridge 3. Define anterior limit of IAC from porus to fundus
4. Completely delineate vertical crest 5. Sharply dissect tumor from medial to lateral
LIMITATIONS The limitations of the EMF approach are dictated by the tumor type and the regional anatomy. The EMF approach has virtually eliminated the barrier imposed by the petrous apex when approaching lesions of the posterior cranial fossa within and anterior to the IAC. This approach is an essential part of tumor removal with hearing and facial nerve functional preservation. Anatomically, the EMF approach provides superb access to lesions of the IAC and prepontine cistern down to the level of the inferior petrosal sinus. Lesions posterior to the IAC are difficult to access through this approach alone and often require a combined petrosal approach. Combining posterior access via a retrolabyrinthine or translabyrinthine approach with an EMF approach
646
OTOLOGIC SURGERY G.g.
Bill’s bar
S.V.N.
P.F.D.
FIGURE 53-7. Ae. G.S.P.N.
C.
D.I.
Ae. C. VII
P.F.D.
FIGURE 53-6.
FIGURE 53-6. Critical microanatomic measurements. Diamond ball no greater than 2 mm can be safely used to remove bone of lateral internal auditory canal. Ae, arcuate eminence; C, cochlea; PFD, posterior fossa dura.
FIGURE 53-7. Essential bone removal at lateral internal auditory canal. Facial nerve followed to geniculate ganglion (Gg) and superior vestibular nerve (SVN) exposed for half the distance of labyrinthine facial nerve. Bill’s bar is clearly defined. Dural incision (DI) is shown along posterior IAC. Ae, arcuate eminence; C, cochlea; GSPN, greater superficial petrosal nerve; PFD, posterior fossa dura.
Area of bone removal
FIGURE 53-8. Boundaries of anterior petrous bone removed for extended middle cranial fossa approach.
Chapter 53 • Extended Middle Cranial Fossa Approach
rovides outstanding access to most areas of the posterior p fossa except the lower clivus, without the need for facial nerve rerouting.
ADVANTAGES The advantage of the EMF approach lies principally in the surgeon’s ability to access larger tumors of the petrous apex and prepontine cistern without the facial nerve rerouting and hearing loss associated with the transcochlear approach. We reviewed 71 patients in whom the EMF approach was used either alone or as part of a combined petrosal approach for meningioma. Historically, these patients might have been treated using a translabyrinthine or transcochlear approach with the attendant facial motor and hearing deficits. Of patients, 85% maintained a House-Brackmann facial nerve grade I postoperatively, and we were able to preserve hearing at the preoperative level in 86% of the patients who presented with class A hearing according to the standards of reporting for the American Academy of Otolaryngology– Head and Neck Surgery.2
CONCLUSION The middle cranial fossa approach is an invaluable approach to tumors of the IAC and lesions of the petrous apex and prepontine cistern. This approach affords access to complex tumors, while potentially preserving hearing. The middle cranial fossa approach to removal of vestibular schwannoma is our approach of choice for patients with small tumors (≤2 cm) and serviceable hearing. We have shown the safety and efficacy of this approach for hearing preservation surgery.
REFERENCES 1. Friedman R A, Pensak M L , Tauber M, et al: Anterior petrosectomy approach to infraclinoidal basilar artery aneurysms: The emerging role of the neuro-otologist in multidisciplinary management of basilar artery aneurysms. Laryngoscope 107:977-983, 1997. 2. Shen T, Friedman R A, Brackmann D E, et al: The evolution of surgical approaches for posterior fossa meningiomas. Otol Neurotol 25:394-397, 2004.
647
3. Drake CG: Bleeding aneurysms of the basilar artery: Direct surgical management in four cases. J Neurosurg 18:230-238, 1961. 4. Parry PH : A case of tinnitus and vertigo treated by division of the auditory nerve. J Laryngol Otol 19:402, 1904. 5. House WF: Surgical exposure of the internal auditory canal and its contents through the middle cranial fossa. Laryngoscope 71:1363, 1961. 6. Dautheribes M, Migueis A, Vital J M, Guerin J: Anatomical basis of the extended subtemporal approach to the cerebellopontine angle: Its value and limitations. Surg Radiol Anat 11:187-195, 1989. 7. Rosomoff H L : The subtemporal transtentorial approach to the cerebellopontine angle. Laryngoscope 81:14481454, 1971. 8. Wigand M E, Haid T, Berg M : The enlarged middle cranial fossa approach for surgery of the temporal bone and of the cerebellopontine angle. Arch Otorhinolaryngol 246:299-302, 1989. 9. Bochenek Z, Kukwa A : An extended approach through the middle cranial fossa to the internal auditory meatus and the cerebellopontine angle. Acta Otolaryngol (Stockh) 80:410-414, 1975. 10. Kawase T, Shibora R , Toya S : Anterior-transpetrosaltranstentorial approach for aneurysms of the lower basilar artery. J Neurosurg 63:857-867, 1985. 11. House WF, Hitselberger WE, Horn K L : The middle fossa transpetrous approach to the anterior-superior cerebellopontine angle. Am J Otol 7:1-4, 1986. 12. Parisier SC : The middle cranial fossa approach to the internal auditory canal: An anatomical study stressing critical distances between surgical landmarks. Laryngoscope 87(Suppl 4):1-20, 1977. 13. Kartush J M, Kemink J L , Graham M D: The arcuate eminence: Topographic orientation in middle cranial fossa surgery. Ann Otol Rhinol Laryngol 94:25-28, 1985.
54
Anterior and Subtemporal Approaches to the Infratemporal Fossa Ricardo L. Carrau, Amin B.����� ������� Kassam, ����������������������������������������� and Moisés A.�������������� ���������������� Arriaga
The infratemporal fossa (ITF) is a potential space bounded superiorly by the greater wing of the sphenoid and the temporal bone. Neurovascular foramina, including the carotid canal, jugular foramen, foramen spinosum, foramen ovale, and foramen lacerum, connect the ITF with the middle cranial fossa. Medially, the ITF is contained by the superior constrictor muscle, the pharyngobasilar fascia, and the pterygoid plates. Medially, the ITF communicates with the pterygopalatine fossa via the pterygomaxillary fissure, which is continuous with the inferior orbital fissure and the orbit. Laterally, the ITF is bounded by the zygoma, mandible, parotid gland, and masseter muscle. The pterygoid muscles constitute the anterior boundary; posteriorly, the ITF is confined by the articular tubercle of the temporal bone, glenoid fossa, and styloid process. Using this definition, the ITF contains the parapharyngeal space (i.e., internal carotid artery [ICA], internal jugular vein, CN IV to XI) and the masticator space (i.e., internal maxillary artery, pterygoid venous plexus, and pterygoid muscles). The presence of neurovascular structures within the ITF (e.g., ICA) or adjacent to it (e.g., CN VII) is the limiting step for designing a surgical approach to the ITF. Surgical approaches often center on the preservation and identification of these neurovascular entities. The first report in the English literature of a surgical approach to the ITF is attributed to Barbosa,1 who in 1961 described his approach for advanced tumors of the maxillary sinus. Transtemporal approaches described by Fisch and preauricular approaches by Schramm and Sekhar are the basis for other modifications.2-8 Subsequent approaches follow the surgical and anatomic principles shown by these authors.
PATIENT SELECTION Tumors may originate within the confines of the ITF, or may invade this area by direct extension from any of its boundaries—the upper aerodigestive tract, the parotid gland, the temporal bone, the greater wing of the sphenoid, and structures within the cranial cavity. Accurate
assessment of the nature, origin, and extension of the tumor is crucial for the therapeutic-surgical plan. Other factors affecting the selection of the surgical approach include patient needs and demands, the biologic behavior of the tumor or other coexistent diseases, and the training and experience of the surgeon. Most pathologies affecting the ITF require a multidisciplinary approach to stage, diagnose, and extirpate the tumor, and at the same time to provide an acceptable cosmetic and functional reconstruction.
PREOPERATIVE EVALUATION Diagnostic and Staging Work-up Owing to the inaccessibility of the ITF to physical examination, radiographic imaging is a vital component of the evaluation. Computed tomography (CT) and magnetic resonance imaging (MRI) provide valuable information and are obtained using standard skull base protocols. CT is superior to MRI, showing the remodeling or erosion of neurovascular foramina or other bones of the skull base. MRI better delineates the soft tissue planes, the tumor–soft tissue interface, and the presence of tumor along neural and vascular structures (Fig. 54-1). CT and MRI are often complementary. Another crucial question is the relationship of the tumor to the ICA. Magnetic resonance angiography (MRA) ���������������������������������������� and computerized tomography angiography (CTA) provide ���������������������������������������������� a noninvasive assessment of the vasculature of the ITF and brain. If preoperative embolization of the tumor is indicated (e.g., juvenile nasopharyngeal angiofibromas, paragangliomas), angiography is preferred over MRA. Angiography provides important information regarding the vascularity of the tumor, its relationship to the ICA, and the cerebral circulation and its collateral blood supply. Neither angiography nor MRA is adequate, however, to predict reliably the adequacy of the collateral intracranial circulation if sacrifice of the ICA is necessary. If the risk for injury or sacrifice of the ICA is high, the collateral cerebral blood flow may be evaluated using 649
650
OTOLOGIC SURGERY TABLE 54-1 Xenon Computed Tomography
FIGURE 54-1. MRI shows soft tissue tumor involving infratemporal fossa. Even malignant tumors of infratemporal fossa may reach a large size before they are visible or palpable on physical examination. Patients with such tumors may present with facial numbness, pain, or difficulty with mastication.
angiography-balloon occlusion with xenon CT (ABOXCT). A nondetachable balloon is inserted in the ICA via the femoral artery. The balloon is inflated for 15 minutes while the awake patient is monitored for any neurologic deficit. If the patient does not develop any deficit, the balloon is deflated, and the patient is transferred to a CT suite. A mixture of 32% xenon/68% oxygen is administered via facial mask for 4 minutes. CT shows the distribution of xenon, which reflects the blood flow within the cerebral tissue, providing a quantitative assessment of milliliters of blood flow per minute per 100 g of brain tissue. The process is repeated after reinflation of the arterial balloon. A computer calculates the differential of the xenon diffusion in the brain before and after the balloon inflation, identifying patients at risk for an ischemic stroke secondary to reduced blood flow after occlusion of the ipsilateral ICA (Table 54-1).9 Despite a negative finding from ABOX-CT testing, patients can sustain ischemic brain injury because of the loss of collateral vessels that are not assessed by balloon occlusion testing (“watershed area”), or because of embolic phenomena. In addition, this test is performed under ideal and controlled circumstances, and does not account for the possibility of episodes of hypoxia, hypotension, or electrolyte and acid-base disturbances that may alter the brain’s hemodynamics. Every effort should be made to preserve or reconstruct the ICA and to diminish the possibility of embolus formation during the surgery. Other techniques that provide information
Cerebral Blood Flow (mL/min/100 g Tissue)
Risk
Implication
>35 21-35
Low Moderate
≤20
High
Carotid may be sacrificed Patient would tolerate occlusion under controlled circumstances; reconstruction is recommended Patient would not tolerate occlusion of internal carotid artery
regarding collateral cerebral blood flow include single photon emission CT (SPECT) with balloon occlusion and transcranial Doppler. Histologic diagnosis should be obtained before the extirpative surgery whenever possible. Tumors amenable to a punch or open biopsy are approached in this manner. Tumors that are in deeper planes may be sampled by fine-needle aspiration biopsy. Rarely, a histologic diagnosis cannot be obtained before the approach because of the intrinsic limitations of fine-needle aspiration biopsy. Under these circumstances, a frozen section analysis, obtained via a skull base approach, may be sufficient to justify the resection of the tumor. Vital neurovascular structures, such as the ICA, the eye, and cranial nerves, should not be sacrificed, however, based on a frozen section analysis. The extent of the evaluation to rule out regional or distant metastasis or to determine that the ITF tumor is a metastasis is dictated by the histologic type and stage of the tumor. CT scan of the neck is more sensitive than physical examination for the detection of regional lymphadenopathy. Patients presenting with tumors that metastasize hematogenously (sarcoma, melanoma) should undergo CT scan of the chest and abdomen and a bone scan. Cerebrospinal fluid (CSF) cytology is advised for patients with tumors that have invaded the dura. These patients are also at risk for “drop metastasis,” which should be ruled out by spinal MRI.
Rehabilitation Considerations Functional or neurologic deficits that are identified preoperatively should be taken into consideration during the surgical planning and during postoperative care. These deficits often have a significant impact on the recovery and functional rehabilitation of the patient. Dysfunction of the trigeminal nerve or the masticator muscles or both is commonly underdiagnosed. Cutaneous and corneal sensation should be assessed preoperatively. Corneal anesthesia associated with concomitant facial nerve palsy requires aggressive measures to prevent corneal injury.
Chapter 54 • Anterior and Subtemporal Approaches to the Infratemporal Fossa
Lateral deviation of the jaw on opening may reflect weakness or paralysis of the ipsilateral pterygoid muscles, invasion of the muscles, or dysfunction of the temporomandibular joint (TMJ). Likewise, trismus may be due to mechanical restriction caused by the bulk of the tumor, ankylosis of the TMJ, scarring, tumor tethering, or pain. The nature of the trismus is an important consideration in the perioperative management of the airway. Trismus secondary to pain resolves with the induction of general anesthesia, allowing safe oral endotracheal intubation. In patients with mechanical trismus, an awake nasotracheal intubation may be performed if it is anticipated that surgery would correct the trismus. Otherwise, a tracheostomy, performed under local anesthesia, is the safest perioperative airway. Neoplastic invasion of the facial nerve may manifest with facial weakness or paralysis, facial spasms, epiphora, facial spasms, and dysgeusia. Significant destruction of the facial nerve by tumor may occur before the patient develops these clinical signs. A gold weight, implanted in the upper eyelid, or surgical tightening of the lower lid may be necessary to protect the cornea. Hearing loss caused by a tumor of the ITF may be either conductive, resulting from eustachian tube dysfunction, or sensorineural, resulting from tumor involvement of the temporal bone or posterior cranial fossa. A myringotomy or amplification or both facilitate communication with the patient. Deficits of the lower cranial nerves (CN IX, X, XI, and XII) are associated with tumors that originate in the parapharyngeal space or tumors that extend to the jugular foramen, or both. Patients with deficits of CN IX, X, and XII present with varying degrees of swallowing or speech problems, such as hypernasal or slurred speech, nasal regurgitation, dysphagia, aspiration, and dysphonia. Findings on physical examination reflect the involvement of specific cranial nerves, and include decreased elevation of the palate, decreased mobility and strength of the tongue with deviation to the involved side on protrusion, decreased supraglottic sensation, pooling of secretions in the hypopharynx, ipsilateral vocal cord paralysis, and decreased bulk and strength of the sternocleidomastoid and trapezius muscles. Patients with partial deficits of the lower cranial nerves (paresis) often experience a complete deficit (paralysis) after surgery, resulting in increased dysphagia and aspiration. Consequently, a tracheostomy for tracheal toilet and a gastrostomy tube for nutrition and hydration are often necessary during the perioperative period. Laryngeal framework surgery (thyroplasty) performed during the extirpative surgery or the early postoperative period improves the glottic closure and decreases the risk for aspiration, often obviating the need for a tracheotomy for the sole purpose of tracheopulmonary toilet.10-12 Laryngeal framework surgery allows the patient to compensate for the deficits using the remaining function (contralateral side) more effectively. Laryngeal
651
f ramework surgery does not restore the motor or the sensory function. These patients remain at a higher risk for aspiration and nutritional deficiencies. Collaboration with an experienced speech-language pathologist, who can assist with the monitoring of the patient and the diet modifications and provide intensive swallowing therapy, is crucial to prevent the pulmonary and nutritional complications of aspiration. In patients with severe deficits or in patients with cognitive problems, strong consideration should also be given to placement of a gastrostomy tube to facilitate postoperative feeding and decrease the risk of prandial aspiration. Velopharyngeal insufficiency may be ameliorated by a palatal lift prosthesis that pushes the soft palate against the posterior pharyngeal wall. Alternatively, a pharyngeal flap or a palatopexy may be performed in patients who do not tolerate the prosthesis.
Reconstructive Considerations Most commonly, a temporalis muscle transposition flap is adequate to separate the cranial cavity from the upper aerodigestive tract and obliterate the dead space. Microvascular free flaps, such as rectus abdominis flap (for soft tissue defects), latissimus dorsi flap (for myocutaneous or massive defects), and iliac composite flap (for defects requiring bone reconstruction), are indicated when the temporalis muscle or its blood supply is sacrificed as part of the oncologic resection, when the patient requires a complex resection involving composite tissue flaps with skin or bone or both, or when the extirpative surgery leads to a massive soft tissue defect and dead space. These needs are usually anticipated during the surgical planning, and the patient and consultants (e.g., the microvascular surgeon) are informed accordingly. Ideally, functional and cosmetic deficits created by the tumor or the surgery should be addressed in a single stage, concomitant with the oncologic resection. When a temporary facial palsy is anticipated, corneal protection using lubricants or a temporary lateral tarsorrhaphy or both is usually adequate. Grafting of the facial nerve involves a longer recovery period, however. Insertion of a gold weight implant into the upper eyelid is advisable. When an immediate reconstruction of the facial nerve is impossible, static fascial slings or muscle transpositions are indicated. Lower cranial nerve deficits may be ameliorated by laryngeal framework surgery, tracheotomy, or laryngotracheal separation, as previously discussed.
Other Perioperative Considerations Preoperatively, the patient’s blood is typed and crossmatched for 2 to 6 U of packed red blood cells, according to the extent and nature of the tumor and surgery. Autologous blood banking is used when feasible, although it is frequently impractical. A Cell Saver autologous transfusion device may be used during the resection of benign vascular tumors.
652
OTOLOGIC SURGERY
Perioperative antibiotic prophylaxis with a wide spectrum against the flora of the skin and upper aerodigestive tract and that exhibits good penetration of the bloodbrain barrier is administered before the surgery and is continued for 48 hours after the surgery. The use of a broad-spectrum cephalosporin with good CSF penetration (e.g., ceftriaxone) seems to be as effective as multiple antibiotic regimens. Somatosensory evoked potential monitoring using the median nerve is indicated whenever surgical manipulation of the ICA is anticipated. Lower cranial nerve monitoring is not routinely employed. It may be useful for the identification and preservation of nerve function when the tumor is in close proximity to these nerves. Conversely, facial nerve monitoring is routinely used for transparotid or transtemporal approaches. Monitoring electrodes and lines for vascular access should be secured with sutures, staples, or adhesive dressings. The choice of anesthetic agent is influenced by the extent of intracranial dissection, potential for brain injury, systemic hemodynamics, the need for monitoring of cortical and brainstem functions (e.g., brainstem-evoked response, somatosensory evoked potentials, electroencephalography), and the need for cranial nerve monitoring (CN VII, and X to XII). All these factors should be thoroughly discussed with the anesthesiologist. When changes of head position during surgery are anticipated, the endotracheal tube should be secured with a circumdental or circum-mandibular wire ligature (No. 26 stainless steel wire). The operating table is positioned perpendicular to the anesthesia staff, and if intradural dissection is anticipated, a spinal drain is inserted and secured with sutures and adhesive dressing (e.g., Tegaderm, OpSite). Other measures to diminish the intracranial pressure, such as hyperventilation, osmotic diuresis, and corticosteroids, are used as needed throughout the surgery. A nasogastric tube and Foley catheter are passed and secured after adequate placement is corroborated. Antiembolic sequential compression stockings are recommended to prevent deep venous thrombosis.
SURGICAL APPROACHES The head of the patient is positioned on a horseshoe headrest or, if necessary for intracranial neurovascular or neurosurgical work, on a three-pin head fixation system. When the horseshoe headrest is used, it is important to use additional “egg-crate” foam padding because the scalp may develop a decubitus ulcer during prolonged surgery. If the ICA is at risk, the head should be positioned in slight extension to provide access to the neck for proximal control of the ICA. Tarsorrhaphy sutures are placed for protection of the eyes. The scalp is shaved following the planned incision line (e.g., bicoronal), and the incision line is infiltrated with a solution of lidocaine and epinephrine (1:100,000 to 1:400,000).
Preauricular (Subtemporal) Approach The preauricular approach is suited for tumors that originate in the ITF and intracranial tumors that originate at the anterior aspect of the temporal bone, or greater wing of the sphenoid bone, and that extend into the ITF.5,13,14 It may also be combined with other approaches, such as a subfrontal approach to expose massive tumors that extend to the anterior and middle skull base. The preauricular approach does not provide an adequate exposure for the resection of tumors that invade the tympanic bone, however, and does not provide control of the intratemporal facial nerve or jugular bulb. An incision, following a hemicoronal or bicoronal line, is carried through the subcutaneous tissue, galea, and pericranium (Fig. 54-2). Over the temporal area, the incision extends down to the deep layer of the temporal fascia. The anterior branches of the superficial temporal artery are preserved, ensuring adequate blood supply to the scalp flap. Ipsilateral to the tumor, the incision is extended following into the preauricular crease down to the level of the tragus. When proximal control of the ICA is warranted, the incision is extended into the neck using a lazy S pattern, or, alternatively, a separate cervical incision is performed. The scalp is dissected following a subpericranial plane, separating the attachments of the pericranium to the deep layer of the temporal fascia. The scalp flap is elevated from the deep temporal fascia using a broad periosteal elevator. Above the zygoma, the deep temporal fascia splits into superficial and deep layers, which attach to the lateral and medial surfaces of the zygomatic arch. To continue the surgical exposure, the superficial layer of the deep temporal fascia is incised following an imaginary line that joins the superior orbital rim to the zygomatic root. The dissection continues deep to this plane, elevating the superficial layer of the deep temporal fascia off the zygomatic arch (see Fig. 54-2). Fascia and periosteum are reflected anteriorly with the scalp flap. This maneuver protects the frontal branches of the facial nerve that are just lateral to the superficial layer of the deep temporal fascia. Elevation of the periosteum from the lateral surface of the zygomatic arch and malar eminence completes exposure of the orbitozygomatic complex. The periorbita is elevated from the lateral orbit using a Penfield No. 1 dissector, exposing the roof and lateral wall of the orbit down to the inferior orbital fissure. The fascial attachments of the temporalis and masseter muscles to the zygomatic arch are transected using electrocautery. The attachments of the temporalis muscle to the cranium are transected with the electrocautery, and the muscle is elevated off the temporal fossa. If the temporalis muscle is to be returned to its original position at the completion of the surgery, a curved titanium plate (1.5 to 1.7 mm) is screwed at the temporal line, leaving some screw holes empty to facilitate suturing from the plate to the muscle (Fig. 54-3). Then the masseteric
Chapter 54 • Anterior and Subtemporal Approaches to the Infratemporal Fossa
653
Hairline
Fat pad
B A
Parotid gland
FIGURE 54-2. A, Bicoronal scalp incision is extended along preauricular skin crease. This incision may be continued into upper cervical region as a lazy S incision, or a separate cervical incision may be made for exposure of vessels and nerves. B, Scalp flap is elevated from underlying cranium, fascia, lateral orbital rim, zygomatic arch, and masseteric fascia. Plane of dissection is deep to superficial layer of deep temporal fascia (incised) and deep to parotid masseteric fascia.
Inferior orbital fissure
FIGURE 54-3. Areas of bone removal are noted with dark stippling. Temporal craniotomy is performed in conjunction with orbitozygomatic osteotomy. Additional exposure of infratemporal skull base may be achieved by removal of subtemporal cranium (striped area) and resection of mandibular condyle (lightly stippled area).
fascia is dissected from the masseter muscle, elevating the overlying parotid gland with a broad periosteal elevator (see Fig. 54-2). To increase the arc of rotation of the scalp flap, any soft tissue anterior to the tympanic bone can be transected from superior to inferior, down to the
level of the facial nerve. The facial nerve is identified and preserved using a standard technique. It is helpful to preserve a cuff of soft tissue around the main trunk of the facial nerve to prevent a traction injury to the main trunk of the facial nerve.
654
OTOLOGIC SURGERY
Using the caudal limb of the incision, the sternocleidomastoid muscle is dissected laterally, and the carotid sheath is exposed. When necessary, the contents of the carotid sheath, including the ICA and common and external carotid arteries, and the internal jugular vein are exposed, dissected, and controlled. CN X to XII are also identified and preserved. Vessel loops are placed around these structures and secured with hemoclips rather than hemostats to avoid inadvertent traction. Orbitozygomatic osteotomies are performed at the zygomatic root posteriorly, the zygomaticofrontal suture superiorly, and the zygomaticomaxillary buttress at the level of the zygomaticofacial nerve medially (see Fig. 54-3). Prior periorbital elevation off the lateral and inferior walls is necessary to identify the inferior orbital fissure and to complete the osteotomies of the orbitozygomatic complex. The tip of the reciprocating saw is placed in the most lateral aspect of the inferior orbital fissure; an osteotomy is performed through the malar eminence following a vertical imaginary line medial to the zygomaticofacial foramen. This osteotomy separates the zygoma from the maxilla. Accidental entry into the maxillary sinus may occur, requiring closure of the defect using fascia or pericranium free grafting or both. These free tissue grafts are held in place by compression against the opening when orbitozygomatic bone graft is replaced and plated at the completion of the surgery. All osteotomies are completed with a reciprocating saw transecting the bone in beveled or V-shaped manner in such a way that maximizes the exposure and facilitates replacement of the bone graft at the completion of the surgery. If tumor involvement of the orbit is present, the osteotomies are modified to secure a complete resection. In cases requiring intracranial and extracranial exposure, the superior and lateral osteotomies are made through the superior and lateral orbital walls after the craniotomy is completed, and the brain is separated from the skull base. This way, the orbital walls can be incorporated in the orbitozygomatic graft. Using intracranial and extracranial exposures, osteotomies are made through the superior and lateral orbital walls to remove the orbitozygomatic bone segment. This approach provides excellent access to the infratemporal skull base, orbital apex, and lateral maxilla. The temporalis muscle is reflected inferiorly until the infratemporal crest is fully visualized. A subperiosteal plane is followed to dissect the soft tissues from the infratemporal cranium. Bleeding from the underlying bone is controlled by the application of bone wax. Fracturing or removal of the coronoid process increases the arc of rotation of the temporalis muscle. Care should be exercised when dissecting the medial aspect of the temporalis muscle, especially near its insertion (coronoid process) because the blood supply to the muscle (deep temporal artery from the internal maxillary artery) penetrates the muscle at this area. Likewise, the soft tissue at the sigmoid notch should be dissected carefully to prevent accidental injury to the internal maxillary artery that travels adjacent to the medial surface of the mandibular ramus.
Dissection of the soft tissues from the infratemporal skull base is usually associated with troublesome bleeding arising from the pterygoid plexus. Bleeding is controlled with the use of bipolar cautery or Cottonoids moistened in oxymetazoline 0.05%, or both. Unipolar cautery is seldom used because it stimulates V3, causing contraction of the mastication muscles and occasional cardiac arrhythmias. A subtemporal craniectomy may aid in the identification of neurovascular structures piercing the infratemporal skull base and to augment the exposure. The lateralmost bone is removed using rongeurs. The origin of the lateral pterygoid plate at the skull base is identified anteriorly. Anatomic relationships that are useful for the identification of infratemporal skull base structures are shown in Figure 54-4, including the posterior curve of the attachment of the lateral pterygoid plate that is in alignment with the foramen ovale, foramen spinosum, and the spine of the sphenoid bone. These structures lie in a straight “line of sight” that is lateral to the canal of the ICA. The inferolateral aspect of the sphenoid sinus may be accessed by removing the bone (i.e., pterygoid plates) between the second and third divisions of the trigeminal nerve. The extirpation of the tumor can now proceed, including the involved soft tissue and bone. To continue the subtemporal exposure, the middle meningeal artery is clipped or cauterized using bipolar electrocautery and transected. Bleeding from the venous plexus that accompanies V3 through the foramen ovale may be controlled with absorbable knitted fabric (Surgicel) packing. Lesions that do not involve the temporal bone or petrous portion of the ICA are adequately exposed with this stepwise approach. Dissection of the petrous ICA is necessary, however; the glenoid fossa is removed as part of the orbitozygomatic bone graft. It is first necessary to perform a temporal craniotomy for exposure of the superior aspect of the glenoid fossa (Fig. 54-5). The capsule of the TMJ is dissected free from the fossa and displaced inferiorly. If possible, the capsule and meniscus are preserved. With use of a reciprocating saw, osteotomies are made through the glenoid fossa, incorporating the lateral two thirds of the fossa (see Fig. 54-5). This maneuver avoids potential injury to the ICA that is located medial to the fossa. In addition, this modification provides stability for the mandibular condyle after reconstruction, although it can be prone to anterior dislocation. Injury to the cochlea is possible if the osteotomies are made too posteriorly. If additional exposure is necessary (i.e., carotid canal and extratemporal ICA), the condylar neck and contents of the condylar fossa can be transected at the level of the sigmoid notch and removed (Fig. 54-6). To dissect the petrous segment of the ICA, it is necessary to transect the mandibular division of the trigeminal nerve at the foramen ovale (see Fig. 54-6). When the ICA is mobilized from its horizontal canal, it can be transposed or retracted to facilitate the resection of tumor, or to gain access to the petrous apex.
Chapter 54 • Anterior and Subtemporal Approaches to the Infratemporal Fossa Foramen spinosum
655
Spine of sphenoid bone
Lateral plate of sphenoid Foramen ovale Glenoid fossa
Scaphoid fossa
Eustachian tube Carotid canal
EAC
B Carotid canal and artery
A
Jugular fossa
FIGURE 54-4. A, Base of skull showing anatomic relationships of carotid canal. Foramen ovale and foramen spinosum are in a direct line from lateral pterygoid plate to spine of sphenoid.
Burr hole
Inferior orbital fissure
FIGURE 54-5. Using intracranial and extracranial approaches, osteotomies of superior orbital roof, lateral orbital wall, and glenoid fossa are performed.
656
OTOLOGIC SURGERY
Zygomaticoorbital osteotomy
Tumor
Internal carotid artery V2 V3
VII Jugular vein XI X
Temporalis muscle Parotid gland IX Glossopharyngeal XII Stump of mandible
FIGURE 54-6. Tumor medial to mandibular division of trigeminal nerve and in close proximity to petrous portion of internal carotid artery. Additional exposure of internal carotid artery and removal of tumor often necessitate transection of mandibular division of trigeminal nerve.
Approaches to the ITF are modified according to the extent of the tumor and other clinical circumstances. Tumors that invade the mandible mandate a partial mandibulectomy to obtain negative margins. In a pediatric patient, the distance from the body of the mandible to the infratemporal skull base is greatly foreshortened. Adequate exposure of the infratemporal skull base can often be achieved using a transcervical approach with superior transposition of the facial nerve. After extirpation of the tumor, it is necessary to close any communication with the upper aerodigestive tract. If viable, a temporalis muscle flap is used to obliterate the dead space and protect the ICA (Fig. 54-7). Because of the branching pattern of the blood supply to the temporalis muscle, the muscle can be divided vertically, and the anterior half of the muscle may be transposed with an intact blood supply. The remaining posterior half of the muscle is transposed anteriorly to fill the temporal fossa defect. Defects of the orbital floor may be reconstructed with titanium mesh, which is then covered with a temporalis muscle transposition flap or temporoparietal fascia flap.
Likewise, defects of the lateral orbital wall can be reconstructed with titanium mesh. In selected patients, anteriorly or posteriorly based pericranial scalp flaps may be elevated to provide protection of the infratemporal skull base. When the temporalis muscle is unavailable, massive soft tissue defects are best reconstructed with microvascular free tissue flaps. The orbitozygomatic bone graft is replaced and fixated in its original position with titanium alloy adaptation plates, wire, or braided nylon sutures. Plating of the bone grafts is preferred because it provides greater stability. To avoid compression of reconstructive flaps, it is sometimes necessary to remove a portion of the zygomatic arch. If resection of the mandibular condyle was necessary to expose the petrous ICA, reconstruction of the TMJ is not attempted. Reconstruction of the TMJ after oncologic exenteration of the ITF does not improve the postoperative function significantly, and may lead to scarring, ankylosis, and trismus. Periosteal and muscular attachments to the craniofacial skeleton must be repaired to prevent retraction or
Chapter 54 • Anterior and Subtemporal Approaches to the Infratemporal Fossa
657
Temporalis muscle in defect Temporalis muscle
FIGURE 54-7. Anterior portion of muscle may be transposed to fill infratemporal skull base defect, as illustrated with this temporalis muscle. Right temporalis muscle has been split vertically with preservation of its axial blood supply.
sagging, or both, of the muscles and other soft tissues. The skin and mucosal incisions are closed using a multilayered technique
Postauricular (Transtemporal) Approach The postauricular approach is designed to expose and resect lesions involving the temporal bone and extending into the ITF.3,14 A question mark–shaped or C-shaped incision is started in the temporal area and extended, postauricularly, into the mastoid region, curving down to follow one of the midneck horizontal skin creases (Fig. 54-8). If the middle ear is to be sacrificed as part of the approach or the tumor resection, and there is a risk of a postoperative CSF leak (intradural work), the external auditory canal (EAC) is closed permanently to prevent CSF otorrhea. The EAC is divided at the bonycartilaginous junction and then closed using everting stitches. This closure is reinforced with a myoperiosteal U-shaped flap based on the posterior margin of the EAC. Alternatively, if the middle ear is to be spared, the canal may be preserved by placing the incisions in the conchal area (Fig. 54-8B). The incision follows the margin of the conchal bowl and tragus so that the scar would be hidden. In the conchal area, the skin, cartilage, and perichondrium are incised to communicate with the retroauricular plane of dissection. These incisions, placed laterally, facilitate the anastomosis of the EAC to the pinna at the end of the extirpative procedure. An incision inside the EAC is not recommended because it is
difficult to suture in a watertight manner, and tends to stenose. A Penrose drain can be inserted through the conchal defect in the skin-auricle flap to facilitate its retraction. Elevation of the cervicofacial flap is carried in a subplatysmal plane in the cervical area, suprasuperficial musculoaponeurotic system plane over the parotid area, and following the deep layer of the deep temporal fascia over the cranium. The main trunk of the facial nerve is identified anterior to the EAC just distal to the stylomastoid foramen, as is described for a parotidectomy. If circumferential mobilization of the main trunk is unnecessary, a cuff of soft tissue is preserved around its main trunk to minimize the possibility of a traction injury when the facial flap is retracted anteriorly. In selected cases, a “tail” parotidectomy may enhance the access to the retromandibular area. A total parotidectomy is indicated when facing an epithelial malignancy of the parotid gland. Skeletonization of the main trunk of the facial nerve and its branches facilitates their retraction and the access to the ITF. Resection of the main trunk of the facial nerve and its branches (radical parotidectomy) is indicated when the nerve is invaded by the tumor. Attention is then directed to the cervical exposure to obtain proximal control of the common, internal, and external carotid arteries, and the internal jugular vein. CN X to XII are identified and preserved. The sternocleidomastoid and digastric muscles are transected at their insertion to the mastoid bone. The stylohyoid and stylopharyngeus muscles are transected, and the styloid process is removed. CN IX usually can be identified at this time, as it crosses lateral to the ICA.
658
OTOLOGIC SURGERY Curvilinear incision
Conchal incision
Conchal incision
A B
FIGURE 54-8. A, Curvilinear incision is made from temporal area to mastoid bone and upper cervical region. Flap is elevated superficial to deep temporal fascia, deep to mastoid periosteum, and deep to platysma muscle. B, Conchal incision is preferable to incision of external auditory canal when permanent obliteration of the ear is not indicated.
A mastoidectomy and dissection of the vertical portion of the facial nerve allows the transposition of the facial nerve, providing a wider access to the infratemporal fossa (see Fig. 54-9). In patients who require a radical parotidectomy, a mastoidectomy provides the means to obtain proximal control of the neural margins and to graft the nerve. It also provides access to the jugular bulb and adjacent lower cranial nerves. Orbitozygomatic osteotomies may be performed as previously described (preauricular approach). After the orbitozygomatic complex is removed, the anterior, superior, medial, and posterior boundaries of the infratemporal fossa are well exposed, and all major vessels are “controlled.” Completion of the infratemporal skull base approach, including a temporal craniotomy, is performed as described in the previous section. The extirpation of the tumor can now proceed, including the involved soft tissue and bone. Reconstruction of the defect follows the principles outlined in previous sections.
Lateral Fisch Infratemporal Fossa Approaches Fisch has described an array of lateral infratemporal fossa approaches that are the prototypic otologic approaches to the ITF (Fig. 54-8,9,10)3. The hallmark of these approaches
is temporal bone management emphasizing facial nerve rerouting and subtemporal dural exposure for wide access to the lateral skull base. Figure 54-10 illustrates the anatomic regions appropriate for the Fisch type A, B, and C approaches. The Fisch type A approach has been described in detail in Chapter 46 regarding its application in glomus jugulare surgery. The Fisch type B and C approaches are designed to approach more anterior pathology involving the petrous apex and clivus. The type C approach is an extension of the type B approach, and is used for lesions of the anterior ITF, sella, and nasopharynx. The Fisch type D is a preauricular ITF approach using an orbitozygotomy and resection of the floor of the middle fossa for medial dural exposure without a lateral temporal craniotomy.
Fisch Type B The principal exposure maneuver in the Fisch type B ITF approach is reflection of the zygomatic arch and temporalis muscle inferiorly and removal of the bone of the skull base floor to provide access to the ITF. The incision is wide and C-shaped beginning at the angle of the mandible and extending retroauricularly and anterior laterally to the eyebrow. The ear canal is transected and closed in the same manner as described in Chapter 47 in a two-layer technique.
Chapter 54 • Anterior and Subtemporal Approaches to the Infratemporal Fossa
659
Raised fascia (deep)
1
Fat pad 2 Facial flap V3
Parotid gland
Ear canal sewn Facial nerve isolated
Temporalis muscle
3
A
B
Temporalis muscle Stump of mandible
FIGURE 54-9. Fisch B approach. A, The external ear canal has been divided and sewn, the facial nerve is exposed, the internal carotid artery is exposed after a TMJ exenteration. B, V3 is exposed after a subtemporal craniotomy. Surgical exposure obtained after temporal craniotomy and reflection of temporalis muscle inferiorly. 1, temporal fat pad; 2, facial nerve.
The main trunk of the facial nerve is identified using standard landmarks at the stylomastoid foramen. The superior division is followed to the level of the frontalis branch. With this direct visualization, the periosteum attached to the zygomatic arch is reflected down to protect the frontalis branch of the facial nerve. At this point, the osteotomies of the zygomatic arch can be completed anteriorly as close as possible to the orbital rim and posteriorly at the root of the zygoma. The masseter muscle is left attached at the zygomatic arch to be reflected inferiorly. The temporalis muscle is completely reflected inferiorly, carefully protecting its blood supply. A key to this extradural exposure is the subtotal “petrosectomy.” This step includes a canal wall down mastoidectomy including complete skeletonization of the labyrinth, facial nerve, sigmoid sinus, middle fossa, and posterior fossa dura and the jugular bulb, and exenteration of all hypotympanic air cells and skeletonization of the ICA. The unique aspect is that by removing the ossicular chain and middle ear structures, the carotid artery can be skeletonized completely beyond the genu to the foramen lacerum. The TMJ is disarticulated by incising the capsule and removing the articular disc. At this point, the bone of the glenoid fossa and the root of the zygoma are completely removed with cutting and diamond burrs. Middle fossa dura in the subtemporal region is completely skeletonized. By placing the infratemporal fossa retractor over the mandibular condyle, additional skeletonization of the
middle fossa dura can be accomplished medially until reaching the middle meningeal artery and V3 at the foramen ovale. Cauterization of the middle meningeal artery and transection of the mandibular nerve permit greater exposure. At this point, further skeletonization of the carotid artery is possible along the lateral and anterior wall of the ICA. Complete exposure of the carotid artery permits its mobilization out of the carotid canal, providing free access to the petrous apex and clivus. The eustachian tube must be sutured closed to prevent infection of the nasal cavity. Although the bone defect is filled with abdominal fat, temporalis muscle is used to cover the fat and is placed inferior to the skeletonized middle fossa dura and mandibular condyle. The zygomatic arch can be secured with microplates, and the skin can be closed in a standard fashion.
Fisch Type C The principal distinguishing feature of the Fisch type C approach compared with the type B approach is resection of the pterygoid plates. This resection permits exposure of the lateral wall of the nasopharynx, eustachian tube orifice, posterior maxillary sinus, and posterior nasopharyngeal wall past the midline. After completion of the type B approach, the lateral surface of the pterygoid process is identified, and soft tissues are elevated. In this manner, the base of medial and lateral plates of the pterygoid processes can be drilled away, exposing the lateral wall of the nasopharynx. The
660
OTOLOGIC SURGERY
FIGURE 54-10. Different exposures obtained with Fisch A, B, and C approaches to infratemporal fossa. EAC, external auditory canal; IAC, internal auditory canal; ICA, internal carotid artery.
nasal cavity can be entered. The exposure permits full visualization of the peritubal area, which can be resected en bloc. Inferiorly, the excision can extend to the upper surface of the palate. Superiorly, the dissection can extend to the carotid artery and the cavernous sinus. The wide communication between the nasopharynx and the operative field makes closure in type C ITF surgery more difficult than in type B. Although mobilization of the entire temporalis muscle into the wound is one technique, vascularized free flaps are often necessary to provide adequate closure. Abdominal fat should be avoided because of the possibility of contamination.
Fisch Type D The Fisch type D ITF exposure is a preauricular modification of the Fisch infratemporal approach. Type D1 addresses tumors of the anterior infratemporal fossa, whereas type D2 is designed for lateral orbital wall lesions and high pterygopalatine fossa tumors. The distinguishing feature of the type D approach from the B and C approaches is that the middle ear and eustachian tube area is not obliterated, and conductive hearing is not sacrificed. In addition, the intratemporal facial nerve
is not rerouted, and the petrous ICA is not fully exposed. Although these preauricular approaches do not include a temporal craniotomy, the floor of the skull base can be drilled away to allow full access to the ITF.
Anterior Transfacial Approach (Facial Translocation) The anterior transfacial technique is best used to approach sinonasal tumors invading the ITF, the masticator space, or the pterygomaxillary fossa, and for tumors of the nasopharynx extending into the ITF (Fig. 54-11).7,13,14 A bicoronal incision with an ipsilateral preauricular extension is performed and extended through the subcutaneous tissue (see the section on preauricular approach). A Weber-Fergusson incision is completed and extended down to the periosteum of the maxilla, nasal bones, and orbital rim. During a “traditional” translocation approach, a horizontal incision is carried over the superior edge of the zygomatic bone, extending into the lateral canthus, to meet the Weber-Fergusson incision (see Fig. 54-11). The frontal branches of the facial nerve are identified and dissected as they cross over the zygomatic arch. They are then entubulated with silicone tubing and transected.
Chapter 54 • Anterior and Subtemporal Approaches to the Infratemporal Fossa
A
Masseter muscle
B
661
C
FIGURE 54-11. Incisions and osteotomies for facial translocation approach as described by Janecka.13,14
These nerve branches are reanastomosed at the end of the case, using an entubulation technique. Subperiosteal dissection of the anterior maxilla exposes the infraorbital nerve, which is then transected and tagged to facilitate its identification and reanastomosis at the end of the case. Then, an inferiorly based flap including the upper third of the upper lip, entire cheek, lower eyelid, parotid gland, and facial nerve is reflected inferiorly. The frontotemporal scalp flap is elevated in a subpericranial plane. This flap is reflected anteriorly, exposing the superior orbital rims (see Fig. 54-11). Alternatively, the exposure can be achieved without the temporal incision by combining the preauricular approach with the anterior exposure provided by the WeberFergusson incision. Orbitozygomatic osteotomies are performed and joined with the maxillary osteotomies to free the anterior face of the ipsilateral maxilla en bloc with the orbitozygomatic complex. Alternatively, the maxillary bone graft can be elevated as a vascularized graft attached to the cheek flap, as described by Catalano and Biller.8 The temporalis and masseter muscles are dissected from the zygomatic bone with electrocautery. Osteotomies are completed, and the bone graft is removed. The temporalis muscle is reflected inferiorly. Removal of the coronoid process increases the caudal arc of rotation of the temporalis muscle. After completion of these steps, the anterior, medial, and lateral boundaries of the ITF are well exposed. In selected cases, the pterygoid plates can be excised to provide further access to the medial ITF or nasopharynx. A temporosubtemporal craniotomy provides additional exposure superiorly and allows dissection of intracranial structures (Fig. 54-12). After the tumor resection, the temporalis muscle may be used to obliterate the surgical defect and provide separation of the cranial cavity from the upper aerodigestive tract, as previously described.
Periosteal and muscular attachments are repaired, and the incisions are closed using a multilayer technique. The conjunctiva is repaired with running 6-0 fastabsorbing suture. The lacrimal canaliculi are stented with Crawford silicone tubing that is tied to itself in the nasal cavity. The eye is closed with a temporary tarsorrhaphy for 10 to 14 days to prevent a lower lid ectropion.
Transorbital Approach In selected cases, a transorbital approach may be used to complement the exposures obtained with one of the previous approaches enhancing the exposure of the orbital apex and cavernous sinus. This approach consists of transection of the orbital tissues posterior to the globe with preservation of the attachments of the orbital soft tissues, including the globe, to the scalp flap. The orbital apex is removed to provide direct anterior access to the cavernous sinus and cavernous ICA. This approach is reserved for patients with benign tumors of the orbita apex and cavernous sinus who have lost vision as a result of tumor growth. It may also be employed for low-grade malignant neoplasms with minimal involvement of the orbital soft apex or optic nerve to obtain complete tumor removal. Extensive involvement of the orbital soft tissues requires an orbital exenteration. The advantages of this approach include improved cosmesis, a result of the preservation of the globe, and excellent anterior and lateral exposure of the cavernous sinus and its associated structures. A preauricular infratemporal skull base approach is performed, as previously described. The periorbita is elevated from the superior, lateral, and inferior walls of the orbit. The periorbita is then incised, and the orbital tissues are transected posterior to the globe using bipolar electrocautery and sharp dissection. A cuff of tissue remains at the orbital apex to provide an adequate tumor margin.
662
OTOLOGIC SURGERY
FIGURE 54-12. Subtemporal craniectomy is performed to provide additional exposure and adequate resection margin of extracranial tumors. Maximal exposure of infratemporal and central skull base is achieved.
The remaining periorbital attachments are elevated medially to allow complete displacement of the globe from the orbital cavity. Over the medial wall, the neurovascular bundles are clipped or cauterized and transected. The lacrimal duct is transected, and the sac is marsupialized. Using rongeurs, bone is removed from the lateral wall of the orbit to the superior orbital fissure. The contents of the superior orbital fissure and optic canal are transected to provide additional exposure of the orbital apex. Because of the loss of orbital bone, enophthalmos results unless the orbital defect is reconstructed with bone grafts or titanium mesh. A temporalis transposition or free tissue transfer provides soft tissue augmentation and protection of the carotid artery.
POSTOPERATIVE CARE After surgery, the patient is transferred to the intensive care unit for continuous cardiovascular and neurologic monitoring. Laboratory tests to rule out postoperative anemia and electrolyte imbalance are performed. Patients who required multiple blood transfusions should be screened for transfusion-induced coagulation disorders. Mild narcotic analgesia is provided, avoiding sedation that could interfere with a detailed neurologic evaluation.
If the ICA is dissected, ligated, or grafted, close monitoring of the patient’s hemodynamic status and fluid balance is essential. When grafting of the ICA is performed, an angiogram is obtained in the early postoperative period to assess the patency of the graft and detect pseudoaneurysm formation. A CT scan of the brain without contrast medium is performed on the 1st or 2nd postoperative day to screen for intracranial complications, such as cerebral contusion, edema or hemorrhage, fluid collections, or pneumocephalus. A compressive dressing is maintained for 24 to 72 hours. When the dressing is removed, the wound is cleaned with normal saline solution and covered with antibiotic ointment three to four times a day. The scalp and other wound drains are kept to bulb suction until the drainage is less than 30 mL/day. The drain is then removed, and the wound is closed using an encircling stitch placed at the time of surgery. If the cranial cavity is entered, wall suction is never used because of the risk of direct negative pressure on the central nervous system. In most cases, the spinal drain is needed only during the surgery and is removed when the procedure is completed. If there is a significant risk of postoperative CSF leak, the spinal drain is kept at the level of the patient’s shoulder, and 50 mL is removed every 8 to 12 hours. The lumbar drain is removed 3 to 5 days after surgery,
Chapter 54 • Anterior and Subtemporal Approaches to the Infratemporal Fossa
and the lumbar puncture site is closed with an encircling stitch (e.g., 2-0 nylon), placed at the time of surgery. Lagophthalmos from weakness or paralysis of the facial nerve may lead to exposure keratoconjunctivitis. Initially, an exposed eye can be protected using artificial tears every 1 to 2 hours, lubrication ointment at bedtime, eye patching, or a moisture chamber. Taping of the eyelids or a temporary tarsorrhaphy is advised if rapid recovery is anticipated. If prolonged paralysis is expected, we prefer a gold weight implant. This procedure can be performed at the time of the original surgery, using a 0.1 to 0.12 g weight (No. 10 or 12), or it may be performed during the early postoperative period. Except in selected cases, we favor the latter because it provides the advantage of being able to establish the exact weight that is needed by the patient. In most cases, the airway can be secured for the short-term using an endotracheal tube (high-volume/ low-pressure cuff). Nevertheless, a tracheotomy is indicated for patients in whom significant edema of the upper aerodigestive tract is anticipated, or if a prolonged mechanical ventilation is anticipated. A tracheotomy also provides better access to the airway for pulmonary toilet for patients with an ineffective cough or severe aspiration. Patients with high vagal lesions or any combination of deficits of CN IX, X, or XII experience severe swallowing difficulty and aspiration. These patients can be assisted with a medialization laryngoplasty, an arytenoid adduction procedure, or an arytenoidpexy with or without a cricopharyngeal myotomy. Patients who continue to aspirate despite all these measures and who develop repeated aspiration pneumonias are managed by a laryngotracheal separation procedure.
PITFALLS AND COMPLICATIONS The most common morbidity associated with surgery of the infratemporal fossa is related to deficits of the trigeminal nerve. Sacrifice of the third, sometimes the second, and rarely the first divisions of the trigeminal nerve may be necessary for surgical exposure or to obtain adequate clear margins of resection. Facial anesthesia may predispose the patient to self-inflicted injuries, including neurotrophic ulcers. The loss of corneal sensation, especially in a patient with paresis of the facial nerve, greatly increases the risk of a corneal abrasion or exposure keratitis. The loss of motor function of the mandibular nerve causes asymmetry of jaw opening and decreased force of mastication on the operated side. Mastication may be impaired further by resection of the TMJ or mandibular ramus. Whenever feasible, sensory and motor divisions of the trigeminal nerve are repaired or grafted after transection for surgical exposure. Permanent deficits (accidental) of the facial nerve or its branches are uncommon. The frontal branches of the
663
facial nerve are at risk of injury during elevation of the temporal scalp flap. Injury is usually the result either of a dissection in a plane that is superficial to the superficial layer of the deep temporal fascia or of compression during the retraction of the flap. To avoid a traction injury, a cuff of soft tissue is preserved around the main trunk of the facial nerve when a preauricular approach is employed. The facial nerve can also sustain an ischemic injury that occurs as a result of devascularization on mobilization of its infratemporal segments or skeletonization of its extratemporal segments. A temporary paresis of the facial nerve is to be expected with mobilization of the mastoid segment of the facial nerve. Close attention to postoperative eye care is necessary in patients with combined deficits of the trigeminal and facial nerves. In most patients, surgical resection of the TMJ is not a major factor in the development of postoperative trismus or difficulties with mastication. Rather, mastication seems to be most affected by loss of function of the mandibular division of the trigeminal nerve. Nevertheless, every effort is made to preserve the TMJ. If resection of the glenoid fossa is necessary, the capsule of the TMJ is displaced inferiorly. If resection of the TMJ is necessary, no attempt is made to reconstruct the joint. These patients experience deviation of the jaw to the unaffected side. This is usually of no major consequence, but some patients may need an occlusal guide to help them when chewing. Postoperative trismus is also a common occurrence because of postoperative pain and scarring of the pterygoid musculature and TMJ. Trismus improves dramatically if patients regularly perform stretching exercises for the jaw. Devices such as the TheraBite appliance are helpful in stretching the scar tissue and forcefully opening the mouth. In severe cases, a dental appliance may be fabricated that is gradually opened by a screw. Infectious complications are rare. Predisposing factors include communication with the nasopharynx, seroma or hematoma, and a CSF leak. Generally, the dead space should be obliterated to prevent fluid collection that subsequently can be infected, and the cranial cavity should be separated from the sinonasal tract. The use of vascularized tissue flaps is preferred, especially when there has been dissection of the ICA or resection of dura. Necrosis of the scalp flap is uncommon because of its excellent blood supply. Poorly designed incisions may result, however, in areas of ischemia, particularly around the auricle, which can make the tissue susceptible to secondary infection. Prolonged use of hemostatic clamps can also lead to necrosis of the wound edges. Neurovascular complications are of the greatest concern. Postoperative cerebral ischemia may result from surgical occlusion of the ICA, temporary vasospasm, and thromboembolic phenomena. Surgical dissection of the ICA can injure the vessel walls, resulting in immediate or delayed rupture and hemorrhage. The ICA is particularly
664
OTOLOGIC SURGERY
vulnerable to injury where it enters the cranial base. Injuries to the ICA should be repaired primarily (or using a vein graft). An angiogram is obtained in the early postoperative period to assess the adequacy of the repair. If a repair of the ICA is impossible, it should be permanently occluded by ligation or by the placement of a detachable balloon or vascular coil. When the artery is to be permanently occluded, the occlusion is performed as distal as possible (near the origin of the ophthalmic artery). The potential for thrombus formation is less with a short column of stagnant blood above the level of occlusion. After occlusion of the ICA, there is a significant risk of immediate and delayed stroke in patients who do not have more than 35 to 40 mL of blood flow per 100 g of brain tissue per minute by ABOX-CT testing. After reconstruction of the ICA with a vein graft, there is a risk of postoperative occlusion because of thrombus formation at the suture line, and torsion or kinking of the graft. Pseudoaneurysm formation and delayed blowout of the graft are also risks, especially in the presence of infection. For this reason, reconstruction of the ICA is usually not indicated in a contaminated field with communication to the upper aerodigestive tract. In such cases, permanent occlusion of the ICA or rerouting of a vein graft posterior to the surgical field is performed. An extracranial-intracranial bypass graft to the middle cerebral artery may be performed before tumor resection when sacrifice of the ICA is anticipated. Patients who undergo surgical manipulation of the ICA may also develop cerebral ischemia at the margins of the vascular territories of the cerebral vessels (watershed areas). This ischemia is of particular concern when there is sacrifice of extracranial-intracranial collateral blood vessels, which are not routinely assessed by ABOX-CT as part of the surgical approach. Decreased oxygen delivery because of hypoxic postoperative anemia or hypotension can result in a cerebral infarct in these watershed areas. A watertight dural closure may be difficult to achieve with large infratemporal skull base defects, particularly around nerves and vessels. An epidural fluid collection may result. In most cases, this fluid collection is contained by the soft tissues and slowly resolves without further intervention. Occasionally, the CSF collection may communicate with the exterior through the EAC, the scalp incision line, or along the eustachian tube to the nasopharynx. Most CSF leaks can be managed nonsurgically by placement of a pressure dressing and a spinal drain to diminish the CSF pressure. Surgical exploration and repair of the dural defect may be necessary if the CSF leak does not resolve within 1 week. A middle ear effusion is often apparent after infratemporal skull base approaches because of dysfunction or interruption of the eustachian tube. Tympanostomy tubes are not placed for at least 6 weeks postoperatively, however, because there is always a risk of CSF communication. We have encountered patients who developed profuse unilateral rhinorrhea in the postoperative period that
was misinterpreted as a CSF leak. These cases all were associated with surgical dissection of the petrous ICA and are probably due to loss of the sympathetic fibers that travel along the ICA in their route to the nasal mucosa. This loss produces vasomotor rhinitis that may be treated with the use of anticholinergic nasal sprays. Testing of the fluid for β2-transferrin is mandatory, however, to rule out a CSF leak. Cosmetic deformities may result from the loss of soft tissue and bone. Transposition of the temporalis muscle results in a depression in the temporal area. This depression can be lessened by placement of a free-fat graft or hydroxyapatite cement in a secondary surgery. If the temporalis muscle is not transposed, the anterior margin of the muscle should be resutured anteriorly and superiorly to prevent its retraction and a resulting depression lateral to the orbital rim. The use of microvascular free muscle flaps, such as the rectus abdominis flap, for reconstruction may necessitate sacrifice of the zygomatic arch to accommodate the additional bulk. As the muscle atrophies, significant depression may occur. It is important to repair all periosteal and muscle attachments around the maxilla, orbital rim, and zygomatic arch to avoid a “cadaveric” look that occurs when the soft tissues over these areas atrophy or retract. Large muscle flaps, such as a latissimus dorsi flap, may swell and compress the brain if the cranial base is not reconstructed.
REFERENCES 1. Barbosa J F: Surgery of extensive cancer of paranasal sinuses: Presentation of a new technique. Arch Otolaryngol 73:129-138, 1961. 2. Terz JJ, Young H F, Lawrence W Jr: Combined craniofacial resection for locally advanced carcinoma of the head and neck, II: Carcinoma of the paranasal sinuses. Am J Surg 140:618-624, 1980. 3. Fisch U: The infratemporal fossa approach for the lateral skull base. Otolaryngol Clin North Am 17:513-552, 1984. 4. Biller H F, Shugar J M A, Krespi YP: A new technique for wide-field exposure of the base of the skull. Arch Otolaryngol 107:698-707, 1981. 5. Sekhar L N, Schramm VL , Jones N F: Subtemporalpreauricular infratemporal fossa approach to large lateral and posterior cranial base neoplasms. J Neurosurg 67:499, 1987. 6. Cocke EW Jr, Robertson J H, Robertson JT, Crooke J P Jr: The extended maxillotomy and subtotal maxillectomy for excision of skull base tumors. Arch Otolaryngol Head Neck Surg 116:92-104, 1990. 7. Janecka I P, Sen C N, Sekhar L N, Arriaga M : Facial translocation: A new approach to the cranial base. Arch Otolaryngol Head Neck Surg 103:413-419, 1990. 8. Catalano PJ, Biller H F: Extended osteoplastic maxillotomy: A versatile new procedure for wide access to the central skull base and infratemporal fossa. Arch Otolaryngol Head Neck Surg 119:394-400, 1993.
Chapter 54 • Anterior and Subtemporal Approaches to the Infratemporal Fossa 9. Snyderman C H, Carrau R L , de Vries E J: Carotid artery resection: Update on preoperative evaluation. In Johnson JT, Derkay C S, Mandell-Brown M K, Newman R K (eds): AAO-HNS Instructional Courses, 6. Chicago, IL, Mosby Year Book, 1993, pp 341-346. 10. Netterville J L , Jackson G, Civantos F: Thyroplasty in the functional rehabilitation of neurotologic skull base surgery patients. Am J Otol 14:460-464, 1993. 11. Carrau R L , Pou A, Eibling D E, et al: Laryngeal framework surgery for the management of aspiration. Head Neck 21:139-145, 1999.
665
12. Pou A, Carrau R L , Eibling D E, Murry T: Laryngeal framework surgery for the management of aspiration in high vagal lesions. Am J Otolaryngol 19:1-8, 1998. 13. Nuss DW, Janecka I P, Sekhar L N, Sen C N: Craniofacial disassembly in the management of skull-base tumors. Otolaryngol Clin North Am 24:1465-1497, 1991. 14. Sekhar L N, Sen C, Snyderman C H, Janecka I P: Anterior, anterolateral, and lateral approaches to extradural petroclival tumors. In Sekhar L N, Janecka I P (eds): Surgery of Cranial Base Tumors. New York, Raven Press, 1993, pp 157-223.
55
Endoscopic Endonasal Approaches to the Skull Base and Paranasal Sinuses Paul A. Gardner, Amin B. Kassam, Carl H. Snyderman, Ricardo L. Carrau, and Daniel M. Prevedello Videos corresponding to this chapter are available online at www.expertconsult.com.
INTRODUCTION/BACKGROUND Approaches to the skull base via the paranasal sinuses were introduced in the late 19th and early 20th century. Tumors of the sellar region were first approached through various incisions in the forehead to gain access to the sphenoid sinus via the ethmoid sinuses. Cushing introduced what became the standard for transsphenoidal surgery in 1910 when he first used a sublabial incision to gain access to the sinuses.1 coincidentally, on the exact same day, June 4, Oscar Hirsch used an endonasal approach to gain access to the sphenoid sinus. This would eventually replace the sublabial approach, but Cushing’s influence delayed this for decades. This transsphenoidal approach to the sella turcica was conceptually unchanged for approximately 70 years. However, the operation and its outcomes were changed greatly during this period by advances in technology which provided improved visualization, intraoperative image-guidance and instrumentation. Dott, Guiot and Hardy added pneumoencephalocisternography, radiofluoroscopic guidance, and the operative microscope. All of this revolutionized the approach and improved outcomes. A similar change in technology began in the 1980s with the development of early intraoperative CT and MR image-guidance systems and the introduction of the endoscope by Carrau.2,3 At the same time, otolaryngologists were developing and refining functional endoscopic sinus surgery (FESS).4-9 It was the collaboration between otolaryngologists and neurosurgeons that led to the expansion of the standard transsphenoidal approach to include the rest of the anterior skull base and beyond.
BASIC ENDOSCOPIC ENDONASAL CONCEPTS The application of endoscopy to the endonasal approach has allowed the expansion of this approach while improving visualization. This is due to one basic optic difference
when compared to a microscope. A microscope visualizes from a distance and delivers light in a cone whose apex is at the target. This requires a superficial exposure which is wider than the deep exposure (“ice cream cone” effect). An endoscope delivers light and provides a view in a cone whose apex is at the tip of the scope. This allows a smaller exposure superficially while allowing a much wider working field in the depth (“flashlight” effect). This becomes a distinct advantage when approaching deep lesions with complex surroundings, such as those involving the skull base and paranasal sinuses. This does create a problem with instrumentation and the modification of existing and development of new instrumentation was necessary and is still ongoing. The loss of three-dimensional visualization is easily overcome using an active, handheld endoscope and proprioceptive cues obtained by keeping one instrument on or near the object of focus. Three-dimensional sensation is recreated via propioception and triangulation of instruments. This is a part of the learning curve, but one which is relatively easy to overcome. Despite these differences, endonasal techniques should not differ from standard microsurgical techniques. Two-handed dissection by the operating surgeon is required and therefore two surgeons are needed. Identification of critical structures, central debulking followed by extracapsular dissection and fine, sharp dissection are all critical components of microsurgery and should be directly translated to endoscopic neurosurgery (endoneurosurgery). One of the main reasons these approaches hold such promise is that many skull base lesions are medial and anterior to the surrounding neurovascular structures. This provides a distinct advantage to a direct anterior, midline approach, as these structures do not have to be displaced or transgressed in order to access them. This principle guides the selection of tumors for EEA. Vascularity, tumor consistency, and size represent important surgical considerations but do not constitute contraindications to an EEA. 667
668
OTOLOGIC SURGERY
approaches. The middle fossa is the most complex and is broken into five “transpterygoid” anatomic zones of approach: medial pterygoid (petrous apex), petroclival junction, quadrangular space or Meckel’s cave, superior cavernous sinus, and the infratemporal fossa (Fig. 55-3). These are all critically dependent upon their relationship with the internal carotid artery (ICA). Finally, the posterior fossa or inferior expanded approaches consist of the transcondylar and parapharyngeal space approaches.
EXPANDED ENDONASAL APPROACHES: ANATOMIC MODULES The key to the development of endonasal approaches to skull base pathology was the collaboration between otolaryngologists and neurosurgeons. The knowledge of the paranasal sinuses developed by otolaryngologists melded with the work neurosurgeons had done in the sella turcica for pituitary tumors. Both were supplemented by knowledge of skull base anatomy as learned through performing traditional, open approaches. Endoscopic endonasal approaches (EEAs) can be divided into the sagittal or rostral-caudal plane (between the carotids) and coronal or paramedian plane (lateral to the carotid). The sagittal plane can be divided from rostral to caudal, anterior to posterior into the transfrontal, transcribriform, transplanum/transtuberculum, transsellar, transclival and transodontoid approaches (Fig. 55-1). The coronal plane is somewhat more complex in that the lateral expanded approaches vary based upon which fossa is involved (Fig. 55-2). The anterior fossa, lateral approaches consist of the supra- and transorbital
Sagittal Plane This represents the region between the internal carotid arteries (ICA) as it rises and courses along the ventral skull base from a caudal to rostral direction as well as the rostral extension of this midline region. The sagittal plane is also referred to as the rostral-caudal plane.
Transfrontal Approach The transfrontal approach represents the most anterior expanded endonasal approach. A transfrontal approach starts with identification of the nasofrontal recesses and
Modular approaches
I II P. 1.
S. S.
V CN 7, 8 CN 9, 10, 11
S.C.
2. 3.
M.C.
4.
I.C.
12
5.
P.
1. 2. 3. 4.
Transcribriform Transplanum Transsellar Transclival: superior (S.C.) middle (M.C.) inferior (I.C.) 5. Transodontoid
S.S .
Tumor
A Figure 55-1. A, Lateral view of the sagittal plane modular endoscopic endonasal approaches (EEAs) as shown in a rostral-caudal axis.
Chapter 55 • Endoscopic Endonasal Approaches to the Skull Base and Paranasal Sinuses
669
1. C.N.I
2. 3.
4. 5. 6.
7. V
8.
B.A.
S. S.
C.E.
Tumor locations 1. Near cribriform plate (ethmoid) 2. Suprasellar near A.C.P. 3. Optic chiasma, and C.N. II 4. Pituitary gland
5. 6. 7. 8.
Dorsum sellae P.C.P. (post.) clinoid process Cavernous sinus Clivus
B Figure 55-1—cont’d B, View of the anterior skull base from above showing sample tumor locations for the various rostral-caudal modules.
continues until bilateral frontal sinusotomies are performed with creation of a superior septal defect and removal of the floor of the frontal sinuses (Draf 3 or Lothrop procedure) if needed. The anterior attachment of the middle turbinate provides a posterior landmark to preserve olfaction and avoid violation of the cribriform plate with resultant cerebrospinal fluid (CSF) leak. This approach can be used in isolation for rare lesions such as nasal dermoid cysts or fibro-osseous tumors and more commonly for frontal sinusitis or mucoceles. In pediatric patients with dermoid cysts, the cyst is drained endonasally but the septum should be resected up to the skull base. The transfrontal approach is also used in conjunction with the transcribriform approach for accessing the anterior extent of lesions such as olfactory groove meningiomas. A key anatomic landmark for this region is the frontoethmoidal recess.
Transcribriform Approach The transcribriform approach is a commonly used approach for anterior skull base tumors. Most often used in our practice for resection of olfactory groove
eningiomas and esthesioneuroblastomas, it is also used m for repair of post-traumatic and iatrogenic CSF leak repairs and has potential for any subfrontal lesion. A complete sphenoethmoidectomy is performed on each side and the nasofrontal recesses are visualized. If needed, exophytic tumor within the nasal cavity can be debulked. The superior nasal septum is transected from the crista galli to the sphenoid rostrum as needed. As mentioned above, a transfrontal approach can be performed to establish an anterior margin. The transcribriform approach provides direct access to the vascular supply of tumors such as olfactory groove meningiomas, allowing early devascularization. The anterior and posterior ethmoidal arteries should be identified early and cauterized or clip ligated. The lateral bony margins are drilled to form “gutters” or osteotomies for the length of the tumor or planned resection as needed. If necessary, the lamina papyracea can be removed as well to allow retraction on the periorbita, thereby displacing the orbital contents for even more lateral access. In fact, the lateral access can extend all the way to the midorbital line at the level of the superior rectus muscle. After the lateral
670
OTOLOGIC SURGERY
argins are drilled, the planum or tuberculum can be m drilled as needed for a posterior margin. Even if necessary for tumor resection, it may be safest to leave the bone of the optic canals intact as long as possible. Finally, the dura is dissected from the crista galli to allow its removal. The remaining bone of the cribriform plate can now be dissected free (Fig. 55-4). If necessary for purely extradural lesions, the dura should be left intact. This may not be possible over the olfactory filaments, which can be cauterized to minimize CSF leakage, but should nonetheless undergo formal repair (see below). Obviously, olfaction (when present) is sacrificed if a bilateral approach is undertaken. The dura is opened in the same fashion as with microsurgery, with a fine blade and scissors. The durotomies are paramedian on both sides of the falx. This is done in order to minimize bleeding from branches of the anterior falcine artery and the inferior sagittal sinus. Bilateral durotomies are made either at the lateral extent of desired resection or overlying tumor. The durotomy is extended to the midline both anteriorly and posteriorly. The falx and inferior sagittal sinus must be systematically addressed next. Pistol-grip, endonasal bipolars are used with one blade on either side of the falx/sinus to coagulate it prior to transecting it to allow anterior tumor or dural release. In addition, anterior falcine branches may be encountered during tumor resection for access to the falx, providing additional devascularization. Dura can be resected en bloc as needed for tumors such as esthesioneuroblastoma. All tissues in such cases must be
1.
2.
3. B.S. S. S.
Ce.
Rectangles designate areas of approach by fossa 1. Anterior 2. Middle fossa 3. Posterior
�
Figure 55-2. Superior view of skull base with general regions of approach by fossa. 1) anterior, 2) middle, 3) posterior.
S.T., S.S.
C.S.
T.L. 4. V3 3.
1. V.N.
C.S.–Cavernous sinus L.P.M. V.N.–Vidian nerve V3–V3 in foramen ovale M.P.M. S.T.–Sella turcica S.S.–Sphenoid sinus T.L.–Temporal lobe L.P.M./M.P.M.–Lat and Medial pterygoid m’s. Np.–Nasopharynx
Np.
T.L. 5.
2.
. I. J . V C. A.
Lateral pterygoid plate Medial pterygoid plate
Figure 55-3. Coronal plane, lateral EEAs by anatomic zones. Zone 1 is the petrous apex, Zone 2 the petroclival junction, Zone 3 the quadrangular space/Meckel’s cave, Zone 4 the superior lateral cavernous sinus, and Zone 5 the infratemporal fossa.
Chapter 55 • Endoscopic Endonasal Approaches to the Skull Base and Paranasal Sinuses
evaluated both intraoperatively and postoperatively for margins. Intradural tumor resection must be performed with caution, especially when approaching the interhemispheric fissure, as there are frequently anterior cerebral artery branches or even the anterior communicating artery on the surface of the tumor (Fig. 55-5). An endonasal ultrasonic aspirator or two suctions can be used depending upon tumor consistency and proximity of involved structures. Microsurgical concepts of preserving
CG
OFim PO
ON
PO
PI tub I C mOCR A sella
ON I C A
I OCR
Figure 55-4. (Reprinted with permission Gardner PA, Kassam AB, Thomas A, Snyderman CH, Carrau RL, Mintz AH, Prevedello DM. Endoscopic endonasal resection of anterior cranial base meningiomas. Neurosurgery 2008;63:36-52.Copyright 2008, Lippincott, Williams & Wilkins.) Intraoperative, endoscopic view of the entire anterior skull base following exposure and removal of necessary bone for approach to an anterior base lesion. (CG = crista galli; ICA = internal carotid artery; lOCR = lateral opticocarotid recess; mOCR = medial opticocarotid recess; Oflm = olfactory filament; ON = optic nerve protuberance; Pl = planum; PO = periorbita; tub = tuberculum.
671
arachnoid planes when possible and sharp dissection of critical structures is maintained throughout. All of this may require the use of angled scopes (45° or 70°).
Transplanum/Transtuberculum Approach Also described as an “extended” approach, the transplanum approach was the first expansion of a traditional transsphenoidal approach.10-13 This approach provides a natural corridor for many suprasellar tumors such as craniopharyngiomas, tuberculum meningiomas and large pituitary adenomas. It may also be used for biopsy or resection of infundibular lesions such as metastases or hypophysitis. This approach is often an integral addition to transsellar and transcribriform approaches. A transsphenoidal exposure (see transsellar approach below) is augmented by a posterior ethmoidectomy. The posterior ethmoidal arteries are a good landmark for anterior extent to preserve olfaction while providing adequate access. The optic canals are obviously critical to identify in order to avoid damage to the nerves. Whenever drilling over or near the optic canals, it is important to constantly irrigate to avoid heat transmission from the drill to the nerves. After the sella is exposed, the planum can be drilled and thinned. This can require an angled endoscope for adequate visualization depending upon the slope of the anterior skull base. Though somewhat counterintuitive, the planum is most easily removed in an anterior to posterior direction after the bone has been adequately thinned, exposing dura at the anterior extent and laterally. The lateral margins are actually the optic nerves which form the sides of a trapezoid which include the anterior planum and tuberculum sellae. Often there is no need to expose the optic nerves. However, in many tuberculum meningiomas, there is extension of disease into the medial optic canals (Fig. 55-6). This disease is
ACA
ACA
OGM
A
B
Figure 55-5. A, Preoperative CT angiography sagittal reconstruction showing the anterior cerebral artery (ACA) involvement with an olfactory groove meningioma (OGM). B, Postoperative, sagittal MRI following endoscopic endonasal approach (EEA) showing complete resection of the tumor with patent and intact ACA.
672
OTOLOGIC SURGERY
A
B
Figure 55-6. (Reprinted with permission Gardner PA, Kassam AB, Thomas A, Snyderman CH, Carrau RL, Mintz AH, Prevedello DM. Endoscopic endonasal resection of anterior cranial base meningiomas. Neurosurgery 2008;63:36-52.Copyright 2008, Lippincott,Williams & Wilkins.) A, Axial, post contrast MRI showing extension of meningioma into the left medial optic canal (arrow). B, Axial, post-contrast MRI following EEA (endoscopic endonasal approach) for meningioma showing resection of tumor from the medial optic canal (arrow).
not easily accessed via a craniotomy and requires release and retraction of an often already compromised optic nerve. In these cases, the bone overlying the tumor in the optic canal should be thinned (“blue-lined”) with a drill and carefully removed with a dissector. When approaching suprasellar lesions, the superior intercavernous sinus (SIS) can often merely be retracted inferiorly rather than being transected. If, however, it must be traversed, it is useful to open the dura of the sella below the SIS in the midline as well as the dura of the planum above the SIS. This provides access on both sides of the sinus to perform a controlled ligation, either with a bipolar or clips. The dura can be reduced bilaterally with bipolar coagulation, using care to not damage the optic nerves. It is also important not to coagulate the superior hypophyseal artery as this can lead to hypopituitarism. Once the dura is opened, tumor resection proceeds carefully, identifying critical structures systematically. First, one supraclinoid internal carotid artery (ICA) is identified. Resection proceeds until the ipsilateral optic nerve is identified, followed by the chiasm, then the contralateral optic nerve and supraclinoid ICA (Fig. 55-7). In order to gain direct and unencumbered access to the paraclinoid ICA it is imperative to remove the medial optico-carotid recess (mOCR). This represents the lateral extension of the tuberculum and the pneumatization of the middle clinoid.14 The mOCR is the key anatomic landmark for this module.
Transsellar Approach Transsellar approaches are well known for access to pituitary adenomas as well as the myriad of sellar pathologies which exist, such as Rathke’s cleft cysts (RCCs), arachnoid cysts and rare craniopharyngiomas. In addition,
ON
ch
ON
Inf
ICA
Figure 55-7. Endoscopic, endonasal view following resection of a s uprasellar meningioma. The tumor is debulked and then the following structures identified in succession: optic nerve (ON), chiasm (ch) and infundibulum (Inf), contralateral optic nerve (ON) and supraclinoid internal carotid artery (ICA).
by utilizing a pituitary transposition, retroinfundibular lesions, such as granular cell tumors and some craniopharyngiomas, can be easily accessed while potentially preserving pituitary function. Endoscopic approaches to the sella are different from standard microscopic approaches in that they both provide more lateral access and require more exposure. Indeed, tumor which extends into the medial cavernous sinus can be accessed effectively using an EEA (Fig. 55-8). The bone overlying the cavernous ICA which guards the medial cavernous sinus must be carefully thinned and removed to allow some displacement of the ICA. However, it is intimate and direct visualization which allows successful resection of tumor
Chapter 55 • Endoscopic Endonasal Approaches to the Skull Base and Paranasal Sinuses
Gland
673
Res Cav
ICA Adenoma
ICA MCW
A
B
Figure 55-8. A, Preoperative, contrast-enhanced coronal MRI showing recurrent pituitary adenoma largely involving the medial cavernous sinus, as evidenced by the enhancing medial cavernous wall (MCW). (ICA = cavernous internal carotid artery). B, Postoperative, contrast-enhanced, coronal MRI showing complete resection of the tumor, including the portion in the medial cavernous sinus. Reminaing pituitary gland and normal cavernous sinus contents are more apparent following tumor resection. (ICA = cavernous internal carotid artery; Res Cav = tumor resection cavity)
in the cavernous sinus. Lateral sphenoidal exposure will be covered in detail in the coronal plane section. Greater exposure is required because room must be made for instruments to work around the endoscope while maintaining visualization. As a result, the sphenoid must be opened completely to work unencumbered in the sella. For example, lesions such as RCCs which are commonly in the pars intermedia can be approached from below, under the anterior gland (“subsellar approach”) rather than through it. This requires drilling the floor of the sphenoid sinus flush with the clival recess to allow this inferior access and visualization that the endoscope provides. The guiding anatomic principle for this module is complete bony removal and exposure of the sella from “blue to blue”; that is, laterally from cavernous sinus to cavernous sinus and from SIS superiorly to the inferior intercavernous sinus (IIS) inferiorly.
Transclival Approaches The transclival approaches may be easier to understand if broken down into superior, middle and inferior clival approaches. The superior clivus extends from the level of the posterior clinoid to Dorello’s canal. The middle clival segment continues from Dorello’s canal to the jugular foramen and the lower third extends from the jugular fossa to the basion. The superior clival approach is usually combined with a transsellar approach and involves a pituitary transposition (see below) and posterior clinoidectomy. The middle clival approach is essentially the bulk of the clivus and has its own potential pitfalls and nuances. The inferior clivus is the approach to the foramen magnum.
Superior Clivus The pituitary transposition is a newly described technique for access to lesions directly behind the pituitary gland or infundibulum, such as retroinfundibular craniopharyngiomas, granular cell tumors and chordomas.15 In addition, clival lesions, such as “chondroid” tumors (e.g. chordomas and chondrosarcomas) often involve this portion of the clivus (Fig. 55-9). While conceptually simple, this is technically demanding. The pituitary gland is dissected free from the sella and lifted to allow access to the space behind it. However, there are many bands of fibrous connective tissue which anchor the gland in the sella, mostly along the lateral walls to the cavernous sinus. These bands must be sharply dissected to free and preserve the gland. There is often some venous bleeding during this portion of the dissection and a two-surgeon, three of four-hand technique becomes critical. The inferior hypophyseal arteries may need to be sacrificed. In our experience, as long as the superior hypophyseal arteries (SHAs) are preserved, gland function will be preserved. After all of the lateral connective bands are released, the aperture in the sellar dura through which the stalk enters should be opened to allow release of the stalk and complete freedom for transposition. This should be done carefully with a fine, pistol-grip scissor with visualization of the SHAs. At this point, the gland can be finally raised from the sella and held superiorly with fibrin glue. This gives direct access to the dorsum sellae and posterior clinoids which can then be drilled and removed for access to the interpeduncular and basilar cisterns. There is more brisk venous bleeding during this drilling, as the clival venous plexus extends into the clinoids and dorsum sellae. This tight space, surrounded by critical structures with often brisk bleeding, requires an experienced team
674
OTOLOGIC SURGERY
scar
gl ch gl
NSF
A
B
Figure 55-9. A, Sagittal post-contrast MRI showing a superior clival chordoma (ch) with extension behind the pituitary gland (gl) into the dorsum sellae. B, Postoperative MRI following EEA for the lesion in (A) showing complete resection of the superior clival lesion which involved a pituitary transposition, as evidenced by the scar tissue posterior to the gland (gl). There is a brightly enhancing nasal septal flap (NSF) in place for reconstruction.
PG MC
MC
A
B
Figure 55-10. A, Preoperative, post-contrast, sagittal MRI showing a large, clival chordoma involving all of the upper and middle clivus (MC) as well as some of the anterior skull base. B, Post-contrast, sagittal MRI of the same patient following EEA resulting in complete resection, including the clival component. (PG = intact pituitary gland; MC = middle clivus)
for a safe and controlled approach. In our opinion, this may be the most complex EEA procedure.
Middle Clivus This approach accesses the majority of the clivus. Tumors such as chordomas and chondrosarcomas can often be relatively easily accessed endonasally (Fig. 55-10).
In addition, we have used EEAs to address large portions of petroclival meningiomas, neurenteric cysts and a clival arteriovenous malformation (AVM). As with any approach it is necessary to understand the anatomic structures involved. The paraclival (vertical portion of the cavernous ICA) creates the lateral boundaries for this approach. It is therefore essential to stay medial to the pterygoid base, beneath which runs the paraclival
Chapter 55 • Endoscopic Endonasal Approaches to the Skull Base and Paranasal Sinuses
675
Kassam line
A
B
Figure 55-11. A, Preoperative, sagittal CT reconstruction showing the ‘Kassam line’ (dotted white line) which is used to determine inferior extent of endonasal resection. The line extends from the tip of the nasal bones to the posterior hard palate and into the depth as needed. B, Postoperative, sagittal CT reconstruction following endoscopic endonasal resection of the odontoid process and pannus. The patient also underwent posterior occipitocervical fusion.
ICA. At times, it is necessary to work behind the paraclival ICA to completely resect clival tumors. In addition, the abducens nerve is always a concern in this area and its course must be understood. The abducens nerve emerges at the level of the vertebrobasilar junction (VBJ), making preferable to open dura or tumor as caudal as possible. Abducens nerve monitoring plus slow, careful dissection toward the superior end of the clivus is critical. In addition, tumors such as petroclival meningiomas, which often have their origin posterior to the VIth nerve can displace the nerve anteriorly, even pressing it against the dura. Careful opening of only the dura when approaching these tumors can prevent inadvertent injury. Perhaps the most challenging aspect of these approaches is the clival venous plexus. Sometimes, the clivus and plexus are full of tumor, thereby thrombosing the plexus. However, it is sometimes the case that the tumor has merely caused venous congestion, engorging the already significant plexus. As a result, the bleeding encountered during this approach can be copious. There are times that the blood loss can be prohibitive and staging of the case may be required after the sinus has been packed, hopefully resulting in thrombosis before the second stage.
Foramen Magnum The foramen magnum is an uncommon location for lesions which require an anterior approach. Foramen magnum meningiomas are the most common tumor we have addressed via this approach. It is often used in conjunction with other approaches for chordomas. We have also resected a recurrent posterior fossa
emangioblastoma and performed a vertebral artery h aneurysmorrhaphy via this approach. The medial condyles and hypoglossal canals create the lateral borders for this approach. Once again, routine monitoring of the XII nerve is valuable. In our experience, resection of the medial half of the condyle (as is the case with the lateral half) does not result in occipitocervical instability. Nevertheless, these patients should be followed long term to ensure that they do not develop delayed instability.
Transodontoid Approach The most common pathology treated in this location is basilar invagination/pannus associated with longstanding degenerative arthritis. Additional pathology includes craniocervical meningiomas and chordomas. The anatomic landmarks for this module are the floor of the sphenoid above and the soft/hard palate below. Laterally, the region is bounded by the Eustachian tubes (ETs) that guard the parapharyngeal ICAs (internal carotid arteries). Inferior extent of endonasal resection can be determined preoperatively by using the ‘Kassam line’, which is drawn from the tip of the bony nasal bridge to the posterior hard palate and extended to the pathology in the depth as needed (Fig. 55-11). The nasopharyngeal mucosa and the longus capitus muscles are resected to expose the basion, arch of C1 and base of the odontoid process. The arch of C1 is removed and the dens resected using an 18 cm highspeed drill. If the lesion extends intradurally, the circular sinus is packed and the dura opened. Intradural resection requires respecting the lateral plane of the vertebral arteries as all of the lower cranial nerves are located laterally at this level.
676
OTOLOGIC SURGERY
Coronal Plane Anterior Fossa Supraorbital Approach An endonasal supraorbital approach (like all of the coronal plane approaches) is a lateral extension of another approach, in this case, a transfrontal or transcribriform approach. Lesions such as fibro-osseous tumors and fibrous dysplasia are often approached with a limited supraorbital approach. As an addition to a midline transcribriform approach, supraorbital approaches provide the lateral EEA exposure for olfactory groove meningiomas, esthesioneuroblastomas and other nasal carcinomas. Many orbital approaches can potentially be performed via only one nostril, but the exposure should not be compromised in order to avoid a binarial approach. The limit for lateral extent is the midorbital line. Reaching this limit requires removal of the lamina papyracea to allow for gentle lateral retraction of the periorbita and orbital contents. Transorbital Approach The orbit can be accessed endonasally via a maxillectomy and ethmoidectomy. ��������������������������������������� Applications include orbital and optic nerve decompressions for Graves’ disease, fibrous dysplasia and optic gliomas as well as resection of intraorbital tumors such as meningiomas and hemangiomas. Imageguidance can be critical for some complex decompressions. In general, for very complex cases such as circumferential fibrous dysplasia, only one nerve is decompressed in each operation. Orbital tumors which are medial to the optic nerve can be addressed with the addition of an ophthalmologist to the team via a corridor between the medial and inferior rectus muscles which are suspended and retracted via an external canthal approach. Transpterygoid Approach These approaches are lateral extensions from the sella and sphenoid. They provide access to five anatomical zones: I) petrous apex, II) petroclival junction, III) quadrangular space (Meckel’s cave), IV) superior lateral cavernous sinus, and V) infratemporal fossa. In general, adequate access to any of these zones requires at least a maxillary antrostomy and may require an extended endoscopic medial maxillectomy (endoscopic Denkers). I. Petrous Apex
Lesions of the petrous apex may be approached medially through the sphenoid sinus. Cholesterol granuloma of the petrous apex with medial expansion into the sphenoid sinus is the most common pathology. Other indications include selected benign and malignant tumors, such as juvenile nasal angiofibromas (JNAs) and chondrosarcomas (which usually require other modules for complete resection). A wide sphenoidotomy is performed and the usual anatomic landmarks identified. Exposure of the sphenoid sinus must extend laterally to include the lateral
recess. Often, this lateral recess is not pneumatized but is nonetheless key to this approach. If the lesion is expansile it may have remodeled the petrous apex thereby protruding into this lateral recess medial to the paraclival ICA. In these cases, the thin cortical bone overlying the lesion can be drilled providing direct access to the target. However, if remodeling has not occurred, the lesion will be located deep and lateral to the paraclival ICA. In these circumstances, the ICA may need to be mobilized laterally to give adequate access. After removal of the sphenoid floor (to the level of the clival recess) and intrasinus septations, the bone overlying the inferior sella and petrous apex is thinned with the drill using vertical strokes, parallel to the paraclival ICA. The ICA can then be skeletonized and mobilized. It may prove useful to thin or remove the clivus medially in order to enlarge the opening into the petrous apex. Tumor or granuloma contents can be removed carefully with two suctions. Angled endoscopes, in addition to introducing instrumentation from the contralateral naris utilizing a binarial approach, help access all of the lesion. In the case of a granuloma, a Silastic stent can be placed into the cavity for to help maintain drainage. II. Petroclival Junction
1. An extension of the petrous apex approach, this zone is usually involved with similar tumors. In addition, debulking of petroclival meningiomas can be achieved via this approach, often in combination with a clival approach. Occasionally, lateralization of the ICA is necessary to increase exposure when there is minimal medial expansion of the lesion. In such cases, the vidian canal is carefully drilled and followed to the 2nd (anterior) genu of the ICA, starting inferomedially to prevent carotid injury. After the bone overlying the paraclival (vertical cavernous) carotid artery is thinned, it can be carefully removed, allowing the artery to be displaced laterally without compressing it on an edge of bone. The relationship between this portion of the carotid artery and the vidian canal16,17 is critical to understand even if a transpterygoid approach is not needed (Fig. 55-12). In addition, the oblique course of the abducens nerve, running from the vertebrobasilar junction (VBJ) to Dorello’s canal, just lateral to the superior-most aspect of the paraclival ICA, to the cavernous sinus, is critical to fully understand in order to prevent injury during these transpterygoid approaches. III. Quadrangular Space (Meckel’s Cave)
The cranial nerves which are contained in the cavernous sinus are crowded into the superolateral cavernous sinus. This allows relatively safe access to the medial cavernous sinus via the previously described transsellar route. Tumor in Meckel’s cave can also be relatively safely accessed, but much greater care must be taken to avoid ophthalmoplegia. Tumors such as pituitary adenomas rarely are found in this location. However, tumors such
Chapter 55 • Endoscopic Endonasal Approaches to the Skull Base and Paranasal Sinuses
677
B Plane of x-section (sagittal) A.C.P.
V2 1.
9.
2.
8. 3.
B Sagittal section
V3 P.C.P.
4A 4B
5.
7.
CN3 CN4 CN6 CN5
9.
CN 3, 4
6.
A 1. 2. 3. 4A 4B 5. 6. 7. 8. 9. 10.
G.g.
5.
CN6 Cavernous sinus Pterygopalatine ganglion Vidian canal Vidian artery Vidian nerve Vertical paraclival I.C.A. Petrous I.C.A. G.P.N. to geniculate ganglion Ophthalmic segment I.C.A. Cavernous I.C.A. segment (parasellar) Maxillary sinus
V V3
1.
To geniculate ganglion
V2
10.
B 6.
4A
4B
3.
2.
4B
Figure 55-12. Illustration demonstrating the relationship of the vidian canal (with nerve and sometimes artery) with the anterior genu of the ICA (internal carotid artery) as it turns from the horizontal (petrous) portion to the paraclival (vertical cavernous) segment. This relationship is critical for transpterygoid approaches.
as adenoid cystic carcinoma and other nasopharyngeal carcinomas can extend via neural spread along trigeminal branches (usually the maxillary [V2] or mandibular [V3]) directly into this space. In addition, this space provides good endonasal access to schwannomas of the cranial nerves in the cavernous sinus (Fig. 55-13). The “quadrangular space” (Fig 55-2) is an area of the skull base which contains the Gasserian ganglion and is bounded inferiorly by the petrous ICA, medially by first the vertical segment of the cavernous ICA (also know as the paraclival ICA), laterally by the maxillary branch of the trigeminal nerve (V2) and superiorly by the abducens nerve (VI). The abducens nerve runs directly under the ophthalmic branch (V1) of the trigeminal nerve and can be easily damaged. Therefore, the ideal way to avoid this is to not cross the superior plane of V2. Again, the vidian canal is the key landmark for accessing this area as it leads to the anterior genu of the ICA which forms the inferomedial corner of the quadrangle. Unless there is wide pneumatization of the lateral sphenoid sinus, bone between the vidian canal and V2 must be removed carefully after the inferomedial aspect of the vidian canal is exposed as described above.
IV.Superior Lateral Cavernous Sinus
In our opinion, there are few indications for addressing tumor in the superior lateral cavernous sinus. With good control of small tumor residuals with other treatments such as radiosurgery, it is uncommon for the risk of ophthalmoplegia to be outweighed by the potential benefits of chasing tumor into this region. However, in the setting of pre-existing ophthalmoplegia, especially when addressing hormonally active pituitary adenomas, it is both reasonable and feasible to access the superior lateral cavernous sinus. Superior extension is usually bounded by the optic nerve and optic strut. Surgical approach is merely an extension of the transsellar and quadrangular approaches discussed above.
Infratemporal Fossa Medial temporal fossa lesions are usually an extension of tumors such as medial sphenoid wing, petroclival meningiomas, and schwannomas as well as nasopharyngeal carcinomas which are addressed as an extension of the approach chosen for these tumors. There are only rare tumors which occur in this region in isolation. However,
678
OTOLOGIC SURGERY
PG ICW schwannoma
PG
QS MCW ICA
A
B
Figure 55-13. A, Preoperative, post-contrast, coronal MRI image showing a schwannoma filling Meckel’s cave (AKA quadrangular space). (ICA = internal carotid artery (cavernous portion); PG = pituitary gland) B, Post-contrast, coronal MRI image following EEA with complete resection of the quadrangular space (QS) schwannoma. (ICW = intercavernous wall, between the medial and lateral cavernous sinuses; MCW = medial cavernous wall; PG = pituitary gland).
ICA ICA
ss
SS
sch
A
TC
B
Figure 55-14. A, Preoperative, post-contrast, coronal MRI showing a giant, recurrent infratemporal fossa trigeminal schwannoma (sch). ICA = internal carotid artery; SS = sphenoid sinus. B, Post-contrast coronal MRI of the same patient as (A) immediately following removal via EEA (endoscopic endonasal approach). ICA = internal carotid artery; SS = sphenoid sinus (filled with packing); TC = tumor cavity.
we have addressed neurenteric cysts and schwannomas in this area effectively via an EEA (Fig 55-14). Often, the inferior tail of the cavernous sinus dura as it ensheathes the trigeminal ganglion and its branches must be transgressed. This is approached initially via a quadrangular region approach. Care should be taken upon opening dura to determine if the ganglion is medial (and
therefore between the surgeon and the tumor) or lateral (and therefore displaced away from the operative trajectory) to the lesion. If the ganglion is medial, a lateral approach such as an anterior transpetrosal could be considered. However, the fibers of the trigeminal nerve may be thinned and splayed thus allowing one to work between them to at least biopsy the lesion. Tumors which
Chapter 55 • Endoscopic Endonasal Approaches to the Skull Base and Paranasal Sinuses
displace the nerves and/or ganglion laterally are ideally suited for an endonasal approach which will provide direct access. A complete understanding of the course of the parapharyngeal, petrous and cavernous ICA is a necessity both for injury avoidance as well as control.
Posterior Fossa The caudal lateral expanded endonasal approaches consist of the transcondylar and infratemporal fossa/ parapharyngeal space approaches. These essentially represent lateral extensions from the lower third of the clivus extending through the medial occipital condyle, the hypoglossal canal and into the jugular foramen.
Transcondylar Approach The medial transcondylar approach (the “far medial” approach) employs the limits of an endonasal approach. Chordomas, meningiomas, schwannomas and even a vertebral aneurysm have all been approached endonasally in part with this corridor. The hard palate becomes a limiting factor with all caudal approaches. An endoscopic medial maxillectomy is also needed to work unencumbered in this location. The complete exposure of the medial condyle is one of the most challenging exposures. In order to access the condyle, the torus of the Eustachian tube must be removed. Appropriate vascular imaging should be used pre- and intraoperatively (image-guidance) to verify the position of the parapharyngeal ICA to avoid injury during this dissection. The cartilage of the Eustachian tube blends with the foramen lacerum and the dissection of this tough material which is intimate with the petrous ICA can be very tedious and potentially dangerous. It may be helpful to identify the ICA in as many areas as possible prior to this dissection. Naturally, hypoglossal nerve monitoring should be employed, especially during drilling. It has been our experience that (as with a lateral approach) half of the condyle can be removed without harming stability (in this case, the medial half). The key to this is to avoid violating the atlanto-occipital joint, thereby preserving stability. Long term follow up is needed to ascertain the durability of this stability.
679
medial maxillectomy. The Eustachian tube must be transected and cartilage of the tarus which blends with foramen lacerum very carefully resected. The IMA branches are identified, coagulated or clip ligated. These can also be coiled using endovascular techniques preoperatively. The parapharyngeal ICA is identified following removal of the pterygoid plates. The lateral pterygoid plate points to the ICA in the depth. Careful, blunt dissection allows identification of the fat pad which protects the ICA. Once the parapharyngeal ICA and petrous bone are identified, resection can proceed. This exposure can be carried further laterally to the level of the posterior ICA canal, providing direct access to the jugular foramen.18
COMPLICATIONS AND LEARNING CURVE As is the case with any new approach, the endoscopic endonasal approaches have a learning curve. It is important to select cases appropriately based upon experience. The senior authors have performed over 1000 cases over the past 10 years. They were able to maintain a low complication rate by very gradually increasing the complexity of cases they attempted. As a result, a set of levels has been developed as a guide. Level I cases are basic sinonasal procedures, which represent an opportunity for the neurosurgeon to gain familiarity with the working corridors and endoscope. Level II cases include simple pituitary adenomas, confined to the sella, and CSF leaks. Neurosurgeons have traditionally done their pituitary tumors without an otolaryngologist and otolaryngologists have always done CSF leaks on their own. However, this is a unique opportunity to learn to work as a team and develop the skills needed to progress to higher level cases. Level III procedures are those which involve the extradural ventral skull base between the plane of the ICAs. Level IV procedures progress to intradural surgery. Level V procedures are those lateral to the plane of the ICA, requiring complete ICA control. CSF leak has been by far the most frequent complication of the expanded endonasal approach. Despite high rates initially, bacterial meningitis rates were low, largely due to rapid reoperation in those patients with identified CSF leaks. The incidence of meningitis in the first 700 patients was 1.2%
Lower Infratemporal Fossa/Parapharyngeal Space 2. Perhaps the most challenging endonasal approach is the most inferior and lateral, the approach to the infratemporal fossa. Filled with muscle and soft tissue interspersed with large vessels including the ICA and IMA (internal maxillary artery) and lacking in clearly defined anatomical landmarks, this region is challenging to navigate. As with the other EEAs, control of the ICA is critical. Tumors such as schwannomas, nasopharyngeal carcinomas, JPAs, paragangliomas meningiomas, and lesions such as meningoencephaloceles can all involve this area. Once again, the majority of the work is done through a
RECONSTRUCTION Initial rates using techniques developed for simple CSF leak repair were as high as 58% for certain tumor types (craniopharyngiomas).19 This prompted us to pursue the development of vascularized reconstructive techniques. As a result, the pedicled, vascularized nasal septal mucosal flap (NSF) was adopted for reconstruction.20 Rates since the adoption of this flap have been reduced to 5.4% (in press), well within the range of traditional approaches. This has been a critical component of the development
680
OTOLOGIC SURGERY
of EEA and has allowed it to become a feasible approach for many tumors of the skull base. This flap is durable, effective and has even proven to be reusable in reoperations, an extremely valuable characteristic. There are patients in whom nasal septal mucosa is not available, either due to prior surgery, radiation therapy, other septal pathology, or tumor involvement. There are also options for vascularized flaps other than the nasal septal flap in those patients who need it . We have used turbinate flaps for repair of small defects in close proximity to a turbinate.21 We have also used a vascularized temporoparietal fascial flap (TPFF) which can be tunneled through the pterygomaxillary fissure via a small lateral canthal incision using a percutaneous tracheostomy tube.22 The advent of these vascularized reconstruction techniques has been critical for endoscopic endonasal skull base surgery.
CONCLUSIONS Endoscopic endonasal surgery for lesions of the paranasal sinuses and skull base has developed over the past decade into a feasible approach. Almost the entire anterior skull base as well as some portions of the middle and posterior fossae are accessible endonasally. The advantage for many lesions is the relationship of critical neurovascular structures, with these approaches providing excellent direct medial corridors to lesions which have the critical neurovascular structures located along their perimeter. These techniques show promise in short term studies and may provide lower morbidity than standard approaches. We are optimistic that these short-term data will be translated into durable long-term results. We believe that the EEA corridors provide an important armamentarium to complement existing lateral external corridors, thereby providing the contemporary skull base surgeon with 360° access to this compact and complex region.
REFERENCES 1. Rosegay H : Cushing’s legacy to transsphenoidal surgery. J Neurosurg 54:448-454, 1981. 2. Carrau R L , Jho H D, Ko Y: Transnasal-transsphenoidal endoscopic surgery of the pituitary gland. Laryngoscope 106:914-918, 1996. 3. Jho H D, Carrau R L , Ko Y, et al: Endoscopic pituitary surgery: Early experience. J Neurosurg 84:744, 1996. 4. Friedrich J P, Terrier G: Chirurgie endoscopique de la sinusite par voie endonasale. Med Hyg 41:3722-3726, 1983. 5. Kennedy DW, Zinreich S J, Rosenbaum A E, et al: Functional endoscopic sinus surgery: theory and diagnostic evaluation. Arch Otolaryngol 111:576-582, 1985. 6. Jankowski R , Auque J, Simon C, et al: Endoscopic pituitary tumor surgery. Laryngoscope 102:198-202, 1992. 7. Messerklinger W: Endoscopy of the Nose. Urban & Schwarzenberg, Baltimore, 1978.
8. Wigand M E : Transnasale endoscopishe chirurgie der nasennebenhohlen bei chronischer sinusitis. HNO 29:215-221, 1981. 9. Stammberger H : Endoscopic endonasal surgery for mycotic and chronic recurring sinusitis. Ann Otol Rhinol Laryngol 119(Suppl):1-11, 1985. 10. Weiss M H : Transnasal transsphenoidal approach. In Apuzzo M L J (ed): Surgery of the Third Ventricle. Baltimore: Williams & Wilkins, 1987, pp 476-494. 11. Mason R B, Nieman L K, Doppman J L , et al: Selective excision of adenomas originating in or extending into the pituitary stalk with preservation of pituitary function. J Neurosurg 87:343-351, 1997. 12. Kouri JG, Chen MY, Watson JC, et al: Resection of suprasellar tumors by using a modified transsphenoidal approach. Report of four cases. J Neurosurg 92:1028-1035, 2000. 13. Kaptain GJ, Vincent D A, Sheehan J P, et al: Transsphenoidal approaches for the extracapsular resection of midline suprasellar and anterior cranial base lesions. Neurosurgery 49:94-100; discussion 100-101, 2001. 14. Rhoton A L Jr: The supratentorial cranial space: microsurgical anatomy and surgical approaches. Neurosurgery 51:S1-385, 2002. 15. Kassam A B, Prevedello D M, Thomas A, Gardner P, Mintz A, Snyderman C, Carrau R : Endoscopic endonasal pituitary transposition for transdorsum sellae approach to the interpeduncular cistern. Neurosurgery. 62(ONS Suppl 1):ONS57-74, 2008. 16. Kassam A B, Vescan A D, Carrau R L , Prevedello D M, Gardner P, Mintz A H, Snyderman C H, Rhoton A L Jr: Expanded endonasal approach: vidian canal as a landmark to the petrous internal carotid artery. J Neurosurg 108:177-183, 2008. 17. Vescan A D, Snyderman C H, Carrau R L , Mintz A, Gardner P, Branstetter B 4th, Kassam AB: Vidian canal: analysis and relationship to the internal carotid artery. Laryngoscope 117:1338-1342, 2007. 18. Kassam A B, Snyderman C, Carrau R , Gardner P, Hirsch B, Mintz A : Endoscopic, expanded endonasal approach to the jugular foramen. Operative Techniques in Neurosurgery 8(1):35-41, 2005. 19. Gardner P, Kassam A, Snyderman C, Carrau R , Mintz A, Grahovac S, Stefko ST: Outcomes following endoscopic, expanded endonasal resection of suprasellar craniopharyngiomas: a case series. J Neurosurg, in press, 2008. 20. Hadad G, Bassagasteguy L , Carrau R L , Mataza JC, K assam A, Snyderman C H, Mintz A : A novel reconstructive technique following endoscopic expanded endonasal approaches: Vascular pedicle nasoseptal flap. Laryngoscope 116(10):1881-1885, 2006. 21. Fortes FS, Carrau R L , Snyderman C H, Prevedello D, Vescan A, Mintz A, Gardner P, Kassam A B : The posterior pedicle inferior turbinate flap: a new vascularized flap for skull base reconstruction. Laryngoscope 117(8):13291332, 2007. 22. Fortes FS, Carrau R L , Snyderman C H, Kassam A, Prevedello D, Vescan A, Mintz A, Gardner P: Transpterygoid transposition of a temporoparietal fascia flap: a new method for skull base reconstruction after endoscopic expanded endonasal approaches. Laryngoscope 117:970976, 2007.
56
Petrosal Approach Todd A. Hillman
Petroclival tumors arise from or involve the petroclival junction cephalad from the jugular tubercle, medial to the trigeminal nerve, and anteromedial to the internal auditory canal (IAC).1,2 Tumors of the petroclival area are a particular challenge because these tumors often involve the middle and posterior cranial fossae, cause significant brainstem compression, invade the cavernous sinus, and abut or surround the upper cranial nerves and basilar artery. The complex anatomy requires an individualized surgical approach for each patient based on tumor origin, area of tumor extension, and preoperative neural function.3 The term combined petrosal approach actually describes numerous surgical approaches to the petroclival area well suited for these tumors. The term combined refers to the fact that this approach combines the middle fossa approach with a variation of a posterior fossa approach. The combination of these approaches allows for excellent exposure of the medial petrous bone and clivus from the cavernous sinus to the foramen magnum. First described by Decker and Malis,4 this combination allows the surgeon to take advantage of the strengths of each approach to minimize brain retraction. The middle fossa approach gives the surgeon good exposure of the petrous bone and clivus superior to the IAC, whereas the posterior approach provides surgical exposure inferior to the IAC. A tumor that is superior to the IAC can be accessed by a middle fossa approach alone. A tumor that is inferior to the IAC and tentorium cerebelli can be addressed with a posterior approach alone. It stands to reason that tumors involving both areas are best treated with a combination of these surgical procedures. The most common tumor of the petroclival area is the meningioma. Less common lesions include chordomas, chondrosarcomas, epidermoids, and vascular abnormalities such as aneurysms and arteriovenous malformations of the basilar system. Tumors that remain intracranial can be treated with a combined petrosal approach. Tumors that extend into the infratemporal fossa require a lateral approach.
COMBINED PETROSAL VARIATIONS All of the combined petrosal approaches include similar middle fossa exposure as a component of the procedure. The variations of the combined petrosal approach are classified by the type of posterior approach used. Traditionally, three variations of the combined petrosal approach were described: the retrolabyrinthine, translabyrinthine, and transcochlear approaches.5,6 The fourth variation described by Sekhar and colleagues,7 which provides exposure in between the retrolabyrinthine approach and the translabyrinthine approach, uses a partial labyrinthectomy for the posterior approach. This approach removes most of the labyrinth, while preserving functional hearing in 80% of patients. These four approaches give different degrees of anterior petrous bone exposure, which changes the level of brainstem and clivus exposure. The narrow angle between the anterior brainstem and clivus is the major factor influencing the level of visualization across both of these structures. Generally, further anterior removal of the otic capsule allows a more direct angle across the clivus, allowing improved medial exposure to the brainstem, basilar artery, and central clival depression. More aggressive anterior otic capsule removal secondarily provides improved access to the ipsilateral medial petrous bone as well (Fig. 56-1A). The choice of posterior approach depends on tumor location and size, preoperative hearing level, and extent of brainstem compression. The translabyrinthine and transcochlear approaches sacrifice residual hearing. The transcochlear approach also includes posterior facial nerve transposition, leading to a temporary facial paralysis and risking permanent injury. The translabyrinthine approach is used when preoperative hearing is poor, and the transcochlear approach is reserved for the largest of tumors that cross the midline of the clivus. Tumor effect on the brainstem is also an important consideration. Sometimes larger tumors need a less radical posterior approach because the posterior compression of the brainstem opens the angle between the clivus and brainstem. 681
682
OTOLOGIC SURGERY
CN 4 Basilar artery
V
CN 5 Tumor
Transcochlear
CN’s 7, 8
Translabyrinthine
CN’s 9, 10, 11
Sig
CN 12
mo
id s
inu
Partial Labyrinthectomy
s
Retrolabyrinthine
A
Cerebellum
FIGURE 56-1. A, Four variations of posterior approach provide unique exposure of anterior and medial petrous bone, and change angle of visualization across brainstem. Retrolabyrinthine variation allows access to petroclival junction, but limited medial access. Partial labyrinthectomy allows exposure up to lateral aspect of clival depression. Total labyrinthectomy allows exposure of anterior brainstem and central clival depression. Transcochlear approach allows access to contralateral clivus as well. B, Large petroclival meningioma. Posterior compression of brainstem keeps angle of approach flat and allows translabyrinthine approach despite contralateral clival involvement. (B Courtesy of Dr. Clough Shelton.)
The tumor in Figure 56-1B was addressed with a translabyrinthine variation of the combined petrosal approach despite involvement of the contralateral clivus. The posterior displacement and contralateral displacement of the brainstem kept the angle between the tumor-involved clivus and brainstem open, obviating the need for a transcochlear approach.
PREOPERATIVE EVALUATION Most patients who present to a neurotologist with a petroclival tumor have had either a computed tomography (CT) scan or magnetic resonance imaging (MRI) of the brain for evaluation of cranial nerve findings or nonspecific complaints. MRI and CT imaging modalities are complementary and obtained for every patient with these complex lesions. MRI needs to be a contrast-enhanced study, but the CT scan does not. MRI
B
is superior for determining the extent and character of the tumor, whereas CT scan gives the bony detail necessary for surgical planning. The surgeon determines from these studies whether the combined petrosal approach is appropriate, and which variation of the combined petrosal approach is needed. High-quality images are required with thin cuts (3 mm for MRI, 1 mm for CT) through the skull base. All images are reviewed with a neuroradiologist and the operating neurosurgeon before any planned intervention. Important temporal bone arterial and venous variations can often be detected from MRI; however, angiography with venous phase should be considered for each patient in whom a combined petrosal approach is being contemplated. Important venous anomalies exist that may result in catastrophe if not recognized before surgery.8 The main venous drainage of the temporal lobe is through the vein of Labbé, a single tributary that runs along the inferior surface of temporal lobe and typically
Chapter 56 • Petrosal Approach
anastomoses into the transverse sinus. The vein of Labbé may run through the tentorium, however, and insert into the superior petrosal sinus, rather than the transverse sinus. Because superior petrosal vein and tentorium transection is a key component of the combined petrosal approach, the vein of Labbé would be at risk if this anatomic variation is present. Accidental transection can lead to venous infarction of the temporal lobe with particularly grave consequences to the patient, especially if it occurs in the dominant hemisphere. A dominant sigmoid sinus should also be recognized before surgery. The surgeon must exercise extra caution not to injure a dominant sigmoid sinus because the contralateral venous drainage may be insufficient or absent, leading to a high risk of venous infarction. A final concerning anatomic variation is a transverse sinus that does not connect at the torcular Herophili. This situation leads to one side of the venous system draining the sagittal sinus system. The sigmoid on the side draining the sagittal system must be preserved to prevent stroke. Angiography easily detects these variations so that the surgeon can optimize the treatment plan. Angiography also gives the surgeon the option of preoperative embolization for particularly vascular tumors.
PREOPERATIVE PREPARATION A neurosurgeon is part of the surgical team in all cases. Patients are given preoperative antibiotics to help prevent postoperative wound infections. A first-generation cephalosporin such as cefazolin is used for non–penicillin-allergic patients, and clindamycin is for patients who have a known penicillin allergy. The patient is anesthetized with the head on the foot of the bed to allow adequate leg clearance for the surgeon. The anesthesiologist is instructed to use short-acting or no paralysis so that detection of facial nerve stimulation and somatosensory evoked potentials are not compromised during the case. Because these cases often take many hours, extra care is taken to pad the patient and bed appropriately to prevent pressure points that can lead to skin breakdown or peripheral neuropathy. The author uses a customized padding system developed by his circulating nurse to keep the legs in a slightly bent position and to pad the elbows generously. The arms are crossed over the patient’s chest, which is a more relaxed arm position than tucking the arms at the side and prevents pressure at the medial epicondyle of the humerus, a common point of ulnar nerve injury. An arterial line and double venous access is placed by the anesthesiologist. The blood pressure cuff is preferably placed on the upper extremity opposite the tumor side. The patient is double or triple strapped to the bed because significant bed tilting is necessary at times. The patient is kept in a supine position with the head turned so that the operative side is up and parallel to the floor. If there is limitation to neck rotation, a shoulder roll may
683
be placed under the shoulder to roll the torso away. The head is placed in a Mayfield head holder and secured to the bed. All retractors are secured to the head holder. The bed is turned 180 degrees away from the anesthesiologist to allow adequate room for the scrub nurse, surgeon, and microscope. A wide head shave is performed with an electric razor, and a small amount of alcohol is used to remove oil from the skin, which is allowed to dry. Benzoin is placed and clear adhesive drapes are placed around the periphery of the planned surgical site. The abdomen should be prepared so that a fat graft can be harvested. CN VII, IX, X, XI, and XII are monitored during the case. For large tumors with brainstem compression, somatosensory evoked potentials are also monitored. For cases in which hearing preservation is being attempted, auditory brainstem responses are monitored in both ears.
SURGICAL TECHNIQUES A large-diameter C-shaped incision is made starting just above the level of zygoma, 2 cm anterior to the zygomatic root. The incision is curved around, staying three fingerbreadths above the auricle, two fingerbreadths posterior to the sulcus, and ending posterior to the mastoid tip. The mastoid incision is kept posterior to allow enough posterior skin retraction for full sigmoid sinus exposure. The temporalis muscle is freed from its temporal bony attachment and reflected anterior and inferior over the zygoma. Skin hooks on rubber bands attached with clamps to the drapes are helpful to provide scalp retraction while maintaining a low profile. An alternative skin incision is a combination of the preauricular incision typically used for a straight middle fossa approach and the incision used for a translabyrinthine approach. Although this incision ends up with a 90 degree angle at the intersection, the author has not noted wound complications with this method, and the scar is less conspicuous. The entire mastoid and squamosal temporal bone must be exposed before proceeding with drilling.
Posterior Exposure The middle fossa craniotomy or the posterior temporal bone approach may be started first; however, the author finds that beginning with the posterior approach facilitates the middle fossa craniotomy by identifying the level of the middle fossa and limiting the number of burr holes required to turn the middle fossa flap (Fig. 56-2A). A complete mastoidectomy is performed with an electric high-speed drill, and all of the bone covering the middle and posterior fossa dura is removed. Suction irrigation is used, and the irrigant should contain bacitracin to prevent infection and the incidence of cerebrospinal fluid
684
OTOLOGIC SURGERY Temporalis m. Craniotomy
Retracted skin dissection over intact canal skin (EAC)
Mastoidectomy
P.C.W. CN VII in Fallopian canal Sigmoid sinus Retrosigmoid bone dissection
B
Superior petrosal sinus
Burr hole
A
Dural incision Partial Labyrinthectomy
Middle fossa dura incised
VII
S
s
inu
id s
o igm
C Posterior fossa dura incised Superior petrosal sinus
(CSF) leak.9 It is important to remove the bone over the sigmoid sinus and at least 5 mm of bone posterior to the sigmoid to aid with posterior retraction of the sinus. The mastoid emissary vein is divided to facilitate the retrosigmoid bone removal. The bone from the distal transverse sinus should be removed for identification, but the surgeon should use caution to preserve the insertion of vein of Labbé. Wide exposure laterally at the sigmoid greatly facilitates instrumentation medially when the exposure is complete. If a retrolabyrinthine approach is chosen, the labyrinth should be skeletonized (Fig. 56-2B). The horizontal canal is identified and followed to the posterior canal.
FIGURE 56-2. Combined retrolabyrinthine petrosal approach. A, Middle fossa and posterior approach outlined. Either approach can be performed first. Burr holes made for middle fossa flap need to be large enough to fit a side-biting drill with footplate attachment. B, Combined retrolabyrinthine approach shown with completed posterior exposure and craniotomy flap turned. Labyrinth is skeletonized to maximize exposure. C, Dural incisions shown with superior petrosal sinus divided. Tentorium is identified and divided. EAC, external auditory canal; PCW, posterior canal wall.
The bone over the superior canal is also removed. The bone covering the posterior fossa dura is removed from the petrosal vein to the jugular bulb. The bone over the middle fossa dura is also removed. The neurosurgeon can perform the middle fossa craniotomy, make the dural incisions, divide the greater petrosal vein, and begin to divide the tentorium to connect the middle and posterior fossae (Fig. 56-2C). If more exposure is needed, a partial labyrinthectomy can be performed by carefully plugging the semicircular canals as they are drilled away (Fig. 56-3). The key is to avoid exposure and traumatic suctioning of the labyrinthine fluids by keeping the labyrinth sealed while removing
Chapter 56 • Petrosal Approach Partial Labyrinthectomy
P.C.W.
M.F.D.
VII
P.F.D.
. S.S
Superior petrosal sinus
FIGURE 56-3. Combined partial labyrinthectomy approach. Semicircular canals are drilled away slowly while advancing bone wax into labyrinth to prevent communication with fluid spaces of membranous labyrinth. MFD, middle fossa dura; PCW, posterior canal wall; PFD, posterior fossa dura; SS, sigmoid sinus.
bone. The superior, horizontal, and posterior canals are skeletonized and blue-lined, but not entered. Bone wax is placed on the end of a 3 mm smooth diamond bit and advanced through the thin bone of the posterior canal to seal it while drilling at slow speed. More wax is placed on the drill bit and advanced inferiorly toward the vestibule, but halted posterior to the facial nerve. The drilling continues superiorly to the common crus. The wax is advanced while drilling at slow speed to the superior semicircular canal. The horizontal canal is also sealed in this fashion. The wax that is placed advances through the membranous labyrinth keeping it sealed as bone is removed. Drilling is stopped short of entering the vestibule, or hearing would be compromised. This technique can give almost the same exposure to the brainstem and clivus as a translabyrinthine approach, although it does not provide access to the IAC. If hearing is poor, or the IAC is significantly involved with tumor, the translabyrinthine approach is the preferable posterior exposure (Fig. 56-4A). Hearing is sacrificed as a result of this procedure. The semicircular canals are removed starting with the horizontal canal, which is followed posteriorly to identify the posterior canal. The anterior half of the horizontal canal is left to protect the second genu of the facial nerve. The common crus is followed to identify the superior semicircular canal. The semicircular canals are followed into the
685
vestibule, which forms the lateral extent of the IAC. The posterior fossa bone is removed off the dura until the IAC is identified. The bone surrounding the IAC is removed with a diamond drill. The IAC should be exposed about 270 degrees, just as for a translabyrinthine approach for an acoustic tumor. The middle fossa craniotomy is performed, and the tentorium divided as shown in Figure 56-4B. Exposure across the brainstem to the contralateral clivus can be maximized by extending the petrosal bone dissection anteriorly with a transcochlear approach. The external auditory canal is transected at the level of the bony-cartilaginous junction, and the canal skin is everted, and then oversewn in two watertight layers. The posterior external auditory canal wall is removed, as are the remaining canal skin, tympanic membrane, and ossicles. The inferior tympanic ring is also removed with the drill. The facial nerve must be mobilized by removing the bone of the IAC, tympanic fallopian canal, and mastoid segment of the fallopian canal. Bone removal should be greater than 180 degrees at all parts of the fallopian canal. The greater superficial petrosal nerve is sectioned anterior to the geniculate ganglion, and the facial nerve is mobilized posteriorly. There is usually bleeding at the canal of the greater superficial petrosal nerve, which must be packed with bone wax. The dura of the IAC is incised whether or not tumor involves the IAC to allow full posterior facial nerve mobilization. The cochlea is drilled away, and the carotid artery is identified (Fig. 56-5). When the carotid canal is exposed, the anterior petrous bone can be drilled away medially. One advantage of the combined petrosal approach is that transcochlear exposure is rarely needed except for the largest tumors. Transposition of the facial nerve has an inherent risk of facial nerve injury.10 This variation is carefully considered only when the translabyrinthine exposure is inadequate.
Middle Fossa Exposure The neurosurgeon usually performs the middle fossa exposure. To facilitate temporal lobe retraction, furosemide, 10 mg, followed by mannitol, 0.5 g/kg, is given intravenously 20 minutes before the middle fossa craniotomy. Burr holes are made as shown in Figure 56-2A, and a side cutting burr with a footplate is introduced into the defect. The craniotomy should be large, following the squamosal suture line. A large craniotomy helps to minimize the temporal lobe retraction required. The craniotomy is centered over the zygomatic arch to ensure adequate anterior exposure. The dura is elevated from posterior to anterior and medially over the remaining bone of the petrous ridge. The middle meningeal artery is coagulated to allow anterior dural elevation. The bone between the trigeminal nerve and IAC is removed (Kawase’s triangle). The dura is divided in the posterior fossa from the jugular bulb and in the middle fossa dura
686
OTOLOGIC SURGERY
M.F.D. Bone over IAC removed and posterior fossa dura exposed
CN. VII in fallopian canal
P.F.D.
ce Displa id o m sig sinus
A Translabyrinthine approach
Superior petrosal sinus
S.P.S.
Infundibulum Tumor CN 5
Tentorium (cut) CN’s 9, 10, 11
CN 4 Brainstem
Cerebellum Sigmoid sinus
CN’s 7, 8
B
FIGURE 56-4. Combined translabyrinthine approach. A, Labyrinth has been removed, and internal auditory canal (IAC) has been exposed. Facial nerve distal to geniculate remains under bone. Dural incisions are made as shown. B, Tentorium is divided parallel to superior petrosal sinus (SPS). Caution needs to be taken medially to recognize and preserve CN IV. Tumor is exposed on clivus and can be removed with careful dissection. MFD, middle fossa dura; PFD, posterior fossa dura.
Chapter 56 • Petrosal Approach
687
Transcochlear approach
MFD Co
VII P.F.D.
S.P.S. . S.S
B.a.
FIGURE 56-5. Combined transcochlear approach. Exter V
VIII
9 10 11 VII Ce
S.S.
to the superior petrosal vein. The superior petrosal vein is divided connecting the middle and posterior compartments, and the tentorium is sectioned from posterior to anterior. CN IV runs with the tentorium medially, and caution is needed to preserve it. When the tumor is exposed, excision proceeds with microdissection technique, debulking the tumor until the capsule can be visualized and safely dissected from cranial nerves, the brainstem, and arteries. The most challenging areas of dissection are the posterior cavernous sinus, tumor-involved upper cranial nerves, and tumor-involved basilar artery.
CLOSURE The middle fossa and presigmoid dura are reapproximated as well as possible. Typically, the middle fossa dura is able to be closed adequately enough to prevent CSF leak, but suturing the presigmoid dura is difficult, and the closure needs to be augmented with fat. Abdominal fat is used for this purpose and is placed though the dural defects in a dumbbell fashion. Prevention of CSF
nal canal is transected and oversewn. Posterior external auditory canal wall is removed. Facial nerve is removed from fallopian canal, and greater superficial petrosal nerve is sectioned to allow nerve to be transposed posteriorly. Removal of anterior and medial petrosal bone facilitates exposure of central clival depression and contralateral clivus. Ba, basilar artery; Ce, cerebellum; Co, cochlea; MFD, middle fossa dura; PFD, posterior fossa dura; SPS, superior petrosal sinus; SS, sigmoid sinus.
leak through the middle ear and eustachian tube varies by what posterior approach was used for the case. Hydroxyapatite cement has been helpful in preventing CSF leak in acoustic tumor surgery, and is used when appropriate for the combined approaches as well.11 If a retrosigmoid posterior approach has been used, temporalis fascia is placed over the antrum, and 5 to 10 mL of hydroxyapatite cement is placed in the posterior defect lateral to the fat graft. The cement should be placed over the antrum and against remaining bone of the posterior external auditory canal and jugular bulb. If a translabyrinthine approach has been used, the incus is removed, the eustachian tube is plugged with muscle, and hydroxyapatite cement is used to fill the antrum and posterior defect. If the transcochlear approach has been used, the use of cement is not advised because there is little bone to hold it in place, the eustachian tube can be aggressively sealed off with soft tissue as is easily seen with this exposure, and the facial nerve is uncovered in the dissection field. When cement is used, a JacksonPratt drain should be placed superficial to the cement and removed overnight because of the higher tendency to
688
OTOLOGIC SURGERY
form seromas. This drain is removed in 24 hours, and the drain hole is sutured closed.
RESULTS Results vary depending on the pathology of the tumor. Meningiomas constitute most of these tumors, and outcomes are well described in the literature. Gross total excision can be accomplished in most cases; however, there are a few reports of good long-term results with subtotal excision to preserve cranial nerves followed by stereotactic irradiation if necessary.12 Gross total excision is preferable if possible because of the difficulty in managing recurrent disease in this area. If tumor invades the cavernous sinus, a small amount can be left because of the morbidity of surgery in this area. If tumor is firmly attached to a cranial nerve, but separated from the brainstem, clivus, and petrous bone, a small piece of the capsule may be left with little concern about growth. Stereotactic irradiation can be used for small residual disease or recurrences that show growth during follow-up.
The most common complication of the combined petrosal approach is cranial nerve injury. The most commonly injured cranial nerves are CN V and VII. CN IV and VI can also be injured resulting in diplopia. Lower cranial nerve injury can result in dysphagia postoperatively. Rarely, preoperative deficits may improve after surgery, typically in nerves with partial dysfunction caused by nerve traction that is relieved by tumor excision. Generally, preoperative deficits do not improve, however. In the University of Utah series, 5 of 19 patients (26%) had at least one new, long-standing postoperative cranial nerve deficit.13 Most patients had preoperative cranial nerve function preserved, however. Gross total excision was accomplished in 86% of all patients. Functional hearing was preserved in 85% of patients. CSF leak is also a common complication of this approach. The rate of CSF leak varies by report, but is approximately 15% to 19%.5,6,13 Most of these cases resolve after placement of a lumbar drain for 3 days. There are not enough data to determine if the success of hydroxyapatite closure in preventing CSF leak after acoustic tumor surgery will translate to the combined petrosal approach.
Venous Drainage 1.
2. 3.
4.
12. 9. 5. 3.
6.
8.
2.
9. 1.
10. 7.
5. 11.
4.
8. 1. 2. 3. 4. 5. 6.
Cavernous sinus 7. Transverse sinus Basilar plexus 8. Confluence of sinuses Superior petrosal sinus 9. Vein of Labbé Inferior petrosal sinus 10. Jugular bulb 11. Mastoid emissary v. Sigmoid sinus 12. Superior sagittal sinus Tentorium (cut) FIGURE 56-6. Typical venous drainage is shown. Variations exist and should be identified preoperatively to prevent accidental trauma and venous infarction.
Chapter 56 • Petrosal Approach
Cerebrovascular accident is a rare, but serious complication of the combined petrosal approach. Venous infarction is a specific complication that can occur if the sigmoid is injured on a dominant or noncommunicating side, or if the vein of Labbé is injured. Avoidance of these complications is maximized by knowledge of the cerebral venous drainage system (Fig. 56-6) and with preoperative imaging to assess the patient’s specific vascular anatomy.
SUMMARY The combined petrosal approach is a safe and effective method to access and remove petroclival tumors and vascular lesions that extend above and below the tentorium cerebelli. Variations of the combined petrosal approach allow the surgeon to tailor the surgical approach for each patient to minimize morbidity, while maximizing postoperative neural function. The exposure obtained from these approaches allows access to an area of the skull base that is otherwise difficult to treat. Selection of the surgical approach is based on the extent of the tumor, predicted morbidity, and status of the patient’s preoperative hearing. The combined petrosal approaches provide a favorable tumor cure or control rate, while maintaining acceptable levels of postoperative morbidity.
REFERENCES 1. Al-Mefty O: Operative Atlas of Meningiomas. Philadelphia, Lippincott-Raven, 1998. 2. Abdel Aziz K M, Sanan A, van Loveren H R , et al: Petroclival meningiomas: Predictive parameters for transpetrosal approaches. Neurosurgery 47:139-152, 2000.
689
3. Arriaga M A, Shelton C, Nassif P, Brackmann D E : Selection of surgical approaches for meningiomas affec ting the temporal bone. Otolaryngol Head Neck Surg 107:738-744, 1992. 4. Decker R E, Malis L I : Surgical approach to midline lesions at base of skull. J Mt Sinai Hosp 37:84-102, 1970. 5. Spetzler R F, Daspit C P, Pappas CT: The combined supra- and infratentorial approach for lesions of the petrous and clival regions: Experience with 46 cases. J Neurosurg 76:588-599, 1992. 6. Daspit C P, Spetzler R F, Pappas CT: Combined approach for lesions involving the cerebellopontine angle and skull base: Experience with 20 cases—preliminary report. Otol Head Neck Surg 105:788-796, 1991. 7. Sekhar L N, Schessel D A, Bucar S D, et al: Partial labyrinthectomy petrous apicectomy approach to neoplastic and vascular lesions of the petroclival area. Neurosurgery 44:537-552, 1999. 8. Erkmen K, Pradvdenkova S, Al-Mefty O: Surgical management of petroclival meningiomas: Factors determining the choice of approach. Neurosurg Focus 19:1-12, 2005. 9. Kartush J M, Cannon SC, Bojrab D I, et al: Use of bacitracin for neurotologic surgery. Laryngoscope 98:10501054, 1988. 10. Selesnik S H, Abraham MT, Carew J F: Rerouting of the intratemporal facial nerve: An analysis of the literature. Am J Otol 17:793-809, 1996. 11. Arriaga M A, Chen D A, Burke E L : Hydroxyapatite cement cranioplasty in translabyrinthine acoustic neuroma surgery-update. Otol Neurotol 28:538-540, 2007. 12. Natarajan S K, Sekhar L N, Schessel D, Morita A : Neuro surgery 60:965-979, 2007. 13. Baugh A, Hillman TA, Shelton C : Combined petrosal approaches in the management of temporal bone meningiomas. Otol Neurotol 28:236-239, 2007.
57
Neurofibromatosis 2 William H. Slattery III
Neurofibromatosis 2 (NF2) is a rare syndrome characterized by bilateral vestibular schwannomas, multiple meningiomas, cranial nerve tumors, spinal tumors, and eye abnormalities. NF2 presents unique challenges to the otologist because hearing loss may be the presenting complaint leading to the diagnosis of the disorder. NF2 is quite invasive, requiring a multispecialist team approach for the evaluation and treatment of the disorder. The primary impairment is hearing loss resulting from bilateral vestibular schwannomas. NF2 must be characterized from neurofibromatosis 1 (NF1); although the names are linked, the disease entities are distinctly different. This chapter reviews the clinical characteristics of NF2, and current recommendations for evaluation and treatment.
NEUROFIBROMATOSIS 2 DIFFERENTIATED FROM NEUROFIBROMATOSIS 1 NF1 has distinctly different clinical characteristics from NF2. NF1 and NF2 have been differentiated as completely different genetic diseases based on the chromosome responsible for the disease. NF1 has been localized to chromosome 17, and NF2 has been localized to chromosome 22. NF1 is a multisystem disorder in which some features may be present at birth and others are age-related manifestations. A National Institutes of Health (NIH) Consensus Development Conference identified the following seven features of the disease, of which two or more are required to establish the diagnosis of NF1: 1. Six café au lait spots equal to or greater than 5 mm in longest diameter in prepubertal patients and 15 mm in longest diameter in postpubertal patients 2. Two or more neurofibromas of any type or one plexiform neurofibroma 3. Freckling in the axilla or inguinal regions 4. Optic glioma (optic pathway glioma) 5. Two or more Lisch nodules (iris, hamartomas)
6. Distinct osseous lesion, such as sphenoid wing dysplasia or cortical thinning of the cortex of long bones with or without pseudarthrosis 7. First-degree relative (parent, sibling, or child) with NF1 according to the above-listed criteria Some patients also manifest learning disabilities or language disorders. A careful examination and a detailed history of the patient’s symptoms help distinguish NF1 and NF2.
CLINICAL CHARACTERISTICS OF NEUROFIBROMATOSIS 2 Definition The NIH Consensus Development Conference also developed guidelines for the diagnosis of NF2. NF2 is distinguished by bilateral vestibular schwannomas with multiple meningiomas, cranial tumors, optic gliomas, and spinal tumors. A definite diagnosis is made on the basis of the presence of bilateral vestibular schwannomas or developing a unilateral vestibular schwannoma by age 30 and a first-degree blood relative with NF2, or developing at least two of the following conditions known to be associated with NF2: meningioma, glioma, schwannoma, or juvenile posterior subcapsular lenticular opacity/juvenile cortical cataract (Table 57-1).1 There may be significant heterogeneity in the presentation of the disease from one individual to the next. Some individuals may have a very mild form of the disease with small vestibular schwannomas manifesting in an older individual. Meanwhile, some children present with multiple intracranial tumors at a very young age. Despite the heterogeneity of the disease within a family, the expression of NF2 tends to be very similar.2 There is a significant genetic component to the disease with much variability within the parameters of the observed phenotype. Studies have shown that a truncating mutation (nonsense and frame shift) may be linked with a more severe form of NF2.3-5 The more severe form of 691
OTOLOGIC SURGERY 80
TABLE 57-1 Neurofibromatosis 2 (NF2) Diagnostic
70
Individuals with the Following Clinical Features Have Confirmed (Definite) NF2 Bilateral VS or family history of NF2 (first-degree family relative) plus Unilateral VS <30 years or Any two of the following: meningioma, glioma, schwannoma, juvenile posterior subcapsular lenticular opacities/juvenile cortical cataract Individuals with the Following Clinical Features Should Be Evaluated for NF2 (Presumptive or Probable NF2) Unilateral VS <30 years plus at least one of the following: meningioma, glioma, schwannoma, juvenile posterior subcapsular lenticular opacities/juvenile cortical cataract Multiple meningiomas (≥2) plus unilateral VS <30 years or one of the following: glioma, schwannoma, juvenile posterior subcapsular lenticular opacities/juvenile cortical cataract VS, vestibular schwannomas.
NF2 is termed Wishart form. Individuals with this severe form present with early onset of the disease with multiple intracranial schwannomas and meningiomas that result in blindness, deafness, paralysis, and death by age 40. Despite the strong genotype-phenotype correlation, individual differences in tumor growth occur within subjects, making it difficult to predict how an individual will change over time even when the genotype is known. The milder form, or Gardner form, of NF2 is less debilitating. The schwannomas may remain stable for many years, few meningiomas develop, and patients may not develop symptoms until later in life and often have fewer disabilities. The genetic basis of the mild form has not been well characterized. Many of these may be mosaic forms of the disease, however.2-4,6-12
Prevalence and Incidence The average age of diagnosis of NF2 is 25 years; however, many patients present with symptoms before the diagnosis. There is an average delay of diagnosis of approximately 7 years (Fig. 57-1). There is no difference in the proportion of men versus women who develop NF2, and no prevalence has been described based on ethnicity. Epidemiologic studies place the incidence of NF2 between 1 in 40,000 live births13 and 1 in 87,410 live births.14
Imaging Studies All patients suspected to have NF2 should have a highquality magnetic resonance imaging (MRI) scan performed with thin cuts through the internal auditory canal (IAC). All patients diagnosed with a unilateral vestibular schwannoma should have a dedicated IAC series to ensure there is not another tumor on the opposite side. Patients diagnosed with NF2 should have a complete
Age of first
Criteria
Age of first Age of diagnosis
80 70
60
60
50
50
40
40
30
30
20
20
10
10
Age of diagnosis
692
0
0 1 13 25 37 49 61 73 85 97 109 121 133 145 157 Patient
FIGURE 57-1. Time from first symptoms to first surgery.
spine series to evaluate the spine and stage the disease. Small spinal tumors commonly may be found in the cauda equina area, and occasionally a large asymptomatic schwannoma or meningioma may be found in the spine that could require treatment. Early treatment of spine tumors can significantly reduce the mortality associated with these tumors. Older patients who present with bilateral IAC tumors must be worked up for other carcinomas. It is unusual for patients older than 40 years to present with NF2, although with more sensitive MRI scanning techniques, more of these individuals are being diagnosed at an older age. Metastasis may rarely manifest with bilateral IAC lesions, and it is important that carcinoma be ruled out in any patient older than 40 years who presents with bilateral IAC lesions. A patient identified with NF2 should have a complete cranial MRI scan with cervical thoracic and lumbar spinal imaging. This scan serves as a baseline. A 6-month follow-up MRI scan is recommended for the intracranial tumors. If the tumors exhibit stability, a yearly MRI scan is performed on the intracranial structures. Large tumors in the spine may be monitored with a similar frequency. If no spinal tumors are present, spinal imaging should be performed if the patient becomes symptomatic. Monitoring spinal tumors every 3 years is recommended when these are present. Intracranial imaging is performed on a yearly basis unless studies over several years indicate stability of the tumors.
Molecular Genetics The NF2 gene was mapped to chromosome 22 q 12-2 in 1993.15-17 The NF2 gene located at chromosome 22 codes for a tumor suppressive protein termed Merlin or Schwannomin. This protein negatively regulates Schwann cell production. The loss of this protein allows overproduction of Schwann cells. The mutation in the NF2 chain predisposes individuals to developing a schwannoma when the second hit occurs to the gene; control of Schwann cells is lost or mutated within the cell. Various types of mutations have been identified, including single base substitutions, insertions, and deletions.4,18-20 The mild, or Gardner, type of NF2 may be associated
Chapter 57 • Neurofibromatosis 2
with missense mutations, whereas associations between the other mutations and phenotypes are not as clear.21 The occurrence of NF2 is not restricted to families known to carry the mutation. Frequently, genetic mosaicism occurs, which may not be detected by common mutation analysis techniques.22 Unilateral vestibular schwannomas may exhibit the same type of genetic markers as NF2.23 The mutations in unilateral vestibular schwannomas are confined to the affected tumor tissue. In patients with NF2, the mutation is present in all cell types.22
Family History NF2 is an autosomal dominant disease, and 50% of children of affected individuals are at risk for developing the disease. Of patients in whom NF2 is diagnosed, 50% present with a family history of NF2. Half of all NF2affected patients have no family history of NF2 and are considered founder cases. NF2 presentation and phenotype tend to be similar within families. The likelihood of NF2 occurring in related individuals who do not exhibit similar clinical symptoms to an affected family member is small. Consideration must still be given to screening these individuals for risks despite the lack of clinical symptoms. Individuals at risk for developing NF2 must be screened to provide an early diagnosis. Individuals at risk include children of NF2-affected patients and their siblings. Fifty percent of all children of NF2 patients are found to have the disease. Siblings of a diagnosed NF2 patient are at risk, especially if the parent also has NF2. The type of screening and timing of screening depend on each NF2 center’s preference. Early screening is advocated so that tumors may be diagnosed pre symptomatically. Screening may occur via MRI or genetic blood testing.
Screening MRI screening of potentially affected individuals uses a postcontrast T1-weighted sequence of the full head with thin cuts through the IAC. A dedicated IAC MRI scan identifies most NF2 patients by showing any vestibular schwannomas. Screening of the spine and ophthalmologic examination should be considered if the cranial MRI scan is positive. An audiogram (pure tone thresh olds) or current clinical standard auditory brainstem response (ABR) testing is likely to miss small vestibular schwannomas. MRI can diagnose presymptomatically. MRI is recommended for at-risk children when this test can be performed without sedation; this usually can be done when the child is 7 to 9 years old. A recommended first step for children younger than 7 years is an audiogram. Any child with an NF2-associated symptom, such as hearing loss or facial weakness, should be screened without regard to the need for sedation or age; MRI should be performed as soon as possible after the symptoms become apparent.
693
Identification of the NF2 gene and chromosome 22 has made genetic testing possible. It is recommended that patients with NF2 see a genetic counselor to discuss the hereditary consequences of the disease. Blood testing for the mutation is able to identify the defect of the NF2 gene in approximately 70% to 75% of patients with a known diagnosis of NF2. If the defect is identified in the affected individual, potential family members may be screened. If the gene is not identified in the affected individuals, blood screening of family members cannot be performed. The use of blood screening for patients without a diagnosis of NF2 or with a suspected diagnosis of NF2 is not recommended. New mutations in patients with mild presentation are most likely to be missense mutations, and this is difficult to identify with genetic testing of NF2 patients.
Tumor Types Bilateral vestibular schwannomas (acoustic neuromas) are benign neoplasms of the acoustic or eighth cranial nerve (Fig. 57-2).24 The tumors are thought to arise at the glioma–Schwann’s cell junction within the internal auditory meatus. The tumors most commonly arise from the superior vestibular nerve, although with NF2, tumors may be found on the cochlear and facial nerves within the internal auditory meatus. The consequences of a vestibular schwannoma are numerous, including hearing loss progressing to deafness, dizziness and balance problems, tinnitus, facial nerve paralysis, brainstem compression, and, if left untreated, death. Despite the strong genetic effect in NF2, there is enormous variability in the number of tumor types, the rate of progression, and the disabilities experienced. This enormous variability is also found in patient presentation. Some patients may be asymptomatic. Patients who have no symptoms when diagnosed have generally been identified on the basis of genetic analysis conducted because a blood relative has NF2 or presymptomatic screening. Although the NIH criteria for NF2 require the presence of bilateral vestibular schwannomas for diagnosis, patients may first develop unilateral schwannomas as a young child with no other tumors, or adult patients may present with multiple meningiomas (cranial and spinal) and no vestibular schwannomas.9,25 Although the NIH criteria for NF2 imply that all NF2 patients develop bilateral vestibular schwannomas, some researchers are not convinced of this.36 Evans and colleagues26 based their conclusion on the observation of a possible variant form of NF2 manifesting with skin and spinal tumors in the absence of vestibular schwannomas. Nonetheless, the phenotype generally is reflective of the underlying disorder. The natural history study of vestibular schwannomas in NF2 conducted at the House Ear Institute showed that 10 of 80 (12.5%) enrolled subjects had no symptoms at diagnosis, and 23 (28.8%) had cranial meningiomas
694
OTOLOGIC SURGERY
A
B
C FIGURE 57-2. Bilateral vestibular schwannomas are characteristics of NF2. A, Small bilateral vestibular schwannomas. B, Medium-size vestibular schwannomas that are compressing the brainstem. C, Giant bilateral vestibular schwannomas that are compressing the brainstem and causing hydrocephalus.
and spinal meningiomas in addition to bilateral vestibular schwannomas. Nearly half (47.5%) had one vestibular schwannoma removed before enrollment. Generally, the tumor resected before enrollment was removed 1.5 years after discovery, and was an average of 2.1 cm at removal. Few patients in the natural history study had spinal tumors or meningiomas removed before enrollment. The preliminary data would indicate that, for this sample of NF2 subjects, the most salient medical issue is the growth of the vestibular schwannomas.
Intracranial Schwannomas Vestibular schwannomas are the most common intracranial schwannoma associated with NF2. The most frequently identified nonvestibular schwannomas are schwannomas of CN III and V. Bilateral CN III or V schwannomas are the most common additional schwannomas seen. It is important to identify these lesions on MRI. Lower cranial nerve schwannomas may also be identified, but are much less frequently seen. A vestibular schwannoma rarely turns malignant, and sometimes the unilateral vestibular schwannoma may regress in size altogether. Growth of the tumors does not seem to be related either to loss of heterozygosity (genetic level of analysis) or to auditory
functioning (phenotype level of analysis). For this reason, it is recommended that a patient have at least yearly MRI scans to track changes in size.33-40 All newly diagnosed patients should have a full head and spine study to stage their disease. After the disease is staged, a 6 month study is performed to determine if the tumor is fast-growing or slow-growing.41 CN V, or trigeminal nerve, schwannomas are the most common type seen after vestibular schwannomas. Oculomotor schwannomas are the third most common schwannomas seen intracranially. Occasionally, it is difficult to distinguish whether these schwannomas have arisen from the oculomotor, trochlear, or obducent nerve, especially when the tumor rises within the cavernous sinus. Trochlear and abducent schwannomas are extremely rare with only a handful of cases reported in the literature. Facial nerve schwannomas may also be seen, although these are difficult to distinguish radiographically from vestibular schwannomas. Some patients may present with small facial schwannomas that are commonly encountered with large tumors where the distinction between the facial nerve and cochlear vestibular nerve cannot be found. CN III and V schwannomas are usually slow-growing and require treatment only when
Chapter 57 • Neurofibromatosis 2
s ignificant growth has occurred, or other intracranial complications are eminent. Lower cranial nerve schwannomas can be quite significant in NF2 because these can lead to speech and swallowing disorders. When bilateral lower cranial nerve schwannomas or jugular foramen meningiomas are associated with these tumors, the patient may develop aspiration problems, which can cause significant morbidity. Glossopharyngeal, vagal, and hypoglossal neuropathies resulting from schwannomas on CN IX, X, and XII may lead to speech and swallowing disorders. Glossopharyngeal schwannomas are the most common schwannoma of the jugular foramen. These may manifest with swallowing difficulty, which may lead to the requirement of gastrostomy feeding tubes for nutritional status. Vagal nerve defects may contribute to swallowing difficulties related to esophageal dysmotility. Vagal nerve deficits may manifest with voice hoarseness owing to vocal cord paralysis, but the most imminent issue is aspiration, which occurs because of loss of sensory innervation to the larynx and loss of the reflexive airway protective mechanisms. Aspiration is often silent in such cases and leads to life-threatening pulmonary complications including pneumonia. Tracheotomy may be required leading to other potential life-threatening complications, including pulmonary infection. It is believed that lower cranial nerve neuropathies may contribute to mortality associated with NF2.
Hearing Changes in Patients with Vestibular Schwannomas Hearing loss is well documented as the most common presenting symptom in patients who have vestibular schwannomas.42-51 Auditory changes over time in vestibular schwannoma patients are less well known. Rosenberg52 studied the natural history of 80 patients with non-NF2 unilateral vestibular schwannomas for an average of 4.4 years. Rosenberg52 found a positive correlation between tumor growth and worsening pure tone average. There was no statistically significant correlation, however, between positive tumor growth and speech discrimination, change in ABR, and bithermal caloric electronystagmography test result over time. Lalwani and colleagues53 reported that pure tone pattern speech receptive thresholds and word recognition scores were significantly worse in NF2 patients who had a mild form of NF2 and large tumors compared with patients with mild NF2 with small tumors. Loss of acoustic reflexes and prolonged wave III and V were also associated with larger tumors. In contrast, patients with severe NF2 showed no relationship among tumor size and pure tone levels, speech discrimination thres holds, or word recognition scores. The lack of association may have been due to complete loss of hearing in severe NF2 patients at the time of the assessment. The larger tumors were also associated with prolonged ABR waves
695
III and V latencies. No data across time were reported. Generally, hearing is progressively impaired with increasing growth of vestibular schwannomas necessitating the need for surgical intervention or medical treatments in NF2 patients. The natural history study evaluated hearing changes prospectively in 63 newly diagnosed NF patients, and found that hearing was stable after diagnosis. During the first 2 years after diagnosis, 27% of patients had a significant change in their hearing, and 73% had very stable hearing during the 2 year period. Patients with a family history of NF2 had more stable hearing after an initial diagnosis compared with patients without a family history. Patients with a family history are usually diagnosed at a younger age; the stability of hearing in this group may represent the younger age group. The better the hearing in the newly diagnosed patients, the more stable the hearing. Hearing changes did not vary between ears with each ear acting independently.54
Other Tumor Types in Neurofibromatosis 2 Meningiomas NF2 has been associated with multiple central nervous system tumors, the most common of which are intracranial meningiomas (Fig. 57-3).55 Nearly all NF2 patients develop these tumors in time. Fifty percent of NF2 patients present with schwannomas and meningiomas; 90% present with spine tumors in addition to schwannomas. The presence of more than one type of tumor in a patient usually indicates a more aggressive disease course. The co-occurrence of vestibular schwannomas and meningiomas has been linked to a synergistic effect of increase in growth rate of the schwannoma and the meningioma beyond that expected of a sporadic schwannoma or meningioma.56,57 Despite the high number of patients with multiple tumors initially, most meningiomas and spinal tumors are asymptomatic and are first seen on MRI. In addition, multiple skin tumors may be found in patients with NF2 (Table 57-3). Meningiomas are monitored in NF2 unless they are quite large. Growing meningiomas may be surgically removed. Growing meningiomas may result in increased intracranial pressure, intractable headaches, hydrocephalus, and seizure disorders.
Spinal Tumors Various spinal tumors may occur in NF2 patients, and can be found in the cervical, thoracic, and lumbar region. These tumors are categorized further as either extramedullary or intramedullary tumors depending on their presentation relative to the spinal cord. Extramedullary tumors are commonly schwannomas or meningiomas, whereas intramedullary tumors are often ependymomas, but these can be astrocytomas or schwannomas.58 Spinal tumors may often be numerous small tumors in the
696
OTOLOGIC SURGERY TABLE 57-2 First Symptoms of Neurofibromatosis 2 Symptoms Neurologic Skin tumor Vision loss Asymptomatic Tinnitus Weakness Vertigo Other/unspecified
Patients (%) 17.5 11.7 10.7 10.7 7.8 2.9 1 4.9
TABLE 57-3 Tumor Type FIGURE 57-3. Multiple meningiomas can be seen in severe forms of NF2. Meningiomas may occur throughout the cranium and skull base area.
cauda equina. Tumors located in the cervical or thoracic region are usually solitary tumors, and these may grow to cause spinal cord compression. These tumors may have a solid or cystic component similar to schwannomas and meningiomas seen intracranially. These tumors may extend out the spinal foramen into the soft tissue causing spinal cord compression.59 As each NF2-related tumor grows and exerts pressure on surrounding structures, treatment encompasses surgical resection, radiation therapy, or both. Another treatment choice early on in the disease course is surgical decompression, in which space is created for the growing tumor, relieving the pressure of the tumor against the nerve. Continued growth of spinal tumors causes loss of motion, numbness, tingling, and, eventually, paralysis.
Eye Findings NF2 subjects tend also to develop cortical and posterior subcapsular cataracts, which can lead to blindness (see Table 57-2).27 Retinal hamartomas have been observed in a few cases,2,28-30 but are not as frequent. Some subjects (2% to 3% of subjects)31,32 present with numbness or tingling in their arms or legs. Almost 30% of NF2 subjects may have surgery to remove spinal tumors, but the progression of spinal tumors associated with NF2 is not well described. At this time, the presence of vestibular schwannoma in NF2 and the consequences of not treating them are well known, and these tumors may be the most debilitating.
TREATMENT OPTIONS FOR VESTIBULAR SCHWANNOMAS IN NEUROFIBROMATOSIS 2 The treatment options for a patient with bilateral vestibular schwannomas vary considerably because of the wide variety of tumor sizes and clinical presentations.
Tumor Type Bilateral vestibular schwannoma Skin Meningioma Spinal
Patients (%) 99 50 46 60
Associated symptoms (brainstem compression or hydrocephalus), loss of useful hearing, and the status of other intracranial tumors all must be considered when discussing treatment intervention.
Hearing Preservation Patients who present with bilateral small tumors (<2 cm in greatest diameter) and good hearing may be candidates for hearing preservation procedures. In these patients, total tumor removal is attempted on the side of the larger tumor or on the side with worse hearing. If hearing is successfully preserved on the first side, contralateral tumor removal may be attempted 6 months later. Hearing preservation rates for small unilateral tumors have approached 70%.51 The results in NF2 patients seem to be worse, however, than the results reported in patients with unilateral vestibular schwannomas.63 Doyle and Shelton64 found that 67% of NF2 patients underwent hearing presentation surgery using the middle fossa approach, and 38% of those had serviceable hearing postoperatively. Improvements in the middle fossa surgical approach were introduced in 1992 that led to a significant improvement in the outcomes of NF2 tumor removal. A more recent review of 18 NF2 patients who had middle fossa removal of their tumors reported that approximately 50% of patients had hearing preserved at the preoperative level; this compares with 25% in the previous study. The overall ability to preserve any hearing was still close to 68%.65 The results of this series have led to a more aggressive attempt to preserve hearing in NF2 patients. Patients who present with small tumors and good hearing are now routinely offered an attempt at hearing preservation. Long-term follow-up is still needed in NF2 patients because additional tumors may arise in the facial or cochlear nerves.
Chapter 57 • Neurofibromatosis 2
A
697
B
FIGURE 57-4. Histology of a vestibular schwannoma. A, Unilateral non-NF2 vestibular schwannoma. Typical schwannoma stain with Bodian silver stain. It stains only the nerve fiber, so the schwannoma cells are not evident. The fibers on the surface of the tumor are visible, however. Nerve fibers of non-NF2 vestibular schwannomas are displaced by the schwannoma cells. B, Another Bodian silver stain with black strands embedded within the tumor, representing the nerve fibers invaded by the tumor. This invasion is different from that seen with non-NF2 solitary vestibular schwannomas. Non-NF2 solitary tumors invade the aggregates of cells and push the fibers aside, rather than invading between the fibers. NF2 vestibular schwannomas are histologically different from non-NF2 sporadic, unilateral tumors.
We have published our results treating children with NF2 with small tumors and good hearing. This retrospective chart review reported on 35 children with NF2 who had undergone middle fossa resection (47 surgeries) between 1992-2004. In 55% of surgeries, hearing of less than or equal to 70 dB pure tone average was maintained postoperatively. It is now our practice to perform middle fossa resection in children with NF2, and the sooner in the disease process, the better. Our results indicate that hearing and facial nerve function can be successfully preserved using this approach. Factors to consider include patient age, severity of additional NF2-related symptoms, and obtaining high-quality, thin-slice MR images before surgery. Bilateral middle fossa resection after hearing preservation on the first side is also successful and potentially preserves hearing in both ears.66
Observation without Surgical Intervention Observation without surgical intervention is the most common treatment option used in patients with NF2 and is used when a small tumor is present in a patient with only one hearing ear or when bilateral tumors are too large for hearing preservation procedures. The patient is assessed routinely to ensure that brainstem compression or hydrocephalus does not result. Initially, MRI is performed 6 months after diagnosis, and then annual MRI scans are performed to document tumor size and determine if intervention is required. Surgical intervention is considered if life-threatening complications occur, the tumors become excessively large (increasing the perioperative morbidity), or the hearing becomes unserviceable.
Middle Fossa Craniotomy and Internal Auditory Canal Decompression without Tumor Removal Middle fossa craniotomy and IAC decompression without tumor removal allows the tumor to grow without causing compression of CN VII and VIII. This procedure is recommended when progression of hearing loss occurs in a patient who is being observed. The bone surrounding the IAC is removed extensively, allowing the entire tumor and CN VII and VIII complex to be decompressed. The tumor itself is not removed because this may increase the risk of hearing loss. Stabilization and improvement of hearing may occur after this procedure.
Retrosigmoid Craniotomy with Partial Removal Retrosigmoid craniotomy with partial removal in NF2 patients carries a significant risk because the cochlear fibers are dispersed throughout the tumor, in contrast to unilateral vestibular schwannomas (Fig. 57-4). The risk of hearing loss with partial removal is much higher, and this procedure is typically not recommended.
Non–Hearing Preservation, Translabyrinthine/Suboccipital Approach and Total Tumor Removal The most common surgical procedure performed in patients with NF2 is non–hearing preservation, translabyrinthine/suboccipital approach and total tumor removal.
698
OTOLOGIC SURGERY
Most patients present when either the tumor is too large for hearing preservation, or the hearing loss is already at a significant level and hearing preservation is not considered. This approach is used for patients with large tumors who have brainstem compression even if serviceable hearing exists. The translabyrinthine or suboccipital craniotomy approaches may be used for this procedure. The risk of a recurrent tumor is slightly higher with the suboccipital approach, however, and with inexperienced surgeons because residual tumor is often left in the lateral aspect of the IAC.
Auditory Brainstem Implant The auditory brainstem implant was developed at the House Ear Institute to allow electric stimulation of the cochlear nucleus after bilateral vestibular schwannoma removal. The device is placed on the brainstem (Fig. 57-5) during translabyrinthine vestibular schwannoma removal. This device is indicated in individuals who have no serviceable hearing and are undergoing vestibular schwannoma removal. Most patients obtain enhanced communication skills with the device.
Stereotactic Irradiation Stereotactic irradiation has been recommended for some NF2 patients, but its use must be carefully considered because radiation exposure may induce or accelerate tumors in a patient with an inactivated tumor suppressor gene. It was reported more recently that two of four patients who had previously received radiation therapy developed a malignancy in the irradiated ear.67 Many algorithms have been proposed as treatment plans, but none are widely accepted among NF2 specialists. The range of treatment options is large—larger than for any other nervous system tumor. The potential benefits of treatment are great (hearing preservation), but the potential risks are also significant, and the riskiest procedures have the greatest potential benefit. Often patients have a window of opportunity when they can choose a risky but potentially beneficial procedure; once the window closes, they cannot go back. Patients, families, and care providers need to know the natural history of these tumors to make rational recommendations and decisions regarding these treatment options better. In addition, noninvasive and tumor-related markers of behavior need to be identified that can tailor further anticipatory guidance for any one patient.
MANAGEMENT OF NEUROFIBROMATOSIS 2 Initial evaluation of an NF2 patient can be very complex because this is a multisystem disease. An inadequate diagnosis may render a patient impaired because early
FIGURE 57-5. CT scan showing placement of auditory brainstem implant within the lateral recess of the fourth ventricle.
diagnosis and treatment may have prevented further impairments. NF2 patients may typically see many different physicians, each with experience in a different field of expertise. NF2 patients require one physician to lead the treatment team—a case manager as it were—to ensure comprehensive care. A neurotologist, geneticist, neurosurgeon, or neurologist may function as the lead physician, depending on the NF2 center. A comprehensive battery of tests is necessary for tumor detection and adequate staging. The initial MRI study showing the presence of bilateral vestibular schwannomas may be inadequate for tumor follow-up. A cranial MRI scan may not have included the IACs, or a spine series may have focused on only one segment of the spine. MRI with gadolinium and thin cuts through the IAC is necessary for the head. Particular attention is focused in the IACs, cavernous sinus, and jugular foramen areas. Any cranial nerve may have tumor formation and should have complete imaging. Auditory assessment is necessary to determine the extent of hearing impairment. At a minimum, this assessment consists of a standard audiogram, with air and bone pure tone thresholds, and speech testing. Some centers prefer additional testing with ABR to assess cochlear nerve function. ABR testing is particularly helpful when considering a hearing preservation procedure. Electronystagmography testing has benefit in determining tumor location; however, its utility for clinical assessment is still under investigation. A complete neurologic examination is required for individuals with suspected NF2. The standard neurologic assessment of dermatomes and muscle strength
Chapter 57 • Neurofibromatosis 2
is required for assessment of potential spinal cord impairment. Cranial nerve testing may find subtle abnormalities for which the patient has slowly compensated. In addition, the patient may not even be aware of his or her own impairment. This is particularly true of the lower cranial nerves. A complete MRI spinal cord survey is required to identify tumors within the spine. The use of spine screening is still under investigation for all NF2 patients. Patients with a significant tumor burden, a family history of spine tumors, or spinal tumor symptoms should definitely have a spine series. A spine series is required in all symptomatic patients. A neuro-ophthalmologic examination is required for all patients with NF2. The potential for deafness in these individuals requires that everything be done to preserve vision. A slit-lamp examination is required. It is preferable that a patient be evaluated by an ophthalmologist familiar with NF2. The initial comprehensive evaluation consists, at a minimum, of MRI of the IAC with gadolinium, auditory assessment, and physical examination. A comprehensive examination includes the previously mentioned MRI with the addition of a full spine series, an ABR test, and an ophthalmologic examination. The timing of follow-up studies is currently inconsistent among NF2 specialty centers. We recommend repeat testing at 6 months and then yearly testing consisting of an MRI of the head and spine, neurology examination, and audiometric testing. When the growth rate from the tumors has been determined, some of these tests may be spread out over time. Spinal tumors tend to be very slow-growing, and after diagnosis may be imaged every 1 to 5 years. The potential for new tumor formation exists especially in patients with severe disease, and it is important that this information be conveyed to the NF2 patient so that comprehensive followup may occur.
GENETIC TESTING Identification of the NF2 gene on chromosome 22 has made genetic testing possible. It is recommended that patients with NF2 see a genetic counselor to discuss the hereditary consequences of this disease. Genetic blood screening is able to identify the defect on the NF2 gene in approximately 70% to 75% of patients with a known diagnosis of NF2. If the defect is identified, potential family members may be screened. If the gene is not identified, blood screening of family members can be performed. The use of blood screening for patients without a diagnosis or with a suspected diagnosis of NF2 is not recommended. New mutations in patients with mild presentation are most likely missense mutations, which are difficult to identify by genetic testing.
699
SUMMARY Care of patients with NF2 requires knowledge of all tumors and symptoms involved with the disorder. It is recommended that patients receive care in a center with expertise in NF2. The role of the neurotologist in this care is determined by the specialty center.
REFERENCES 1. Gutmann D H, Aylsworth A, Carey JC, et al: The diagnostic evaluation and multidisciplinary management of NF1 and NF2. JAMA 278:51-57, 1997. 2. Parry D M, Eldridge R , Kaiser-Kupfer M I, et al: Neurofibromatosis 2 (NF2): Clinical characteristics of 63 affected individuals and clinical evidence for heterogeneity. Am J Med Genet 52:450-461, 1994. 3. Evans DG, Trueman L , Wallace A, et al: Genotype/ phenotype correlations in type 2 neurofibromatosis (NF2): Evidence for more severe disease associated with truncating mutations. J Med Genet 35:450-455, 1998[published erratum appears in J Med Genet 361(1):87, 1999]. 4. Kluwe L , Mautner VF: A missense mutation in the NF2 gene results in moderate and mild clinical phenotypes of neurofibromatosis type 2. Hum Genet 97:224-227, 1996. 5. Ruttledge M H, Andermann A A, Phelan C M, et al: Type of mutation in the NF2 gene frequently determines severity of disease. Am J Hum Genet 59:331-342, 1996. 6. Baser M E, Mautner VF, Ragge N K, et al: Presymptomatic diagnosis of neurofibromatosis 2 using linked genetic markers, neuroimaging, and ocular examinations. Neurology 47:1269-1277, 1996. 7. Bijlsma E K, Merel P, Fleury P, et al: Family with neurofibromatosis type 2 and autosomal dominant hearing loss: Identification of carriers of the mutated NF2 gene. Hum Genet 96:1-5, 1995. 8. Gardner WJ, Frazier C H : Bilateral acoustic neurofibromas: A clinical study and field survey of a family of five generations with bilateral deafness in thirty-eight members. Arch Neurol Psychiatry 23:266-302, 1930. 9. Mautner VF, Baser M E, Kluwe L : Phenotype variability in two families with novel splice-site and frameshift NF2 mutations. Hum Genet 98:203-206, 1996. 10. Sainio M, Strachan T, Blomstedt G, et al: Presymptomatic DNA and MRI diagnosis of neurofibromatosis 2 with mild clinical course in an extended pedigree. Neurology 45:1314-1322, 1995. 11. Welling D B : Clinical manifestations of mutations in the neurofibromatosis type 2 gene in vestibular schwannomas (acoustic neuromas). Laryngoscope 108:178-189, 1998. 12. Wishart J H : Case of tumors of the skull, dura mater and brain. Edinburgh Med Surg J 18:393-397, 1822. 13. Evans DG, Huson S M, Donnai D, et al: A genetic study of type 2 neurofibromatosis in the United Kingdom, I: Prevalence, mutation rate, fitness, and confirmation of maternal transmission effect on severity. J Med Genet 29:841-846, 1992.
700
OTOLOGIC SURGERY
14. Antinheimo J, Sankila R , Carpen O, et al: Populationbased analysis of sporadic and type 2 neurofibromatosisassociated meningiomas and schwannomas. Neurology 54:71-76, 2000. 15. Rouleau G A, Wertelecki W, Haines J L , et al: Genetic linkage of bilateral acoustic neurofibromatosis to a DNA marker on chromosome 22. Nature 329:246-248, 1987. 16. Rouleau G A, Merel P, Lutchman M, et al: Alteration in a new gene encoding a putative membrane-organizing protein causes neurofibromatosis type 2. Nature 363:515521, 1993. 17. Trofatter J A, MacCollin M M, Rutter J L , et al: A novel moesin-ezrin-, radixin-like gene is a candidate for the neurofibromatosis type 2 tumor suppressor. Cell 72:791800, 1993. 18. Merel P, Haong-Xuan K, Sanson M, et al: Predominant occurrence of somatic mutations of the NF2 gene in meningiomas and schwannomas. Genes Chromosomes Cancer 13:211-216, 1995. 19. Merel P, Hoang-Xuan K, Sanson M, et al: Screening for germ-line mutations in the NF2 gene. Genes Chromosomes Cancer 12:117-127, 1995. 20. Welling D B, Guida M, Goll F, et al: Mutational spectrum in the neurofibromatosis type 2 gene in sporadic and familial schwannomas. Hum Genet 98:189-193, 1996. 21. Welling D B : Clinical manifestations of mutations in the neurofibromatosis type 2 gene in vestibular schwannomas (acoustic neuromas). Laryngoscope 108:178-189, 1998. 22. Wu C L , Thakker N, Neary W, et al: Differential diagnosis of type 2 neurofibromatosis: Molecular discrimination of NF2 and sporadic vestibular schwannomas. J Med Genet 35:973-977, 1998. 23. Irving R M, Harada T, Moffat D A, et al: Somatic neurofibromatosis type 2 gene mutations and growth characteristics in vestibular schwannoma. Am J Otol 18:754-760, 1997. 24. Cushing H : Bilateral Acoustic Tumors, Generalized Neurofibromatosis and the Meningeal Endotheliomata. Tumors of the Nervous Acoustics and the Syndrome of the Cerebellopontine Angle. Philadelphia, Saunders, 1963 (1917 original edition). 25. Mautner VF, Lindenau M, Koppen J, et al: Type 2 neurofibromatosis without acoustic neuroma. Zentralbl Neurochir 56:83-87, 1995. 26. Evans DG, Lye R , Neary W, et al: Probability of bilateral disease in people presenting with a unilateral vestibular schwannoma. J Neurol Neurosurg Psychiatry 66:764767, 1999. 27. Bouzas E A, Freidlin V, Parry D M, et al: Lens opacities in neurofibromatosis 2: Further significant correlations. Br J Ophthalmol 77:354-357, 1993. 28. Good WV, Erodsky MC, Edwards M S, Hoyt WF: Bilateral retinal hamartomas in neurofibromatosis type 2. Br J Ophthalmol 75:190, 1991. 29. Landau K, Yasargil G M : Ocular fundus in neurofibromatosis type 2. Br J Ophthalmol 77:646-649, 1993. 30. Ragge N K, Baser M E, Klein J, et al: Ocular abnormalities in neurofibromatosis 2. Am J Ophthalmol 120:634641, 1995. 31. Evans DG, Huson S M, Donnai D, et al: A clinical study of type 2 neurofibromatosis. Q J Med 84:603-618, 1992.
32. Lim DJ, Rubenstein A E, Evans DG, et al: Advances in neurofibromatosis 2 (NF2): A workshop report. J Neurogenet 14:63-106, 2000. 33. Mathies C, Samii M, Krebs S : Management of vestibular schwannomas (acoustic neuromas): Radiological features in 202 cases—their value for diagnosis and their predictive importance. Neurosurgery 40:469-481, 1997. 34. Burkey J M, Rizer FM, Schuring AG, et al: Acoustic reflexes, auditory brainstem response, and MRI in the evaluation of acoustic neuromas. Laryngoscope 106:839841, 1996. 35. Curati WL , Graif M, Kingsley D P, et al: MRI in acoustic neuroma: A review of 35 patients. Neuroradiology 28:208-214, 1986. 36. Duffner PK, Cohen M E, Seidel FG, Shucard DW: The significance of MRI abnormalities in children with neurofibromatosis. Neurology 39:373-378, 1989. 37. Levine SC, Antonelli PJ, Le CT, Haines SJ: Relative value of diagnostic tests for small acoustic neuromas. Am J Otol 12:341-346, 1991. 38. Lhullier FM, Doyon D L , Halimi PM, et al: Magnetic resonance imaging of acoustic neuromas: Pitfalls and differential diagnosis. Neuroradiology 34:144-149, 1992. 39. Long S A, Arriaga M, Nelson R A : Acoustic neuroma volume: MRI-based calculations and clinical implications. Laryngoscope 103:1093-1096, 1993. 40. Modugno GC, Pirodda A, Ferri GG, et al: Small acoustic neuromas: Monitoring the growth rate by MRI. Acta Neurochir 141:1063-1067, 1999. 41. Irving R M, Moffat D A, Hardy DG, et al: A molecular, clinical, and immunohistochemical study of vestibular schwannoma. Otolaryngol Head Neck Surg 116:426-430, 1997. 42. Strasnick B, Glasscock M E III, Haynes D, et al: The natural history of untreated acoustic neuromas. Laryngoscope 104:1115-1119, 1994. 43. Fucci M J, Buchman C A, Brackmann D E, Berliner K I : Acoustic tumor growth: Implications for treatment choices. Am J Otol 20:495-499, 1999. 44. Brackmann D E, Owens R M, Friedman R A, et al: Prognostic factors for hearing preservation in vestibular schwannoma surgery. Am J Otol 20:495-499, 1999. 45. Briggs R J, Brackmann D E, Baser M E, Hitselberger WE : Comprehensive management of bilateral acoustic neuromas: Current perspectives. Arch Otolaryngol Head Neck Surg 120:1307-1314, 1994. 46. Doyle K J, Nelson R A : Bilateral acoustic neuromas (NF2). In House WF, Luetje C M, Doyle K J (eds): Acoustic Tumors: Diagnosis and Management. San Diego, Singular Publishing Group, 1997. 47. Evans DG, Huson S M, Donnai D, et al: A genetic study of type 2 neurofibromatosis in the United Kingdom, I: Prevalence, mutation rate, fitness, and confirmation of maternal transmission effect on severity. J Med Genet 29:841-846, 1992. 48. Gadre A K, Kwartler J A, Brackmann D E, et al: Middle fossa decompression of the internal auditory canal in acoustic neuroma surgery: A therapeutic alternative. L aryngoscope 100:948-952, 1990. 49. Kesterson L , Shelton C, Dressler L , Berliner K I : Clinical behavior of acoustic tumors: A flow cytometric analysis. Arch Otolaryngol Head Neck Surg 119:269-271, 1993.
Chapter 57 • Neurofibromatosis 2 50. Saunders J E, Luxford WM, Devgan K K, Fetterman B L : Sudden hearing loss in acoustic neuroma patients. Otolaryngol Head Heck Surg 113:23-31, 1995. 51. Slattery WH III, Brackmann D E, Hitselberger W: Middle fossa approach for hearing preservation with acoustic neuromas. Am J Otol 18:596-601, 1997. 52. Rosenberg S I : Natural history of acoustic neuromas. L aryngoscope 110:497-508, 2000. 53. Lalwani A K, Abaza M M, Makariou EV, Armstrong M : Audiologic presentation of vestibular schwannoma in neurofibromatosis type 2. Am J Otol 19:352-357, 1998. 54. Masuda A, Fischer LM, Oppenheimer ML, et al. 2004 55. Bouzas E A, Parry D M, Eldridge R , Kaiser-Kupfer M I : Visual impairment in patients with neurofibromatosis 2. Neurology 43:622-623, 1993. 56. Pallini R , Tancredi A, Cassalbore P, et al: Neurofibromatosis type 2: Growth stimulation of mixed acoustic schwannoma by concurrent adjacent meningioma: Possible role of growth factors. Case report. J Neurosurug 89:149-154, 1998. 57. Antinheimo J, Haappasalo H, Haltia M, et al: Proliferation potential and histological features in neurofibromatosis 2-associated and sporadic meningiomas. J Neurosurg 87:610-614, 1997. 58. Patronas NJ, Courcoutsakis N, Bromley C M, et al: I ntramedullary and spinal canal tumors in patients with neurofibromatosis 2: MR imaging findings and correlation with genotype. Radiology 218:434-442, 2001.
701
59. Gillepsie J E : Imaging in neurofibromatosis 2: Screening using magnetic resonance imaging. Ear Nose Throat J 78:102-103, 1999. 60. Delleman JW, De Jong JG, Bleeker G M : Meningiomas in five members of a family over two generations, in one member simultaneously with acoustic neurinomas. Neurology 28:567-570, 1978. 61. King A, Gutmann D H : The question of familial meningiomas and schwannomas: NF2B or not to be? Neurology 54:4-5, 2000. 62. Wiebe S, Munoz DG, Smith S, Lee D H : Meningioangiomatosis: A comprehensive analysis of clinical and laboratory features. Brain 122:709-726, 1999. 63. Slattery WH III, Brackmann D E : Hearing preservation and restoration in CPA tumor surgery. Neurosurg Q 7:169-182, 1997. 64. Doyle K J, Shelton C : Hearing preservation in bilateral acoustic neuroma surgery. Am J Otol 14:562-565, 1993. 65. Slattery WH, Brackmann DE, Hitzelburger W: Hearing Preservation in NF-2. Am Journal Otology 19:638-643, 1998. 66. Slattery WH, Fischer LM, Hitzelburger W, et al: Hearing Preservation Surgery for NF-2 related Vestibular Schwannomas in Pediatric Patients. J Neurosurg 106: 255-260, 2007. 67. Baser M E, Ragge N K, Riccardi VM, et al: Phenotype variability in monozygotic twins with neurofibromatosis 2. Am J Med Genet 64:563-567, 1996.
58
Auditory Implants for the Central Nervous System Steven R. Otto, Jose N. Fayad, Robert V. Shannon, Derald E. Brackmann, and William E. Hitselberger Videos corresponding to this chapter are available online at www.expertconsult.com.
Loss of auditory nerve integrity, as often occurs after removal of vestibular schwannomas in neurofibromatosis type 2 (NF-2), for many years left patients completely deafened. Sign language, lipreading, and vibrotactile aids provided some communication assistance but could not restore useful auditory sensations. The development of the auditory brainstem implant (ABI), and more recently variations like the penetrating ABI (PABI) and auditory midbrain implant (AMI), provided a means of bypassing the cochlea and auditory nerve to directly stimulate the more central auditory pathways, thereby giving sound sensations to otherwise deaf patients. This chapter updates and discusses the clinical and surgical aspects of ABI, PABI, and AMI electrode array placement and perceptual performance. The techniques are derived from experience in the implantation of nearly 250 patients with various devices since 1979 at House Ear Clinic and Institute (HEI, Los Angeles), and elsewhere. U.S. Food and Drug Administration (FDA) approval was obtained in October 2000 for the multichannel ABI manufactured by Cochlear Limited (Sydney, Australia). Typically, ABI recipients are now being implanted with a 21-electrode ABI. ABIs also have been produced by other implant manufacturers including Med-El, Digisonics, and Advanced Bionics. General technical and theoretical considerations of central auditory implantation and stimulation have been reviewed elsewhere.1, 2
PATIENT SELECTION Patients originally received the ABI under a protocol monitored by the FDA. The criteria for implantation are listed in Table 58-1. The device originally was designed for patients with NF-2 manifesting bilateral vestibular schwannomas, although others with compromised auditory nerves were considered to be eventual ABI candidates. At least 90 per cent of NF-2 patients exhibit bilateral eighth nerve neuromas.3 An unpublished review
of patients with NF-2 seen at the House Ear Clinic revealed that two thirds had bilateral internal auditory canal–cerebellopontine angle (CPA) tumors alone or with one other tumor as the only central nervous system manifestation of their disease. The patients were young (average age, 28 years). With improvements in medical care and surgical techniques, the life span of most of these patients has been significantly prolonged. Restoration of even rudimentary auditory function can enhance their quality of life and ability to function in a hearing world. Our results have shown that the multichannel ABI has the potential of offering even greater benefit. The current NF-2 protocol allows implantation at the time of first-or second-side acoustic neuroma removal or in patients whose tumors have previously been removed. Implantation during removal of the first tumor has allowed experience with the device and may enhance performance when the patient loses all hearing. Also, implantation on the first side gives the patient two chances at obtaining an optimally functioning system should the procedure in the first side not be successful. The management of bilateral acoustic neuromas should be highly individualized.4 Hearing preservation remains an ideal goal in the management of these tumors in patients with NF-2, and early identification and treatment have permitted this in a number of cases. An intact auditory system is highly desirable in preference to an artificial means of restoring hearing. Therefore, preserving as much of the patient’s own hearing as possible is paramount. Patients meeting the criteria listed in Table 58-2 may be considered and observed accordingly. The availability of the ABI provides an alternative to a desperate attempt to preserve nonserviceable hearing when large tumors are removed and hearing conservation is unlikely. The anticipated applications of the ABI and similar devices now has included bilateral temporal bone fractures, cochlear ossification, cochlear nerve avulsion, and anatomical birth anomalies, The ABI also may be beneficial in cases of demyelinating diseases affecting the eighth cranial nerve but sparing at least one cochlear nucleus. 703
704
OTOLOGIC SURGERY
TABLE 58-1 Criteria for Implantation Evidence of bilateral seventh and eighth cranial nerve tumors involving the internal auditory canal or cerebellopontine angle Language competency Age 12 years or older Psychologic suitability Willingness to comply with research follow-up protocol Realistic expectations
PREOPERATIVE EVALUATION AND COUNSELING The goal of implantation is to place a safe and stable device that provides the patient with some degree of environmental sound awareness and recognition and also improves communication in conjunction with lipreading without side effects. Prospective patients are apprised of the goals, limitations, and risks of the ABI during two or three preoperative evaluation and counseling sessions. It is important to impress on the potential candidate that, although the ABI is similar to a cochlear implant, it has provided generally lower levels of performance with more gradual improvement over time. The implant candidate’s expectations are carefully evaluated, and informed consent is obtained. The importance of an experienced multidisciplinary implant team including the neurotologist, neurosurgeon, audiologist, neuro/ auditory physiologist, anatomist, radiologist, and others cannot be overemphasized. Several factors contribute to a successful result from implantation. Experience of team members can greatly influence outcomes. Chief among these factors are the correct identification of the implantation site and the achievement of a stable placement of the electrode array. This, of course, is essential to obtain auditory sensations from stimulation and to optimize performance. The overall results of tumor removal and postoperative recovery also play a role. For example, factors such as eye dryness related to postoperative facial nerve function may affect lipreading ability and communication using the ABI. General health, social activity level, and presence of a support group also can affect ABI use and benefit. Patient expectations are important and may be influenced by publicity about cochlear implants. Assessing expectations for the ABI and ensuring informed consent prior to implantation are highly important but may be complicated in candidates overwhelmed by a plethora of preoperative concerns. A thorough and frank appraisal preoperatively of the potential benefits, limitations, and requirements for adjusting to the device will help increase the likelihood of a satisfied user in the long run. At this time, the ABI requires a certain level of acceptance, motivation, and commitment from the recipient to maximize benefit; therefore, the device may not be for everyone.
TABLE 58-2 Criteria for Observation in Auditory
Brainstem Implant Candidacy in Patients with Neurofibromatosis Type 2
Second tumor in an only-hearing ear Any tumor in a hearing ear that measures >2 cm in the largest diameter (hearing preservation unlikely with removal) Short life expectancy due to other tumors, medical problems, or advanced age Serviceable hearing with a tumor that shows no significant growth by sequential magnetic resonance scans and stable hearing by serial audiograms
DEVICE The ABI hardware has evolved through a number of modifications since the original ball electrode was inserted by Drs. William Hitselberger and William House in 1979.1, 5 Significant design changes have involved transitioning from a percutaneous connector to a transcutaneous coil link to the implant, converting from ribbon electrodes to .7-mm-diameter disk electrodes, and fabrication of a semiflexible silicone electrode carrier (2.5 × 8.5 mm) with a specialized mesh backing to stabilize placement. The first 25 ABI recipients were fitted with a single-channel sound processor and a 2-or 3-electrode array. From 1992 until about 2000, the electrode array used in most patients employed eight platinum disks in a perforated silicone and mesh carrier connected to an implantable receiver/ stimulator. From about 2000 to the present, the 21 electrode Nucleus ABI24 (see Fig. 58-1) has been used. The external device in the 8 electrode patients consisted of a post-auricular microphone, a transcutaneous transmitter coil, and a sound processor (Spectra Model, Cochlear Corporation). Signal-processing strategies have evolved in an effort to improve performance.6 Present patients receive a 21-electrode array (Fig. 58-1A and B) interfaced with the Freedom postauricular sound processor and ear-level or body-worn controller (Cochlear Corporation, Englewood, CO). Since 2003, use of a hybrid ABI array consisting of a version of the present surface electrode and a 10-electrode penetrating array also has been studied in an FDA clinical trials. Implementation of this system in laboratory animals originally demonstrated the capability for improved microstimulation of auditory neurons7 and possibly improved perceptual performance in humans.
ANATOMIC CONSIDERATIONS The target of the ABI electrodes is the cochlear nucleus complex—dorsal and ventral cochlear nuclei. In humans, the cerebellar peduncle that forms the base of the pons covers the auditory nuclei. This means that the nuclei are not visible to the surgeon and must be located from surface
705
Chapter 58 • Auditory Implants for the Central Nervous System
landmarks. Figure 58-2 illustrates the major structures of the pontomedullary junction region with the translabyrinthine approach surgical field of view within the dashed lines. The terminus of the sleeve-like lateral recess forms the foramen
A
B
of Luschka. Just inferior to the foramen is the root of the glossopharyngeal (ninth) nerve. Superior to the foramen lie the root entry and exit zones of the vestibulocochlear and facial nerves. This area is frequently distorted by the tumor. The cochlear nuclei come closest to the surface of the brainstem within the medial and superior aspect of the lateral recess.8, 9 The main target for stimulation is the ventral cochlear nucleus, which forms the main relay for cochlear nerve input and the greater part of the ascending auditory pathway. Placement of the electrodes within the recess has resulted in the fewest side effects and preserves auditory stimulation even though some part of the electrode array lies adjacent to the dorsal cochlear nucleus.2 Also, the disadvantage of lack of exposure is partially offset by positional stability provided to the electrode carrier by the limited space in the lateral recess.
SURGICAL CONSIDERATIONS
FIGURE 58-1. A, Auditory brainstem implant 24 receiver/stimulator with 21-electrode array and remote ball ground electrode. B, Freedom auditory brainstem implant processor with ear-level controller and transmitter coil, and the body-worn controller for longer battery life. (A and B, Courtesy of Cochlear Corporation, Englewood, CO.)
The surgical approach for tumor removal in ABI cases at HEI has been exclusively via translabyrinthine craniotomy (see Chapter 50). The translabyrinthine route
II
III
V
VII VIII
V Foramen of Luschka
IX
VI
Flocculus VII
X
VIII IX
Choroid plexus Tonsil 1. Medial vestibular nuclei 2. Inferior 3. Inferior cerebellar peduncle 4. Dorsal cochlear nuclei 5. Ventral 6. Glossopharyngeal n. 7. Olive 8. Pyramid
4.
X
5.
XI 3.
2.
1.
4. 5. VIII
FIGURE 58-2. Schematic of cochlear nuclei region demonstrating relative location of various landmarks. Dashed area represents approximate surgical view. Electrode is fully inserted into proper position.
706
OTOLOGIC SURGERY
Implant site
Mastoidectomy
Incision
0.5 cm
1.5 cm
2 cm
FIGURE 58-3. Location of incision with respect to planned site of receiver/stimulator.
has been found to provide the most direct access to the lateral recess and surface of the cochlear nuclei.10 Until the actual placement of the device, the surgery proceeds as in any other translabyrinthine acoustic neuroma excision, with the following exceptions. Electrodes are placed for recording electrically evoked auditory brainstem responses (EABRs) and for monitoring cranial nerves VII and IX. The postauricular incision used is shown in Figure 58-3. It is important that the incision not cross the receiver/stimulator package. Electrophysiologic monitoring is performed during implantation to help ensure that the electrode array placement is correct for activating the auditory system, and also to detect possible activation of nonauditory brainstem structures. There may be considerable uncertainty about the correct position for the electrode array when a large tumor has distorted the anatomic landmarks at the brainstem. To aid in placing the electrode array, EABRs are recorded. A repeatable EABR indicates that stimulation of the auditory system is occurring. Intraoperative EABRs obtained with electrical stimulation of
the cochlear nucleus differ considerably from brainstem responses routinely recorded with acoustic stimulation (ABRs) in normal-hearing individuals.11 An experienced electrophysiologist interprets these waveforms at the time of implantation based on data collected from previously implanted patients (Figure 58-4). For recording EABRs, subdermal needle electrodes are inserted at the vertex of the head, over the seventh cervical vertebra in the neck, and at the hairline on the neck prior to the draping of the sterile field. After the receiver/stimulator of the implant has been fastened to the skull and the electrode array has been placed in the brainstem, the transmitter coil is placed over the receiver antenna. The stimuli for evoking responses are biphasic current pulses. Scalp-recorded evoked potentials are sampled and averaged by computer following suitable amplification and filtering. Electrophysiologic monitoring also helps determine the electrode array position that minimizes nonauditory side effects. In addition to monitoring the facial nerve in standard manner,12 bipolar electrodes are inserted in the
Chapter 58 • Auditory Implants for the Central Nervous System
ipsilateral pharyngeal (soft palate) muscles to monitor activation of cranial nerve IX. If the electromyographic recordings reveal activation of nonauditory centers during stimulation through the implant, or if a muscle evoked potential is seen in the averaged waveform, the electrode array is repositioned.
Near CN
Lateral to CN
Return to CN
2 �V Facial nerve
Return to CN
0
ms
10
FIGURE 58-4. Electrically evoked auditory brainstem responses. CN, cochlear nucleus.
707
IMPLANTATION TECHNIQUE Tumor dissection proceeds normally via a translabyrinthine craniotomy. After tumor removal and hemostasis, an area of cortical bone posterior to the mastoid is flattened, and a trough to accept the wires from the electrodes to the receiver/stimulator is created in a manner similar to that of cochlear implantation. Using a replica of the receiver/stimulator as a guide, a circular area of bony cortex posterosuperior to the mastoid defect is drilled with cutting burrs (Fig. 58-5). A specially designed butterfly bit or other cylindrical bits associated with high-speed drills may be employed. Using a replica of the receiver/ stimulator as a guide, the surgeon drills four holes into the bone to accept the tiedown suture. The receiver/ stimulator is fixed with nylon suture prior to electrode array positioning so that the manipulation of the leads does not alter the electrode placement (Fig. 58-6). Because only bipolar cautery may be used after the electrode array is inserted to minimize the risk of current shunting through the device into the brainstem, meticulous hemostasis of the entire wound and CPA is ensured prior to implantation. Anatomic landmarks lead the way to the surface of the cochlear nuclei (ABI placement video clip). Normally intact choroid plexus marks the entrance to the lateral recess (foramen of Luschka), and the taenia obliquely traverses the roof of the lateral recess, marking the surface of the ventral cochlear nucleus. These structures may not be clearly visible, however, when a large tumor has significantly distorted the lateral aspect of the pons and medulla. Following the stump of the eighth cranial
I.A.C. Facial n. V. Co. n.
Choroid plexus
Trough
Drill trough for electrode
FIGURE 58-5. Surgical view of completed translabyrinthine craniotomy, trough for wires, and receiver/stimulator site.
708
OTOLOGIC SURGERY
Temporalis m. Electrode Implant
FIGURE 58-6. ABI receiver/stimulator and electrode array in place.
Facial n.
nerve usually leads to the opening of the lateral recess in these cases. The ninth cranial nerve can also be used as a reference point for the lateral recess. A concavity sometimes visualized between the eighth and ninth nerves should not be confused with the introitus of the recess. The location of the lateral recess may be confirmed by noting the egress of cerebrospinal fluid as the anesthesiologist induces a Valsalva maneuver in the patient. This technique should be reserved as a final check after the opening to the recess has been located by standard landmarks because cerebrospinal fluid will be drained quickly and the advantage of this technique will be lost with multiple Valsalva maneuvers. After identifying the foramen of Luschka, a Rosen needle is used to insert the electrode array into the lateral recess with the electrodes facing superiorly (Fig. 58-7). With experience, we have found that the system functions better, with fewer side effects, when the entire electrode array is placed just inside the lateral recess.2 After placement, selected electrodes in the array are activated to confirm their position over the nucleus. They are tested for the presence of EABRs, stimulation of adjacent cranial nerves (VII and IX), and changes in vital signs. The position of the
Cochleovestibular nerve stump Lower cranial nerves
Electrode Choroid plexus
Ventral VII
VIII Lateral recess Foramen of Luschka
FIGURE 58-7. Surgical view of the electrode array placed into the lateral recess (magnified view of the dashed area in Figure 58-6) and cross-section through the brainstem in the region of the electrode array.
Axial crosssection
Electrode
Dorsal
Chapter 58 • Auditory Implants for the Central Nervous System
electrode array usually needs some adjustment to maximize the EABRs and minimize electromyographic responses from the other nerves. If stimulation of the IX nerve occurs, the electrode is separated from it with teflon felt. The electrode array is secured by a small piece of Teflon felt packed into the meatus of the lateral recess. Fibrous tissue eventually stabilizes the array in position. The wires are positioned in the mastoid cavity and the bony trough previously drilled (Fig. 58-8). Abdominal fat obliterates the mastoid defect. The incision is closed in three layers, and care is taken not to disturb the wires. The wound is not drained routinely. A large mastoidtype dressing is left in place for 4 days.
POSTOPERATIVE CARE The postoperative care after implantation shares many of the features of routine translabyrinthine tumor resections (see Chapter 50). A similar schedule for advancing patient activity and decreasing the level of intensity of nursing care is maintained. A mastoid dressing should remain in place for at least 4 days. Careful attention to any moisture on the bandages allows prompt identification of cerebrospinal fluid leak through the postauricular wound. Intravenous antibiotics are administered prophylactically 1 day preoperatively and continued through the fifth day postoperatively. While we originally attempted testing of ABI recipients’ devices within days after surgery, we no longer do so. Swelling of the skin flap covering the receiver/ stimulator may prevent an adequate signal from reaching the implant and preventing device power-up. Instead, the device is typically activated for the first time about 4 to 8 weeks after implantation.
Temporalis muscle Fat
Implant
FIGURE 58-8. Implant, wires, and fat in place prior to skin closure.
709
Because the magnet is typically removed from the receiver/stimulator so that patients may continue to have magnetic resonance imaging (MRI), there may be difficulty in identifying the location of the antenna of the receiver/stimulator at the time of initial stimulation. If the antenna cannot be located, the transmitter coil may not be properly positioned at initial stimulation, and it may mistakenly appear that the device is nonfunctional or the patient is nonstimulable. While normally it is possible to palpate the scalp for the location of the receiver antenna, this may not be possible in patients with thicker skin. Therefore, consideration should be given to this potential difficulty at the time of implantation and appropriate steps taken (such as possible thinning of the skin, or by marking the antenna location with a tattoo) prior to the initial stimulation session. In actual use, ABI patients must shave this area and apply a thin tape and metal disk to which the magnetic transmitter coil can adhere. The patient, or a companion, must be trained to ensure proper and consistent positioning of the transmitter coil over the implant receiver antenna.
POSTOPERATIVE COMPLICATIONS The most significant complication in the immediate postoperative period is cerebrospinal fluid leak. Unlike routine translabyrinthine surgery, in which the fluid usually takes the nasal route via the eustachian tube, the ABI electrode and wires provide a path along which cerebrospinal fluid can travel beneath the skin flap. We have noted a marked reduction in the rate of leak after transitioning over to the fully implantable receiver from a percutaneous connector used with the early single-channel ABI. Prevention of a leak begins with meticulous dural approximation and packing of the eustachian tube and mastoid cavity with various materials. Although the dural opening cannot be closed in a watertight manner, it should be approximated as closely as possible to minimize the opening. A dumbbell-shaped graft of fat will plug the residual space. Muscle and oxidized cellulose (Surgicel) commonly are employed for eustachian tube closure, and autologous fat works well in the mastoid. Multilayered closure for the wound decreases pathways for cerebrospinal fluid egress. Despite these precautions, patients with the ABI appear more prone to cerebrospinal fluid leak than do those undergoing translabyrinthine procedures without implantation. Leaks from the nose and wound usually respond to reapplication of mastoid pressure dressing and bed rest. A lumbarsubarachnoid cerebrospinal fluid drain is added for persistent leaks. Finally, surgical exploration and repacking of the wound can be employed for leakage unresponsive to more conservative measures. Meningitis can occur either spontaneously or as a result of postoperative cerebrospinal fluid leak. This unusual complication, when identified promptly, res ponds to antibiotics and cessation of the leak.
710
OTOLOGIC SURGERY
Normal healing to a stable implant situation usually takes 4 to 6 weeks, after which initial activation of the device occurs. Multichannel ABI recipients now have experienced up to 15 years of trouble-free use of the device. The first patient ever to be implanted with an ABI in 1979 continues to use her single-channel implant with benefit on a daily basis.
RESULTS Over 200 patients with NF-2 have been implanted with the Nucleus multichannel ABI system at HEI between 1992 and the present. Data from the earliest group of implantees comprised a large portion of a clinical trials submission to the FDA.13 In this chapter, we present results from speech and environmental sounds testing of a large group of patients experienced in using their ABIs. Since performance with the ABI improves more gradually than in cochlear implants, results from experienced users are more representative of longer term benefit. Early results on many of these patients have been presented elsewhere.6 Even 2 years of experience should not be considered sufficient to reach asymptomatic performance. Although improvements are generally greatest during the first year, many patients have continued to improve even after 10 years of use. All patients used the SPEAK (spectral maxima) speech processing strategy.14 Figure 58-9 shows mean scores on a portion of the multichannel ABI perceptual test battery. The lowest
FIGURE 58-9. Mean speech perception test scores of auditory brainstem implant patients with 2 or more years of experience (N = 25). MTS-W, Monosyllable, Trochee, Spondee Word Score; MTS-S, Monosyllable, Trochee, Spondee Stress Score; SERT, Sound Effects Recognition Test; NUCHIPS, Northwestern University Children’s Perception of Speech Test; CID (S), Central Institute for the Deaf Sentence Test in sound only; CUNY (S), CUNY (V), and CUNY (SV), City University of New York Sentence Tests in sound only, vision only, and sound plus vision, respectively.
scores shown are for the CID and City University of New York (CUNY) sentence tests, which are presented in sound only. These tests are difficult for the majority of ABI recipients, and they reflect the generally limited capability of the ABI in the absence of lipreading cues. The CUNY sentence scores in lipreading alone and in sound plus lipreading modes are higher than those administered with sound alone because of the visual cues provided. Obviously lipreading cues are highly important in faceto-face communication, and a large average increase (39 per cent) in sentence recognition occurs when ABI sound is added to lipreading as indicated by the lipreading only and sound plus lipreading CUNY scores. The Monosyllable, Trochee, Spondee (MTS) test, Sound Effects Recognition Test (SERT), and Northwestern University Children’s Perception of Speech (NUCHIPS) test are all “closed-set” tests in which the individual has to select the correct answer from a limited set of options. Therefore, these tests are somewhat easier than “open-set” (essentially unlimited set) tests, but they nevertheless represent a challenging auditory discrimination task to ABI recipients. In the MTS Word (MTS-W) test and the NUCHIPS test the listener selects a word from a set of alternatives. In the case of NUCHIPS, they are rhyming words. In the SERT, the patient selects the correct sound from a set of four pictured alternatives. The highest mean score on all the perceptual tests occurs on the MTS Stress (MTS-S) test, which is a derived score from the MTS-W test. The patient does not have to correctly identify the word—only the correct stress pattern of the word presented, making it one of the easiest tasks. Tests
Chapter 58 • Auditory Implants for the Central Nervous System
such as the MTS-S and SERT provide rather immediate evidence to new ABI recipients that the auditory cues they receive from their implants are indeed useful. With notable exceptions, these results indicate that ABI performance generally has not reached the high levels typically seen with multichannel cochlear implants. At least five patients have shown high levels (≥50 per cent) of open-set speech recognition ability on soundonly sentence tests, and about 16% of our ABI recipients have shown significant (at least 20 per cent correct) ability in this area. There is reason to be hopeful that ABI performance in general will improve. Many ABI recipients experience electrode-specific pitch sensations similar to cochlear implant recipients, and it may be possible to increase these cues with improved microstimulation systems such as the penetrating ABI (PABI). Capitalizing on these cues by carefully assessing auditory percepts from ABI stimulation can be time consuming, but is a necessary part of programming the speech processor to optimize performance.13 Perceptual test scores of ABI recipients presently indicate a significant ability to discriminate many environmental sounds as well as enhance sentence recognition ability over lipreading only. Several of the best performers use the telephone with familiar speakers in controlled conditions. Nevertheless, as is true of hearing aids and cochlear implants, the ABI cannot be expected to be highly beneficial for every potential candidate. Because sound from the device is most effective in combination with lipreading cues, patients with limited vision have generally experienced relatively less communication benefit. Severe visual impairment also greatly complicates the process of accurately testing and programming the ABI sound processor, and special techniques and procedures must be used. Patients with limited social contact may find fewer occasions to use the ABI, which may inhibit their progress with it. Also, we have noted some difficulties with acceptance and use of the ABI by teenagers who may have special cosmetic concerns in addition to problems related to dealing with NF2. Some patients (about 9 per cent in our hands) have not received auditory sensations at all, instead experiencing only mild or moderate nonauditory side effects when their device was activated. Many of these cases were noted to involve anatomic difficulties at the time of implantation. Preoperative MRIs may signal potential problems leading to nonstimulation, such as a large lateral recess or tumor damage to the cochlear nucleus region. Speech processing for brainstem stimulation has profited from research in cochlear implants. Interestingly, similar strategies used in cochlear implants also have worked well with the brainstem implant.6 Flexibility of the ABI programming system has been essential to accommodate anatomic variations and the range of auditory and nonauditory sensations that can result from stimulation. Proper assessment and use of this
711
information in configuring ABI speech processors can have a significant effect on speech perception performance. Poor selection or misalignment of frequency bands to electrode channels in speech processor programming can limit performance.13 Particularly in patients with a greater incidence of nonauditory sensations, experience and flexibility in the clinician’s approach can sometimes mean the difference between use and nonuse of a device. Future studies regarding stimulation rates, frequency assignment of channels, and methods of coding speech cues will contribute to improvements in speech processing strategies for the ABI.
OTHER RESEARCH Niparko and associates15 demonstrated the feasibility of implanting and stimulating within the substance of the cochlear nucleus in guinea pigs. In a related study, el-Kashlan and colleagues16 compared the effectiveness of surface electrodes with those placed into the nucleus. They found lower thresholds and a wider dynamic range in animals with penetrating electrodes than in those with surface placement. These results were influential in the development of the penetrating ABI system that employs an array of needle microelectrodes. McCreery and coworkers7 demonstrated the efficacy of such a system for activating discrete populations of tonotopically tuned neurons within the substance of the cochlear nucleus. This was achieved without significant risk to tissue or blood supply in longer term preparations with properly constructed and inserted electrodes. The needle-type electrodes with a somewhat blunt-tip configuration (Fig. 58-10) were atraumatically inserted on-axis with a specialized spring-powered tool.
Results Using the Penetrating Electrode ABI The PABI was developed in an effort to improve the precision of stimulation of brainstem auditory neurons, and also hopefully to improve speech recognition. In actuality, patients have generally performed best on speech perception tests when using a speech processor program that combines both surface and penetrating electrodes. The two types of electrodes seem to work synergistically, and each offers advantages. Surface electrodes generally create a larger current field that increases the likelihood of activating auditory neurons and ultimately resulting in beneficial hearing sensations. Penetrating electrodes have generally provided auditory sensations at lower current levels (1-2 nC) than surface electrodes, and they have resulted in a wide range of pitch percepts. In comparison with the larger surface electrodes, however, the incidence of failing to achieve auditory sensations has been higher with penetrating electrodes. Also, in some instances, stimulation from penetrating electrodes has
712
OTOLOGIC SURGERY
FIGURE 58-10. Schematic illustration (right) of an early penetrating auditory brainstem implant (PABI) electrodearray and implant site (left) accessing tonotopic gradient of fibers in ventral cochlear nucleus (VCN). DCN, dorsal cochlear nucleus. The present PABI uses 10 penetrating microelectrodes with the depth of the stimulating surfaces ranging from 1 to 2.5 mm (Courtesy of Huntington Medical Research Institute, Pasadena, CA.)
reached the maximum charge limit without achieving a comfortable level of sound. The combination of arrays in the PABI was very valuable to one recipient who did not experience any auditory sensations on his surface electrodes. He did receive auditory sensations on 6 of 10 of his penetrating electrodes, and he is now able to use his PABI with benefit. It is clear that placement of the penetrating electrode array is more critical than the surface electrode array and requires considerable accuracy. A slight deviation (only a mm or so) from the target region can result in no auditory responses on penetrating electrodes. The electrical ABR monitoring that is used intraoperatively to assist with placement of the surface ABI array has been of little use in penetrating electrode placement because the microelectrodes do not typically generate a sufficient neural response for detection by scalp monitoring electrodes. Also, neural response telemetry (NRT), that has been useful in the near-field detection of cochlear nerve action potentials generated by cochlear implants, does not appear to be useful in ABI (or PABI) implantation because of difficulty differentiating auditory from non-auditory action potentials. In postoperative testing with awake patients, NRT waveform morphology often appeared the same regardless of whether patients reported hearing sensations, non-auditory side effects, or no sensations at all.17
Auditory Midbrain Implants (AMI and ICI) It is possible that ABI patients who lose their auditory nerves due to tumors have limited performance with the ABI because of damage to the brainstem region, either due to the tumor or during tumor removal. If this is the case, then it may be advantageous to bypass this damaged region to provide these patients with good speech recog-
nition. Two recent projects have investigated the possibility of prosthetic stimulation of the inferior colliculus (IC), an auditory nucleus in the midbrain. One approach, the auditory midbrain implant (AMI), places a surface electrode on the IC [48]; and the other, the inferior colliculus implant (ICI), uses a penetrating electrode array to access the deep layers of tonotopic organization within the IC [49]. Both projects have implanted patients as of Spring 2007, and the preliminary results (R. Shannon, personal communication) show that auditory sensations can be elicited. Patients hear different pitch sensations on the different electrodes and can use the sounds from the devices to supplement lipreading. At the present time, these patients are not able to understand speech without lipreading, but it is still too early to assess the long-term capabilities of a midbrain implant.
SUMMARY A multiple-electrode array for electrical stimulation can be safely and reliably placed on the brainstem of patients and chronically stimulated to produce useful auditory sensations by selective activation of the cochlear nucleus. Few side effects and minimal morbidity typically characterize the clinical course of patients with these implants. Speech perception performance varies and, with some exceptions, does not reach the high levels typically seen with modern cochlear implants. In combination with lipreading cues, however, ABI sound has proven to be highly beneficial for the majority of device recipients. Further improvements in the hardware (including possibly the PABI, AMI and ICI) and in sound processing methods could give patients improved speech understanding in the future.
Chapter 58 • Auditory Implants for the Central Nervous System
Acknowledgments The authors are grateful to Michael Waring for editorial assistance on electrophysiology, Butch Welch for graphics assistance, and the patients of the House Ear Clinic for their time and effort in laboratory testing.
References 1. Brackmann D E, Hitselberger WE, Nelson R A, et al: Auditory brainstem implant: I. Issues in surgical implanta tion. Otolaryngol Head Neck Surg 108:624-633, 1993. 2. Shannon RV, Fayad J, Moore J K, et al: Auditory brainstem implant: II. Postsurgical issues and performance. Otolaryngol Head Neck Surg 108:634-642, 1993. 3. Riccardi VM : Neurofibromatosis. Neurol Clin North Am 5:337-349, 1987. 4. Briggs R J, Popovic E A, Brackmann D E : Recent advances in the treatment of neurofibromatosis type II. Adv Otolaryngol Head Neck Surg 9:227-245, 1995. 5. Hitselberger N, House WF, Edgerton B S, Whitaker S: Cochlear nucleus implant. Otolaryngol Head Neck Surg 92:52-54, 1984. 6. Otto S R , Shannon RV, Brackmann D E, et al: The multichannel auditory brainstem implant (ABI): Results in 20 patients. Otolaryngol Head Neck Surg 118:291-303, 1998. 7. McCreery DG, Shannon RV, Moore J K, et al: Accessing the tonotopic organization of the ventral cochlear nucleus by intranuclear microstimulation. IEEE Trans Rehabil Eng 4:1-9, 1998. 8. Terr L I, Edgerton B J: Surface topography of the cochlear nuclei in humans: Two-and three-dimensional. Hear Res 17:51-59, 1985.
713
9. Sinha VK, Terr L I, Galey FR , Linthicum FH : Computer-aided threedimensional reconstruction of the cochlear nerve root. Otolaryngol Head Neck Surg 113:651-655, 1987. 10. Monsell E M, McElveen JT, Hitselberger WE, House WF: Surgical approaches to the human cochlear nucleus complex. Am J Otol 8:450-455, 1987. 11. Waring M D: Refractory properties of auditory brainstem responses evoked by electrical stimulation of human cochlear nucleus: Evidence of neural generators. Electroenceph Clin Neurophysiol 108:331-344, 1998. 12. Niparko J K, Kileny PR , Kemink J L , et al: Neurophysiologic intraoperative monitoring: II. Facial nerve function. Am J Otol 10:55-61, 1989. 13. Otto S R , Ebinger K, Staller SJ: Clinical trials with the auditory brainstem implant. In Waltzman S B, Cohen N L (eds): Cochlear Implants. New York, Thieme, 2000, pp 357-365. 14. McDermott H J, McKay C M, Vandali A E : A new portable sound processor for the University of Melbourne/ Nucleus multielectrode cochlear implant. J Acoust Soc Am 91:3367-3371, 1992. 15. Niparko J K, Altschuler R A, Xue X L , et al: Surgical implantation and biocompatibility of central nervous system auditory prostheses. Ann Otol Rhin Laryngol 98:965-970, 1989. 16. El-Kashlan H K, Niparko J K, Altschuler R A, Miller J M : Direct electrical stimulation of the cochlear nucleus: Surface versus penetrating stimulation. Otolaryngol Head Neck Surg 105:533-543, 1991. 17. Otto S R , Waring M D: Kuchta, J: Neural response telemetry and auditory/non-auditory sensations in 15 recipients of auditory brainstem implants. J Am Acad Audiol 16:219227, 2005.
59
Extreme Lateral Infrajugular Transcondylar Approach for Resection of Skull Base Tumors Debara L. Tucci, Thomas M. Pilkington, and Takanori Fukushima
The extreme lateral infrajugular transcondylar (ELITE) approach was born out of the anatomic constraints and operative challenges presented by the cranial base. The transcondylar approach was first devised as an avenue for access to the foramen magnum and ventral medulla.1,2 In 1986, Heros3 described the extreme lateral inferior suboccipital approach for access to vertebral and vertebrobasilar artery lesions. The ELITE approach is a special transcondylar and transjugular tubercle skull base approach developed and elaborated in Germany and Japan in 1987.4,5 After years of refinement, this has become a very useful and well-described approach for this region of the skull base. The standard ELITE approach involves drill resection of the medial and superior medial one third of the occipital condyle and jugular tubercle. This is an excellent approach for vertebral artery and posterior inferior cerebellar artery aneurysms; vertebrobasilar junction aneurysms; and more complex foramen magnum, infrajugular, and anterior brainstem tumors. An anterolateral approach modification includes addition of a high cervical exposure, which is ideal for large glomus jugulare tumors. Other modifications include a limited transcondylar approach for some smaller vertebral artery and posterior inferior cerebellar artery lesions, and an extensive ELITE approach involving partial resection of the occipital condyle and resection of the C1 arch for access to chordomas, chondrosarcomas, and high cervical spine lesions. These modifications are beyond the scope of this chapter. Paragangliomas of the temporal bone were first identified as a unique entity in 1945.6 Attempts at characterization and ultimate surgical resection were initially fraught with high morbidity and mortality because of their location amidst complex anatomic structures. Primary treatment with radiation therapy soon became the mainstay for therapy with the hopes of limiting complications. Radiation therapy has been shown to be superior to observation of these lesions; however, the long-term risks of malignancy and radiation morbidity limit its success.7-10 Definitive surgical resection remains the mainstay for attempted cure.
Attempts at surgical resection and cure led to the development of tumor classification strategies to help guide surgical planning and to improve communication of results. Two main classification schemes were devised and remain in use today, primarily in the otology/neurotology literature—the Fisch11 and Glasscock-Jackson classifications.12 The Fisch scheme is as follows: Type A: Tumors limited to the middle ear cleft Type B: Tumors limited to the tympanomastoid area Type C: Tumors involving the infralabyrinthine region Type D1: Tumors with an intracranial extension less than 2 cm in diameter Type D2: Tumors with an intracranial extension greater than 2 cm in diameter The Glasscock-Jackson scheme is divided for tympanicum and jugulare tumors. The following glomus jugulare scheme includes a superscript for degree of intracranial extent (e.g., a paraganglioma type IV2.0 is a lesion with 2 cm of intracranial extension): Type I: Small tumor involving the jugular bulb, middle ear, and mastoid process Type II: Tumor extending under the internal auditory canal; may have intracranial extension Type III: Tumor extending into the petrous apex; may have intracranial extension Type IV: Tumor extending beyond the petrous apex into clivus or infratemporal fossa; may have intracranial extension Glomus jugulare tumors with advanced stage (Fisch class C and D, Glasscock-Jackson class II-IV) typically are treated surgically with a classic Fisch infratemporal fossa approach (types A or B). Key to surgical treatment of glomus jugulare tumors is safe access to surgical margins while preserving vital nerves and vessels. Classic approaches described have surgical morbidity, including facial nerve dysfunction owing to translocation, sacrifice of the external auditory canal, hearing loss, anterior dislocation of the mandible to gain access to the petrous 715
716
OTOLOGIC SURGERY
portion of the carotid artery, and cerebrospinal fluid (CSF) leak. Additionally, some tumors with intracranial extent can require a two-stage procedure to minimize complications. The authors have used a simpler classification of tumors in this location, and modifications and additions to the ELITE approach—including transsigmoid, trans jugular, infrajugular, and high cervical approaches—are tailored to provide needed exposure for tumor resection (Fig. 59-1): Type A: Main tumor mass is intradural—uses primarily an ELITE approach Type B: Main mass is combined intradural and intrajugular —uses transjugular and ELITE approaches Type C: Main mass is intradural, intrajugular, and high cervical, typical of extensive glomus jugulare tumor— uses ELITE, transsigmoid, transjugular, infrajugular, and high cervical exposure The ELITE approach provides a one-stage trans jugular posterior infratemporal fossa approach to resect complex glomus jugulare tumors and other lesions of the jugular foramen. This method can avoid rerouting of the facial nerve, and may preserve the external auditory canal and hearing function. The petrous portion of the carotid is accessible without anterior dislocation of the mandible. Two modifications of this approach are described. The dorsolateral ELITE approach is used for intradural, type A, tumors involving the vertebral artery, posterior inferior cerebellar artery aneurysms, foramen magnum meningiomas, and glomus jugulare tumors with a large intradural component. The anterolateral ELITE approach is used for type B and C tumors, including extensive glomus jugulare tumors. The approach can be broken down into a series of maneuvers, as follows: 1. Postauricular infratemporal incision 2. Retrolabyrinthine mastoidectomy 3. High cervical exposure 4. Skeletonization and (if needed) anterior translocation of the facial nerve 5. Lateral suboccipital and transcondylar-transtubercular exposure 6. Removal of the internal jugular vein (IJV), jugular bulb, and sigmoid sinus 7. Intradural exposure
SURGICAL PROCEDURE A preoperative arteriogram and embolization are performed 48 hours before the surgical procedure.13 Facial nerve monitoring is carried out in all cases. The surgeon may consider additional monitoring, including auditory brainstem evoked potentials, somatosensory evoked potentials, motor evoked potentials, and monitoring of
CN X, XI, and XII. An electromyographic endotracheal tube can be used for CN X monitoring, and electrodes placed directly into the sternocleidomastoid muscle and the tongue can be used for CN XI and XII monitoring. Before the incision, perioperative antibiotics and corticosteroids are administered. Patient positioning and incision differ for the dorsolateral and anterolateral modifications. For the dorsolateral ELITE procedure, the patient is placed in a lateral decubitus position (Fig. 59-2A and B). For the anterolateral approach, the patient is positioned in a supine fashion with the head displaced away from the side of the lesion. A shoulder roll is placed to aid with the high cervical portion of the procedure and to raise the shoulder on the side of the lesion. Obese patients and patients with very short necks may be placed in a lateral position if necessary. The patient should be padded and secured for ease of rotation of the table throughout the case. Incisions are shown in Figure 59-2C. For the dorsolateral approach, a lazy S incision is used, 1 to 2 cm posterior to the mastoid bone, and extending inferiorly along the hairline. For the anterolateral ELITE procedure, a retroauricular curvilinear C-shaped, or question mark– shaped, skin incision is begun approximately 2 to 3 cm posterior to the upper border of the ear. Inferiorly, this incision is carried down into the neck, traversing the border of the sternocleidomastoid muscle (SCM) and running parallel to the body of the mandible, approximately two fingerbreadths below. The skin flaps are raised. For the dorsolateral approach, the SCM is retracted anteriorly; for the anterolateral approach, the SCM is retracted posteriorly. The inferior edge is raised in the subplatysmal plane. Superiorly, the temporoparietal fascia is elevated as a separate flap under the skin flap, after it is dissected off the bone with the periosteum. The anterior flap is now reflected, and the posterior auricular muscle should be visible behind the external auditory canal. Inferiorly, the greater auricular nerve should be seen crossing the SCM approximately 3 cm below the mastoid tip. This nerve can be preserved for use as an interposition graft, if needed. The posterior flap is displaced, and the posterolateral neck muscles are reflected to expose the mastoid more fully. This group of muscles is composed of three layers (Fig. 59-3). The superficial layer is composed of the SCM and splenius capitis muscles. The middle layer consists of the longissimus capitis and semispinalis capitis. These muscles are reflected to expose the deep layer, which consists of the rectus capitis posterior major, the obliquus capitis superior, and the obliquus capitis inferior muscles. These muscles form the suboccipital triangle (Fig. 59-4A). The styloid diaphragm also is identified in this region. Beneath that diaphragm runs the occipital artery as it courses beneath the posterior belly of the digastric muscle (Fig. 59-4B). A mastoidectomy is performed. The sigmoid sinus and jugular bulb must be thoroughly skeletonized. Mastoid cells should be removed to visualize the middle fossa
Chapter 59 • Extreme Lateral Infrajugular Transcondylar Approach for Resection of Skull Base Tumors
717
Type A Intradural Tumor A. B. C. D. E. F.
V VII VIII
F. E.
JB– Jugular bulb Ce.–Cerebellum Vv.– Vagal vein Va.– Vertebral a. Ba.– Basilar a.
AICA Ba. IX
Ce.
Atlas C1 sup. articular facet Brainstem Occipital condyle Dura Choroid plexus Flocculus
X XI Vv. XII
PICA
JB
C.
Va. A. V3 Type B “Dumbbell” Shaped Tumor
Type C Tumor with High Cervical Extension
V 7, 8
F
VII
V3
VII
C1
IX XII IX
C2
C.C. XI
J.V.
XII
C1 X XI
X C2
FIGURE 59-1. Intradural tumor, type A (top). Tumor mass intradural and transjugular, type B (bottom left). Tumor mass intradural, transjugular, and high cervical, type C (bottom right).
dura, presigmoid dura, superior petrosal sinus, and retrosigmoid dura. The digastric ridge must be widely exposed and identified. The bony labyrinth should be skeletonized for ease of retrolabyrinthine dissection. The facial nerve
should be skeletonized. The retrofacial air tract must be drilled to expose the jugular bulb completely. The bone between the sigmoid sinus and the jugular bulb can be quite adherent to these structures, and the dissection is prone to bleeding.
718
OTOLOGIC SURGERY Ear, mastoid highest
Vertex down
A
B
Asterion
2
1
V3 SCM
C1 C2 C3 C4
C FIGURE 59-2. A and B, Patient positioning for dorsolateral ELITE procedure, lateral decubitus position. C, Incisions. 1 shows lazy S incision for dorsolateral approach; 2 shows question mark–shaped incision for anterolateral ELITE approach. SCM, sternocleidomastoid muscle.
For resection of glomus jugulare tumors, a high cervical approach is necessary to identify the lower cranial nerves (CN IX-XII), and to gain lower control of the internal carotid artery and IJV. When the subplatysmal flaps are raised, the posterior angle of the mandible is identified as the anterior limit of dissection. The mastoid tip is defined as the posterior limit. Dissection is carried anterior to the SCM to identify the posterior belly of the digastric muscle. The stylomastoid diaphragm is removed, and the occipital
artery, running just below, is controlled. The digastric muscle can be displaced superoanteriorly to cover the facial nerve and gain access to the transverse process of C1. The accessory nerve can be seen coursing posteroinferiorly from the lateral point (defined as 3 to 15 mm inferolateral to the transverse process of C1). The accessory nerve is usually seen running lateral to the IJV, medial to the digastric. The hypoglossal nerve is identified as it courses over the IJV. The carotid sheath is opened to identify the vagus nerve. The stylohyoid, stylopharyngeus, and styloglossus muscles lie anterior to the carotid artery. Detaching the styloid proc ess from the base of the cranium improves exposure of the carotid canal at the skull base. The glossopharyngeal nerve is identified 1 to 2 cm from this point, crossing the carotid inferomedially. Gently retracting the IJV can help expose the carotid branch of CN IX. The facial nerve is skeletonized in its vertical portion, and the mastoid tip is removed. The styloid proc ess (already detached) is reflected anteriorly to expose the infratemporal carotid artery. Facial nerve rerouting is rarely performed. Instead, a limited anterior translocation (approximately 5 mm) may be performed to gain anterior exposure in cases of tumor extension to the internal carotid artery (C7 portion). Preserving the periosteum of the facial nerve can minimize the risk of postoperative paresis with this procedure. The carotid canal is drilled out to expose the petrous carotid. CN IX exits dorsal to the carotid in this location and can be injured without proper care. Jacobsen’s nerve can be seen branching from CN IX before reaching the jugular bulb. It is essential to identify the V3 segment of the vertebral artery to avoid injury during this exposure. V3 can be identified in one of three ways: by dissection of the suboccipital triangle, by identification of the vertebral sulcus or J groove of the C1 lamina, or by use of a Doppler probe (Fig. 59-5). For identification of the suboccipital triangle, the deep layer of muscles mentioned previously serves as the guide to this anatomic region. The rectus capitis posterior major inserts superiorly on the nuchal line and attaches inferiorly on the spinous process of C2. The obliquus capitis inferior muscle inserts superiorly on the transverse processes of C1 and inferiorly on the spinous process of C2. The obliquus capitis superior attaches superiorly at the temporo-occipital suture and inferiorly on the transverse process of C1. The suboccipital triangle is opened by detaching the superior and inferior oblique muscles off of the transverse process of C1 and reflecting them posteriorly. The rectus capitis posterior major is reflected posteriorly as well, after freeing it from the inferior nuchal line; this helps identify the C1 lamina and the vertebral artery. Alternatively, the vertebral artery can also be identified by following the C1 lamina from the posterior point along its superior edge until the vertebral sulcus or J groove is identified. The vertebral artery is encased by a venous plexus as it exits the foramen transversarium of C1 and then travels in the vertebral sulcus until its medial
Chapter 59 • Extreme Lateral Infrajugular Transcondylar Approach for Resection of Skull Base Tumors
719
Posterolateral approach
SNL
M.
6. 5.
INL
M.
FM 4.
G.O.N. SCM
1.
C1
2. 3. C2
Splenius capitis m.
C3
Greater occipital n.
A
B
1. C1 transverse process 2. I.J.V. 3. Levator scapula m. 4. Longissimus m. 5. Splenius capitis m. 6. S.C.M.
FIGURE 59-3. A, Superficial dissection. B, Middle layer. FM, foramen magnum; GON, greater occipital nerve; IJV, internal jugular vein; INL, inferior nuchal line; M, mastoid process; SCM, sternocleidomastoid muscle; SNL, superior nuchal line.
turn into the atlanto-occipital membrane. The venous plexus can be avoided by dissecting in the subperiosteal plane as the vertebral artery is freed from the sulcus. Venous plexus bleeding can be controlled by using bipolar cautery or absorbable knitted fabric (Surgicel). Care should be taken to avoid injuring an extradural posterior spinal artery or the posterior inferior cerebellar artery. The atlanto-occipital membrane is incised to expose the craniocervical dura. The lateral suboccipital craniotomy is performed with a craniotome. The boundaries of the craniotomy are about 3 cm posterior to the retrosigmoid line, superiorly to the nuchal line (transverse sinus), inferiorly to the posterior rim of the foramen magnum, and lateral to the occipital condyle. Exposure of the jugular bulb and sigmoid sinus is crucial. The posterior condylar emissary vein is encountered as it exits the jugular bulb and can be controlled by bipolar cautery and Surgicel packing. In most cases, the lateral edge of the foramen magnum needs to be totally removed. The posterior and medial one third of the occipital condyle is removed in front of the C1 dura, using a highspeed diamond drill. Resecting more than half of the condyle can cause instability of the craniovertebral junction, and stabilization should be considered on resection. As the
condyle is removed, the surgeon proceeds through a cortical layer leading to a cancellous bony layer. Venous bleeding may be present and should be controlled with bone wax and Surgicel before proceeding. The next cortical layer overlies the hypoglossal canal. The hypoglossal nerve, branch of the ascending pharyngeal artery, and venous plexus of the canal are located within the canal. Because the canal sits superior to the condylar facet and inferior to the jugular tubercle and is directed slightly cephalad at a 60 degree anterolateral plane, skeletonization of the canal to its lateral extent indicates a one third resection of the superomedial portion of the occipital condyle. Removal of this portion of the condyle provides easier access to the ventral foramen magnum (Fig. 59-6A and B). Dissection continues superiorly toward the jugular tubercle. The jugular tubercle is a smooth osseous protrusion on either side of the occipital bone.14 The base of the tubercle is defined by the hypoglossal canal, and it extends anteriorly 20 mm toward the clivus. Reduction of the tubercle permits access to the anterior clivus. Because the tubercle lies medial to the jugular foramen, CN IX, X, and XI are at risk as the tubercle is drilled out. Risk of injury can be minimized by drilling within a core of the tubercle and working anteriorly, deeper toward the clivus, minimizing heat and potential stretch injury to
720
OTOLOGIC SURGERY
5.
6.
6.
6.
5.
5. 14.
5.
13. 4.
12.
9. M.
9.
10.
7.
11.
15. 11. 8.
1. 10.
8.
4.
3.
4.
3.
4. 7. 1. 2.
2.
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.
A
C1 transverse process I.J.V. Levator scapulae m. Longissimus m. Splenius capitis m. S.C.M. Great occipital n. Inferior obique n. Superior oblique m. C1 lamina V3 & C1 nerve root in subocciptal triangle
6.
12. 13. 14. 15. 16. 17. 18. 19. 20. 21.
Semispinalis m. Rectus capitis major m. Rectus capitis minor m. Posterior atlanto-occipital membrane C1 condyle Occipital condyle Condylar fossa Foramen magnum C1 root C2 root
6.
5.
9.
9.
13.
14. 19.
18.
15.
V3
13, 14. 20.
10.
7.
17.
21.
1. V4
2.
8. C2
B FIGURE 59-4. A, Deep muscle layer, including suboccipital triangle. B, Dissection deep to muscle. IJV, internal jugular vein; M, mastoid process; SCM, sternocleidomastoid muscle.
721
Chapter 59 • Extreme Lateral Infrajugular Transcondylar Approach for Resection of Skull Base Tumors
. S.S
C.
Dural incision
M.
Inferior retrosigmoid point
17. V3 20.
21.
Fibrous ring of VA
A Craniotomy/mastoidectomy site
C1 transverse process V3
B
10.
1.
C1 posterior point
V4
J Groove
C FIGURE 59-5. A-C, Identification of V3 segment of vertebral artery (VA), including use of J groove in dissection. SS, sigmoid sinus.
the nerves. Drilling should be carried 20 to 25 mm deep toward the clival junction. A 3 mm smooth diamond burr is ideal here, and can then be sized down to a 2 mm burr. Remaining thin bone can be dissected; inner cortical bone can be preserved to avoid damage to CN IX and X. It is important to have flat access from the lateral edge of the foramen magnum under CN IX, X, and XI to the inferior clivus and ventral medullary area (Fig. 59-6C-E). Between the jugular bulb and the C1 transverse process, the small rectus lateralis muscle must be removed. Occasionally, the C1 transverse process is shaved to expose the V3-V4 genu for wider exposure. In this process, care should be taken not to injure CN XI. Adequate exposure of the sigmoid sinus, jugular bulb and IJV, and all lower cranial nerves should allow for preservation of vital functions and easy identification of the tumor within these structures. Figure 59-7 shows the exposure of these structures provided by this approach. The approach used for resection of large glomus jugulare tumors includes (1) the suprajugular, retrofacial, infralabyrinthine
approach; (2) the infrajugular-transcondylar approach; and (3) the high cervical approach. Before tumor resections, it is important at this point to control the arterial feeders to the tumor. The IJV is then ligated inferiorly. The sigmoid is occluded above the tumor mass. The IJV lateral wall can be incised, and the tumor specimen can be resected up to the jugular bulb. The medial wall of the jugular bulb is preserved, and the tumor is excised from the pars nervosa. The dural sheath over the lower cranial nerves should be left intact. Bleeding should be controlled with gentle packing because bipolar cautery can damage the cranial nerves. The intradural portion of the tumor is addressed next. The dural incision is curvilinear and is placed several millimeters posterior to the sigmoid sinus. It is carried inferiorly to the point where the vertebral artery pierces the dura. The dura is reflected back with multiple tacking sutures. With sufficient drilling of the occipital condyle and jugular tubercle, the accessory nerve is readily identifiable. Dissection can expose the medullary
722
OTOLOGIC SURGERY
S.S.
J.B. O.C. C1
T.T. – Tubercular triangle C.T. – Condylar triangle H.C. – Hypoglossal canal and C.N. XII Cl – Condyle (atlas) O.C. – Occipital condyle S.O.D. – Suboccipital dura
V3
A
S.O.D.
S.S.
J.B.
T.T. C.T. O.C.
H.C.
B.S.
S.S.– Sigmoid sinus B.S.– Brainstem J.B.– Jugular bulb
C1 Dura
V3
B
C1 groove
FIGURE 59-6. A and B, Dissection of occipital condyle. C-E, Further dissection toward foramen magnum and clivus, with exposure of ventral medullary area.
branch of the accessory nerve, the vagal vein and vagus nerve, and the glossopharyngeal nerve. The tumor is dissected free from the cranial nerves after the jugular foramen is inspected for further extension. Blood supply from the posterior inferior cerebellar artery can be cauterized (Fig. 59-8).
The lower clivus can be accessed, if necessary, by inferior translocation of the lower cranial nerves and anterior translocation of the vertical petrous segment of the internal carotid artery. The dural sleeve over the lower cranial nerves can protect them from trauma. Resection of the clivus can continue toward the petrous apex.
Chapter 59 • Extreme Lateral Infrajugular Transcondylar Approach for Resection of Skull Base Tumors Removal of Jugular Tubercle
Ba. T. 9, 10, 11
A
7, 8 F
C D
Va.
J.T. XII
J.F.
O.C.
S.S.
E
C1 Condyle B Synovial capsule C1
D
C
J.F.– Jugular foramen J.T.– Jugular tubercle O.C.– Occipital condyle
A–B : JT length 1.2–3.0 (1.65) cm C–D : JT height 0.7–1.7 (1.05) cm E–F : JT thickness 0.2–1.0 (0.61) cm Ba.– Basilar a. Va.– Vertebral a. Removal of bone surrounding the jugular bulb
MFD S. S.
PFD VII IX
E FIGURE 59-6, cont’d.
C. C. J.V.
V3 1 C
XI
XII X
723
724
OTOLOGIC SURGERY
The wound is irrigated, and reconstruction of the cranial base is performed to obtain a watertight dural closure. Options include fascia for smaller defects, a pericranial flap, or a free flap reconstruction for larger defects. The use of vascularized myofascial flaps for reconstruction is thought to reduce the incidence of postoperative CSF leak.15 Abdominal fat should be harvested to fill the mastoid defect. Lumbar drainage is used in most cases.
COMPLICATIONS Complications of cranial base surgery for tumors of the jugular foramen are well described. Several series have reported common complications, including CSF leak and wound infection, with an incidence of approximately 5% to 10% for each case.16,17 Facial nerve and hearing deficits are common, and outcomes vary significantly depending on preoperative function, approach, and tumor extent. Lower cranial nerve deficit, often producing aspiration and voice disorder, was the most common surgical complication in one large series, at almost 10%.15 Other complications include cerebrovascular accident, meningitis, pulmonary embolus, hydrocephalus, and death. These more severe complications are rare, occurring in about 1% of cases.16,17 Craniovertebral instability is a concern with over-resection of the condyle; however, with preservation of at least 50% of the condyle, stabilization should be unnecessary.18
RESULTS Liu and colleagues18 reported surgical outcomes from resection of glomus jugulare tumors in 30 patients using the ELITE procedure. Mean follow-up was 28.5 months. Total resection was achieved in 73%, and near-total resection was achieved in 20% of patients. Eight patients (27%) had worsened facial nerve palsy postoperatively, and four of those patients maintained a permanent deficit. Two patients with permanent facial paralysis underwent partial anterior transposition of CN VII, and two required an interposition cable graft because of tumor invasion of the nerve. Four patients developed sensorineural hearing loss; all were patients with significant intradural tumor. Cochlear function was thought to be compromised secondary to any of various factors, including traction on the cerebellum or CN VIII, or violation of the cochlea during bone dissection or tumor resection. Of patients, 30% had worsened CN IX and X function, six with permanent deficit, and three patients went on to laryngoplasty to improve swallowing and voice function. There were five CN XI and three CN XII pareses. There were no CSF
V VII, VIII
Surgical approaches to the jugular foramen D. AICA branch IX
F. S.P.S.
Ba.
E. Ce.
S.S.
X XI Vv. XII
1. VII
ICA
PICA
2.
C.
Va.
V3 C1
J.B.
A. V3
3. C1 C.C.
I.J.V.
FIGURE
59-7. Exposure
afforded by various approaches. 1, suprajugular, retrofacial, infralabyrinthine approach; 2, infrajugulartranscondylar approach; 3, high cervical approach.
A. B. C. D. E. F.
Atlas C1 sup. articular facet Brainstem Occipital condyle Dura Choroid plexus Flocculus
FIGURE 59-8. ELITE exposure including high cervical dissection. AICA, anterior inferior cerebellar artery; Ba, basilar artery; Ce, cerebellum; JB, jugular bulb; PICA, posterior inferior cerebellar artery; Va, vertebral artery; Vv, vagal vein.
Chapter 59 • Extreme Lateral Infrajugular Transcondylar Approach for Resection of Skull Base Tumors
leaks and no operative mortality. No patients required stabilization of the occipitocervical junction. One advantage of the ELITE surgical approach is tumor access with minimal manipulation of the facial nerve. It is recognized that long facial nerve rerouting results in interruption of the deep petrosal artery, leading to perigeniculate ischemia. There is also a chance of mechanical trauma with mobilization of the nerve from the narrow labyrinthine segment and manipulation of the epineurium in the internal auditory canal.19 The key to preserving facial function is minimal dissection, and translocation only as necessitated by the tumor and the approach. Preservation of the periosteum of the bone surrounding the facial nerve can help to preserve the stylomastoid artery, which supplies the mastoid segment of the facial nerve. We describe a select anterior translocation only as necessary to gain access to the vertical C7 (infratemporal) portion of the carotid artery. Although the approaches used by many surgeons require facial nerve rerouting, more conservative facial nerve management has been advocated, at least for some tumors without extensive anterior involvement. Pensak and Jackler20 described the fallopian bridge technique, and reported usage in a large percentage of glomus jugulare tumors in their practices. In this technique, the facial nerve is skeletonized and left in situ, and the external auditory canal is not violated. This approach cannot be used when there is anterior extension of tumor to the carotid genu, but is not contraindicated by the presence of an intradural tumor component. They report outcomes in 35 patients using this technique, with a House-Brackmann grade I result in 92% at 6 months after surgery. Results of glomus jugulare tumor resection using the ELITE approach may be compared with reports of series using other approaches. Pareschi and coworkers21 described their experience with 52 glomus jugulare tumors. Of those tumors, 42 were at least Fisch classification type C, and 37 underwent surgery, with an overall resection rate of 96%. Overall, 33 patients required an infratemporal fossa approach with 5 requiring A and B types. Two patients required second-stage petro-occipital transsigmoid approach. The external and middle ear was sacrificed in all cases. Facial nerve function at 1 year was House-Brackmann grade I or II in 51% of patients, and grade III or better in 84%. The facial nerve was transposed in 70% of patients, and preserved anatomically in 80%. The glossopharyngeal and vagus nerves were preserved anatomically in 48% and 35%. The authors noted that most patients have some difficulty with dysphagia, but there were no reported pneumonias. Twelve patients required vocal cord medialization. CSF leaks were not reported; however, three patients developed a subcutaneous CSF collection that was managed conservatively. Jackson and colleagues22 reported results for 182 lateral skull base surgeries for glomus jugulare tumors. Of those patients, 21% were class I, 20% were class II, 35% were class III, and 23% were class IV. Total surgical
725
control was reported as 85%. Their series included 66 patients with intracranial extent, 44 of which had pars nervosa involvement requiring resection of CN IX through XII. Jackson and colleagues22 noted that one or more new cranial nerve deficits arose in 59% of patients. Complete tumor removal was possible without any cranial nerve resection in 31%. CSF leak was reported in 3 of 66 cases after 1987, when complex reconstruction of the cranial base began in their practice. The Otology Group16 described their management of the facial nerve in lateral skull base surgery for benign tumors. Their outcomes are reported as good (HouseBrackmann grade I-II), fair (House-Brackmann grade III-IV), and poor (House-Brackmann grade V-VI). The patients in whom the nerve was left in situ had normal function. Procedures with short translocation preserved good function in 92%, while procedures with long translocation preserved good function in 66%. Extreme translocation resulted in good function in only 38% of patients. Patients with resected and grafted nerves had much poorer results, as would be expected. Oghalai and associates23 reported on results of surgical treatment using a transjugular craniotomy for resection of a series of 28 jugular foramen tumors with intracranial extent, including 6 glomus jugulare tumors. A fallopian bridge technique of facial nerve dissection, which leaves the vertical portion of the facial nerve in situ, was used when possible. Of the 24 patients with normal or near-normal preoperative facial nerve function, function was preserved (House-Brackmann grade I-II) postoperatively in 22. Patients who did not require facial nerve translocation did not experience deterioration of function. Tumor access required ear canal closure in a few (21%) cases. Of patients, 60% had good preoperative hearing levels (defined as American Academy of Otolaryngology–Head and Neck Surgery class A or B), and 41% had good hearing postoperatively. All nine patients who were spared destructive otologic procedures (ear canal closure, translabyrinthine or transcochlear procedures) maintained good hearing postoperatively. A 53% incidence of complete lower cranial nerve palsy was identified in the patient population preoperatively, and new lower cranial nerve deficits were noted in 27%. In the series reported by Fisch and colleagues and cited by Moe and colleagues,17 132 patients with glomus jugulare tumors were reviewed. Most (n = 119) of the patients had advanced tumors (class C and D); 81 had intradural extension. Total tumor removal was reported in 83% of the total population. Of patients, 84% required an advanced infratemporal fossa approach, 72% required an anterior displacement of the facial nerve, and approximately 15% required facial nerve resection, with either grafting or anastomosis. Results with anterior displacement were favorable, with good function (House-Brackmann grade I-II) retained in 88% of patients. Preservation of the lower cranial nerves for class C and D tumors varied depending on class and nerve. CN VIII and IX were most at risk and had
726
OTOLOGIC SURGERY
reservation rates between 20% and 50%. CN X, XI, and p XII fared better, with preservation rates of 60% to 80%. In the reported follow-up, 10% of patients had aspiration. Ten percent were noted to have CSF leaks postoperatively.
DISCUSSION The ELITE approach to glomus jugulare tumors and other tumors of the jugular foramen provides safe, effective access to the cranial base. It is ideal for tumors around the jugular foramen, lower clivus, and high cervical region. Resection results and postoperative function compares favorably with other large series in the literature as shown previously. Good facial nerve function is important for quality of life, and injury to the facial nerve carries significant morbidity. Classically, large glomus jugulare tumor resection required rerouting of the facial nerve and possibly resection of the nerve to complete surgical resection. Numerous techniques have been described to translocate the facial nerve.19 The ELITE approach may use a slight anterior translocation if needed to gain access to the petrous portion of the carotid artery. Anterior dislocation of the mandible can produce symptoms of trismus, pain, malocclusion, and first bite syndrome.16 The transcondylar transtubercular approach allows the surgeon to work posterior to the angle of the mandible without displacement, and the petrous portion of the carotid artery can still be accessed without the corresponding morbidity. In a similar manner, this exposure prevents violation of the external ear canal, sparing the patient a conductive hearing loss. Access to the lower clivus is difficult to achieve. It is necessary to displace the petrous portion of the carotid anteriorly and translocate the lower cranial nerves inferiorly. Lower cranial nerve morbidity is common during glomus jugulare surgery because these nerves are intimately related to the tumor.24 Preserving the dural covering of the lower cranial nerves as they are displaced can help minimize this morbidity. The ELITE approach is a versatile one-stage approach to difficult tumors of the jugular foramen. Surgical results are favorable compared with results described in the literature. It is a useful tool in the armamentarium of the neurotologic surgeon when planning management of these complex tumors.
REFERENCES 1. Seeger W: Atlas of Topographical Anatomy of the Brain and Surrounding Structures. Vienna, Springer-Verlag, 1978. 2. Bertalanffy H, Seeger W: The dorsolateral, suboccipital, transcondylar approach to the lower clivus and anterior portion of the craniocervical junction. Neurosurgery 29:815-821, 1991. 3. Heros RC : Lateral suboccipital approach for vertebral and vertebrobasilar artery lesions. J Neurosurg 64:559562, 1986.
4. Fukushima T, Day J D: Manual of Skull Base Dissection. Pittsburgh, AF-Neurovideo Inc, 1996. 5. Fukushima T, Sameshima T, Friedman A H (eds): Exercise 10. In Manual of Skull Base Dissection. 2nd ed. R aleigh, NC, AF-Neurovideo, 2004. 6. Rosenwasser H : Carotid body-like tumor of the middle ear and mastoid bone. Arch Otolaryngol 41:64-67, 1945. 7. Gottfried ON, Liu JK, Couldwell WT: Comparison of radiosurgery and conventional surgery for the treatment of glomus jugulare tumors. Neurosurg Focus 17:E4, 2004. 8. Pollock B E : Stereotactic radiosurgery in patients with glomus jugulare tumors. Neurosurg Focus 17:E10, 2004. 9. Lim M, Gibbs IC, Adler J R Jr, Chang S D: Efficacy and safety of stereotactic radiosurgery for glomus jugulare t umors. Neurosurg Focus 17:E11, 2004. 10. Carrasco V, Rosenman J: Radiation therapy of glomus jugulare tumors. Laryngoscope 103(11 Pt 2 Suppl 60) 23-27, 1993. 11. Fisch U, Mattox D: Classification of glomus temporale tumors. In Microsurgery of the Skull Base. New York, Thieme Medical Publishers, 1988, pp 149-153. 12. Jackson CG: Glomus tympanicum and glomus jugulare tumors. Otolaryngol Clin 34:5, 2001. 13. LaRouere M J, Zappia JJ, Wilner H I, et al: Selective embolization of glomus jugulare tumors. Skull Base Surg 4:21-25, 1994. 14. Mintelis A, Sameshima T, Bulsara K R , et al: Jugular t ubercle: Morphometric analysis and surgical significance. J Neurosurg 105:753-757, 2006. 15. Ramina R , Maniglia JJ, Fernandes YV, et al: Tumors of the jugular foramen: Diagnosis and management. Operative Neurosurg 57(1 Suppl):59-68, 2005. 16. Manolidis S, Jackson G, Von Doersten PG, et al: Lateral skull base surgery. The Otology Group experience. Skull Base Surg 7:129-137, 1997. 17. Moe K S, Li D, Linder TE, et al: An update on the surgical treatment of temporal bone paraganglioma. Skull Base Surg 9:185-194, 1999. 18. Liu J K, Sameshima T, Couldwell WT, Fukushima T: The combined transmastoid retro- and infralabyrinthine transjugular transcondylar transtubercular high cervical approach for resection of glomus jugulare tumors. Operative Neurosurg 59:115-125, 2006. 19. Parhizkar N, Hiltzik D H, Selesnick S H : Facial nerve rerouting in skull base surgery. Otolaryngol Clin North Am 38:685-710, 2005. 20. Pensak M L , Jackler R K : Removal of jugular foramen t umors: The fallopian bridge technique. Otolaryngol Head Neck Surg 117:586-591, 1997. 21. Pareschi R , Righini S, Destito D, et al: Surgery of glomus jugulare tumors. Skull Base Surg 13:149-157, 2003. 22. Jackson CG, McGrew B M, Forest J A, et al: Lateral skull base surgery for glomus tumors: Long-term control. Otol Neurotol 22(3):377-382, 2001. 23. Oghalai J S, Leung M K, Jackler R K, McDermott MW: Transjugular craniotomy for the management of jugular foramen tumors with intracranial extension. Otol Neurotol 25:570-579, 2004. 24. Sen C, Hauge K, Kacchara R , et al: Jugular foramen: Microscopic anatomic features and implications for neural preservation with reference to glomus tumors involving the temporal bone. Neurosurgery 48:838-848, 2001.
60
Management of Postoperative Cerebrospinal Fluid Leaks Derald E. Brackmann, Grayson K. Rodgers, and Eric P. Wilkinson Videos corresponding to this chapter are available online at www.expertconsult.com.
Egress of cerebrospinal fluid (CSF) from the subarachnoid space into surgical wounds may result in leakage of CSF from the wound, the ear canal (if the tympanic membrane is not intact), or the nose (via the eustachian tube). CSF follows the path of least resistance. Any procedure that encounters the subarachnoid space can be complicated by a postoperative CSF leak. These leaks result from failure to obtain watertight dural closure or from an inadequate seal of dural defects. CSF leaks are a concern because the defect provides a potential portal of entry for infection to seed the leptomeninges. Meningitis in this setting is accompanied by significant morbidity and mortality. CSF leaks should be corrected promptly to avoid more serious complications. This chapter describes the various techniques for treating postoperative CSF leaks.
PRESSURE DRESSING For cases in which an abdominal fat graft has been used during closure to plug dural defects, a pressure dressing can be applied to control CSF leaks. The pressure dressing works by pushing the fat back into the dural defect, sealing off the subarachnoid space. In cases in which fat has been used, most leaks stop with a pressure dressing. If a subcutaneous CSF collection (pseudomeningocele) is present, it may be aspirated before the placement of the pressure dressing. If performed, aspiration should be done with sterile technique. Cases involving incisional wound leaks may be treated by a simple suture overclosure of the leaking portion of the incision and a pressure dressing.
Technique The dressing is applied in similar manner to the initial postoperative dressing. First, a vertical gathering tie is placed in the temporal fossa. Dressing sponges (4 × 4) are folded in half and placed directly over the fat graft and in the postauricular sulcus to support the auricle (Fig. 60-1A). Next, fluffed Kerlix is placed over the 4 × 4 sponges and the auricle (Fig. 60-1B). A tight wrap of
roller gauze is applied (Fig. 60-1C). The direction of the wrap should be from the ear toward the occiput, which ensures that the auricle is not damaged by anterior folding. Also, the vertical tie must be as lateral as possible in the temporal fossa to avoid a pressure point on the forehead. Pressure necrosis of forehead skin can develop easily if attention is not given to this point. The last layer of this dressing is a 3-inch elastic bandage, which provides the final compression (Fig. 60-1D). The elastic bandage should be wrapped firmly, but patient comfort must be accommodated. Usually, the last several turns can be altered to adjust the exact amount of compression. Ideally, a pressure dressing should be left in place for 4 or 5 days, which allows time for healing of the CSF leak site to occur. A minimum of 48 hours without leakage must pass before the dressing is removed. In addition to the pressure dressing, other conservative actions should be undertaken. These measures all are directed at decreasing CSF pressure. Straining (Valsalva maneuver) is strictly avoided, and the patient is kept at bed rest with the head elevated 45 degrees. Limited activity, such as bathroom privileges or brief periods of sitting up in a chair, is at the surgeon’s discretion. Stool softeners and cough suppressants can be used. Acetazolamide (Diamox) may be given to decrease CSF production.
LUMBAR DRAIN The next step in CSF leak treatment is a lumbar subarachnoid spinal fluid drain. By removing CSF from the system at a site away from the dural defect, CSF pressure is decreased, and healing can occur. Some surgeons routinely place a lumbar drain at the time of surgery to assist with intraoperative CSF removal and decompression in the postoperative period.
Technique To place the catheter, a lumbar puncture is performed at the L4-L5 level. The patient is placed in either a sitting or a lateral decubitus position. The spinous processes 727
728
OTOLOGIC SURGERY
A
B
C
D
FIGURE 60-1. A, Folded 4 × 4 dressing sponges placed over fat graft site after translabyrinthine acoustic tumor removal. Sponges are also placed in the postauricular crease to support the auricle. B, Fluffed Kerlix in place over 4 × 4 sponges. C, Appearance of dressing after roller gauze wrap. Tape can be applied to secure the position of the dressing; this is especially helpful on the forehead, where the dressing may slide inferiorly onto the brow. Note laterally placed vertical gathering tie. D, Ace wrap in place. Tape can be helpful to stabilize the elastic dressing.
are palpated, and L4-L5 is identified at the level of the iliac crest. The patient is asked to flex the back and bring the knees and chin to the chest. A prepared catheter kit contains all the necessary supplies for preparing, draping, and anesthetizing the site. A 14 gauge Tuohy needle is introduced in the midline with a slightly superior angle between the spinous processes (Fig. 60-2). The obturator is removed at intervals so that the surgeon can look for a flow of CSF. When a good flow of CSF is established, the epidural catheter is introduced through the needle and threaded into the epidural space. The opened side of the bevel of the needle faces the patient’s
left or right side on penetrating the spinous ligaments and arachnoid. Before the catheter is threaded, the bevel is turned to open superiorly, facilitating directing the catheter cephalad. With a flow of CSF established, the needle is withdrawn, and the connector is placed on the end of the catheter so that a connection to intravenous tubing can be made. An empty intravenous fluid bag is attached to the tubing, and all connections are secured with tape (Fig 60-3A). Several methods of regulating CSF output exist. This regulation is important because, if CSF is removed too rapidly, tension pneumocephalus and brain herniation
Chapter 60 • Management of Postoperative Cerebrospinal Fluid Leaks
A
729
B
FIGURE 60-2. A, Tuohy needles used for lumbar puncture. B, Close-up of end of Tuohy needle.
A
B
FIGURE 60-3. A, Site of lumbar puncture at L4-L5 with epidural catheter secured in place by Tegaderm. This provides an occlusive seal over the catheter site and allows for easy inspection of the site. B, IMED pump set up to withdraw cerebrospinal fluid from the patient and into the c ollection bag.
can occur.1-3 Alternatively, if an inadequate amount of CSF is removed, the purpose of the drain is defeated. Humans produce approximately 18 mL of CSF per hour. The natural mechanisms of CSF absorption reduce some CSF; the lumbar drain should remove less than 18 mL of CSF per hour. Common neurosurgical practice is to order 50 mL of CSF removed every 8 hours. Gravity is the standard method of effecting CSF drainage. The collection bag is placed at the level of the patient’s heart or below to create increasing output of CSF as needed. This method requires extremely attentive nursing care to ensure the proper amount of drainage. Drains placed to respond to gravity have highly variable output, and a change in patient position may significantly increase or decrease flow. Establishing an even flow of CSF is difficult, and usually a bolus of fluid is removed, after which the drain is clamped. The system is at risk of occluding during these clamped periods. If flow becomes impaired, removing the amount of CSF ordered becomes impossible. To avoid some of these problems, we place the intravenous tubing in reverse direction through an intravenous infusion pump. The pump is set at a rate of 6 to
10 mL/hr, and a slow, controlled, continuous flow of CSF is withdrawn from the patient (Fig. 60-3B). Because this system is not gravity dependent, patients may move about without fear of a rapid discharge of CSF. Better patency of the catheter system is maintained because the flow is never stopped. If any occlusion of the system or air in the line is sensed, an alarm is sounded as the pump stops. Other physicians have employed flow-regulated systems,4,5 but the system described here is a definite improvement over those described. We have employed this system in a large series of patients and have found it vastly superior to gravity drainage. While the lumbar drain is in place, the patient must be closely observed for signs of infection. The temperature, surgical wound, lumbar drain site, and white blood cell count must be monitored. Samples of CSF may be examined at any time. Headache during CSF drainage is common, but meningismus should not be present. Prophylactic antibiotics are controversial, but are administered by some surgeons. As with a pressure dressing, the lumbar drain should be left in place for 4 or 5 days. We often shut the pump off for the last 24 hours so that if the leak recurs, the drainage can be resumed.
730
OTOLOGIC SURGERY
Chapter 60 • Management of Postoperative Cerebrospinal Fluid Leaks
731
FIGURE 60-4. A, Postauricular approach and transection of ear canal skin and dissection of cartilage from canal skin. B, Everting stitches placed in superior and inferior canal skin. These sutures are grasped with a hemostat placed through the canal and then pulled through, everting the canal skin. C, Everted canal skin is oversewn. FIGURE 60-5. A, Mastoid periosteal flap is developed for second layer of closure and is pedicled just posterior to meatus. B, Flap is rotated anteriorly and secured with absorbable sutures.
FIGURE 60-6. Middle fossa exposure with removal of tegmen tympani and bony eustachian tube (ET) roof. Cochlea, vestibular labyrinth, facial nerve, and internal auditory canal are also shown. FIGURE 60-7. A and B, Obliteration of bony eustachian tube with bone wax, bone pâté, and muscle. Inset (B) is a cross section of eustachian tube (ET) showing relationship of internal carotid artery (ICA), tensor tympani muscle, and greater superficial petrosal nerve.
WOUND EXPLORATION AND RECLOSURE When leaks do not resolve within 48 hours with a pressure dressing or lumbar drain, wound exploration with repacking of the abdominal fat and reclosure is indicated. In wounds in which a dural defect was packed with fat, dislodging of the fat from the defect may cause a CSF leak. When the wound is explored, the fat is removed and repacked (with additional fat as required) into the defect, as described in Chapter 49.6 Some surgeons employ autologous fibrin glue as an adjunct to their closures.7,8 This material adds an additional seal to supplement the fat graft or, in other cases, a muscle or pericranial flap. Collagen matrix (DuraGen, Integra city) may also be used to augment the dural closure. A search should be made for opened mastoid air cells that can serve as pathways for CSF. These air cell tracts can be occluded with bone wax. In cases of severe hearing loss, the eustachian tube should be packed off with absorbable knitted fabric (Surgicel) and temporalis muscle after removal of the incus and cutting of the tensor tympani tendon. Although a return to the operating room seems aggressive and is accompanied by the small risks brought by further surgery and anesthesia, this course of action has provided the most expedient control of CSF leaks that fail conservative therapy. The expeditious closure of a leak plays an important part in the prevention of infection. The longer a leak remains open, the greater the chance that meningitis has to develop. In a review of CSF leaks after translabyrinthine acoustic tumor removal at the House Ear Clinic, no statistically significant association between postoperative CSF leaks and the development of meningitis was found.9 One explanation for this is that leaks are aggressively treated and stopped quickly, limiting the time available for contamination to occur.
REFRACTORY CEREBROSPINAL FLUID LEAKS The techniques discussed earlier used separately or in combination stop most CSF leaks. Rarely, a leak is refractory to these measures. These leaks involve CSF that tracks through temporal bone cell tracts, finds its way to the middle ear, and subsequently discharges from the
nose (eustachian tube) or the ear canal (tympanic membrane not intact). With an intact tympanic membrane, the eustachian tube is the final common pathway of most leaks. Although many methods involving repair of the source of the leakage are described, perhaps an easier and less risky approach is to block the final pathway. Our management of these difficult leaks depends on the status of the patient’s hearing and the tympanic membrane. If no serviceable hearing exists, or if the tympanic membrane is not intact, blind sac closure of the ear canal and obliteration of the middle ear and eustachian tube are performed. If the tympanic membrane is intact, and the patient has serviceable hearing, middle fossa closure of the eustachian tube is the procedure of choice.
Technique Ear Canal Closure with Eustachian Tube and Middle Ear Obliteration In cases with no residual hearing, blind sac closure is the technique of choice. Blind sac closure of the ear canal is accomplished through a postauricular incision with transection of the ear canal. The canal skin is separated from the cartilage of the lateral canal and everted through the external auditory meatus. This everted canal skin is oversewn with a nonabsorbable suture that is left in place for 10 days postoperatively (Fig. 60-4). A flap of mastoid periosteum is developed on a pedicle just posterior to the external auditory canal. This flap is rotated anteriorly and secured as a second layer of closure for the meatus (Fig. 60-5). Next, all canal skin, the tympanic membrane, the malleus, and the incus are removed. To ensure complete removal of squamous epithelium and to enlarge the canal, a canalplasty is performed with a cutting burr. The eustachian tube is curetted and packed with Surgicel, temporalis muscle, and bone wax. Before wound closure, the middle ear and remaining canal are packed with additional muscle. A pressure dressing is placed for 4 postoperative days; a lumbar spinal fluid drain can also be used in the initial postoperative period. In cases of CSF leak after middle fossa craniotomy in which the tympanic membrane is not intact, one closure option is blind sac closure of the ear canal followed by placement of a bone-anchored cochlear stimulator
732
OTOLOGIC SURGERY
(Baha implant, Cochlear). Baha placement in this instance would overcome the maximal conductive hearing loss induced by a blind sac closure.
Middle Fossa Obliteration of the Eustachian Tube In cases in which the patient has good hearing, and the tympanic membrane is intact, closure of the eustachian tube can be accomplished via the middle fossa.10 This procedure is approached as a middle fossa craniotomy, as outlined in Chapter 48. The bone flap is removed, and the dura is elevated from posterior to anterior. The arcuate eminence, greater superficial petrosal nerve, and middle meningeal artery are identified, and the middle fossa retractor is placed. A diamond burr is used to remove bone from the tegmen tympani and to expose the head of the malleus and the body of the incus (Fig. 60-6). Identification of the tensor tympani tendon and the cochleariform process allows identification of the eustachian tube, which is directly anterior. Also, the eustachian tube is just lateral to the greater superficial petrosal nerve. The bony eustachian tube is unroofed, and the mucosa is carefully curetted from the tube. The surgeon must be aware that the internal carotid artery can be dehiscent in the medial or inferior eustachian tube. The tube is packed first with bone wax, then bone pâté, and finally temporalis muscle (Fig. 60-7). A split-thickness piece of bone from the bone flap is placed over the tegmen defect to avoid fixation of the ossicles against the dura. The wound is closed in the usual manner. In this setting, a pressure dressing is not likely to be helpful, but a lumbar drain could be placed for several postoperative days.
New Strategies More recent strategies to reduce the rate of postoperative CSF leak after skull base surgery include the use of various materials for cranioplasty. At the House Ear Clinic, titanium mesh cranioplasty is used to secure the fat graft into the mastoid. In a series of 324 patients undergoing titanium mesh cranioplasty after translabyrinthine acoustic neuroma removal, the CSF leak rate was 3.3%.11 Hydroxyapatite compounds such as Bone-Source (Leibinger) have also been employed. A consecutive series of 108 translabyrinthine acoustic tumors at Pittsburgh Ear Associates showed a CSF leak in 3.7% of the 54 defects closed with hydroxyapatite cement versus 12% of the 54 cases closed with abdominal fat alone.12 In that series, a small quantity of abdominal fat was placed over the dural defect, and the mastoid cavity was filled with the hydroxyapatite cement. The ultimate role of these materials will be influenced by success rates, long-term costs, and ease of use.
SUMMARY Postoperative CSF leak is a potential complication of any cranial base surgical procedure that violates the meninges. Although a CSF leak by itself is not a problem, it provides a portal of entry for bacteria to seed the meninges. Because postoperative meningitis is a serious complication, CSF leaks should be treated aggressively. This chapter discussed the methods for closing CSF leaks in a sequential manner from conservative to aggressive. Using this approach, the skull base surgeon should be able to seal all CSF leaks.
REFERENCES 1. Snow R B, Kuhel W, Martin S B : Prolonged spinal drainage after the resection of tumors of the skull base: A cautionary note. Neurosurgery 28:880-883, 1991. 2. Effron M Z, Black FO, Burns D: Tension pneumocephalus complicating the treatment of postoperative CSF otorrhea. Arch Otolaryngol Head Neck Surg 107:579-580, 1981. 3. Graf C J, Gross C E, Beck DW: Complication of spinal drainage in the management of cerebrospinal fluid fistula: Report of three cases. J Neurosurg 54:392-395, 1981. 4. Swanson S E, Kocan M J, Chandler WF: Flow-regulated continuous spinal drainage: Technical note with case report. Neurosurgery 9:163-165, 1981. 5. Swanson S E, Chandler WF, Kocan M J: Flow-regulated continuous spinal drainage in the management of cerebrospinal fluid fistulas. Laryngoscope 95:104-106, 1985. 6. House J L , Hitselberger WE, House WF: Wound closure and cerebrospinal fluid leak after translabyrinthine surgery. Am J Otol 4:126-128, 1982. 7. Epstein G H, Weisman R A, Zwillenberg S, Schreiber A : A new autologous fibrinogen-based adhesive for otologic surgery. Ann Otol Rhinol Laryngol 95:40-45, 1986. 8. Sierra D H, Nissen A J, Welch J: The use of fibrin glue in intracranial procedures: Preliminary results. Laryngoscope 100:360-363, 1990. 9. Rodgers G K, Luxford WM : Factors affecting the development of cerebrospinal fluid leak after translabyrinthine acoustic tumor surgery. Laryngoscope 103:959-962, 1993. 10. Friedman R A, Cullen R D, Ulis J, Brackmann D E : Management of cerebrospinal fluid leaks after acoustic tumor removal. Neurosurgery 61:35-39, 2007. 11. Fayad J N, Schwartz M S, Slattery WH, Brackmann D E : Prevention and treatment of cerebrospinal fluid leak after translabyrinthine acoustic tumor removal. Otol Neurotol 28:387-390, 2007. 12. Arriaga M A, Chen D A : Hydroxyapatite cement cranioplasty in translabyrinthine acoustic neuroma surgery. Otolaryngol Head Neck Surg 126:512-517, 2002.
61
Care of the Eye in Facial Paralysis Robert E. Levine
Rehabilitation of a patient with facial paralysis depends on restoration of optimal lid position and function.1-3 This chapter summarizes techniques that I have found to be most helpful in achieving that goal, based on more than 5000 patients with facial paralysis on whom I have operated during the past 3 decades. The chapter is divided into sections for lid reanimation procedures, lower lid reapposition procedures, and ancillary procedures. In practice, a combination of these techniques may be performed during the same operation. When the procedures are combined, the reanimation procedure is performed first because it is most influenced by the lid swelling that occurs during the course of the surgery. The lower lid reapposition procedure is performed next. Upper lid entropion correction is performed just before the upper lid incision is closed, and brow elevation is performed last. The final section of the chapter explains two useful temporizing procedures.
CRITERIA FOR SURGERY The following three groups of patients with facial paralysis require lid surgery for functional reasons: 1. Patients who are either symptomatic or who show signs of conjunctival or corneal injury, or both, despite maximum tolerated medical therapy. 2. Patients who require rapid ocular rehabilitation to resume their usual occupation and responsibilities. Keeping the eye full of ointment might protect the cornea adequately, but would not be a realistic option for a monocular patient or one who works as a pilot. 3. Patients whose ocular status is currently stable, but who are at high risk of corneal complications. Patients with diminished corneal sensitivity or totally anesthetic corneas secondary to associated CN V involvement are the prime candidates in this group. If fifth and seventh cranial nerve deficits are present, even minimal lagophthalmos is a risk factor for corneal breakdown. Poor Bell’s phenomenon or the absence of tears may complicate the picture further.
Patients with short-term problems (≤3 months to anticipated recovery of orbicularis oculi function) can usually be treated by conservative means. Patients with significant paralytic deficits who require 6 months or more to recover, or who are not expected to recover, are generally best served by early surgical intervention. The group of patients whose prognosis is unclear, such as patients who might improve in 3 months, but could conceivably require 6 months or more to recover, poses the greatest challenges in surgical selection. In such patients, criteria such as the reliability of follow-up; the accessibility of medical care; the ability of the patient or family to care for the eye; and the patient’s own needs, desires, and lifestyle all play a role in decision making. When it is safe and feasible, a prolonged trial of conservative management may allow the patient and the physician to decide if they are on the correct course.
PREPARATION OF THE PATIENT The surgery is preferably done on an eye (or head and neck) gurney. Two advantages of this approach are ease of access to the eye area by the surgeon and the assistant, and ability to crank up the bed so that the patient is brought to the seated position, enabling the brow and lids to be checked with gravity operative. A doughnut is used to stabilize the head, and a nasal cannula with air is added to provide adequate circulation under the drapes. Some anesthesiologists prefer a carbon dioxide exhaust line as well. Air is used routinely instead of oxygen to eliminate the possibility of an accident resulting from the mixture of oxygen and cautery. If oxygen is required at any time, cautery is withheld until after the oxygen has been turned off.
ANESTHESIA AND SURGICAL PREPARATION In lid reanimation procedures, in which the patient’s cooperation is required, any medication that might make the patient drowsy and unable to cooperate fully 733
734
OTOLOGIC SURGERY
t hroughout the operation is not used. Short-acting intravenous medication, such as propofol (Diprivan), methohexital sodium (Brevital Sodium), or a similar agent is given at the beginning of the surgery in amounts just adequate to cover the discomfort of the local injection. Lidocaine 2% (Xylocaine) with epinephrine (unless contraindicated by hypertension or cardiac problems), to which sodium bicarbonate 7.5% (Neutra-Caine) has been added (1 part sodium bicarbonate to 9 parts lidocaine with epinephrine) is used at the beginning of the surgery. Bupivacaine 0.5% (Marcaine) is used at the end of the surgery to reduce pain during the immediate postoperative period. It may also be included in the initial injection, using equal parts of bupivacaine and lidocaine with epinephrine. Local infiltration is placed in the areas to be operated, such as along the upper lid fold and along the lateral orbital rim for spring implantations, and at the canthi and brow areas, if surgery is to be performed there. Excessive infiltration should be avoided because it paralyzes the levator, impairs extraocular motility (making it harder to judge lid position), and distorts lid anatomy. The eyelids,
both sides of the face above the mouth, and the forehead are prepared with green soap and then with povidoneiodine (Betadine), which is washed off.
DRAPING The hair is covered with a small drape formed into a turban and secured with a clamp. A second small sheet is incorporated with that drape to cover the superior end of the table. A body sheet is also placed. The eyelids and brow are isolated by means of two No. 1000 Steri-Drapes cut in half (Fig. 61-1). Each of the four drapes forms a border of the surgical field, which includes both eyes and the forehead area. Before the Steri-Drape is placed over the nose, a thin cloth towel is placed over the nose to avoid the suffocating feeling resulting from plastic over the nose. In addition, the plastic drape over the towel inferior to the adhesive area is excised to prevent moisture accumulation. Care is taken to avoid distorting the lower lid anatomy or brow areas by undue traction from the drapes.
FIGURE 61-1. Draping the patient. After hair has been covered with a turban, which is secured with a towel clip, two No. 1000 Steri-Drapes are cut in half. The first half-drape is placed with the edge as close to the hairline as possible so as not to interfere with judgment of brow position. A thin cloth towel is placed over the nose, and a half-drape is placed to secure it into position, with the sticky end of the drape bridging the skin and the superior end of the towel. The nonsticky portion of the drape is cut away and discarded, leaving only the towel over the nose. If the drape is not cut away, excess moisture builds up. A third half-drape is placed laterally, as far lateral to the lateral canthus as is practical, and a fourth half-drape is placed similarly on the contralateral side.
Chapter 61 • Care of the Eye in Facial Paralysis
The body drape is fastened to the head drape on both sides with a towel clip so that it does not slip down when the patient is brought to the seated position. The patient is secured on the table with a safety belt, and the belt is positioned to provide access for loosening it as needed when the patient is brought to the seated position. The Mayo stand is brought over the drapes for easy access to the instruments. Although the tent effect obtainable by draping over the Mayo stand would be desirable, it does not lend itself readily to moving the stand away when the patient needs to be brought to the seated position.
GENERAL CONSIDERATIONS Procedures are generally performed on an outpatient basis, unless the patient is already hospitalized because of the neurotologic or head and neck surgery. Bipolar or unipolar cautery may be used for hemostasis. Unipolar cautery should not be used if the patient has an auditory brainsterm implant (ABI) or cochlear implant. Cutting cautery with a fine needle tip is useful for performing the lid dissection. The eye is constantly protected with a scleral shell. At the end of the procedure, antibiotic ophthalmic ointment is applied to the wounds. Ophthalmic ointment is used because it is not irritating if any gets into the eye. The lids are not bandaged. An ice pack is applied to the closed lids and kept in place for 48 hours, after which time warm tap water compresses are used for at least 20 minutes four times a day until the swelling subsides. The antibiotic ointment is applied to the wounds twice a day until they are healed, and appropriate lubricating drops are prescribed for the eye.
UPPER LID REANIMATION PROCEDURES The four procedures that I believe are the most useful in reanimating paralyzed lids are the following, in order of preference: 1. Enhanced palpebral spring implantation 2. Palpebral spring implantation 3. Gold weight implantation 4. Silicone rod prosthesis implantation In all of these procedures, the principle is to create an external force that opposes the levator palpebrae superioris, the opening muscle of the lid. The respective forces are spring tension, gravity acting on the gold weight, and elasticity of the silicone rod. The relative merits and limitations of the various procedures are related to how they develop external closing forces and the consequences of increasing those forces. The greater the force required by any device, the greater the pseudoptosis (lid droop in the primary position of gaze that results from the implant). If the palpebral spring, the gold weight, and the silicone rod prosthesis all are adjusted to provide the same closing force in a given patient, the
735
pseudoptosis should be the same with each device. In the enhanced palpebral spring implantation procedure, the levator muscle is strengthened to balance the spring force, and less pseudoptosis is possible with the same closing force compared with any of the other three procedures. Similarly, by tightening the levator and using a stronger spring force, the surgeon makes increased blink speed possible. Because the gold weight is gravity dependent, very large and unsightly gold weights may be required in lids that need a strong closing force. Also because the gold weight is gravity dependent, lid closure may not be guaranteed when the patient is supine, as in sleep. Blink speed is limited with the gold weight. Patients whose work or lifestyle requires their being in a very cold or hot environment may experience pain from the cold or hot gold weight in their eyelid. A surgeon who does these procedures only infrequently can more easily master the techniques of the gold weight implant and silicone rod prosthesis, however, than the technique of the palpebral spring implant or the enhanced palpebral spring implant. The silicone rod prosthesis has the advantage of providing support for the lower lid as well. Its major disadvantage is its inevitable loss of elasticity over months or years. If facial nerve function does not recover before the prosthesis runs out of elasticity, it will need to be replaced. By contrast, a gold weight or palpebral spring can function over many years. A small percentage of springs may fail over time because of fatigue and breakage, necessitating replacement. An additional advantage of the palpebral spring is that the tension on the wire can be adjusted postoperatively, either externally or through a small incision. This technique allows loosening of the spring when the patient recovers partial function, but is not yet well enough to have the spring removed. All prosthetic devices are subject to the potential hazards of extrusion or infection over the long-term. With the techniques currently in use, these complications have been sufficiently infrequent, however, as not to limit the usefulness of the devices. In recent years, since I devised the enhanced palpebral spring procedure, I have used it with increasing frequency instead of nonenhanced spring implantation. The extra surgical effort is usually well rewarded by the diminished pseudoptosis and increased blink speed obtained by this procedure. I prefer the enhanced palpebral spring implantation in most cases of significant upper lid closure deficits. When the patient has a strong levator or when the spring is being used only as a short-term remedy, the nonenhanced palpebral spring procedure may be used. Surgeons who are just beginning to undertake palpebral spring implantation should start off with the nonenhanced procedure and then move on to the enhanced procedure. I find the gold weight beneficial to patients whose closure problem is minimal, but nevertheless just exceeds the limits of conservative management. Patients who require definitive, reliable closure, such as patients with coexistent poor Bell’s phenomenon or CN V involvement, are
736
OTOLOGIC SURGERY
Chapter 61 • Care of the Eye in Facial Paralysis
737
FIGURE 61-2. Palpebral spring implantation. A, With protective scleral shell in place, incision is made along the lateral two thirds of the lid crease and is carried across the orbital rim laterally. Dissection is carried downward at the medial end of the incision to expose tarsal plate. Dissection is also carried upward and laterally to expose orbital rim. B, A 22 gauge blunted spinal needle with the stilette in place is passed from the medial end of the dissection to emerge laterally in the plane between orbicularis and the tarsus. Passage should be carried out overlying the midtarsus, and needle is angulated slightly downward at its lateral extent. Exit of the needle tract should be close to lateral orbital rim periosteum. The lid is everted to confirm that needle has not inadvertently perforated the tarsus. The previously prepared wire spring, sterilized in low-temperature chemical sterilizer or gas sterilizer, is passed through the needle, and needle is withdrawn. C, Cross section of lid illustrates placement of needle over the midtarsus in the plane between tarsus and orbicularis. Wire spring should be resting on the epitarsal surface. D, Scleral shell is removed, and fulcrum of the spring is brought into the desired position along the orbital rim. The spring should be placed in a position in which its curves conform perfectly to the eyelid contour. (Inset: Fulcrum of spring is secured to lateral orbital rim periosteum with three 4-0 Mersilene sutures, and an extra bite of the periosteum is taken with each stitch.) Loops are fashioned at each end, and the spring is cut to size. Loops should be flat and tightly closed to leave no sharp edges. The medial loop is enveloped in 0.2 mm thick polyester (Dacron) patch material, to which it is secured by means of three 8-0 nylon sutures tied internally. Polyester patch is creased in Gelfoam press before surgery and is autoclaved with the other instruments. The folded polyester envelope is cut to size at surgery. The crease in the patch material should be directed downward so that the spring and patch together provide a smooth inferior surface. The loop at the end of the inferior arm is directed upward for the same reason. Suturing of the loop to the polyester is facilitated by resting the polyester on a retractor. E, The end of the spring with its polyester envelope is replaced into the lid between tarsus and orbicularis. In time, the end of the spring becomes fixed to tarsus by granulation tissue integrating into the polyester patch. Securing the patch to the tarsus directly with an additional running 8-0 nylon suture helps to provide fixation until connective tissue grows into the polyester. Tension on the spring is checked, with the patient in the upright and supine positions. The tension can be adjusted by grasping the upper end of the spring with forceps and changing its position. When the correct tension has been determined, the upper loop of the spring is secured to the orbital rim periosteum with a 4-0 Mersilene suture. An extra bite of the periosteum may be taken in the stitch before it is tied. When sutures are placed to secure either the fulcrum or the upper loop of the spring to the orbital rim periosteum, it is safer to sew in the direction away from the globe. Spring tension is checked again with the patient seated and supine. Additional adjustments can be made by bending the wire or repositioning the loop. When the adjustments are completed, two additional 4-0 Mersilene sutures are placed through the upper loop in a manner similar to that of the initial suture. Deeper tissues overlying the spring are closed with 5-0 plain gut suture to ensure that the spring and Mersilene sutures are well covered. Skin and muscle are closed with running 6-0 plain gut fast-absorbing suture. F, The end of the spring should be at the pupillary axis, with eyes in the primary position of gaze. (From Tse D, Wright KW [eds]: Oculoplastic Surgery. Philadelphia, Lippincott, 1992.)
better protected with springs than with weights. Patients whose ocular management failed with weights in place have been successfully treated by removing the weights and replacing them with springs. The silicone elastomer (Silastic) elastic prosthesis is most useful in patients with an excellent prognosis for recovery in about 6 months in whom significant lower lid lagophthalmos coexists because stretching of the prosthesis over time may decrease its function after 6 months.
Palpebral Spring Implantation The palpebral spring4-14 is built preoperatively either in the office or at the bedside. Building the spring is timeconsuming, and the surgeon and the patient should be comfortable during the procedure. Good light must be available, and if possible, the patient should be seated so that lid movement can be best evaluated. Each spring is constructed from a plain piece of wire that is shaped to conform to the individual lid anatomy of the patient. Generally, a 0.011 inch wire provides suitable tension for most patients. Patients with very strong levators may require the use of 0.011 inch or 0.012 inch wire, and patients with weak levators (especially when the levator is not going to be tightened) may benefit from the use of 0.009 inch or 0.008 inch wire. A new wire used in pacemaker leads, the alloy 35NLT (Fort Wayne Metals), has been made with sufficient tensile strength to replace the previous alloy. With this wire, the 0.011 inch wire diameter often does need to be varied. Because of its track record in biomechanical applications, it is anticipated to have excellent longevity.
The construction is begun by forming an 8 mm loop at what is to become the fulcrum of the spring. The posterior aspect of that loop should be the superior arm of the spring. Because loosening the spring intraoperatively is easier than tightening it, the two arms should form an angle of about 120 degrees as they leave the fulcrum. The fulcrum is placed over the lateral orbital rim and held in position by the surgeon’s fingers. Curves are created in the lower arm to match the patient’s lid anatomy. Curvature is also provided to accommodate the fact that the upper eyelid opens up and back, not straight up and down. Slight variations in spring position and curvatures may enhance its effect; these factors should be varied in the evaluation of the spring preoperatively. Sometimes, making more than one spring with slightly different curvatures is useful in determining which model would work best. Usually, the fulcrum should be placed as far laterally as possible without lengthening the spring so much that its design and placement are difficult. The completed spring is stored until the day of surgery, when it is placed on a gauze pad to prevent loss and sterilized in a low-temperature chemical unit and not a steam autoclave, to subject the wire to less heat stress. Alternatively, the wire may be sterilized in gas sterilizer 24 hours or more before surgery, and retained in a sterile envelope.
Surgical Technique The eye is protected with a scleral shell. An incision is made along the lid fold at the junction of the medial one third and lateral two thirds of the palpebral aperture and is carried across the orbital rim (Fig. 61-2). Dissection is
738
OTOLOGIC SURGERY
carried superolaterally, in the plane between the septum and orbicularis, to expose the orbital rim. Dissection is carried downward at the medial aspect of the incision to expose the tarsus. A blunted 22 gauge spinal needle is passed, beginning in the area of the exposed tarsus, 5 mm superior to the lid margin. The needle passes in the plane between the orbicularis and the tarsus to a point 2 mm above the lid margin at the lateral aspect of the lid. It continues until it emerges at the anterior aspect of the lateral orbital rim. The stylette is then removed. The undersurface of the lid is inspected to ensure that the needle has not inadvertently perforated the tarsus. The end of the lower arm of the previously prepared palpebral spring is passed into the needle, and the needle and spring are withdrawn medially, bringing the spring into the lid. The scleral shell is removed, and the spring is positioned so that its previously determined curvature conforms to the lid anatomy. The upper arm of the spring is placed in position, and its length is determined. It usually needs to be about 3⁄4 inch long. A loop is fashioned at the point that is to become the end of the upper arm, and the wire is cut to size. The loop is closed to leave no sharp ends. The upper loop is made at a 90 degree angle to the fulcrum loop so that the upper loop can be tucked under the superior orbital rim. The loop is primarily held in place by pushing against the bone, decreasing the role of the sutures and helping to prevent late slippage. The loop at the fulcrum is held in place with forceps, and the patient is asked to open and close the eye. The position at which the spring curvature best conforms to the lid anatomy is then found, with the eye opened and closed, and the loop at the fulcrum is sutured in place. The suturing is accomplished with 4-0 Mersilene suture, and an extra bite of periosteum is taken with each stitch. Three such sutures are generally placed for the nonenhanced procedure, and five are placed for the enhanced procedure because of the greater tensions involved. The lower arm of the spring is cut to size, and a loop (which is also meticulously closed) is formed on its end. The loop should be formed upward to maintain a smooth inferior surface to the spring. The end of the spring should be at the pupillary axis, with the eyes in the primary position of gaze. Before the end of the spring is covered with polyester (Dacron) patch material, the angulation of the loop should be checked with the patient’s eyes open and closed to ensure that the spring tracks well with lid movement, and that the loop stays relatively parallel to the tarsus during opening and closing. A piece of 0.2 mm polyester patch material, which has been creased by its placement in a press that is used for compressing absorbable gelatin sponge (Gelfoam) before it is autoclaved, is cut to size to fit over the inferior loop. This piece is converted into a pouch by closure of the sides with 8-0 nylon sutures tied internally. The creased side is directed downward. The open lateral side is slipped over the spring, to which it is secured with an
8-0 nylon suture beginning within the pouch. The suture is passed through the spring loop and the posterior end of the pouch and is terminated by passing through the anterior side of the pouch. The knot is tied internally to prevent erosion. The polyester envelope is secured to the tarsus with one or more 8-0 nylon sutures, as needed, to prevent slippage of the polyester until granulation to the tarsus occurs. Spring tension is adjusted to just close the eye, by moving the upper arm of the spring closer or farther from the orbital rim. At the desired tension, the upper arm is bent so that the loop can be tucked under the orbital rim; it is secured with 4-0 Mersilene sutures to periosteum, taking an extra bite of periosteum with each stitch. Tension is checked with the patient seated and supine. Additional adjustments can be made by bending the wire of the upper arm to loosen or tighten it. Bending the wire of the lower arm near the fulcrum should be avoided at this time because such adjustments may be required during the postoperative period, and excess bending of the wire may increase its chance of breakage. When the final position of the upper loop has been determined in the nonenhanced procedure, two additional 4-0 Mersilene sutures are placed, and an extra bite of periosteum is taken with each stitch. Four additional sutures are used in the enhanced procedure. The deeper aspect of the wound overlying the orbital rim is closed with 5-0 plain gut suture to cover the spring and Mersilene sutures at the upper loop and fulcrum. The lid fold incision is closed with running 6-0 plain gut suture. The eye is dressed with antibiotic ointment and an ice pack.
Enhanced Palpebral Spring Implantation The enhancement of the palpebral spring operation consists of tightening of the levator during the same procedure (Fig. 61-3). The spring is prepared similarly except that wire lighter than 0.010 inch is not used. I primarily use the 0.011 inch 35NLT (Fort Wayne Metals) alloy. A scleral shell is placed. The initial skin fold incision is the same as that described earlier. Dissection is carried upward until preaponeurotic fat can be visualized through the septum. The septum is opened, and dissection is carried down to expose the levator. Dissection is carried inferiorly to expose the tarsus centrally. A 5-0 Mersilene suture is placed through midtarsus where the loop-polyester complex overlies the tarsus. The lid’s undersurface is inspected to be sure the suture has not perforated the tarsus. Both suture arms pass through the polyester, and the medial arm passes through the loop as well, before completing the suture through the levator, to emerge just above the point where aponeurosis ends and levator muscle is visualized. This serves the purposes of directly opposing the maximal force of the spring (at its distal end) with the levator force and of securing the polyester further until granulation tissue fixes it in place.
Chapter 61 • Care of the Eye in Facial Paralysis
739
FIGURE 61-3. Enhanced palpebral spring implantation. A, Levator aponeurosis (LA) and inferior aspect of the muscular portion of the levator are exposed. Centrally, the superior portion of the tarsus is also exposed. A double-armed 5.0 Mersilene suture is placed through midtarsus. B, Each arm of the suture is brought superiorly through the levator to emerge just above the point at which the aponeurosis meets the levator muscle (LM). Temporary knots are tied. If necessary, an additional lateral suture and possibly an additional medial suture are placed in a similar manner. Inset: Course of suture is illustrated in cross section. The surgeon should check to ensure that the suture has not perforated tarsus or conjunctiva. PAF, preaponeurotic fat; S, orbital septum; STL, superior transverse ligament.
740
OTOLOGIC SURGERY
A temporary knot is placed, and the patient is asked to open the eye. The scleral shell is removed, and the extent of levator tightening is evaluated. If necessary, an additional lateral suture and possibly an additional medial suture are placed in a similar manner to achieve desired lid strengthening and maintenance of proper upper lid curvature. If the surgeon is unsure whether additional sutures are necessary, their placement can be deferred until after the spring has been placed, and the overall effect of the spring and the initial suture can be evaluated. Regardless of how many sutures are used, they are adjusted after the spring is in place, at the same time that the spring itself would otherwise be adjusted. The levator is tightened to a point at which maximum strengthening is achieved without inducing cicatricial lagophthalmos from an overly shortened levator. The stronger the levator can be made, the greater the tension possible on the spring, and the more rapid the blink. The nuances of adjustment of the levator and the spring can be appreciated only with experience. Nevertheless, when the general principles are understood, excellent results can be obtained (Fig. 61-4).
Gold Weight Implantation The size of the weight is selected preoperatively.15,16 With the patient seated, a gold weight (or the comparable weight from a sizing set) that is estimated to be suitable for the degree of lagophthalmos is selected and is secured to the patient’s upper lid with cyanoacrylate glue, doublestick tape, or a temporary lid suture. The patient is asked to open and close the eye, and the surgeon determines whether the weight of the gold is correct. The evaluation is repeated with the patient supine. Weights of 1.2 to 1.5 g are suitable for use in most patients. The weight may be fixated either supratarsally (Fig. 61-5) or tarsally. Supratarsal fixation is preferable unless the weight is so large that this is impractical. A scleral shell is placed. An incision is made in the lid fold, and dissection is carried upward to expose the orbital septum. The preaponeurotic fat can be seen through the septum, which is opened, and the weight is secured to the levator with a single 5-0 polyester suture placed through the holes in the weight. The knot is buried. The function of the weight is tested with the patient in the seated and supine positions. If the desired effect is not obtained, a different-sized weight may be tried. Overlying skin muscle is closed with running 6-0 plain gut suture. An alternative gold weight technique used by Michael Roberts MD, is presented.17-19 It is based on 36 cases performed; no follow-up data are available. The size of the weight is selected preoperatively. With the patient seated, a gold weight, or the comparable weight from a sizing set, is secured to the patient’s upper lid with cyanoacrylate glue, double-stick tape, or a temporary lid suture. Optimal position is 2 to 3 mm
above the lash margin and centered horizontally within the lid where levator function is maximal. The patient is asked to open and close the eye, and the surgeon determines whether the weight of the gold is correct. Ideally, the paretic eyelid should close completely and remain equal to or 1 mm lower than the normal eyelid with both eyes open, as levator function improves in response to presence of the gold weight. The evaluation is repeated with the patient supine. Weights of 1 to 1.8 g are suitable for use in most patients. Similar weight sizes in platinum are also available for patients with a gold allergy. Special attention is needed for patients with proptotic eyes. In these patients, the globe partially supports the weight, and additional weight may be required for full lid closure.
Surgical Technique The upper eyelid crease is marked, and a scleral shell is placed. A curvilinear incision is made along the marked crease through skin and orbicularis muscle. In Asian patients, the incision is placed 3 to 4 mm above the lash margin because of a lower insertion of the orbital septum onto the levator aponeurosis. The weight may be fixated to the septum, tarsus, or levator aponeurosis posterior to the septum. The advantage of septal fixation is the gold weight is less visible postoperatively than when sutured to the tarsus, traditionally a thinner area of the eyelid. Fixation to the tarsus has three advantages: (1) the suture is less likely to cheese-wire through the tarsus than the septum; (2) levator attachments to the anterior tarsal surface are stripped during exposure, creating a modest levator recession preventing postoperative lid retraction; and (3) there is less potential for inferior migration of the gold weight because the lid margin acts as a barrier. Fixation to the levator alleviates the need for sizing of the weight, but requires additional surgical skill because the septum is incised before placement of the weight, then sewn back after the implant is in position. After the initial incision is made, blunt dissection is performed posteriorly to expose either the tarsus or the orbital septum. When these are identified, dissection is continued medially and laterally until a sufficient pocket is created for placement of the weight and centration over the pupil. Levator fibers are gently stripped from the tarsus in cases of tarsal fixation. The weight is then fixed to the septum or the tarsus with a 5-0 nonabsorbable suture placed through the holes in the weight. In cases of fixation to the levator, the septum is opened, and the weight is secured to the underlying levator with a 5-0 nonabsorbable suture. The septum is closed with a 6-0 nonabsorbable suture. Some surgeons test function of the weight with the patient in the seated and supine positions. If the desired effect is not obtained, a different sized weight may be inserted. Overlying orbicularis muscle and skin are closed in layers with 6-0 absorbable sutures.
Chapter 61 • Care of the Eye in Facial Paralysis
741
FIGURE 61-4. The patient is a 49-year-old woman with facial paralysis secondary to an acoustic neuroma. She had undergone two tarsorrhaphy procedures before she was first seen for evaluation. A, Eyes open, but eye function is largely blocked by tarsorrhaphy. Note also the brow droop. B, Attempted closure. Note that tarsorrhaphy, although extensive, fails to protect cornea well. C, Eyes open. Tarsorrhaphy has been opened, and lid margins have been reconstructed. The patient also has had enhanced palpebral spring implantation, medial and lateral canthoplasties, correction of upper lid entropion, and elevation of the brow. D, Attempted closure. Note excellent protection of the cornea by the spring. FIGURE 61-5. Gold weight. The levator is exposed, and gold weight is secured to it with single 5-0 polyester suture placed through the holes in the weight. The knot is buried. Larger weights may be placed pretarsally in a similar manner. LA, levator aponeurosis; LM, levator muscle; T, tarsus.
742
OTOLOGIC SURGERY
Table 61-1 compares the advantages of enhanced palpebral springs and gold weights.
Silicone Rod Prosthesis Implantation The prosthesis to be used in silicone rod prosthesis implantation20-22 is commercially available as a 1 mm diameter rod. While the eye is protected with a scleral shell, a curvilinear incision is made overlying the medial canthal tendon (Fig. 61-6). The incision should be just lateral to the angular vein to avoid the vein in the course of the dissection. Dissection is carried posteriorly to expose the origin of the tendon. The prosthesis is threaded on a large, noncutting needle (e.g., No. 5 Mayo needle), and is sewn through the tendon twice. The lower arm should emerge from the posterior aspect of the tendon, which facilitates holding the lower lid against the globe. A second incision is made at the lateral orbital rim, and dissection is carried to the periosteum. By use of sharp dissection, a tunnel is started in the plane between the orbicularis and the tarsus at the lateral aspect of the lower lid. A special introducer is passed as close to the lid margin as possible across the lid to emerge medially close to where the silicone rod has been sewn through the tendon. The lower arm of the rod is threaded onto the introducer, and the introducer is withdrawn laterally, bringing the rod through the lid. In a similar manner, the upper arm of the prosthesis is brought through the upper lid except that passage is accomplished over the midtarsus.
TABLE 61-1 Comparison of Enhanced Palpebral
Springs and Gold Weights*
Advantages of Spring Closes eye at night when patient is supine Increased blink speed Adjustability without removing device Can protect cornea in severe cases of combined CN V and VII involvement Increased protection, comfort, and return to full functional activity in patients with active lifestyle (e.g., skiing, surfing, flying, athletics) More esthetic than a large gold weight Comfort in patients exposed to extreme heat or cold environments Facilitates use of adjunctive bandage contact lenses Increases possibility of successful corneal surgery in patients with previously damaged corneas *Based
Advantages of Gold Weight Simple technique Works well in cases that have only minimal lagophthalmos
on data ����� over ����������������������������������������������������� >30 years�������������������������������������������� from patients who have had weights removed and springs implanted instead.
The desired point of anchorage for the lower end of the prosthesis is determined, and a suture loop is formed at that point through periosteum by use of a 2-0 polypropylene (Prolene) suture sewn through the periosteum twice at the orbital rim. The prosthesis must be secured at the inner aspect of the lateral orbital rim to pull the lid posteriorly. The point selected should be just above the horizontal raphe to draw the lid upward as well. In a similar manner, the upper arm of the prosthesis is secured just below the horizontal raphe, passing anterior to the lower arm. With the patient in the seated position, suitable tension is placed on the lower arm to secure the lower lid in the desired position. (For additional discussion on how to best position the lid, see the section on assessing lid position.) Similarly, the tension on the upper lid is adjusted to permit good opening and closing of the lid, and final knots are tied. Each arm of the prosthesis is secured further with additional suture bites through the lateral rim periosteum. A heavy suture, such as 2-0 Prolene, is selected to avoid cutting through the prosthesis. Deep tissues are closed with 5-0 plain gut suture medially and laterally, and the skin is closed with 6-0 plain gut suture.
LOWER LID REAPPOSITION PROCEDURES Canthoplasty Surgery may be performed at either the medial or the lateral canthus to tighten and elevate the lower eyelid. Operating at the ends of the tarsus has the advantage of not creating an irregular lid margin or interfering with the visual field. Although tightening the lid only at the lateral canthus is usually the procedure of choice in patients with nonparalytic lid laxity, the paralyzed lid usually requires medial canthal tightening as well. Otherwise, tightening the lid only laterally may result in either marked displacement of the inferior punctum laterally or failure to elevate the lid. In some cases, tightening the lid only laterally can cause the lid to act as a shorter chord beneath the globe—lowering the lid rather than raising it. If only a limited amount of lid tightening and elevation is required, medial canthoplasty alone may be the procedure of choice.
Assessing Lid Position Assuming that extraocular motility has not been impaired by excess local anesthesia infiltration, the position of the lid relative to the limbus (corneoscleral junction) can be used as a guideline in the adjustment of lid position. The gaze of the patient should be directed such that the limbus of the contralateral lid is placed just tangential to the inferior limbus. The lid being operated on can be tightened to a position in which it, too, is tangential to the limbus, or it can be slightly overcorrected to allow for postoperative loosening. Matching this tangential position is
Chapter 61 • Care of the Eye in Facial Paralysis
743
FIGURE 61-6. Silicone rod prosthesis. A, Incision is made medially just lateral to angular vein; a second incision is made laterally over orbital rim. B, Laterally, dissection is carried down to expose orbital rim. Medially, the medial canthal tendon is exposed, and the prosthesis is sewn through it by use of a large noncutting needle. C, Prosthesis is sewn further through the tendon so that the lower arm emerges posterior to tendon, facilitating holding the lower lid against the globe. D, Each arm of the prosthesis is ready to be engaged on the introducer. E, With a special introducer, upper arm of the prosthesis has been passed between the orbicularis and tarsus in the upper lid, at the level of midtarsus. The introducer is shown in the lower lid, in preparation for passing the lower lid of the prosthesis. Prosthesis must be as close to lid margin as possible to prevent ectropion of the lid. F, Lower arm of prosthesis is secured to orbital rim periosteum with 2-0 Prolene suture. G, Both arms of prosthesis are sutured to the periosteum further with 2-0 Prolene. Inferior arm is posterior to superior arm because lower lid must be pulled posteriorly.
744
OTOLOGIC SURGERY
Chapter 61 • Care of the Eye in Facial Paralysis
745
FIGURE 61-7. Medial canthoplasty. A, While globe is protected with scleral shell, and canaliculi are protected with probes, an incision is made along the mucocutaneous junction and continued downward at a point 2 mm medial to the punctum. B, Inferior flap is elevated, which exposes inferior arm of medial canthal tendon. Tendon is pulled medially to expose its junction with lateral aspect of tarsus (insertion of the tendon), and a 5-0 polyester double-armed suture is placed through insertion. C, Each arm of the suture is woven through the tendon and brought through the origin of the tendon. The two sutures are tied under appropriate tension, tightening lower lid. Mucocutaneous junction is reconstructed with 8-0 Vicryl suture, leaving medial canthus with a normal appearance. D, If additional lower lid inversion is required, a second incision is made in upper lid along mucocutaneous junction, directed upward at a point 2 mm medial to punctum. E, Superior flap is elevated, which exposes superior arm of canthal tendon. F, A 5-0 polyester suture is placed as a horizontal mattress suture. The suture passes near the lower end of the lower tendon and near the upper end of the upper tendon, inverting the lids. G, Edges of the skin flaps are closed to each other with 6-0 plain gut suture. The steps shown in D-G may be performed independently when lid inversion is more important than lid tightening, or they may be combined with the steps shown in A-C. When steps D-G are required, the result is a more blunted medial canthal angle, with some shortening of horizontal fissure. If the procedure shown in A-C suffices, it is preferable. Nevertheless, the additional steps may be required in patients with marked medial lid laxity or eversion.
more precise than judging the amount by which the lid crosses the cornea or comes inferior to it compared with the other eye.
Medial Canthoplasty Medial canthoplasty23 consists of exposing the lower arm of the medial canthal tendon and the origin of the common tendon and tightening the lower arm. If additional effect is required, the upper arm of the tendon may also be exposed so that it, too, can be included in the surgical procedure.
Surgical Technique To expose the medial canthal tendon, the globe is protected with a scleral shell, and the canaliculi are protected with probes. An incision is made at the mucocutaneous junction, beginning 2 mm medial to the lower punctum, along the mucocutaneous junction to the medial canthus, and for an additional 2 mm beyond the canthus (Fig. 61-7). A skin/muscle flap is elevated with scissors, and hemostasis is achieved with bipolar cautery. The insertion of the lower arm of the medial canthal tendon is grasped with forceps and drawn superonasally, permitting the placement of a 5-0 double-armed polyester suture through the insertion. Each arm of the suture is woven through the tendon and sewn through the origin of the tendon. Temporary knots are tied. The patient is brought to the seated position, and the apposition of the lid and position of the punctum are evaluated. If a lateral canthoplasty is also being performed, sitting the patient up can be deferred, and tension on both sets of canthoplasty sutures can be adjusted simultaneously after completion of the lateral canthoplasty. If only a medial canthoplasty is being performed, the tension on the lid is adjusted, and final knots are tied. If the punctum is inadequately inverted by the suture used to tighten the medial canthal tendon, the canthal tendon is exposed in the upper lid, and a second 5-0 polyester suture is placed. Each arm of that suture starts at the lower edge of the inferior canthal tendon and ends at the upper edge of the superior canthal tendon, inverting the lower punctum further.
The canthoplasty using a suture only in the lower lid is completed by reforming the mucocutaneous junction with 6-0 plain gut (on a tiny needle) running or interrupted sutures. These sutures may be allowed to dissolve or may be removed in 5 to 7 days. When a second canthal suture is required, the lid flaps are closed to each other with 6-0 plain gut sutures. Because a second canthal suture results in the blunting of the canthal angle and horizontal shortening of the palpebral fissure, such a suture should not be used unnecessarily.
Lateral Canthoplasty The object of lateral canthoplasty24,25 is to pull the lid superoposteriorly against the globe. In lieu of the various lid-shortening procedures that have been used in the past, the currently preferred method is to create a new lateral canthal tendon from the tarsus itself, and to secure it to the orbital rim.
Surgical Technique A hemostat is placed across the lateral canthal angle and removed (Fig. 61-8). The clamped area is cut with scissors, accomplishing a canthotomy. The inferior crux of the lateral canthal tendon is cut (inferior cantholysis) with scissors. At this point, the lateral aspect of the lower lid is freely movable. The lid is pulled taut across the globe. If a medial canthoplasty has been performed, suitable tension should be placed on that canthoplasty suture before this maneuver. The amount of excess lid is marked, and the skin/ muscle lamina is separated from the tarsus lateral to this mark. The lid margin in this area is also removed. Some surgeons prefer to remove the conjunctiva at this point, leaving a pure tarsal tongue. I have not found such removal necessary, and the tarsus may be damaged during such attempts. A double-armed 5-0 polyester suture is passed as a mattress suture beginning on the anterior surface of the tarsus, 3 mm from the lateral end, with each needle emerging through the cut end. A second pass of each suture arm is made, and each arm of the suture is locked. The lower arm of the suture is marked with a marking pen for future reference.
746
OTOLOGIC SURGERY
Chapter 61 • Care of the Eye in Facial Paralysis
747
FIGURE 61-8. Lateral canthoplasty. A, Hemostat is used to clamp lateral canthal angle for hemostasis. B, Clamp is removed, and a lateral canthotomy is performed. C, Inferior canthal tendon is cut so that lower lid is freely movable. D, Lower lid is approximated to desired position with slight overcorrection. E, Excess skin muscle and mucocutaneous junction tissue are excised, leaving a tarsal tongue. F, Double-armed 5-0 polyester suture is passed through tarsal tongue, and one arm of the suture is marked with a marking pen for subsequent identification. A tunnel is created under the superior arm of the lateral canthal tendon. A clamp is passed into the tunnel, and the sutures are grasped and withdrawn laterally. The tarsal tongue is brought into the tunnel and secured under the desired tension to the lateral orbital rim periosteum. By use of identifying markings previously placed, superior suture is kept superior, and inferior suture is kept inferior, preventing twisting of the tarsus. Suture is tied under appropriate tension. Canthoplasty is completed with 5-0 plain gut suture to re-establish the canthal angle and 6-0 plain gut suture to close skin muscle. G, When lateral canthoplasty is combined with spring implantation, and only a moderate amount of lateral canthal tightening is required, the procedure can be accomplished through the prior lid fold incision. H, Tunnel is made under superior arm of the lateral canthal tendon. Forceps is introduced into the tunnel, and lateral aspect of the tarsus is grasped. I, Lateral end of tarsus is drawn into the tunnel, and a doubled-armed 5-0 polyester suture is used to secure it. J, Each arm of 5-0 polyester suture is secured through orbital rim periosteum under suitable tension. OR, orbital rim; ST, superior tendon; T, tarsus.
A tunnel is made underneath the upper arm of the lateral canthal tendon, and a clamp is passed into the tunnel. The sutures are grasped within the clamp and brought through the tunnel. The desired location of the lid is determined, and the sutures are placed appropriately, with care taken to obtain a bite of orbital rim periosteum at the inside aspect of the rim, slightly superior to the desired lid position. In this manner, the lid is drawn up and posteriorly. The previously placed mark on the suture helps to avoid confusion about the location of the inferior arm of the suture. A temporary knot is tied, and the patient is brought to the seated position. Adjustments in lid position and tension can be made, and final knots can be tied. The surgeon completes the canthoplasty by trimming excess skin and muscle, re-establishing the canthal angle with a 5-0 plain gut suture, and closing the skin with 6-0 plain gut suture. In situations in which lateral canthoplasty is combined with palpebral spring implantation, and the amount of lid tightening required is not too great, a variant of the technique may be used. In such circumstances, a separate canthal incision is not required. The upper arm of the lateral canthal tendon may be approached from the existent extended lid fold incision. A tunnel is made under the upper arm of the tendon. With forceps placed into the tunnel, the lateral aspect of the tarsus of the lower lid is grasped and brought into the tunnel. A 5-0 polyester double-armed suture is placed in the tarsus. This stitch is used to recreate the lateral canthal tendon in a manner similar to that described previously. If a great deal of lid tightening is required, however, it is impossible to omit the steps of canthotomy and inferior cantholysis, and still adequately mobilize the lid to achieve the desired position.
Fascia Lata Suspension of Lower Lid In patients in whom medial and lateral canthoplasty together are inadequate to elevate the central portion of the lid, the lid may be supported by a fascia lata suspension. In this procedure, the fascia lata is anchored at
each end of the lid and acts as a hammock to support the central lid. Either autologous or banked fascia lata can be used. The preserved fascia lata is more subject to resorption over time; the autologous fascia lata is more likely to give a reliable long-term result. When autologous fascia lata is used, it is obtained from the leg by a general surgeon while the lid is being prepared for its placement. Nevertheless, because of the morbidity from the leg incision, I currently prefer to use preserved fascia.
Surgical Technique A strip of fascia 3⁄16 inch wide and approximately 4 inches long is used. The technique for placing the fascia within the medial canthal tendon and in the lower lid is similar to the technique described in the section on the silicone rod prosthesis. The medial end of the fascia is anchored with a 5-0 polyester suture passed twice through the fascia and the medial canthal tendon (Fig. 61-9). The knot of the suture should be well buried to avoid later erosion. The lateral end of the tendon is secured to the inner aspect of the lateral orbital rim with 5-0 polyester suture in a manner similar to that described for lateral canthoplasty, and is adjusted in a similar way before final knots are tied. The suture should encircle the fascia, however, rather than go through it to avoid tearing it. When the fascia has been fixed in its position, additional bites through the periosteum and around the fascia are desirable to prevent slippage. Performing canthotomy and inferior cantholysis may be necessary to mobilize the lid adequately so that the fascial suspension holds it in the desired position.
Lid Stents An alternative to supporting the central portion of the lower lid with a sling is to use a stent. In the past, auricular cartilage grafts (which did not have curvature that matched the curvature of the lid) and hard palate grafts (which created significant morbidity at the donor site) were used. Because of the limitations cited, they were not used extensively. More recently, the availability of
748
OTOLOGIC SURGERY
Chapter 61 • Care of the Eye in Facial Paralysis
749
FIGURE 61-9. Fascia lata suspension. This technique is analogous to that shown for placing the silicone rod prosthesis. Fascial strip is placed on a large needle and sewn through medial canthal tendon, emerging behind tendon. The end is locked with 5-0 polyester suture. With the special introducer, the fascia is brought laterally through the lid, where it is secured to orbital rim periosteum with 5-0 polyester suture in a manner similar to that described in the section on lateral canthoplasty. Excess fascia lata is secured to the periosteum further by continuation of the same sutures after initial knots are tied. These sutures encircle the fascia rather than perforate it, to avoid tearing it.
FIGURE 61-10. Brow elevation. A, If skin muscle excision is not planned, incision is made as close as possible to the brow over the area that needs to be suspended. Usually, this area consists of approximately the central two thirds of the brow. B, When skin muscle excision is planned, ellipse to be excised is marked, and skin incision is performed. C, Skin and muscle have been removed. D, Regardless of whether skin muscle excision is required, suspension sutures are placed in a similar manner. Dissection is carried superiorly to expose periosteum. Sutures of 4-0 Novafil are placed through periosteum, and through subcuticular tissue at the lower end of the wound. After these have been tied under appropriate tension, the brow is closed in layers.
FIGURE 61-11. Correction of upper lid entropion. Series of 6-0 Vicryl horizontal mattress sutures are placed between lower edge of levator a poneurosis (LA) and subcuticular tissue close to inferior edge of wound. Tightening these sutures under appropriate tension rotates lashes outward, and creates a pleasing lid fold. LM, levator muscle.
a lloplastic materials (e.g., porous high-density polyethylene [Medpor]) have made lid stents a much more viable possibility. I have devised a new technique for placing such a stent in which it can be placed over a guide from a small lateral incision. I use it frequently in conjunction with enhanced palpebral spring implantation.
Midface Support The constant drag of the weight of the paralyzed face on the lower lid can undo the most expert lower lid repair. To prevent this from happening, it is important to stabilize or elevate the midface when planning lower lid reconstruction. Many ways of supporting the midface are described in the plastic surgical literature, and these descriptions are beyond the scope of this chapter. Currently, I have devised a technique that uses an Endotine ribbon secured to the lateral orbital rim.
ANCILLARY PROCEDURES Patients with facial paralysis frequently manifest brow droop and entropion of the upper lid. These problems can be addressed surgically at the same time that upper lid reanimation surgery is undertaken. Elevation of the ptotic brow in a patient with facial paralysis should not be undertaken independent of a procedure to enhance lid closure. The droop of the brow tends to push the lid shut, and ameliorates the upper lid lagophthalmos. Correcting the brow position without concomitantly improving upper lid closure may significantly worsen the patient’s lagophthalmos.
Brow Elevation An eyebrow can be elevated23 in the following ways: 1. Excising the skin and muscle just above the brow 2. Suspending the brow from the periosteum above it
3. Lifting the forehead by a coronal approach 4. Endoscopically raising the brow The coronal forehead lift requires extensive dissection, does not give as much effect as a direct lift, and may be difficult to control to elevate only one brow. A direct approach to brow elevation is generally preferable in patients with facial paralysis. Skin/muscle excision alone is less effective in the paralyzed frontalis than in a normally innervated frontalis. I have found combining skin/muscle excision with brow suspension, and brow suspension alone to be two simple, useful brow-elevating procedures. Endoscopic brow elevation is more timeconsuming, and may be harder to do in a patient who is undergoing multiple procedures. It is an excellent procedure, however. The surgical technique of endoscopic brow lift is beyond the scope of this chapter. The choice of the procedure depends on several factors. First, is frontalis function expected to return? If so, it is best not to excise tissue, but only to suspend the brow. The second factor to consider is the effect of raising the brow without tissue excision. In some patients, this process induces a few brow wrinkles, which are welcome from an appearance standpoint in a previously abnormally smooth paralyzed area. In other patients, only a bulge of tissue is created by elevation of the brow without tissue excision. In such patients, skin/muscle excision can be performed. Elevating the entire brow is not always necessary. The point of maximum brow elevation should be noted on the contralateral side. A line drawn downward through this point usually passes at or near the lateral canthus. The exact position of such an imaginary line should be noted, and a comparable line should be marked on the side to be operated. Some patients have a different brow configuration and require more medial elevation. When the point of maximum brow elevation is determined and marked, the brow is raised at that point. How much of the brow must be elevated to obtain a desirable contour can then be seen. The extent of required brow elevation can also be judged by drawing a line tangential
750
OTOLOGIC SURGERY
to the point of maximum brow elevation on the contralateral side. This line is drawn perpendicular to a vertical line bisecting the nose. Corresponding points on each brow are marked and measured relative to the horizontal reference line to determine the extent of brow elevation required at each point on the paralyzed side. If it cannot be determined with certainty whether tissue excision would be required, the extent of the anticipated excision should be marked in advance of lid infiltration, so that the determination is not distorted by the swelling induced by the injection. The brow can be suspended without tissue excision. If the result is not pleasing, skin/muscle excision can be carried out. The role of the ptotic brow in aiding lid closure should be assessed. If a paretic brow that is assisting closure is raised without simultaneously performing a procedure such as spring implantation to enhance lid closure, the result may be worsened lagophthalmos.
Surgical Technique The area in which brow elevation, tissue excision, or both are required is marked as close to the superior extent of the brow as possible (Fig. 61-10), which helps conceal the resultant scar. The skin is incised perpendicular to the skin surface with a scalpel blade until muscle is reached. Blunt dissection is carried out at each site where a suspension suture is deemed necessary. The brow is elevated during the dissection so that the frontalis periosteum is encountered superior to the brow. A 4-0 Novafil suture on a very curved needle is passed through the periosteum, and then through the dermis at the lower aspect of the wound as a horizontal mattress suture. One to four such sutures may be required, depending on the contour of the brow. When all of the planned sutures have been placed, and temporary knots have been tied, the patient is brought to the seated position, and the brow contour is adjusted to match that on the contralateral side in the primary position of gaze. The normally innervated, nonfixated brow moves downward on lid closure—the suspended brow is incapable of doing this. Care must be taken not to elevate the brow so much that a cicatricial lagophthalmos is induced. Especially in patients in whom lid skin is in short supply (e.g., in patients who have undergone a previous blepharoplasty), fully elevating the paretic brow may be impossible without adversely affecting lid closure. In such cases, it is better to place the brow slightly lower than the contralateral side, rather than induce lagophthalmos. Final knots are tied. If skin/muscle excision is required, it can be carried out. The Novafil sutures are rotated to bury the knots deeply. The brow should be closed in layers to minimize the scar. Depending on the thickness of the tissue, one or two rows of deep 5-0 or 6-0 polyglactin 910 (Vicryl) sutures are placed before skin closure with interrupted 5-0 plain gut sutures and a Steri-Strip.
Two alternative ways of elevating the brow are a vailable—internally suspending the brow with suture or with a special Endotine prosthesis designed for this purpose. Both are performed through a lid fold incision, and may be excellent choices when brow elevation is being combined with eyelid surgery. In the case of suspending the brow with suture, dissection is carried under the brow, from the lid fold incision to a point above where the brow is to be raised. One or more 4-0 Novofil sutures are placed in periosteum and continued into deep tissue below the brow to elevate and fixate the brow. In the case of the Endotine prosthesis designed to be used just above the orbital rim, similar principles apply. Dissection is carried from the lid fold incision to just above the superior orbital rim, at which point periosteum is incised and elevated. Adjacent tissue is dissected to allow the brow to be mobilized. A hole is drilled in the line of maximal brow elevation, 15 mm above the superior orbital rim. A special drill is provided to use with the prosthesis. A 3-0 Prolene suture is loaded in the prosthesis before prosthesis placement. This suture is used to secure further the overlying tissue to the prosthesis after the brow has been positioned.
Correction of Upper Lid Entropion Correction of upper lid entropion is most easily carried out in combination with enhanced or nonenhanced spring implantation. This correction can be carried out as a separate procedure by opening the lid in the lid fold or, in combination with gold weight implantation, by extending the lid fold incision across the entire lid. A series of 6-0 Vicryl sutures is placed across the lid (Fig. 61-11). Each of these sutures begins supratarsally, in the levator, and continues as a horizontal mattress suture through subcuticular tissue just inferior to the lower aspect of the skin incision. Tightening these sutures rotates the lid margin outward, correcting the entropion resulting from the facial paralysis or from the downward pressure from an implanted prosthetic device. Tension on the sutures is adjusted at the time of surgery to give a slight overcorrection. The sutures also create a pleasing lid fold.
TEMPORIZING MEASURES Two simple surgical techniques are described to protect the eye before a decision to undertake definitive surgery is made.
Lid Suture Taped to Cheek A lid suture can be placed easily at the conclusion of a neurotologic or head and neck procedure in which the function of CN V or VII is anticipated to be compromised postoperatively. The suture protects the eye during the
Chapter 61 • Care of the Eye in Facial Paralysis
751
FIGURE 61-12. Lid suture. A 4-0 or 5-0 nonresorbable monofilament suture is passed through skin and orbicularis muscle (OM), which have been pulled away from the tarsus (T) with forceps. The two arms of the suture are tied together with multiple knots to prevent slippage underneath the tape, and are secured to the cheek with a strip of tape. The suture is brought upward and locked with a second piece of tape. FIGURE 61-13. Temporary tarsorrhaphy suture. A, Each arm of a double-armed nonresorbable monofilament suture is passed through skin and orbicularis of the central third of the upper lid, beginning 5 mm above lid margin and exiting at gray line of lid margin. B, Sutures continue to the gray line of the lower lid, exiting through orbicularis and skin 5 mm below the lid margin. Two arms of the suture are placed 1 cm apart. C, Cotton bolsters are placed between suture and skin before suture is tied. Suture is tied with a bow, and the ends are left long.
752
OTOLOGIC SURGERY
immediate postoperative period without impairing the ability to check for pupillary or other neurologic signs involving the eye. If the patient subsequently requires temporary eye protection involving less than round-theclock eye closure, the suture may be taped out of the way (to the forehead) during part of the day and used to close the eye at other times. The eye is protected with a scleral shell. A 4-0 or 5- 0 nonresorbable monofilament suture is passed through the skin and orbicularis, which have been pulled away from the tarsus with forceps (Fig. 61-12). The needle is passed parallel to the tarsus. The two arms of the suture are tied together with multiple knots to prevent slippage underneath the tape and are secured to the cheek with a strip of tape. The suture is brought upward and locked with a second piece of tape. A third strip locks the suture further and keeps it out of the way. When it is necessary to inspect the eye or to check for pupillary signs, all three pieces of tape can be lifted together with the suture away from the cheek and then replaced. By placing antibiotic ointment at the suture sites in the lid, one can usually maintain a suture in the lid for 2 to 3 weeks without undue lid induration.
road to recovery, lid surgery greatly advances the patient’s chances for successful overall rehabilitation. The purpose of this chapter is to present procedures as I currently do them. Many of the modifications are my own, and tracing the historical evolution of procedures is not always possible. I would be remiss, however, if I did not credit the individuals who initially devised, pioneered, or popularized these procedures; Table 61-2 provides a summary of that information.
Temporary Tarsorrhaphy Suture
REFERENCES
Temporary tarsorrhaphy suturing allows the lids to be kept securely closed over several weeks without damage to the lid margins from the creation of a true tarsorrhaphy. It also permits one to inspect the eye at intervals by untying the suture, inspecting the eye, and retying the suture without having to replace it. It is particularly useful when there is a significant lower lid laxity component to the exposure problem. The eye is protected with a scleral shell. Each arm of a double-armed, monofilament, nonresorbable suture is passed through the skin and orbicularis of the central third of the upper lid, beginning 5 mm above the lid margin and exiting at the gray line of the lid margin (Fig. 61- 13). The sutures continue into the gray line of the lower lid, exiting through the orbicularis and skin 5 mm below the lid margin. The two arms of the suture are placed 1 cm apart. Cotton bolsters are placed between the suture and skin before the suture is tied. The suture is tied like a shoelace, with a bow (leaving the ends of the suture long), to facilitate untying and retying in the future. The bow and ends of the suture are taped out of the way to the lid. Antibiotic ointment is applied to the suture sites twice daily, allowing the suture to be maintained for several weeks without undue induration of the lid.
CONCLUSION By appropriate selection of the procedures presented, most patients with facial paralysis can be helped to obtain markedly improved lid position and function. Because eye problems frequently present major hurdles on the
TABLE 61-2 Eyelid Surgical Procedures Procedure
Author
Palpebral spring
Morel-Fatio and Lalardrie,4,5 Levine et al6-14 Levine13,14 Jobe,15 May16 Arion,20 Marrone and Soll,21 Levine7,11,13 Beard23 Tenzel et al24,25
Enhanced palpebral spring Gold weight Silicone rod prosthesis Medial canthoplasty and brow lift Lateral canthoplasty
1. Jelks GW, Smith B, Bosniak S: The evaluation and management of the eye in facial palsy. Clin Plast Surg 6:397419, 1979. 2. Rosenstock TG, Hurwitz JJ, Nedzelski J M, Tator C H : Ocular complications following excision of cerebellopontine angle tumours. Can Ophthalmol 21:134-139, 1986. 3. Seiff S R , Chang J: Management of ophthalmic complications of facial nerve palsy. Otolaryngol Clin North Am 25:669-690, 1992. 4. Morel-Fatio D, Lalardrie J P: Palliative surgical treatment of facial paralysis: The palpebral spring. Plast Reconstr Surg 33:446, 1964. 5. Morel-Fatio D, Lalardrie J P: Le ressort palpebral: Contribution a l’etude de la chirurgie plastique de la paralysie faciale. Neurochirurgie 11:303, 1965. 6. Levine R E, House WF, Hitselberger WE : Ocular complications of seventh nerve paralysis and management with the palpebral spring. Am J Ophthalmol 73:219, 1972. 7. Levine R E : Management of the eye after acoustic tumor surgery In House WF, Luetje C M (eds): Acoustic Tumors. vol. 2 Baltimore, University Park Press, 1979, pp 105-149. 8. Levine R E : Management of the ophthalmologic complications of facial paralysis. Trans Pac Coast Ophthalmol Otolaryngol Soc 61:85-93, 1980. 9. Levine R E : Protection of the exposed eye. In Brackmann D E (ed): Neurological Surgery of the Ear and Skull Base. New York, Raven Press, 1982, pp 81-87. 10. Levine R E : Protection of the exposed eye in facial paralysis. In Graham M D, House WF (eds): Disorders of the Facial Nerve. New York, Raven Press, 1982, pp 336-375. 11. Levine R E : Eyelid reanimation surgery. In May M (ed): The Facial Nerve. New York, Thieme-Stratton, 1985, pp 681-694.
Chapter 61 • Care of the Eye in Facial Paralysis 12. Levine R E : Palpebral spring for lagophthalmos due to facial nerve palsy. In Wesley R E (ed): Techniques in Ophthalmic Plastic Surgery. New York, John Wiley & Sons, 1986, pp 424-427. 13. Levine R E : Management of lagophthalmos with palpebral spring and Silastic elastic prosthesis. In Hornblass A (ed): Ophthalmic and Orbital Plastic Reconstructive Surgery. vol 1. Baltimore, Williams & Wilkins, 1989, pp 384-392. 14. Levine R E : Lid reanimation with the palpebral spring. In Wright K, Tse D (eds): Color Atlas of Ophthalmic Surgery. Philadelphia, Lippincott, 1992, pp 231-238. 15. Jobe R P: A technique for lid-loading in the management of lagophthalmos in facial paralysis. Plast Reconstr Surg 53:29-31, 1974. 16. May M : Surgical rehabilitation of facial palsy. In May M (ed): The Facial Nerve. New York, Thieme-Stratton, 1985, pp 695-777. 17. Choo PH, Carter S R , Seiff S R : Upper eyelid gold weight implantation in the Asian patient with facial paralysis. Plast Reconstr Surg 105:855-859, 2000.
753
18. Seiff S R , Boerner M, Carter S R : Treatment of facial palsies with external eyelid weights. Am J Ophthalmol 120:652-657, 1995. 19. Tower R N, Dailey R A : Gold weight implantation: A better way? Ophthalmic Plast Reconstr Surg 20:202-206, 2004. 20. Arion HG: Dynamic closure of the lids in paralysis of the orbicularis muscle. Int Surg 57:48, 1972. 21. Marrone AC, Soll D: Modification of the Arion encircling silicone spring. Thesis for membership in the American Society of Ophthalmic Plastic and Reconstructive Surgery, 1977. 22. Wood-Smith D: Experience with the Arion prosthesis. In Tessier P (ed): Symposium on Plastic Surgery in the Orbital Region. St. Louis, Mosby, 1976. 23. Beard C : Canthoplasty and brow elevation for facial palsy. Arch Ophthalmol 71:386-388, 1964. 24. Tenzel R R : Treatment of lagophthalmos of the lower lid. Arch Ophthalmol 81:366-368, 1969. 25. Tenzel R R , Buffam FV, Miller G R : The use of the lateral canthal sling in ectropion repair. Can J Ophthalmol 12:199-202, 1977.
62
Hypoglossal Facial Anastomosis Eric P. Wilkinson and William M. Luxford Videos corresponding to this chapter are available online at www.expertconsult.com.
Facial nerve paralysis is a debilitating problem cosmetically and functionally. During the course of otologic and neurotologic surgery, sacrifice of the facial nerve is sometimes necessary, and may occur inadvertently, despite meticulous surgical technique. In these cases, one must be prepared to rehabilitate and restore function as much as possible. Direct repair of the injured nerve is currently the best option available to re-establish facial function. If a direct approximation of the nerve ends is impossible, a graft connecting the two ends is the next best choice. There are situations in which neither of these options is feasible. Perhaps the most common of these involves the extirpation of cerebellopontine angle tumors, in which the facial nerve is severed at the brainstem, and there is no proximal stump present for connection of a graft. There are also cases in which a very attenuated but intact nerve regains no function. In these situations, an alternative to direct repair and nerve grafting is required. Ideally, facial nerve restoration procedures should provide normal facial tone and symmetry, strong volitional and emotional facial movement, protection of the eye, facilitation of mastication, avoidance of dyskinesias, and no additional motor deficits. Even immediate direct anastomosis cannot fulfill these criteria, however. Several methods to restore some facial function have been developed that require neither direct repair nor grafting. These include cross-facial nerve grafting; nerve/muscle pedicle grafts; and nerve substitutions, such as phrenic, accessory, hypoglossal, and ansa cervicalis. Connection of a graft with the normal facial nerve followed by redirection of some of these fibers to the paralyzed side is termed cross-facial grafting. This method can provide some symmetry of movement, while avoiding other motor deficits. This procedure partially compromises the normal nerve, however, and provides a scant supply of neural elements to the recipient muscles, leading to inconsistent results.1,2 As a consequence, this procedure has not met with widespread acceptance.3,4 Combinations of cross-facial grafting and traditional nerve substitutions may also be employed.5-8 Nerve/muscle pedicle grafts such as temporalis muscle transfer have been used with some success, but have also
yielded inconsistent results.9 Because the results of nerve/ muscle pedicle grafts and cross-facial grafts have been disappointing, modifications of these procedures have been grafting and developed that use combination cross-facial�������������� microvascular free muscle flaps for facial reanimation.9,10 More recently, a modification of the traditional temporalis transfer has been used that incorporates the temporalis tendon alone instead of a portion of the muscle.11 Nerve substitution procedures provide a large supply of axons to the recipient muscles, and are generally technically facile. The results are consistent and predictable. The major disadvantages include the lack of emotional facial function and the donor nerve deficit. Because of the close relationship of tongue movement to facial movement, the hypoglossal/facial (CN XII/VII) anastomosis has proved to be a useful procedure in cases of facial paralysis in which a direct repair or graft is impossible. This procedure produces varying degrees of tongue dysfunction, based on the amount of hypoglossal nerve donated for transposition to the facial nerve. Newer techniques have been designed with the goal of preserving tongue function. This chapter focuses on the hypoglossal/ facial anastomosis.
PATIENT SELECTION Patients with facial paralysis must receive a detailed evaluation to determine the etiology of the paralysis. In some cases, the cause is obvious, as in resection of the nerve in the course of removal of a neoplasm. The evaluation of facial palsy of unknown cause is beyond the scope of this chapter. Careful evaluation should precede any reanimation procedure, however, to avoid missing treatable disease, and to avoid destruction of a nerve that has potential for return of function. In cases of known facial nerve discontinuity in which direct repair or grafting is impossible, the CN XII/VII anastomosis should be performed as soon as reasonably possible. Muscle atrophy and degeneration proceed rapidly after denervation.12 Early repair provides axonal growth to the muscles and limits the amount of muscle degeneration. 755
756
OTOLOGIC SURGERY
The severed nerve also begins to experience fibrosis.13,14 In early anastomoses, new axons fill the nerve sheath before fibrosis and potentially allow a greater supply of axons to the muscles. Although earlier anastomosis gives a better functional result, the CN XII/VII anastomosis is also effective after a prolonged denervation and should be considered up to 2.5 years after injury.4,15 Return of function can occur 4.5 years after injury.16 In other patients in whom the continuity of the nerve is in question, including patients who sustain trauma, have idiopathic palsy, and have nerves damaged in surgery, it is prudent to wait at least 1 year to ensure that no return of function occurs. Electrophysiologic testing is helpful in determining the innervation and viability of facial muscles. A positive response to electroneuronography or evoked electromyography indicates that at least some motor end plates are functional. These patients should be given the longest possible time to show improvement in function. Some of these patients have so few remaining neural elements, however, that they never regain any useful function. In this situation, a CN XII/VII graft helps provide a sufficient amount of neuronal input to the muscles. Electromyography helps detect polyphasic action potentials indicative of reinnervation, and fibrillation potentials indicative of denervation. In cases of longstanding paralysis (>2.5 years), a muscle biopsy in addition to electromyography may be useful to determine viability, atrophy, and fibrosis. In cases of severe muscle atrophy and neural fibrosis, the results of any reinnervation procedure would be poor, and muscle transfers and other augmenting procedures should be considered. The clinician should also consider the status of the contralateral twelfth cranial nerve when deciding on the CN XII/VII transposition. Contralateral hypoglossal paralysis is a contraindication to the standard CN XII/VII transposition, as are multiple lower cranial nerve deficits that already compromise swallowing and speech.
also be used for the superior aspect of the incision. This is particularly true if a mobilization of the facial nerve out of the fallopian canal is planned. Skin flaps are raised anteriorly and posteriorly. The parotid is mobilized from the anterior border of the sternocleidomastoid muscle and from the external auditory canal. The angle formed by the cartilage of the anterior external canal, known as the tragal pointer, is followed medially to the stylomastoid foramen, where the facial nerve exits the temporal bone. The nerve is dissected from the parotid gland to expose the pes anserinus and free the main trunk from the gland. The nerve is transected at the stylomastoid foramen. The hypoglossal nerve is identified by retracting the sternocleidomastoid muscle posteriorly and exposing the great vessels of the neck. The posterior belly of the digastric muscle is retracted superiorly, and the hypoglossal nerve is found coursing inferiorly with the great vessels and then turning anteriorly as it supplies the ansa cervicalis, which descends in the carotid sheath (Fig. 62-2). The hypoglossal nerve is followed anteriorly and medially as it enters the tongue muscle. The nerve is freed from its fascial attachments in the neck. The network of veins and arteries entering the internal jugular vein and external carotid artery should be controlled during this maneuver. After the nerve is freed from its attachments, it is divided as far anteriorly as is possible to gain sufficient length. The free hypoglossal nerve is rotated superiorly. Directing the nerve medial to the digastric muscle in this rotation gives the most length, but is unnecessary for a satisfactory anastomosis. There are many ways to anastomose the ends of the nerves, including collagen trays and fibrin glue, vein sheaths, laser welding, and various suture techniques. It
SURGICAL TECHNIQUE In addition to standard head and neck surgical instrumentation, hypoglossal facial anastomosis requires jeweler’s forceps to handle the nerve ends, a Castroviejo needle holder, and microforceps for knot tying. A sterile tongue blade is useful to improve visibility when the ends of the nerves are anastomosed under microscopic vision. After satisfactory general endotracheal anesthesia has been obtained with the patient in the supine position, the neck is extended, and the face is turned toward the side opposite the paralysis. The ear, face, and neck are prepared and draped using sterile technique. A standard lazy S parotidectomy incision is made in the preauricular crease and extended behind the lobule and then anteriorly about 2 cm below the angle of the mandible (Fig. 62-1). In cases where the patient has had a prior postauricular incision for craniotomy, the postauricular incision may
Incision
FIGURE 62-1. Lazy S standard parotidectomy incision is used in this procedure. Scar is well hidden in preauricular crease and in natural skin crease in submandibular area.
Chapter 62 • Hypoglossal Facial Anastomosis
is beyond the scope of this chapter to describe the various methods; however, several principles are almost universally agreed on. The two ends of the nerves should be free of all tension. This requirement is usually not a problem if the technique described earlier is followed. The ends should be cut sharply to provide a flush connection. The anastomosis should be as atraumatic as possible, yet provide strength to prevent disruption. It is also important to ensure, using frozen section histologic evaluation, that the distal facial nerve has not totally fibrosed in cases of long-standing paralysis. A conventional suture technique that yields reliable results is as follows: 1. By use of the operating microscope, the distal end of the facial nerve and proximal end of the hypoglossal nerves are stripped of the epineurium 2 to 3 mm from the cut ends.
757
2. The ends are freshened with a sharp, clean, perpendicular cut to provide a good flush connection. 3. The perineurium is approximated with two or three 9-0 nylon sutures (Fig. 62-3). 4. The wound is closed in layers over a Penrose drain. 5. A fluffed, snug parotidectomy dressing is applied, and the drain is removed the next day. Perioperative antibiotics are unnecessary.
NERVE GROWTH FACTORS AND CONDUITS Nerve grafts are the traditional technique to bridge the gap between two ends of nerve to be anastomosed. Because nerve repairs need to be tension-free, grafts are required if the ends of the nerve cannot be freely coapted. If a way could be developed to bridge this gap without
FIGURE 62-2. Parotid gland is mobilized anteriorly and superiorly as sternocleidomastoid muscle is retracted posteriorly, exposing facial and hypoglossal nerves.
FIGURE 62-3. Proximal end of hypoglossal nerve is anastomosed to distal end of facial nerve. Care is taken to use maximal length of each nerve to achieve tension-free anastomosis.
758
OTOLOGIC SURGERY
VII VII XII
XII Distal XII
Graft
Needle through XII
XII
Ansa hypoglossi n.
Ansa hypoglossi n. “Jump” graft partial XII to end of VII
FIGURE 62-4. In CN XII/VII jump graft, greater auricular or sural nerve graft is harvested. Facial nerve is cut at stylomastoid foramen. Hypoglossal nerve is cut across half its diameter using Keith needle and scalpel blade (see inset). Tension-free anastomoses are created between jump graft, facial nerve, and hypoglossal nerve.
using a nerve graft, that would be a potential improvement. Artificial conduits are one potential replacement for nerve grafts.17,18 Collagen tubules have been used to provide a pathway for axonal growth to occur. Other substances that have been used to create conduits include hydrogels19 and collagen-filled vein grafts.20 Nerve ends to be repaired are placed inside these conduits and either directly coapted in the conduit or placed at some distance from one another. One manufacturer (Integra, Plainsboro, NJ) recommends that if the nerve ends are not directly in contact, saline should be instilled to fill the tubule. The nerve ends may be attached to the tubule by one epineurial suture. Several different nerve growth factors have been used to augment nerve graft repair and in collagen tubules. Brainderived neurotrophic factor is an agent that has shown promise when used in collagen tubules or when applied directly to nerve repairs.21,22 Other agents that have been employed include neurotrophin-3 and fibroblast growth factor 1.23 Ciliary neurotrophic factor, a neurocytokine, has also been used in nerve repairs and may increase the rate of recovery over brain-derived neurotrophic factor alone.24
NEWER MODIFICATIONS OF HYPOGLOSSAL-FACIAL ANASTOMOSIS One modification of the standard CN XII/VII crossover graft uses a jump graft between the CN XII and VII. This procedure, as described by May and others,25,26 succeeds
in preserving tongue function and reinnervating the facial muscles. The exposure of CN XII and VII is identical to that described earlier (see Fig. 62-2). A 5 cm length of great auricular nerve is harvested. CN XII is cut through on half of its diameter in a beveled manner. The incision should be made in a portion of the nerve distal to the divergence of the ansa cervicalis to avoid tapping the fibers that constitute this nerve (Fig. 62-4, inset). Stimulation of CN XII with a nerve stimulator proximal to the partial transection should confirm preservation of tongue function. The great auricular graft and the distal segment of the facial nerve are prepared for anastomosis in the manner described for the standard CN XII/VII crossover. The graft is sutured to the proximal segment of the partially severed CN XII. The other end of the graft is sutured to the prepared distal end of the facial nerve (see Fig. 62-4). Two or three sutures are used for this connection. Enough length of graft must be used to avoid tension at the sites of anastomosis. The wound is closed as in the standard procedure. The patient is usually kept in the hospital overnight and discharged the following morning after removal of the drain. Although the patient may have some trouble with pooling of food in the ipsilateral oral vestibule, no special diet is necessary. Pitfalls to avoid in this surgery include use of the ansa cervicalis branch of CN XII instead of CN XII itself. This particular method leads to a much weaker result. The surgeon should also ensure that the distal facial nerve is not fibrosed, and that the facial muscles are still viable.
Chapter 62 • Hypoglossal Facial Anastomosis
759
“Jump” graft partial XII to end to VII with Neurogen tubule. Epineurium Fascicles Perineurium Capillary
Nerve trunk VII
Axons
Arteriole venule Intrafascicular epineurium
Axon
Neurogen tubule Graft Distal XII
X
Ansa hypoglossi Neurogen tubule
Saline
FIGURE 62-5. Collagen tubules may be used as adjunct to coaptation of jump graft and facial nerve or hypoglossal nerve. With collagen tubules, one simple epineurial suture is used between nerve and tubule. Tubule may be filled with saline, or ends of nerve may be directly coapted inside tubule.
In another variation of the jump graft hypoglossal/ facial anastomosis, a collagen tubule may be used as an adjunct to the anastomosis between the facial nerve and the jump graft (Fig. 62-5). Collagen tubules, described previously, may be used to facilitate direct coaptation of the nerve ends, or saline may be placed in the tubule between the nerve ends. A single epineurial suture is used to connect the nerve graft to the collagen tubule (see Fig. 62-5, inset).
End-to-Side Anastomosis via Translocation of Facial Nerve from the Fallopian Canal A more recent technique consists of rerouting the facial nerve from the fallopian canal and connecting it directly to the hypoglossal nerve through an end-to-side anastomosis.27 In this technique, the facial nerve is decompressed from the stylomastoid foramen inferiorly to the geniculate ganglion superiorly. The greater superficial petrosal nerve is transected at the geniculate ganglion, and the facial nerve is mobilized inferiorly. The nerve itself is traced through the periosteum of the stylomastoid foramen until the pes anserinus is reached (Fig. 62-6).
When the pes is reached, the nerve is of sufficient length for the proximal end of the nerve to be anastomosed to a partial (“end-to-side”) transection of the hypoglossal nerve, either through a direct anastomosis or with a collagen tubule (see Fig. 62-6). This technique usually provides enough length to avoid the use of a jump graft. The primary advantage of this technique is a single facial/hypoglossal anastomosis with the ability to use a partial hypoglossal donor nerve. In this manner, an efficient anastomosis may be achieved with minimal tongue weakness.
RESULTS A patient with facial paralysis who is considered a candidate for CN XII/VII crossover should be counseled about the expected results from this procedure. The patient cannot expect normal facial function for several reasons.3,4,28-32 As in a direct nerve repair, the axons directed to specific muscles find a random path. The muscles of the face include agonists and antagonists for various expressions and movements. When agonist and
760
OTOLOGIC SURGERY
Facial n. completely mobilized
Swingdown of facial nerve
Facial n. and divisions Parotid gland VII
XII
XII
Ansa hypoglossi
FIGURE 62-6. Translocation of intratemporal facial nerve with direct anastomosis between facial nerve and split end of hypoglossal nerve (end-toside anastomosis). Collagen tubules (not shown) may also be used to facilitate anastomosis.
antagonist muscles are simultaneously stimulated, the result is a canceling effect. This effect is similar to that occurring during stimulation of a flexor and extensor muscle at the same time. The hypoglossal nerve controls different muscle groups, allowing training of facial movement as the patient attempts various tongue motions. The training is successful, but is not helpful in emotional facial response. This procedure does not reproduce the blink reflex, even though some investigators have shown a trigeminal-hypoglossal reflex.28 Problems with xerophthalmia and exposure keratitis may require adjunctive lid procedures, such as a palpebral spring or gold weight lid implant. Within 4 to 6 months, the patient begins to see tone in the muscles and a resting symmetry.3,4,13,32 With a rehabilitation program, volitional movement is possible, allowing the patient to smile with tongue movement. Electromyographic feedback–enhanced rehabilitation has shown some additional benefits.1,13,33 Because many facial movements are also coordinated with oral function, the CN XII/VII anastomosis improves the patient’s ability to eat by providing tension to the buccal area and keeping the bolus in the oral vestibule. The natural interaction between the facial and hypoglossal nerves in eating, swallowing, and speaking facilitates the rehabilitation seen with this crossover graft as opposed to the accessory or phrenic nerve. Because of the nonselective nature of reinnervation, movement of the face results in synkinesis and mass movement that varies from patient to patient. Synkinesis can be reduced by exercise and biofeedback early in recovery.33 Selective section of branches of the facial
TABLE 62-1 F acial Nerve Function after Cranial
Nerve XII-VII Anastomosis: Quality of Return Criteria before HouseBrackmann Grading Scale
Poor Fair Good Excellent
Tone without symmetry or movement Tone, symmetry, limited movement Tone, symmetry, fair movement, moderate synkinesis Tone, symmetry, good movement, mild synkinesis
From Luxford WM, Brackmann DE: Facial nerve substitution: A review of sixty-six cases. Am J Otol Suppl 55-57, 1985.
nerve is also useful in severe cases, as is the use of selective botulinum toxin injections.34 Because the function gained by reinnervation procedures cannot compare with normal facial function, a different method of grading facial function is used in evaluation of the results of the CN XII/VII crossover. A grading system used in a prior analysis of CN XII/VII crossover patients in our facility is presented in Table 62-1. Although the methods used to evaluate results vary from study to study, we have extrapolated the grading system in Table 62-1 to provide the results from several sizable studies (Table 62-2). This analysis should give the reader a good idea of reasonable expectations from this procedure. More recent studies, particularly studies reporting on results for jump interposition grafts or translocation of the facial nerve with end-to-side anastomosis, use
761
Chapter 62 • Hypoglossal Facial Anastomosis TABLE 62-2 Results of Cranial Nerve XII-VII Anastomosis from Selected Studies Study al30
Sabin et Pensak et al29 Luxford and Brackmann32 Gavron and Clemis37 Conley and Baker4 Immediate Delayed
N
No. Follow-up
134 61 54 36
13 0 6 6 NA NA
94 43
Poor (%)
Fair (%)
Good (%)
9 10 8 7
48 48 32 20
43 39 35 33
5 30
18 29
77 41
Excellent (%) *
3 25 40 * *
NA, not available. * These studies did not use an “excellent” designation.
TABLE 62-3 Results of More Recent Studies on Variations of Hypoglossal-Facial Anastomosis Study
N
Type
May et al25 Hammerschlag26 Atlas and Lowinger27 Sawamura and Abe38 Darrouzet et al39 Rebol et al40 Godefroy et al41
20 17 3 4 6 5 7
Jump* Jump* E-S E-S E-S E-S E-S
H-B I (%)
H-B II (%)
H-B III (%)
23†
80 65 75 83 40 86
H-B IV (%) 15 12 100 25 17 20 14
H-B V (%)
H-B VI (%)
5
20
20
Mild Tongue Dysfunction (%) 15 0 0 0 33 0 0
*Jump
= interposition graft. in the chapter text. Because of significant synkinesis with nerve grafting, a grade II result is unlikely. E-S, end-to-side direct anastomosis of transposed facial nerve to partial hypoglossal transection; H-B, House-Brackmann grade.35 †Explained
the House-Brackmann grading scale for result reporting.35 Table 62-3 shows results from more recent studies using either jump grafting or end-to-side anastomosis. The patients in these studies were divided between early intervention (<12 months) and late intervention (>12 months, but generally <24 months), and had similar outcomes. Generally, the best result to be expected on the House-Brackmann grading scale with any nerve grafting procedure is a grade III result. This result requires movement of the forehead. Almost all patients have significant synkinesis after hypoglossal/facial anastomosis, rendering grade III the best result to be expected. The deficit incurred in the sacrifice of one hypoglossal nerve is easily overcome by most patients.4,13,32 Initially, some pooling of food in the lingual sulcus is problematic. As the buccal musculature regains tone, and as the ipsilateral tongue atrophies, this problem lessens. In Conley and Baker’s large series,4 about one fourth of the patients experienced severe or minimal atrophy, and the remaining half experienced moderate atrophy. Very few patients had trouble with speech. The best results in regaining facial function and overcoming the CN XII deficit were seen with early anastomosis compared with procedures performed on patients with long-standing paralysis.4 More recent reports indicate that the choice of technique does not cause a difference in facial functional outcomes.36 Methods that preserve tongue function may
become preferable to the standard hypoglossal/facial anastomosis.
SUMMARY The CN XII/VII crossover graft is a easy, reliable procedure in the rehabilitation of facial paralysis. A thorough preoperative evaluation is required, as is proper timing. The patient can expect return of tone and symmetry and synkinesis and mass movement. The donor deficit is not significant when measured against the benefits gained from the procedure, particularly when partial hypoglossal nerve transection and jump graft or end-to-side anastomosis is performed. Patients who are given realistic expectations are pleased with the improvement seen from this procedure.
REFERENCES 1. O’Brien B M, Pederson WC, Khazanchi R K, et al: Results of management of facial palsy with microvascular free-muscle transfer. Plast Reconstr Surg 86:12-22, 1990. 2. May M : Management of cranial nerves I through VII following skull base surgery. Otolaryngol Head Neck Surg 88:560-575, 1980.
762
OTOLOGIC SURGERY
3. Chuang DC, Wei FC, Noordhoff MS: “Smile” reconstruction in facial paralysis. Ann Plast Surg 23:56-65, 1989. 4. Conley J, Baker DC : Hypoglossal-facial nerve anastomosis for reinnervation of the paralyzed face. Plast Reconstr Surg 63:63-72, 1979. 5. Yamamoto Y, Sasaki S, Sekido M, et al: Alternative approach using the combined technique of nerve crossover and cross-nerve grafting for reanimation of facial palsy. Microsurgery 23:251-256, 2003. 6. Ueda K, Akiyoshi K, Suzuki Y, et al: Combination of hypoglossal-facial nerve jump graft by end-to-side neurorrhaphy and cross-face nerve graft for the treatment of facial paralysis. J Reconstr Microsurg 23:181-187, 2007. 7. Hadlock T, Sheahan T, Heaton J, et al: Baiting the crossface nerve graft with temporary hypoglossal hookup. Arch Facial Plast Surg 6:228-233, 2004. 8. Yoleri L , Songur E, Mavioglu H, et al: Cross-facial nerve grafting as an adjunct to hypoglossal-facial nerve crossover in reanimation of early facial paralysis: Clinical and electrophysiological evaluation. Ann Plast Surg 46:301307, 2001. 9. Tucker H M : Restoration of selective facial nerve function by the nerve-muscle pedicle technique. Clin Plast Surg 6:293-300, 1979. 10. Harrison D H : The treatment of unilateral and bilateral facial palsy using free muscle transfers. Clin Plast Surg 29:539-549, 2002. 11. Byrne PJ, Kim M, Boahene K, et al: Temporalis tendon transfer as part of a comprehensive approach to facial reanimation. Arch Facial Plast Surg 9:234-241, 2007. 12. Belal A : Structure of human muscle in facial paralysis: Role of muscle biopsy. In May M (ed): The Facial Nerve. New York, Thieme, 1986, pp 99-106. 13. Pitty L F, Tator C H : Hypoglossal-facial nerve anastomosis for facial nerve palsy following surgery for cerebellopontine angle tumors. J Neurosurg 77:724-731, 1992. 14. Ylikoski J, Hitselberger WE, House WF, et al: Degenerative changes in the distal stump of the severed human facial nerve. Acta Otolaryngol 92(3-4):239-248, 1981. 15. Kunihiro T, Kanzaki J, Yoshihara S, et al: Hypoglossalfacial nerve anastomosis after acoustic neuroma resection: Influence of the time anastomosis on recovery of facial movement. ORL J Otorhinolaryngol Relat Spec 58:32-35, 1996. 16. Hitselberger WE : Hypoglossal-facial anastomosis. In House WF, Luetje C M (eds): Acoustic Tumors, vol 2: Management. Baltimore, University Park Press, 1979, pp 97-103. 17. Archibald S J, Shefner J, Krarup C, et al: Monkey median nerve repaired by nerve graft or collagen nerve guide tube. J Neurosci 15(5 Pt 2):4109-4123, 1995. 18. Archibald SJ, Krarup C, Shefner J, et al: A collagenbased nerve guide conduit for peripheral nerve repair: An electrophysiological study of nerve regeneration in rodents and nonhuman primates. J Comp Neurol 306:685-696, 1991. 19. Belkas J S, Munro C A, Shoichet M S, et al: Peripheral nerve regeneration through a synthetic hydrogel nerve tube. Restor Neurol Neurosci 23:19-29, 2005. 20. Lee DY, Choi B H, Park J H, et al: Nerve regeneration with the use of a poly(l-lactide-co-glycolic acid)-coated collagen tube filled with collagen gel. J Craniomaxillofac Surg 34:50-56, 2006.
21. Lewin S L , Utley DS, Cheng ET, et al: Simultaneous treatment with BDNF and CNTF after peripheral nerve transection and repair enhances rate of functional recovery compared with BDNF treatment alone. Laryngoscope 107:992-999, 1997. 22. Barde YA, Edgar D, Thoenen H : Purification of a new neurotrophic factor from mammalian brain. EMBO J 1:549-553, 1982. 23. Midha R , Munro C A, Dalton PD, et al: Growth factor enhancement of peripheral nerve regeneration through a novel synthetic hydrogel tube. J Neurosurg 99:555-565, 2003. 24. Lewin S L , Utley DS, Cheng ET, et al: Simultaneous treatment with BDNF and CNTF after peripheral nerve transection and repair enhances rate of functional recovery compared with BDNF treatment alone. Laryngoscope 107:992-999, 1997. 25. May M, Sobol S M, Mester SJ: Hypoglossal-facial nerve interpositional-jump graft for facial reanimation without tongue atrophy. Otolaryngol Head Neck Surg 104:818825, 1991. 26. Hammerschlag PE: Facial reanimation with jump interpositional graft hypoglossal facial anastomosis and hypoglossal facial anastomosis: Evolution in management of facial paralysis. Laryngoscope 109(2 Pt 2 Suppl 90):1-23, 1999. 27. Atlas MD, Lowinger DS: A new technique for hypoglossalfacial nerve repair. Laryngoscope 107:984-991, 1997. 28. Stennert E: I. Hypoglossal facial anastomosis: Its significance for modern facial surgery. II. Combined approach in extratemporal facial nerve reconstruction. Clin Plast Surg 6:471-486, 1979. 29. Pensak M L , Jackson CG, Glasscock M E III, et al: Facial reanimation with the VII-XII anastomosis: Analysis of the functional and psychologic results. Otolaryngol Head Neck Surg 94:305-310, 1986. 30. Sabin H I, Bordi LT, Symon L , et al: Facio-hypoglossal anastomosis for the treatment of facial palsy after acoustic neuroma resection. Br J Neurosurg 4:313-317, 1990. 31. Chang CG, Shen A L : Hypoglossofacial anastomosis for facial palsy after resection of acoustic neuroma. Surg Neurol 21:282-286, 1984. 32. Luxford WM, Brackmann D E : Facial nerve substitution: A review of sixty-six cases. Am J Otol Suppl:55-57, 1985. 33. Hammerschlag PE, Brudny J, Cusumano R , et al: Hypoglossal-facial nerve anastomosis and electromyographic feedback rehabilitation. Laryngoscope 97:705709, 1987. 34. Dressler D, Schonle PW: Botulinum toxin to suppress hyperkinesias after hypoglossal-facial nerve anastomosis. Eur Arch Otorhinolaryngol 247:391-392, 1990. 35. House JW, Brackmann D E : Facial nerve grading system. Otolaryngol Head Neck Surg 93:146-147, 1985. 36. Guntinas-Lichius O, Streppel M, Stennert E : Postoperative functional evaluation of different reanimation techniques for facial nerve repair. Am J Surg 191:61-67, 2006. 37. Gavron J P, Clemis J D: Hypoglossal-facial nerve anastomosis: A review of forty cases caused by facial nerve injuries in the posterior fossa. Laryngoscope 94(11 Pt 1): 1447-1450, 1984.
Chapter 62 • Hypoglossal Facial Anastomosis 38. Sawamura Y, Abe H : Hypoglossal-facial nerve side-toend anastomosis for preservation of hypoglossal function: Results of delayed treatment with a new technique. J Neurosurg 86:203-206, 1997. 39. Darrouzet V, Guerin J, Bebear J P: New technique of side-to-end hypoglossal-facial nerve attachment with translocation of the infratemporal facial nerve. J Neurosurg 90:27-34, 1999.
763
40. Rebol J, Milojkovic V, Didanovic V: Side-to-end hypoglossal-facial anastomosis via transposition of the intratemporal facial nerve. Acta Neurochir (Wien) 148: 653-657, 2006. 41. Godefroy WP, Malessy M J, Tromp A A, et al: Intratemporal facial nerve transfer with direct coaptation to the hypoglossal nerve. Otol Neurotol 28:546-550, 2007.
63
Facial Reanimation Techniques Barry M. Schaitkin
The care of a patient with facial paralysis requires careful examination of the overall clinical picture, including the patient’s age, medical status, etiology of the facial paralysis, other cranial nerve defects, prior reanimation surgery, goals, and expectations. The procedures offered run the gamut from complex dynamic procedures employing microvascular techniques and tissue transfer to simpler static procedures, and frequently a combination of techniques is necessary to achieve the best results. Protection of the integrity of the cornea is paramount throughout the duration of the patient’s facial paralysis. Generally, the upper and lower face receive separate consideration regarding functional and cosmetic procedures for the eye, and recreation of symmetry and function for the mouth. The reader is directed to Chapter 61 for a comprehensive discussion of eye reanimation for these patients and to Chapter 62 for an in-depth discussion of hypoglossal facial anastomosis.
DYNAMIC PROCEDURES FOR FACIAL REANIMATION Nerve Grafting Procedures Patient Selection The ideal reanimation procedure in any patient with disruption of the facial nerve is reconstitution of the nerve by primary repair or by grafting to re-establish continuity between the facial nerve nucleus and facial musculature. The procedure should be done in a clean field as soon as possible after the injury. Factors affecting results include the condition of the nerve, the time after onset of paralysis, the presence of tumor in the nerve, and the adherence to microsurgical techniques of nerve repair. Ideally, the procedure should be done in the first 30 days; if not, and the results of repair are disappointing after 1 year.1
Surgical Technique After trimming the ends of the nerve to achieve a healthy nerve for anastomosis devoid of trauma, debris, or neoplasm, the surgeon must assess the need for a graft.
A primary repair offers no advantage if the two nerves are under tension, and this can be assessed if the nerves stay in good approximation without sutures. The fewest number of sutures necessary to achieve stable coaptation of the nerve suffice. Intracranially, sutures may be limited technically to one or two 9-0 or 10-0 monofilament sutures; in the temporal bone, no suturing is required; and extracranially, two or three sutures are sufficient. Nerve repair is traditionally either by epineurial or by fascicular repair techniques, and studies have not supported superiority of one over the other. Likewise, reversing grafts and clipping nonessential branches have not provided advantages.1 The use of graft is dictated by the deficit present. Most defects for the otologist can be accommodated by use of the greater auricular nerve. This branch of C-2 and C-3 is a good size match for the facial nerve, and provides two reliable branches for intraparotid use. The total length available is 7 cm. The nerve is generally close to a perpendicular line drawn at the midpoint of an imaginary line connecting the mastoid tip to the angle of the mandible. Patients should be advised of the sensory loss that accompanies this or any proposed nerve grafts that are selected for use. The medial branchial cutaneous nerve is selected when the greater auricular is unavailable, is too short, or lacks the additional necessary branching pattern.2 This graft is a sensory nerve of the arm, provides a good size match for the facial nerve, has a potential length of 20 cm, and usually has at least four branches. In comparison, the sural nerve, found behind the lateral malleolus in relation to the lesser saphenous vein, offers 35 cm in length, and an adequate branching pattern, but is generally larger in diameter than the facial nerve. We have successfully harvested the sural nerve using the minimally invasive endoscopic technique being employed for saphenous vein harvest. We have been able to achieve a fair to superb facial nerve outcome in our nerve graft patients 80% of the time. Patients with paralysis caused by malignant tumors, older patients, and patients who had delay in repair tended to have poorer outcomes.3 Wax and Kaylie4 had no difference in facial nerve outcome, however, in patients who 765
766
OTOLOGIC SURGERY
had a positive margin in the reconstructed nerve. In their series of 19 patients, they had a grade III or IV result in 50% of the patients in each group.
Nerve Substitution Procedures Patient Selection Patients may have either a cross facial nerve graft or a hypoglossal nerve substitution if they are not candidates for a primary repair. Ideally, these procedures also should be performed in the first 30 days after injury. Results with a hypoglossal/facial jump graft are acceptable when performed 1 year after injury, and a full hypoglossal transfer can be performed 2 years after injury, although the synkinesis and tongue weakness must be considered and discussed extensively.5
Surgical Technique Hypoglossal jump grafting is performed via the surgeon’s preferred standard parotid incision; either a facelift or modified Blair is acceptable. Standard landmarks of the tragal pointer and digastric muscle are used to locate the facial nerve. An additional useful landmark is the mastoid tip. The facial nerve generally is located in a superoinferior aspect at the midpoint of the surgeon’s finger when placed on the mastoid tip at the depth of the digastric muscle. The facial nerve is isolated from this point out to the level of the first bifurcation with meticulous hemostasis. Transection of the nerve should be done without transecting the epineurium on the posterior aspect of the nerve. Preserving the epineurium at this point allows for ease of microscopically performed neurorrhaphy, preventing the retraction of the distal nerve into the parotid tissue. The hypoglossal nerve is found in relation to the posterior aspect of the digastric muscle. Visualization of the nerve usually requires retraction of the digastric superiorly. The nerve always passes lateral to the carotid artery, and retrograde dissection from this point is sometimes helpful, especially in the presence of significant adipose tissue. Using the nerve just distal to the ansa cervicalis allows for better therapy to retrain the patient’s smile because only tongue motion fibers are redirected up to the facial nerve. A small Penrose drain can be passed behind the nerve at this location and tightened using a self-retaining retractor. This drain gently elevates the hypoglossal nerve into the field and avoids obscuring tissue fluid at the time of repair. This procedure is being discussed in greater detail in Chapter 62. The cross-face anastomosis was originally described by Scaramella6 and modified over the years by many surgeons. It is most useful when powering a free muscle graft for recreating a smile. It has been of more limited usefulness when trying to reanimate the entire facial nerve from a branch or branches of the contralateral side, which is why my preference has been for the jump graft in most patients.
Temporalis Muscle Transposition Patient Selection Patients who may be candidates for temporalis muscle transposition include patients (1) who have absent or poor facial function, either with spontaneous recovery 2 years after the onset of paralysis or 2 years after nerve repair or nerve grafting; (2) who are not candidates for or refuse facial nerve repair or grafting or facial/hypoglossal nerve grafting; (3) who have neurofibromatosis, ipsilateral CN X paralysis, or another condition that is a contraindication to facial/hypoglossal nerve grafting; and (4) who have undeveloped facial nerves or facial musculature, such as may occur with Möbius syndrome.3,4 The procedure has also been recommended to give reasonable function and symmetry while awaiting the results of a nerve graft to occur, and to augment the results of facial nerve repair or grafting.7
Patient Evaluation Systematic assessment of facial function in a patient who has elected to undergo temporalis muscle transposition includes evaluation of all areas of the face at rest and with smiling. The positions of the nasal alae, depths of nasolabial creases, and nasal airway structures are evaluated. The appearance of nasolabial structures should be considered in the planning for temporal muscle transposition. Airway structures are evaluated because nasal valve collapse may have occurred, and if a nasal septal deformity is also present, nasal obstruction could result. Nasal obstruction may indicate the need for a nasoseptoplasty procedure to be performed at the time of temporalis muscle transposition. The patient’s smile on the unaffected side is classified, as described by Rubin,8 as (1) corner-of-the-mouth, or “Mona Lisa” (67% of the population); (2) canine, or “Jimmy Carter” (31% of the population); or (3) fullmouth, or “Lena Horne” (2% of the population).8 The appropriate smile can be partly recreated on the affected side by temporalis muscle transposition with careful consideration of how the various muscles of the mouth contract to form each type of smile. Finally, to make an informed decision for surgery, the patient must understand what is realistically achievable in his or her case. The surgeon needs to discuss with the patient possible results of facial reanimation surgery. Spontaneous mimetic expression can be restored only with facial nerve reinnervation; however, temporalis muscle transposition can provide significant improvement in appearance and function of the paralyzed face.
Surgical Technique Traditional temporalis muscle transposition is performed with the patient under general anesthesia. Small series have been reported with nontransposed temporalis
Chapter 63 • Facial Reanimation Techniques
muscle.9 The patient is positioned supine for surgery with the head in a doughnut head holder and turned so that the affected side is exposed. Hair is trimmed with scissors on either side of the part, and the entire surgical site is prepared. The scalp and lip/cheek incision sites are infiltrated with a solution of 1% lidocaine with 1:100,000 epinephrine to improve hemostasis. The initial incision is made in the scalp with a blade and continued through subcutaneous tissue and loose aponeurotic tissue with cutting, needle-tip cautery. After the temporalis muscle fascia has been identified, it is widely exposed from the zygomatic arch to just above the superior temporal line. A 4 cm wide segment (about two fingerbreadths) of the midportion of the muscle is outlined with the cautery (Fig. 63-1). A heavy periosteal elevator is used to elevate the muscle off the squamous portion of the temporal bone, beginning superiorly and moving inferiorly to the level of the zygomatic arch (Fig. 63-2). Care must be taken as the medial aspect of the muscle is elevated inferiorly to preserve its neurovascular supply from deep temporal nerves and vessels. Next, a tunnel is made into which the temporalis muscle is transposed. The tunnel is begun by developing a plane deep to the hair follicles and superficial to the superficial musculoaponeurotic system, in the direction of the corner of the mouth. Remaining superficial to the superficial musculoaponeurotic system protects underlying facial nerve branches, which is particularly important in patients with some intact facial function or in whom a chance exists for spontaneous recovery or for whom a facial nerve reinnervation procedure is planned. The lip/ cheek incision is then made. This incision can be made in the nasolabial crease or in the vermilion/cutaneous border, depending on surgeon preference. The pockets off the scalp and lip/cheek incisions are connected to form a tunnel large enough to accommodate two of the surgeon’s fingers. After the tunnel has been made, the temporalis muscle flap is bisected longitudinally, creating two 2 cm wide pedicles. A 2-0 polypropylene (Prolene) suture is placed through each pedicle in a figure-eight, and the needle is left on the suture (Fig. 63-3). Large clamps are used to pull the needles with sutures and attached muscle pedicles through the subcutaneous tunnel (Fig. 63-4). The pedicles of the temporalis muscle are sutured to facial muscle, if present, and submucosal layers so that one slip is above the oral commissure and one slip is below the commissure (Fig. 63-5). Additional sutures are used to secure the muscle so that the corner of the mouth is pulled toward the angle between the two pedicles to create a lateral smile that is overcorrected to show the first molar (Fig. 63-6). Overcorrection is crucial because a certain degree of settling occurs in the first few weeks. The resultant tissue defect above the zygoma can be ameliorated with commercially available or harvested tissue.
767
Results The results of temporalis muscle transposition begin to be evident 3 to 6 weeks postoperatively, with the appearance of facial symmetry and resolution of the overcorrected smile. At 6 weeks postoperatively, patients are instructed to create a smile on the affected side by biting down. They learn to balance this voluntary smile with the smile on the unaffected side by practicing in front of a mirror. In some cases, these efforts can be enhanced by motor sensory re-education, a biofeedback technique in which a therapist uses electromyography to help the patient identify which muscles are being activated by voluntary effort.10 With time, the amount of conscious effort involved in creating a balanced smile decreases. The results of temporalis muscle transposition continue to improve for about 1 year after the procedure. Results are judged to be (1) excellent, if voluntary smiling results in the ability to show teeth; (2) good, if voluntary smiling moves the corner of the mouth; (3) fair, if the face is symmetric at rest; and (4) poor, if no improvement is noted. Good-to-excellent results may be expected in about 85% of patients; 10% of patients have some spontaneous emotion movement of the face. In addition, myoneurotization of denervated facial muscles may occur via trigeminal nerve fiber extension. Fair (10% of cases) or poor (5% of cases) results of temporalis muscle transposition may often be improved with revision surgery.
Complications The main complications are hematoma and infection. There is a 20% incidence, however, of dehiscence of the connection of the muscle to the corner of the mouth, which can be successfully revised. Small series are available using a nontransposed temporalis muscle that avoids the defect in the temporal fossa and bulge of the transposed muscle over the zygoma. This technique has been described using either an intraoral�9 or external�11 lyexposure. A fascia graft is generally required to extend from the coronoid process to the mouth. The remainder of the procedure is the same.9,11
Free Muscle Flaps Thompson�12 first described the use of free (nonneurovascularized) autogenous muscle transplants to reanimate the paralyzed face; denervated muscle was placed in direct contact with muscle on the nonparalyzed side of the face. Subsequently, Frielinger13 introduced the use of free nonvascularized muscle grafts innervated by cross-face grafting. The techniques described by Thompson and Frielinger had limited success, but spurred interest in use of revascularized and innervated free muscle flaps. Harii and associates14 reported using a free gracilis muscle graft to reanimate the chronically paralyzed face.
768
OTOLOGIC SURGERY
Chapter 63 • Facial Reanimation Techniques
769
770
OTOLOGIC SURGERY
The vascular supply to the graft was provided by microvascular anastomosis to the superficial temporal vessels. At first, innervation was supplied by anastomosis of the graft nerve to the deep temporal nerve, but later, crossface grafting was performed instead to provide the possibility of symmetric, mimetic facial function.15 Numerous donor muscles have been proposed for free muscle graft rehabilitation of the paralyzed face, including the gracilis, rectus abdominis, serratus anterior, latissimus dorsi, and pectoralis minor.15 Ideally, the donor muscle has (1) a long neurovascular pedicle, (2) a cross-sectional area adequate to provide a flap of the width needed, (3) fiber length suitable to reproduce muscle action on the unaffected side, and (4) anatomy and physiology that permit harvesting with minimal morbidity at the donor site and small enough to avoid undesirable bulging in the face.16 Currently, an ideal candidate for a free muscle graft procedure is a young person with chronic facial paralysis who is not a candidate for facial nerve repair, nerve grafting, or a muscle transposition procedure. Candidates include patients with developmental facial paralysis and patients with atrophy or fibrosis of the distal facial nerve or musculature in whom standard reinnervation procedures are unlikely to succeed. Free muscle grafting has the disadvantage over muscle transposition of usually requiring several procedures, which also lengthens the time to final results. Harii and associates17 reported a one-stage procedure using the latissimus dorsi muscle and connecting the thoracodorsal nerve via the upper lip to contralateral facial nerve branches. In most centers, free muscle transfer has replaced regional muscle transposition for many facial nerve patients, particularly younger patients, patients who can tolerate a longer anesthetic, and patients with long survival expectations. The amount of movement with these techniques continues to disappoint in many cases, and extensive patient counseling is required.
STATIC PROCEDURE FOR FACIAL REHABILITATION Static Slings A static sling or a muscle plication procedure may be performed to elevate paralyzed lower face tissues. Although it does not affect facial function, plication of the angular elevator muscles of the mouth may improve facial symmetry at rest. Patients who may be candidates for such a procedure are patients in whom a facial reinnervation or other facial reanimation procedure has failed, and patients who are not candidates for a dynamic procedure. A static sling is performed through a nasolabial or vermilion/cutaneous incision. Fascia lata grafts or palmaris longus muscle tendon are used and continue to be superior to commercially available products. The grafts are attached to suspend the corner of the mouth and
c ollapsed nasal ala from the zygomatic arch in a procedure similar to that for temporalis muscle transposition. We have also used Gore-Tex patch for static facial suspension, but have found an extrusion rate of 10% to 30%. The grafts are first fixed to the muscle and submucosal tissue around the mouth, and the mouth or ala is elevated and slightly overcorrected by pulling the grafts toward the malar bone. The tendon or fascia grafts are fixed to the zygomatic arch with either sutures to the temporalis fascia or a miniplate and screws. In addition to this technique to suspend deeper muscles, a standard rhytidectomy procedure can be performed simultaneously to suspend sagging skin.
Lower Lip Rehabilitation Many procedures have been developed to depress the lower lip during smiling to create a “full-mouth” smile. Patients who have complete facial paralysis are not candidates for such a procedure because depression of the lower lip would decrease oral competence, particularly if a procedure was also performed to elevate the corner of the mouth. Rehabilitation of the paralyzed lower lip is appropriate, however, when this is an isolated problem. One method for rehabilitating the lower lip is to transpose the tendon of the anterior belly of the digastric muscle to the paralyzed orbicularis oris muscle.18 A tunnel is created between the tendon of the anterior belly of the digastric muscle and the lower lip depressor muscles. The anterior belly of the digastric muscle is left attached to the mandible, and the tendon is brought through the tunnel and attached to the lip depressor muscles. This procedure provides a symmetric smile in a patient with isolated lower lip paralysis because downward pull of the digastric muscle tendon counteracts upward pull of muscles elevating the lips. In patients with oral incompetence, a procedure to reduce the size of the oral sphincter and transpose innervated muscle from the normal side to the denervated side can improve oral sphincter function. One such cheiloplasty procedure is V-wedge excision of a portion of the paralyzed lower lip; another is commissure Z-plasty.1
Surgical Management of Hyperkinesis Some degree of synkinesis, hypokinesis, and hyperkinesis accompanies reinnervation of the face, whether nerve regeneration occurs with nerve grafting or nerve substitution techniques or with spontaneous recovery from a denervating injury. Synkinesis can be improved by sensorimotor re-education, in which the patient practices in front of a mirror, with the help of electromyography, to separate facial muscle activities.19,20 Hyperkinesis may be treated medically or surgically.21 Botulinum toxin injected into muscles involved in hyperkinesis causes temporary paralysis and temporary relief from hyperkinesis.
Chapter 63 • Facial Reanimation Techniques
When the effects of the botulinum toxin dissipate (3 to 6 months after injection), injection can be repeated. Surgically selective neurolysis or regional myectomy can provide longer lasting treatment for hyperkinesis. Selective neurolysis involves weakening or paralyzing innervation to the hyperkinetic muscle. The results of neurolysis are difficult to predict, however, and hyperkinesis may return, even after excision of a segment of nerve. For these reasons, regional myectomy is the currently preferred surgical technique for management of hyperkinesis in patients who fail or who are unwilling to use botulinum toxin. Hyperkinesis of the oral levator muscles, which results in pulling of the mouth to the affected side, can be improved by selective resection of the zygomaticus and levator labii superioris muscles. The problem with paralyzing or resecting these muscles is oral commissure drooping; such a resection must be done conservatively. Chin spasm may be improved by mentalis myectomy, which is performed through a submental incision. Platysma hyperkinesis, which results in unsightly cords being evident in the neck, usually can be treated satisfactorily by excision of a portion of this muscle through a horizontal cervical incision.
SUMMARY Surgical rehabilitation of the paralyzed face is a challenging, yet rewarding, area of specialization. When a patient is not a candidate for a standard facial reinnervation procedure, or when such a procedure has failed, a combination of static and dynamic procedures may successfully improve the appearance and function of the face. Facial symmetry, eyelid closure, and balanced smile usually can be restored by standard techniques, and recent innovations and future developments in this field promise the possibility of re-establishing spontaneous mimetic motion of the face.
REFERENCES 1. May M, Schaitkin B (eds): May’s the Facial Nerve, 2nd ed. New York, Thieme, 2000. 2. Haller J R , Shelton C : Medial antebrachial cutaneous nerve: A new donor graft for repair of facial nerve defects at the skull base. Laryngoscope 107:1048-1052, 1997. 3. Bascom D A, Schaitkin B M, May M, Klein S : Facial nerve repair: A retrospective review. Facial Plast Surg 16:309-313, 2000.
771
4. Wax M K, Kaylie D M : Does a positive neural margin affect outcome in facial nerve grafting? Head Neck 29: 546-549, 2007. 5. May M, Sobol S M, Mester SJ: Hypoglossal-facial nerve interpositional jump graft for facial reanimation without tongue atrophy. Otolaryngol Head Neck Surg 204:818826, 1991. 6. Scaramella L F: Cross-face nerve anastomosis: Historical notes. Ear Nose Throat J 75:347-352, 1996. 7. May M : Muscle transposition for facial reanimation. Arch Otolaryngol Head Neck Surg 110:184-189, 1984. 8. Rubin L : Reanimation of the Paralyzed Face. St Louis, Mosby, 1977. 9. Croxson G R , Quinn M J, Coulson S E : Temporalis muscle transfer for facial paralysis: A further refinement. Facial Plast Surg 16:351-356, 2000. 10. Sobol S M, May M, Mester S : Early facial reanimation following radical parotid and temporal bone tumor resections. Am J Surg 160:382-386, 1990. 11. Byrne PJ, Kim M, Boahene K, et al: Temporalis tendon transfer as part of a comprehensive approach to facial reanimation. Arch Facial Plast Surg 9:234-241, 2007. 12. Thompson N: Autogenous free grafts and skeletal muscle. A preliminary experimental and clinical study. Plast Reconstr Surg 48:11, 1971. 13. Frielinger G: A new technique to correct facial paralysis. Plast Reconstr Surg 56:44-48, 1975. 14. Harii K, Ohmori K, Torii S: Free gracilis muscle transplantation with microneurovascular anastomoses for the treatment of facial paralysis. Plast Reconstr Surg 57:133143, 1976. 15. O’Brien B M, Pederson WC, Khazanchi R K, et al: Results of management of facial palsy with microvascular free-muscle transfer. Plast Reconstr Surg 86:12-22, 1990. 16. Wells M D, Manktelow RT: Surgical management of facial palsy. Clin Plast Surg 17:645-653, 1990. 17. Harii K, Asato H, Yoshimura K, et al: One-stage transfer of the latissimus dorsi muscle for reanimation of a paralyzed face: A new alternative. Plast Reconstr Surg 102:941-951, 1998. 18. Conley J, Baker DC, Selfe TW: Paralysis of the mandibular branch of the facial nerve. Plast Reconstr Surg 70:569576, 1982. 19. Diels H J: New concepts in nonsurgical facial nerve reha bilitation. In: Meyers E, (ed.) Advances in Otolaryngology/Head and Neck Surgery. St Louis, Mosby—Year Book, 1995, pp 289-315. 20. Diels H J: Facial paralysis: Is there a role for a therapist? Facial Plast Surg 16:361-364, 2000. 21. May M, Croxson G R , Klein S R : Bell’s palsy: Management of sequelae using EMG rehabilitation, botulinum toxin, and surgery. Am J Otol 10:220-229, 1981.
64
Intraoperative Neurophysiologic Monitoring Adrien A. Eshraghi, Sarah S. Connell, Ray C. Chang, and Fred F. Telischi
Intraoperative neurophysiologic monitoring is a valuable tool for improved patient outcomes because it permits the surgeon to evaluate functional changes in structures at risk. Facial nerve monitoring has reached a level of consistency that makes it a state-of-the-art adjunct to lateral and posterior skull base approaches. Techniques for monitoring auditory function continue to evolve.
MONITORING FACIAL NERVE History Krause1 first described facial nerve monitoring in 1912 using a faradic stimulation during cochlear nerve section for tinnitus. Twitching of the ipsilateral facial muscles during stimulation helped him preserve the facial nerve, and the patient had transient facial weakness postoperatively. In the 1960s, dedicated facial nerve monitoring systems were developed. The Hilger stimulator2 was used principally in the assessment of facial paralysis, but was also used during surgery. Further developments in facial nerve monitoring occurred in the 1970s and 1980s. Delgado and colleagues3 described the use of electromyography (EMG) monitoring in cerebellopontine angle (CPA) surgery. Moller and Jannetta4 combined the specificity of EMG recording with the advantage of acoustic feedback to the surgeon.
Indications The underlying principle for intraoperative monitoring is that some types of injury can be reversed. Although introduction of facial nerve monitoring has improved functional outcomes further, especially in medium and large CPA tumors, facial nerve monitoring should never replace anatomic knowledge, technical skill, and clinical judgment. Facial nerve monitoring can be useful to identify the facial nerve when it is not clearly visible in the surgical field. Other benefits include localizing distal nerve fragments in trauma cases, and identifying sites of nerve compression, as long as wallerian nerve degeneration has not completely occurred. Otologic procedures in which the nerve is at risk include cochlear implantation, revision
tympanomastoidectomy, and repair of external auditory canal bony stenosis. Monitoring may augment safety in cases where the anatomy is altered by infection, trauma, or congenital malformation. It may also be beneficial in training centers where some portions of operations are performed by less experienced surgeons.
Physiology Stimulation of a motor nerve by electric, mechanical, or thermal means results in depolarization of the nerve. A compound muscle action potential is the composite electric activity within the target muscle resulting from synchronous activation of a group of motor neurons within a nerve bundle. EMG monitoring essentially measures this electric activity in the portion of the muscle near the recording electrodes and converts the electric activity to sound via a loudspeaker with or without visual oscilloscope display. Synchronous activity initiated by electric stimulation produces a biphasic and well-defined waveform. Asynchronous activity, as is generally produced by mechanical stimulation, produces a polyphasic pattern. When these response patterns are converted to sound, electric stimulation results in well-defined pulsed sounds, whereas mechanical stimulation yields a rough, almost scratchy, burst of acoustic energy.
Electromyography Subdermal needle electrodes are most commonly used for facial nerve monitoring. They have the advantages of ease of use, low impedance, and stability (less likely to be displaced than surface discs or cups). Monitoring more than one muscle provides additional sensitivity and redundancy.5 For two-channel bipolar recording, a typical sensing montage includes a pair of electrodes in the orbicularis oculi approximately 1 cm apart and another pair in the orbicularis oris (Fig. 64-1). The ground electrode is placed in the forehead, and the anode for the monopolar nerve stimulator is inserted at the ipsilateral shoulder. The operating room is a noisy place with abundant 773
774
OTOLOGIC SURGERY Vertex
VII Orbicularis oculi m.
VII Orbicularis oris m.
Trapezius m. SCM
FIGURE 64-1. Electrode placement for facial nerve intraoperative monitoring. SCM, sternocleidomastoid.
electric interference. The connections are checked by tapping the electrodes and observing an audible and oscilloscopic response. This mechanical compound muscle action potential results in a characteristic sound from the monitor. A second check, when available, involves testing the impedance of the inserted electrodes. Individual impedance values should be less than 5 kilohm (kΩ) to avoid electromagnetic interference from other devices in the operating room. Current instrumentation uses differential amplification techniques that improve signal-to-noise ratio optimally when impedance imbalance is less than 1 kΩ between the electrode pairs. Ideally, the imbalance-to-impedance ratio should be less than 10%. Personnel who perform intraoperative monitoring must be present in the operating room to ensure all equipment is functioning properly. Many facial nerve monitoring systems are available commercially, including the Nerve Integrity Monitor (Xomed, Inc., Jacksonville, FL) (Fig. 64-2), Neurosign 100 (Smith & Nephew Richards, Inc., Memphis, TN), Brackmann II (WR Medical Electronics Co., Stillwater, MN), and NEI (Grass Instrument Co., Quincy, MA). All of these devices use EMG. The Silverstein Facial Nerve Monitor (WR Medical Electronics Co.,) is an example of a motion detector device. Some systems, such as the Silverstein Monitor, include the ability to electrify instruments to aid with monitoring.
Stimulation Pulsed stimulation may be safer and more efficacious.6 The parameters of safe nerve stimulation are 100 to 250 μs pulses with a range of 0.05 to 0.5 mA.5 The upper
limit of safe nerve stimulation varies along the course of the nerve, but animal studies have shown myelin and axonal injury using 2 mA stimulus for 3 seconds.7 A 0.5 mA stimulus applied briefly 50 times was found to manifest no functional or histologic evidence of injury in a mouse model.8 Most normal facial nerves should be stimulated with direct contact of the probe using a 100 μs pulse of 0.05 mA. Settings of 0.05 to 0.1 mA are recommended when working close to the nerve. Farther from the nerve, currents of 0.2 to 0.3 mA may be used. Higher settings may be required when the nerve is covered by bone, connective tissue, or granulation tissue. Cerebrospinal fluid or blood may shunt current from a stimulator probe. In these cases, stimulation with a constant voltage may be used. In an attempt to determine the threshold necessary to detect a surgical dehiscence of the facial nerve electrically, Choung and associates9 prospectively estimated the minimal threshold of electric current needed to change the EMG of facial muscles using the Nerve Integrity Monitor (NIM)-2 in 100 patients. They found that 43% of ears with surgical dehiscence responded to electric stimulation of 0.7 mA or less. The mean threshold of minimal electric stimulation was 0.29 mA for tympanic segments and 0.41 mA for mastoid segments. It is crucial that patients be adequately grounded to the monopolar electrocautery unit to permit safe current return. Otherwise, the electrocautery current may find a route through the nerve monitoring electrodes, causing severe burns. All patient connections must be optically and/or electrically isolated to prevent patient injury. This isolation separates the patient from power line voltages and currents. Some monitoring devices achieve this using battery power.
Chapter 64 • Intraoperative Neurophysiologic Monitoring EMG
Electrodes
Settings
Patient
775
may be obtained, however, when drilling close to or after complete transection of the nerve. Lack of burst activity during dissection may be associated with minor manipulation of a healthy nerve, significant manipulation of an already injured nerve, or a problem with the monitoring connections and instruments. Electrically stimulating the nerve at this point verifies the integrity of the nerve.
Trains
A EMG
Electrodes
Settings
Patient
EMG
Electrodes
Settings
Patient
B
Episodes of repetitive EMG activity may occur several seconds to minutes after the stimulus, making it difficult to identify the initiating factor or to modify dissection technique. As seen in Figure 64-2, trains are caused by prolonged depolarization of the nerve beyond its threshold for developing an action potential. Subsequent repetitive firing continues until the nerve repolarizes, or can no longer sustain the repetitive activation.5 The most common initiating factor is traction on the nerve. Trains may indicate significant trauma has occurred, although this is not always the case. Changes in temperature around the nerve may also precipitate train activity. When caused by cool irrigating fluid, the spontaneous activity usually subsides with warming. When a train pattern develops after laser application or cautery, however, thermal damage should be suspected. Elevated stimulation thresholds after repetitive nerve activity suggest significant injury. Higher frequency train activity (>50 Hz) with an amplitude greater than 250 μV is associated with a more ominous outcome; in a series of 51 patients, Nakao and coworkers10 found that 86% of patients with train activity having an amplitude greater than 250 μV had severe facial nerve dysfunction. More than 10 seconds of cumulative train time is associated with postoperative facial paresis.11
Anesthetic Issues
C FIGURE 64-2. Facial nerve monitor screen (NIM Response 2.0 Nerve Integrity Monitor; Medtronic ENT, Jacksonville, FL). Two-channel modes are typical for mastoid surgery. A, Burst activity, occurring with mechanical stimulation or brief electric stimulation. B, Pulsatile stimulus. C, Train activity.
Burst Activity During the course of surgery, many burst potentials may be observed and are usually not associated with significant trauma to the nerve. Bursts are brief discharges in which the stimulus and response are simultaneous (see Fig. 64-2). Burst activity occurring with gentle manipulation suggests a healthy nerve. Very similar responses
It is common practice that when facial nerve monitoring is employed, neuromuscular blocking agents are avoided after induction of anesthesia. In 2003, Kizilay and associates12 examined the effects of different levels of neuromuscular blockade (NMB) on electric stimulation thresholds of the facial nerve during otologic surgery. Minimal facial nerve stimulation causing EMG responses in the facial musculature was measured during recovery from the effects of muscular relaxants and with 25%, 50%, 75%, and 100% levels of NMB. All of the patients had detectable EMG responses of the facial musculature at the 50% and 75% levels of NMB in response to the electric stimulation (mean 0.1 mA) of the facial nerve. No responses were measured in 31% of the patients when the level of peripheral NMB was 100%. The investigators concluded that a regulated 50% level of peripheral NMB provides reliable intraoperative EMG monitoring of the facial musculature in response to electric stimulation and adequate anesthesia, with full immobilization of
776
OTOLOGIC SURGERY
the patient. Chronic injury resulting from compression from a tumor may make the facial nerve more sensitive to the effects of NMB, however.13 Spontaneous repetitive responses may be observed when the level of anesthesia becomes inadequate, and the facial muscles contract. This activity is commonly the first indication that the patient is awakening and the prelude to larger movements, such as coughing or bucking. Migration of inadequately secured electrodes may create auditory artifacts that mimic the spontaneous activity of light anesthesia.
Practical Application in Surgery Locating the Facial Nerve Monopolar stimulation is especially useful for mapping the nerve throughout its course.4 Monitoring becomes more important in cases of large tumors.5 A stronger stimulus may need to be used to confirm that the nerve is not in close proximity to an area of tumor to be removed. False-positive responses may be obtained from stimulation of the superior vestibular and cochlear nerves owing to electric current dispersion, especially within the internal auditory canal where the nerves lie in close proximity to one another. Stimulation of the trigeminal nerve may elicit motor activity that can be interpreted as facial nerve stimulation. False-negative responses are usually related to technical errors such as failure to connect the stimulus probe, anesthetic-induced muscle paralysis, or impedance imbalances. The electric current dispersion is most effectively overcome by using bipolar stimulation or a flush-tip monopolar stimulator on the lowest possible setting. A useful method to check the monitoring system is to stimulate the nerve where it is known to be available and intact. If no response can be obtained at that location, the entire setup should be inspected from the electrode placements in the facial muscles to the recording instrument. When a reliable response is elicited from the nerve, stimulation may be resumed with confidence.
Avoiding Trauma Continuous EMG monitoring provides real-time feedback, enabling the surgeon to alter technique immediately when a burst response occurs. When performing lateral skull base surgery, it is useful to maintain monitoring during wound obliteration and closure to avoid trauma to the nerve.
Postoperative Prognosis and Prognosis of Acute Injury The threshold for stimulation has been shown to correlate with postoperative facial nerve function. When using constant current stimulation with a 50 μs pulse, if the
threshold for facial nerve stimulation at the brainstem was 0.1 mA or lower, 90% of patients exhibited HouseBrackmann grade I or II function at 1 year after CPA surgery.14 Thresholds of 0.1 to 0.2 mA were associated with grade I or II function in 77% of cases.15 The amount of energy delivered to the nerve using a 50 μs pulse of 0.1 mA is equal in magnitude to a 100 μs pulse of 0.05 mA as delivered by many commercial facial nerve monitors. When testing the nerve at the root entry zone of the brainstem to determine threshold at the conclusion of the procedure, it is important to contact the nerve long enough to confirm the precisely timed pulses of electric stimulation. A single noise could be a mechanically evoked burst, which may give the surgeon a falsely optimistic estimate of nerve function. In 2005, Grayeli and colleagues16 looked at the short-term facial prognostic value of a four-channel facial EMG device in vestibular schwannoma surgery. In 89 patients, EMG detection was performed in frontal, orbicularis oculi, orbicularis oris, and platysma muscles. Postoperative facial function at 6 months was assessed as House-Brackmann grade I or II in 80%, as grade III or IV in 16%, and as grade V or VI in 4% (n = 80). A proximal threshold between 0.01 and 0.04 mA had a positive predictive value of 94% for good facial function (grade I or II). The proximal threshold was lower in patients with improving or stable facial function compared with patients with a delayed deterioration between days 8 and 30. The maximal EMG response was detected in the frontal muscle or the platysma in 27% of cases and in orbicularis oris and oculi in 73% of cases. Also in 2005, Neff and coworkers17 sought to evaluate prospectively whether the intraoperative stimulus threshold and response amplitude measurements from facial EMG can predict facial nerve function at 1 year after vestibular schwannoma resection. In 74 consecutive patients, the minimal stimulus intensity and EMG response amplitude were recorded during stimulation applied to the proximal facial nerve after vestibular schwannoma removal. Of the 74 patients, 66 of 74 (89%) had House-Brackmann grade I or II facial nerve function, and 8 of 74 (11%) had House-Brackmann grade III-VI function at 1 year after surgery. With intraoperative minimal stimulus intensity of 0.05 mA or less and response amplitude of 240 μV or greater, the authors were able to predict a House-Brackmann grade I or II outcome in 56 of 66 (85%) patients at 1 year after surgery. With these same electrophysiologic parameters, only 1 of 8 (12%) patients with HouseBrackmann grade III-VI also met this standard and gave a false-positive result. Logistic regression analysis of the data showed that a stimulus threshold of 0.05 mA or less and a response amplitude of 240 μV or greater predicted a House-Brackmann grade I or II outcome with a 98% probability. Stimulus threshold or response amplitude alone had a much lower probability of the same result, however.
Chapter 64 • Intraoperative Neurophysiologic Monitoring
777
TABLE 64-1 Studies Using Facial Nerve Monitoring to Predict Facial Nerve Function after Acoustic
Neuroma Surgery
Study
N
Lin et al, 2006
38
Grayeli et al, 200516
89
Neff et al, 200517
74
Isaacson et al, 2003
229
Fenton et al, 2002
67
Goldbrunner et al, 2000 Nissen et al, 1997
137 116
Zeitouni et al, 1997
109
Selesnick et al, 199615
49
Taha et al, 1995
20
Lalwani et al, 1994
129
Prasad et al, 1993
77
Niparko et al, 1989
29
Result With 0.3 mA stimulus, proximal-distal amplitude >50% had PPV 93% for HB grade II or better immediately postoperative Stimulation threshold of 0.01-0.04 mA had PPV 94% for HB grade II or better 180 days postoperative using four-channel EMG For <0.05 mA stimulus and response amplitude of >0.240 V, authors could predict 85% of patients with HB grade II or better 1 yr postoperative Using proximal-distal amplitude and stimulation threshold, authors developed regression function with sensitivity 89%, specificity 83%, PPV 94% in predicting HB grade III or worse in immediate postoperative period Using stimulation current and tumor size, authors developed regression function correctly describing 93% of patients at 2 yr follow-up With proximal-distal amplitude >80%, 98.4% had HB grade I at 6 mo Median threshold for HB grade II or better group was 0.100 V versus 0.725 V for HB grade III or worse group HB grade III or worse long-term was related to higher minimum stimulation current At 1 yr after surgery, 90% of patients stimulating at 0.1 mA had HB grade II or better 100% HB grade III or better initial and grade I long-term (≤28 mo) function occurred if proximal-distal amplitude >67% Of patients with stimulus of <0.2 V, 98% had HB grade II or better 1 year postoperative versus 50% if stimulus was >0.2 V When threshold after tumor removal was ≤0.2 V, 93% patients had early, and 85% had late postoperative HB grade I or II 83% HB grade I 1 wk postoperative if proximal-distal amplitude = 1 with 88% of these patients HB grade I 1 yr after surgery
HB, House-Brackmann; PPV, positive predictive value.
A similar study by Isaacson and colleagues18 looked at two independent intraoperative monitoring parameters in predicting long-term facial nerve function in 60 patients undergoing resection of vestibular schwannomas. They found that 5 of 60 (8.3%) patients showed significant long-term weakness (i.e., House-Brackmann grade III or worse). Intraoperative monitoring parameters (proximal stimulation threshold, proximal-to-distal response amplitude ratio) were accurate in predicting increased risk of long-term facial nerve dysfunction when used in a logistic regression model. Table 64-1 summarizes a selection of studies using facial nerve monitoring to evaluate postoperative prognosis in acoustic neuroma surgery38–47.
Artifactual Responses Artifacts (i.e., monitor activity that is not due to facial nerve stimulation) are common and may cause confusion. Electrocautery obliterates facial nerve responses by saturating the monitor with electric noise. Consequently, monitors have a muting function while electrocautery is in use. During muting, monitoring is disengaged. Ultrasonic aspirators also cause a large electric artifact, and may trigger a muting circuit.14 When accurate monitoring cannot be performed, many surgeons engage a member of the team to observe visually or palpate the patient’s face.
Troubleshooting the Facial Nerve Monitor Although the facial nerve monitor can give the surgeon confidence to work safely and expeditiously by verifying anatomy, when the monitor does not respond as expected, the surgeon should proceed cautiously and initiate a sequence of steps systematically to ensure that the equipment is functioning and providing adequate monitoring sensitivity. Manufacturer manuals serve as a resource for troubleshooting specific to the equipment. Direct communication with the anesthesiologist to confirm complete neuromuscular reversal is warranted. Checking along the electric circuit should be done, including electrode placement in the skin, electrode connections to the monitor, and probe connections. The volume of the monitor might have been turned down. A typical event threshold setting is 100 μV. Event threshold settings set too high may lead to a lack of a response. Conversely, false-positive results can be reduced by increasing the event threshold. The stimulus intensity should be confirmed and can be incrementally increased by 0.05 to 0.1 mA. In the middle ear, high sensitivity can be achieved with stimulus up to 0.7 mA.9 If a continuous current is used, stimulus should be changed to pulsed mode. Selesnick19 calculated the optimal stimulus duration of 50 μs, although many commercial monitors use a default pulse of 100 μs.
778
OTOLOGIC SURGERY
Fixed voltage stimulation is an option when the surgical field is not dry to compensate for current leak from the stimulator. The stimulator probe can be replaced with a stand-alone nerve stimulator. When the face is not covered, visual verification of stimulation is possible and can help in troubleshooting. Alternatively, the face may be manually felt to move under surgical drapes during stimulation. Final options include rebooting the monitor or changing the monitor counsel. Ultimately, surgical judgment should prevail; it behooves the surgeon not to perform irreversible steps, unless these potential sources of error and corrective measures are considered.
Antidromic Facial Nerve Monitoring When a nerve is stimulated, in addition to conduction distally toward the neuromuscular junction (orthodromic), an action potential is conducted proximally (antidromically). The orthodromic wave produces the M wave on muscle contraction. The antidromic electric activity can be measured and used to assess neural function as the signal is reflected back toward the muscle resulting in a more subtle F wave. Wedekind and Klug20 prospectively evaluated F wave monitoring comparing it with intraoperative EMG, and found it useful monitoring facial nerve injury. They found that transient loss of F waves portends imminent severe facial dysfunction, and reported a 100% positive predictive value for unfavorable facial outcome with a permanent loss of F waves. A major advantage of antidromic monitoring is that it may be performed under NMB. Arriaga and associates21 described another use for antidromic potentials to locate the geniculate ganglion during middle fossa craniotomy.
MONITORING HEARING Hearing depends on the integrity of the peripheral and central auditory structures and their vascular supplies. The goal of intraoperative monitoring of auditory function is to preserve hearing, not just anatomy.
Pathophysiology Operating in the internal auditory canal and CPA poses two potential mechanisms for hearing loss: interruption of blood flow to the cochlea and the cochlear nerve, and injury to neural structures of the auditory pathway.22 In some cases, preoperative audiometric tests suggest a site of the lesion for the hearing loss. An inappropriately large reduction in speech discrimination score with good pure tone average suggests tumor effect on CN VIII or brainstem, rather than on the cochlea. Intact otoacoustic emissions (OAE) with poor auditory brainstem response (ABR) suggest the same. Retraction of the cerebellum during the early stages of the operation may stretch and damage CN VIII, which
is particularly vulnerable in the CPA. Surgeons need an objective measure of hearing preservation because visual confirmation of intact anatomic structures alone is inadequate to ensure postoperative auditory function.
Preoperative Testing A standard audiogram displays auditory thresholds at frequencies important to language (500 to 8000 Hz). Various word list tests may be administered to corroborate pure tone threshold averages and to quantify speech discrimination ability. Acoustic reflex tests examine the integrity of the CN VII and VIII reflex arc. Cochlear potentials, ABR, and OAE are useful to evaluate the integrity of the auditory pathways further and to establish a baseline for intraoperative monitoring. It is important to know preoperatively whether the physiologic measures to be monitored are abnormal.
Indications Selecting appropriate cases for monitoring involves analysis of preoperative hearing, disease prognosis, and surgical approach. The role of monitoring hearing is to decrease the risk to the auditory nerve in CPA operations. Hearing can be monitored directly from the auditory nerve, and auditory evoked potentials (electrocochleography [ECoG] potentials) can be used to check the endolymphatic system in endolymphatic sac decompression operations. Operations in which monitoring of auditory function has been reported include microvascular decompression of cranial nerves, vestibular nerve section,21 and removal of vestibular schwannomas and other CPA masses. The most common application for monitoring of auditory function is the removal of vestibular schwannomas. Hearing preservation rates for surgical extirpation of vestibular schwannomas vary inversely with tumor size. There seems to be little reason to monitor hearing during planned total removal of a tumor 4 cm or larger. Conversely, auditory monitoring is ideal when removing smaller tumors or sectioning the vestibular nerve in the face of serviceable hearing. Many other factors may suggest a better prognosis for hearing preservation, including lack of tumor extension into the lateral internal auditory canal and erosion of bony walls, low behavioral thresholds (i.e., good hearing), normal ABR, and reduced caloric response on electronystagmography (i.e., superior vestibular tumor). The generally accepted limits of serviceable hearing have been 50 dB HL pure tone average threshold and 50% speech discrimination score. Tumors in only hearing ears and bilateral tumors warrant special consideration. Worse hearing may indicate a more severe insult of auditory structures by disease and poor prognosis for hearing conservation. Exceptions exist, however, for hearing improvement when tumor removal alleviates a CN VIII conduction block.
Chapter 64 • Intraoperative Neurophysiologic Monitoring
779
In 2006, Samii and coworkers23 published their outcomes in a retrospective review of 200 consecutive vestibular schwannoma resections. They found that anatomic preservation of the facial nerve was possible in 98.5% of patients. By the last follow-up examination, excellent or good facial nerve function had been achieved in 81% of the cases. In patients with preserved hearing, the rate of anatomic preservation of the cochlear nerve was 84%. The overall rate of functional hearing preservation was 51%.
General Considerations Many lateral skull base approaches involve temporal bone dissection. Fluid and bone dust may accumulate in the middle ear and external auditory canal, resulting in a conductive hearing loss. ABR, ECoG, and direct CN VIII potentials are delayed in the presence of conductive hearing loss. OAE generally cannot be measured in the presence of significant conductive hearing loss. Whenever possible, the middle ear and external canal should be isolated from manipulation when monitoring auditory function. None of the methods for measuring auditory function during surgery should be affected by the use of routine anesthetics, including NMB. Patient factors, such as core temperature, muscle activity, blood pressure, and oxygenation, may alter electrophysiologic responses, however. Small decreases in core body temperature to 34.5° C result in delays of ABR waves.23 Similarly, increased muscle activity (e.g., fasciculations in a nonparalyzed patient) or hypotension and hypoxia may interfere with accurate measurements.
Auditory Evoked Brainstem Response ABR is considered a far-field response because it represents activity measured between scalp electrodes that are placed at relatively large distances from CN VIII and brainstem generator sites. Brainstem auditory evoked potentials consist of five to seven peaks representing electric activity of auditory nerve, nuclei, and fiber tracts of the ascending auditory pathways. Peak I refers to the distal portion of the auditory nerve. Peak II represents the central portion of the auditory nerve. Peak III represents the cochlear nucleus. Peak V represents the termination of the lateral lemniscus in the inferior colliculus. Waves III, IV, and V are created by multiple generators. Intraoperative ABR monitoring setup is similar to the setup for routine office measurement (Fig. 64-3). Subdermal needle electrodes are used for recording brainstem auditory evoked potentials. Insert ear plugs are suitable for delivering sound for the recording of auditory evoked potentials. Recorded potentials should be compared with baseline recording before the beginning of the operation. To reduce the time to obtain interpretable evoked potentials, the following actions are recommended:
FIGURE 64-3. Setup montage for auditory brainstem response, including ear canal inserts, active earlobe electrode, reference electrode on low forehead, and ground electrode on high forehead.
1. Reduce electric interference from reaching recording electrodes 2. Filter recoded potentials to attenuate background noise 3. Optimize stimulus repetition rate and strength 4. Optimize electrode placement of recording electrodes to decrease electrode impedance 5. Use methods for quality controls that do not require record replication In pathologic ears, ABR morphology is likely to eteriorate and become even more variable. Amplid tude is likely to decrease, especially in the presence of peripheral hearing loss, and latencies may increase based on the location and size of the lesion. Consequently, during monitoring of a previously impaired ear, changes from a well-established baseline are generally of more use than comparison with a norm. Because wave I is less likely to be present in an impaired ear, monitoring decisions may often depend on wave V. Enhancement of the wave I response can be achieved through ECoG recordings. Selection of testing parameters for intraoperative ABR must be made with the goal of optimizing the recording of the desired response, while minimizing the interference inherent in the electrically hostile operating room environment. Click stimuli should be present at sufficiently high rates and intensities. Wherever possible, background noise must be eliminated at the source. Finally, state-of-the-art signal processing techniques should be employed to enhance the signal-to-noise ratio and minimize the number of responses averaged to obtain an adequate waveform.
780
OTOLOGIC SURGERY
Interpretation of Results During posterior fossa surgery, damage is most likely to occur at the cochlea, CN VIII, or its root entry zone at the brainstem. An increase in peak V latency with or without a decrease in peak V amplitude and no change in peak III amplitude indicates the brainstem has been manipulated. Changes in interpeak latency I-III indicate stretching or compression of the auditory nerve. Disruption of the cochlear blood supply produces a quick and complete loss of ABR. Rapid response by the surgeon to restore the blood supply, if possible, is vital to prevent permanent damage. Other types of insults, such as stretching, compression, manipulation, or heat from the drill, can affect CN VIII and its root entry zone. ABR responses to these insults are neither as quick nor as easily interpretable as the responses produced by the loss of blood flow. Much depends on obtaining an optimal preoperative baseline response before entering the operating room and again after induction of anesthesia before initiation of the procedure. Criteria for judging that a change has occurred include increases in absolute latency of wave V from 0.5 to 2 ms.23 The strongest change is the loss of a repeatable wave V after all potential sources of interference have been ruled out. Harper and Slavit24 used criteria of greater than a 0.1 ms increase in latency or greater than a 50% decrease in amplitude to identify a change in the response.
Electrocochleography ECoG is a method of measuring the most peripheral of the auditory evoked responses. ECoG response consists of three primary components: cochlear microphonic, summating potential, and action potential. The action potential is an alternating current potential that is associated with the synchronous discharge of numerous neural fibers located in the basal region of the cochlea. The action potential represents the activity similar to wave I of ABR. The action potential component is most useful for intraoperative monitoring. ECoG has the advantage of being a near-field recording, and as such requires fewer averages and less time to obtain a response. In addition, the response obtained is larger and easier to interpret. Noninvasive extratympanic recordings require 250 to 1200 samples over a 12 to 60 second time course, whereas transtympanic needle electrodes require only 40 to 100 sweeps over a 2 to 5 second time period. The stimulus should be a broad-band rarefaction click of high intensity (85 to 95 dB normalized HL) with a rate of 21.1/second. An impedance of 100 kΩ may be acceptable in a transtympanic montage compared with the need for extremely low impedance of less than 5 kΩ required for surface electrodes. Recording parameters differ from those of ABR primarily in that the positive recording site is the ipsilateral promontory as opposed to a distant surface placement. Negative and
ground electrodes remain the same. Another difference is that the number of samples may be reduced to less than 100 compared with the 1500 suggested for ABR.
Interpretation of Results As with ABR, an adequate preoperative baseline response must be obtained against which to judge changes observed during the procedure. ECoG has been used to ascertain whether the goal of decompression operation of the endolymphatic shunt has been achieved. It has been thought that summating potential potentials normalize when pressure imbalances of the cochlea have been eliminated. Zappia and colleagues25 suggested that latency changes of greater than 1 ms and an amplitude decrease of greater than 50% should be considered significant changes in the response, although any measurable deviation should be reported immediately to the surgeon. Significant changes in the action potential indicate either direct or ischemic cochlear injury. After interruption of cochlear blood flow, 20 seconds or more may elapse before changes in ECoG response can be detected. This delay presumably occurs as a result of metabolic reserves that sustain cochlear function until their depletion from prolonged ischemia causes failure of electrophysiologic activity. It has been suggested that simultaneous recording of ABR and ECoG would result in the most effective monitoring system. ECoG provides an interpretable wave I in cases where it cannot be identified in ABR. Using the responses in combination, it would be possible to monitor changes in interpeak latencies, which could not be done using either alone. ECoG alone is not reflective of activity in the proximal region of CN VIII or the brainstem. It is possible to record a normal-appearing ECoG in the presence of significant CN VIII dysfunction. Monitoring of ABR wave V provides information related to the more central areas of auditory processing.
Direct Eighth Cranial Nerve Recording It was found that potentials measured directly from the cochlear nerve required significantly fewer averages than those of an ABR (0 to 100 versus hundreds to thousands) to display a recognizable wave pattern. This finding suggested that responses could be updated every few seconds under ideal circumstances. In addition, cochlear nerve action potentials were found to be present when ABR and ECoG recordings had been eroded by tumor or other pathology. The disadvantage of this type of recording in vestibular schwannoma surgery became evident during procedures to treat larger tumors that extended to or into the brainstem. There must be some portion of uninvolved cochlear nerve on which to place the electrode. Practically, many larger tumors are unsuitable for hearing conservation surgery anyway. Recording electrodes are most commonly placed on or around the intracranial segment of the cochleovestibular
Chapter 64 • Intraoperative Neurophysiologic Monitoring
nerve. A cotton or fibrous wick sutured to the tip of a polytef (Teflon)-insulated wire secures the electrode to the nerve atraumatically. The locations of the reference and ground electrodes are similar to those described for recording ABR. The response waveforms obtained from direct recordings of the cochlear nerve typically exhibit triphasic patterns.26 The initial positive deflection (downward) represents nerve activity approaching the recording site. The generally larger negative (upward) deflection occurs as the impulse passes under the electrode. As the depolarization moves away (more centrally) from the recording electrode, another positive wave results and completes the triphasic complex. In the case of CN VIII monitoring, baseline recordings are obtained after craniotomy, but before tumor dissection.
Interpretation of Results Tumor dissection causing pressure of the cochlear nerve results in immediate loss of compound action potentials (CAP) wave. Stretching the cochlear nerve causes increased latency and broadening of the response waveforms. Partial block of conduction, as may occur with direct pressure by a surgical instrument on the nerve, results in lower amplitude of the prominent negative peak. Only the initial positive deflection is seen after complete conduction block or transection of the nerve.
Otoacoustic Emissions OAE are low-level sounds measured from the ear canals of humans and animals with intact cochlear function. Transient-evoked OAE and distortion-product OAE may be useful in differentiating sensory (cochlear) from neural (retrocochlear) hearing losses. Poor behavioral hearing thresholds and good emissions suggest a retrocochlear site of lesion. Reduced OAE indicates at least cochlear dysfunction. Distortion-product OAE have been shown to be very sensitive to hypoxia and interruption of inner ear blood flow. Being sound waves, OAE are described by amplitude, frequency, and phase measures. Amplitude and phase are affected by manipulations that alter cochlear (and presumably outer hair cell) function. Subtle early changes in amplitude were measurable in some subjects within the first 10 seconds after internal auditory artery (IAA) occlusion. More recently, phase changes have been noted within 2 to 3 seconds of IAA compression, suggesting that phase measures may be the most sensitive OAE parameter for monitoring cochlear function during surgery.27 Emissions were stable during various procedures over many hours. Inhalation and intravenous (narcotic) anesthetics did not seem to affect the recording of emissions. Acoustic noise has the potential to undermine the recording of OAE. This effect was particularly prominent when measuring emissions at frequencies less than 2 kHz.28
781
Based on numerous recordings from various operating rooms, most background acoustic noise energy seemed consistently concentrated at less than 2 kHz. The authors recommended choosing patients for surgical OAE monitoring with preoperatively intact hearing and emissions greater than 2 kHz.28 Some of the noise can be eliminated from the ear canal by sealing the probe and the meatus. A surgical scrub sponge has been used with some success.29 Silicone, wax, and other substances also have been employed in an attempt to solve this problem. Distortion-product OAE have been monitored during loud suctioning of cerebrospinal fluid (relatively high frequency) from the operative field. Emissions could not be monitored during drilling (large-amplitude noise of low and high frequency). Mechanical interference with the conductive hearing mechanism would affect measurement of OAE. Fluid or debris such as bone dust in the external auditory canal or middle ear would interfere with emissions. For this reason, OAE monitoring is unsuitable for many transmastoid surgical approaches. Transient-evoked OAE result in an amplitudeweighted frequency spectrum that may be repeated throughout the critical portion of the operation.28 Generally, reliable transient-evoked OAE can be obtained in 30 to 40 seconds in ideal conditions and longer intervals in noisy situations. Using a customized software PC-based system, distortion-product OAE have been monitored during surgery every 2 seconds. Distortion-product OAE seem to be more robust than other types of emissions and are expected to be present in ears with mild to moderate sensory hearing loss (pure tone thresholds ≤45 dB). Also, monitoring programs can be designed to monitor a single or numerous frequency locations of the cochlea. The best frequency for monitoring may change during the case, and the program can be modified to reflect this. Because of these properties, distortion-product OAE at this time seem to be the best OAE suited for intraoperative monitoring. Because OAE reflect only cochlear function. ABR or direct CN VIII techniques would be appropriate to combine with emissions monitoring. Distortion-product OAE and ABR have been measured during vestibular schwannoma surgery using the same acoustic probe (ER 10-B, Etymotic Research Corp., Elk Grove Village, IL), which contained two speakers and a microphone.29 Rapid switching between the two recording instruments allowed comparisons of the responses. The probe should be secured in the ear canal at the level of the meatus. The pinna, probe, and tubing may be prepared out of the field by folding and securing (with tape or suture) the pinna anteriorly. The tubing should be secured to the patient or operating table in a location to prevent pulling or other disturbance. Baseline measurements obtained after probe insertion and after sterile draping of the operative field help ensure successful monitoring.
782
OTOLOGIC SURGERY
Laser-Doppler Cochlear Blood Flow Cochlear blood flow can be measured using laser Doppler flowmetry.30 The technique involves positioning the end of a needle probe, housing emitter and collector optical fibers, against the cochlear promontory in the middle ear. The angle of the probe is adjusted to obtain maximal flow readings. Although the potential application of this technology is exciting, transferring laser Doppler flow measurement techniques developed in the animal model to human surgery is being done in research trials and is to be studied further before it can be practically applied during tumor removal operations.
Monitoring Other Cranial Nerves Skull base tumors may involve many cranial nerves. CN III, IV, and VI, which control extraocular muscles, are at risk in tumors of the cavernous sinus. Motor portions of trigeminal nerve may be involved, and CN IX, X, XI, and XII can be involved in large skull base lesions. Nerves are localized using the same technique as used for the facial nerve electrode, employing subdermal needle electrodes placed with great care. Subdermal needle electrodes can be placed percutaneously in extraocular muscles that are innervated by a respective cranial nerve. Great care is needed to prevent injury to the globe. It is important to secure the electrodes well so that the location of the needle does not change when the patient is repositioned. The opposite forehead serves as a good location for reference electrode to avoid contamination with EMG potentials from ipsilateral facial muscles.31 EMG from soft palate (CN IX),32 false vocal cords or laryngeal muscles (CN X),33 sternocleidomastoid or trapezius muscle (CN XI), or lateral tongue (CN XII) can be used for the lower cranial nerves. CN XII monitoring can be done by placing recording electrodes spaced 1 cm apart in the lateral tongue.34 Each EMG potential is monitored on separate channels and clearly labeled so that the oscilloscope loudspeaker can be set to monitor the channel of importance at a particular time in a case.
CONCLUSION Intraoperative monitoring of CN VIII has become the standard of care for skull base and CPA surgery. Costeffectiveness analysis supports facial nerve monitoring in mastoid and middle ear surgery.35 Innovative methods of cranial nerve monitoring using techniques such as intraoperative F wave20 and transcranial electric motor evoked potential measurement36,37 are being investigated, but have yet to be widely adopted in otolaryngology. The latter technique is used in neurosurgery and vascular surgery, and provides the advantage of not relying on irritating stimulus of the nerve, but rather stimulation of the
c erebral cortex. Their use may be limited by the requirement for only intravenous anesthetic agents because of suppression of the signal when inhalational agents are used. For monitoring hearing, the monitoring paradigm must provide the surgeon with timely and accurate information to prevent, or repair, reversible injuries to auditory structures and to identify the maneuvers that cause hearing loss so that they may be avoided or modified in the future. Generally, the loss of a physiologic response correlates with poor postoperative hearing. Regardless of the monitoring technique used, the presence of an unchanged or reduced response is not predictive of hearing, which limits the prognostic accuracy of current monitoring techniques. There seems to be a consensus to use auditory monitoring during nerve section and microvascular decompression procedures where hearing preservation rates are quite high. Refining techniques of intraoperative auditory monitoring may ensure better hearing outcomes after acoustic neuroma removal. It is important to separate and distinguish changes in sensory and neural auditory responses to understand how a particular surgical manipulation would affect hearing. In the future, probably a combination of two or more of the methods described in this chapter will be the standard. Ultimately, facial nerve function and hearing outcome depend on many factors, including the surgeon’s experience, tumor size and location, tumor invasiveness, surgical approach, and preoperative function. Refining techniques of CN VIII monitoring will ensure better outcomes in the future.
REFERENCES 1. Krause F: Surgery of the Brain and Spinal Cord. New York, Rebman, 1912. 2. Hilger J A : Facial nerve stimulator. Trans Am Acad Ophthalmol Otolaryngol 68:74-76, 1964. 3. Delgado TE, Buchheit WA, Rosenholtz H R , Chrissian S: Intraoperative monitoring of facial muscle evoked res ponses obtained by intracranial stimulation of the facial nerve: A more accurate technique for facial nerve dissection. Neurosurgery 4:418-421, 1979. 4. Moller AR, Jannetta PJ: Preservation of facial function during removal of acoustic neuromas: Use of monopolar constant-voltage stimulation and EMG. J Neurosurg 61:757-760, 2984. 5. Pratt R L : Iatrogenic facial nerve injury: The role of facial nerve monitoring. Otolaryngol Clin North Am 29:265275, 1996. 6. Hughes G, Bottomy M, Dickins J, et al: A comparative study of neuropathologic changes following pulsed and direct current stimulation of the mouse sciatic nerve. Am J Otolaryngol 1:378-384, 1980. 7. Hughes G B, Bottomy M B, Jackson CG, et al: Myelin and axon degeneration following direct current peripheral nerve stimulation: A prospective controlled experimental study. Otolaryngol Head Neck Surg 89:767-775, 1981.
Chapter 64 • Intraoperative Neurophysiologic Monitoring 8. Chase SG, Hughes GB, Dudley AW Jr: Neuropathologic changes following direct-current stimulation of the rat sciatic nerve. Otolaryngol Head Neck Surg 92:615-617, 1984. 9. Choung YH, Park K, Cho M J, et al: Systematic facial nerve monitoring in middle ear and mastoid surgeries: “Surgical dehiscence” and “electrical dehiscence.” Otolaryngol Head Neck Surg 135:872-876, 2006. 10. Nakao Y, Piccirillo E, Falcioni M, et al: Electromyographic evaluation of facial nerve damage in acoustic neuroma surgery. Otol Neurotol 22:554-557, 2001. 11. Prell J, Rampp S, Romstock J, et al: Train time as a quantitative electromyographic parameter for facial nerve function in patients undergoing surgery for vestibular schwannoma. J Neurosurg 106:826-832, 2007. 12. Kizilay A, Aladag I, Cokkeser Y, et al: Effects of partial neuromuscular blockade on facial nerve monitorization in otologic surgery. Acta Otolaryngol 123:321-324, 2003. 13. Blair E A, Teeple E Jr, Sutherland R M, et al: Effect of neuromuscular blockade on facial nerve monitoring. Am J Otol 15:161-167, 1994. 14. Lacombe H, Keravel Y, Eshraghi AA : Interest of monitoring of facial nerve on facial function in translabyrinthine surgery of acoustic neuroma. Ann Otolaryngol. Chir Cervicofac 111:89-93, 1994. 15. Selesnick S H, Carew J F, Victor J D, et al: Predictive value of facial nerve electrophysiologic stimulation thresholds in cerebellopontine-angle surgery. Laryngoscope 106: 633-638, 1996. 16. Grayeli A B, Guindi S, Kalamarides M, et al: Fourchannel electromyography of the facial nerve in vestibular schwannoma surgery: Sensitivity and prognostic value for short-term facial function outcome. Otol Neurotol 26:114-120, 2005. 17. Neff B A, Ting J, Dickinson S L , Welling D B : Facial nerve monitoring parameters as a predictor of postoperative facial nerve outcomes after vestibular schwannoma resection. Otol Neurotol 26:728-732, 2005. 18. Isaacson B, Kileny PR , El-Kashlan H K : Prediction of long-term facial nerve outcomes with intraoperative nerve monitoring. Otol Neurotol 26:270-273, 2005. 19. Selesnick S H : Optimal stimulus duration for intra operative facial nerve monitoring. Laryngoscope 109: 1376-1385, 1999. 20. Wedekind C, Klug N: Facial F wave recording: A novel and effective technique for extra- and intraoperative diagnosis of facial nerve function in acoustic tumor disease. Otolaryngol Head Neck Surg 129:114-120, 2003. 21. Arriaga M, Haid R , Masel D: Antidromic stimulation of the greater superficial petrosal nerve in middle fossa surgery. Laryngoscope 105:102-105, 1995. 22. Levine R A, Ojemano RG, Montgomery WV, McGaffigan PM : Monitoring auditory evoked potentials during acoustic neuroma surgery: Insights into the mechanism of the hearing loss. Ann Otol Rhinol Laryngol 93:116123, 1984. 23. Samii M, Gerganov V, Samii A : Improved preservation of hearing and facial nerve function in vestibular schwannoma surgery via the retrosigmoid approach in a series of 200 patients. J Neurosurg 105:527-535, 2006.
783
24. Harper C M, Harner SG, Slavit D H, et al: Effect of BAEP monitoring on hearing preservation during acoustic neuroma resection. Neurotology, 42(8):1551-3, 1992. 25. Zappia JJ, Wiet R J, O’Connor C A : Intraoperative monitoring in acoustic neuroma surgery. Otolaryngol Head Neck Surg 115:99-106, 1996. 26. Schwartz D M, Bloom M I, Dennis I M : Perioperative monitoring of auditory brainstem responses. Hear J 38: 9-13, 1985. 27. Telischi FF, Stagner B B, Widick M P, et al: Distortionproduct otoacoustic emission monitoring of cochlear blood flow. Laryngoscope 108:837-842, 1998. 28. Telischi FF, Widick M P, Lonsbury-Martin B L , McCoy M I : Monitoring cochlear function intraoperatively using distortion product otoacoustic emissions. Am J Otol 16: 597-607, 1995. 29. Cane M A, O’Donoghue U M, Lutman M E : The feasibility of using oto-acoustic emissions to monitor cochlear function during acoustic neuroma surgery. Scand Audiol 21:173-176, 1992. 30. Nakashima T, Naganawa S, Sone M, et al: Disorders of cochlear blood flow. Brain Res Brain Res Rev 43:17-28, 2003. 31. Sekiya T, Hatayama T, Iwabuchi T, Maeda SH: A ring electrode to record extraocular muscle activities during skull base surgery. Acta Neurochir (Wien) 117:66-69, 1992. 32. Yingling C D: Intraoperative monitoring in skull base surgery. In Jackler R K, Brackmann D E (eds): Neurotology. St Louis, Mosby–Year Book, 1994, pp 967-1002. 33. Stechison MT: Vagus nerve monitoring: A comparison of percutaneous versus vocal fold electrode recording. Am J Otol 16:703-706, 1995. 34. Moller AR: Intraoperative monitoring of evoked potentials: An update. In Wilkins RH, Rengachary SS (eds): Neurosurgery Update I: Diagnosis,��������������������������� Operative Technique, and Neuro-Oncology. New York, McGraw-Hill, 1990, pp 169-176. 35. Wilson L , Lin E, Lalwani A : Cost-effectiveness of intraoperative facial nerve monitoring in middle ear or mastoid surgery. Laryngoscope 113:1736-1745, 2003. 36. Akagami R , Dong CC, Westerberg B D: Localized transcranial electrical motor evoked potentials for monitoring cranial nerves in cranial base surgery. Neurosurgery 57 (1 Suppl):78-85, 2005. 37. Liu BY, Tian YJ, Liu W, et al: Intraoperative facial motor evoked potentials monitoring with transcranial electrical stimulation for preservation of facial nerve function in patients with large acoustic neuroma. Chin Med J 120: 323-325, 2007. 38. Lin VY, Houlden D, Bethune A, et al: A novel method in predicting immediate postoperative facial nerve function post acoustic neuroma excision. Otol Neurotol 27:10171022, 2006. 39. Isaacson B, Kileny PR , El-Kashlan H, Gadre A K : Intraoperative monitoring and facial nerve outcomes after vestibular schwannoma resection. Otol Neurotol 24:812817, 2003. 40. Fenton J E, Chin RY, Fagan PA, et al: Predictive factors of long-term facial nerve function after vestibular schwannoma surgery. Otol Neurotol 23:388-392, 2002.
784
OTOLOGIC SURGERY
41. Goldbrunner R H, Schlake H P, Milewski C, et al: Quantitative parameters of intraoperative electromyography predict facial nerve outcomes for vestibular schwannoma surgery. Neurosurgery 46:1140-1146, 2000. 42. Nissen A J, Sikand A, Curto FS, et al: Value of intraoperative threshold stimulus in predicting postoperative facial nerve function after acoustic tumor resection. Am J Otol 18:249-251, 1997. 43. Zeitouni AG, Hammerschlag PE, Cohen N L : Prognostic significance of intraoperative facial nerve stimulus thresholds. Am J Otol 18:494-497, 1997. 44. Taha J M, Tew J M Jr, Keith RW: Proximal-to-distal facial amplitude ratios as predictors of facial nerve function after acoustic neuroma excision. J Neurosurg 83:994998, 1995.
45. Lalwani A K, Butt FY, Jackler R K, et al: Facial nerve outcome after acoustic neuroma surgery: A study from the era of cranial nerve monitoring. Otolaryngol Head Neck Surg 111:561-570, 1994. 46. Prasad S, Hirsch B E, Kamerer D B, et al: Facial nerve function following cerebellopontine angle surgery: Prognostic value of intraoperative thresholds. Am J Otol 14:330-333, 1993. 47. Niparko J K, Kileny PR , Kemink J L , et al: Neurophysiologic intraoperative monitoring, II: Facial nerve function. Am J Otol 10:55-61, 1989.
65
Stereotactic Radiosurgery of Skull Base Tumors P. Ashley Wackym, Christina L. Runge-Samuelson, and David R. Friedland
Management of many skull base tumors has shifted in recent years away from surgical resection and towards control of growth. This is particularly true for vestibular schwannomas, i.e., acoustic neuromas, and is increasingly applicable to glomus jugulare tumors. The principal modality for such treatment is gamma knife surgery although other conformal radiation treatment systems are available. Gamma knife surgery is advantageous in requiring a single session for treatment of most skull base lesions, which increases its appeal to both surgeon and patient. This chapter will focus primarily on the methods used in treating skull base tumors with gamma knife surgery. Gamma knife surgery, similar to microsurgery, has advantages and disadvantages which must be thoroughly discussed with the patient.1,2 For the patient it is alluring to undergo an outpatient procedure rather than microsurgical management that requires a much longer period of care. Further, gamma knife outcomes show excellent tumor control and, with current methods, low cranial nerve morbidity. Gamma knife surgery is a viable treatment modality for the appropriate patient as defined by age, medical history, tumor characteristics and physical findings. As such, many neurotologists now offer gamma knife surgery as part of their armamentarium for managing vestibular schwannomas and glomus tumors.3 Several institutions world-wide offer training courses for physicians and radiation physicists at centers having the Leksell Stereotactic System or Leksell Gamma Knife. To date more than 1500 neurotologists, neurosurgeons, physicists, and radiation oncologists have received such training. In addition, the parent company, Elekta Instrument AB (Stockholm, Sweden), offers basic and advanced training courses and workshops. Courses typically consist of didactic lectures, observation of patient treatment, and practical hands-on training. Further, all new installations of Leksell Gamma Knife are accompanied by a one-week on-site start-up training for the neurotologists, neurosurgeons, radiation oncologists, and physicists comprising the gamma knife treatment team.
PATIENT SELECTION Opting for gamma knife surgery over observation or microsurgical resection is a complex decision. There are the preferences of the informed patient, the comfort and experience of the surgeon, the patient’s medical history and condition, and the characteristics of the tumor. While there are no definitive measures defining or restricting the use of gamma knife surgery, particular guidelines can inform the decision making process. Although a tissue diagnosis is not typically acquired prior to gamma knife treatment, radiographic and clinical diagnoses of vestibular schwannoma and glomus jugulare are sufficient to initiate a discussion of gamma knife surgery. Other potential neoplasms amenable to gamma knife treatment by the neurotologist are cerebellopontine angle meningiomas, posterior fossa and jugular foramen non-vestibular schwannomas, temporal bone metastatic lesions and primary vascular neoplasms. An absolute contraindication to gamma knife treatment would be tumors extending too far inferiorly to enable placement into the centrum of the collimator helmet. Gamma knife surgery is also contraindicated in large tumors causing life-threatening brainstem and central aqueduct compression. Such large tumors, in the absence of clinically significant problems, provide a relative contraindication to gamma knife surgery as post-treatment swelling may cause obstructive hydrocephalus requiring emergent intervention. Typically, vestibular schwannomas greater than 2.5 cm in the cerebellopontine angle should be cautiously approached if gamma knife proves the best option given other medical concerns. Most surgeons will not treat vestibular schwannomas greater than 3.0 cm in maximum axial dimension within the cerebellopontine angle because of the risk of post-treatment obstructive hydrocephalus. Other guidelines for gamma knife surgery require clinical judgment as to the medical condition of the patient, the expected growth and potential morbidity of the tumor, the functional status of the patient, audiometric and vestibular performance, age and expected life-span of the patient. Individualized treatment plans 785
786
OTOLOGIC SURGERY
depend on a frank and thorough dialogue between physician and patient as to the options available, risks and benefits of each approach, and expected outcomes based upon evidence-based reviews or an analysis of each institution’s outcomes.
PREOPERATIVE COUNSELING Informed consent for gamma knife surgery requires the surgeon to discuss alternative options such as observation and microsurgical resection.2 The risks and benefits of these alternatives should be frankly described and compared to gamma knife treatment. Many patients have received information from the Internet or from physicians with limited experience with gamma knife and may have erroneous information. Common misconceptions include the expectation that gamma knife surgery completely removes the tumor and that hearing will improve, or conversely that cranial nerve morbidities are significant. These need to be addressed with evidence-based reports and information. One statistic, which is particularly alarming to patients considering gamma knife surgery is that there have been eight cases of malignancy within vestibular schwannomas (as of 2002).4 Four of these patients had been previously treated with radiosurgery. While it remains possible that these four malignancies developed after the radiation treatment, it is more likely that these malignant tumors were misdiagnosed as benign at the outset of evaluation and treatment. Delayed development of radiation-induced neoplasms was addressed by Pollock and colleagues in 1998.5 They reviewed more than 20,000 patients treated with radiosurgery worldwide and found no increased incidence of new neoplasm development (i.e., benign or malignant). A retrospective cohort study comparing the Sheffield, England radiosurgery patient database with the national mortality and cancer registries identified a single new astrocytoma among those treated.6 Based on their national incidence figures, 2.47 cases would have been predicted. The risk of radiosurgery induced malignancy in patients with neurofibromatosis type 2 (NF2) and von Hippel-Lindau disease was similarly studied.7 Of 118 NF2 and 19 von Hippel-Lindau disease patients, totaling 906 and 62 patient-years of follow-up data, respectively, only two cases of intracranial malignancy were found. Both of these were in NF2 patients. One was thought to have arisen before the radiosurgery; the other was a glioblastoma diagnosed three years after radiosurgery. Gliomas may occur in as many as 4% of NF2 patients and the single case may not represent an increased risk. It was suggested that the late risk of malignancy arising after irradiation must be put in the context of the condition being treated, the treatment options available to these individuals, and their life expectancy.
Despite the findings of the studies just reviewed, it is important to counsel patients about the possibility of malignant transformation or induction. A handful of tumors suggestive of radiation induced malignancy have been reported among the tens of thousands who have undergone gamma knife treatment. Lustig and colleagues reported the development of a squamous cell carcinoma following radiation treatment of vestibular schwannoma.8 Hanabusa and colleagues reported the malignant transformation of a vestibular schwannoma following gamma knife surgery.9 There was histologic evidence of vestibular schwannoma following a retrosigmoid resection. Four years after this resection, recidivistic tumor was identified, and the patient was subsequently treated with gamma knife surgery. Six months post-treatment, the tumor had grown, and the patient underwent surgical resection via a combined retrosigmoid-translabyrinthine approach. Abnormal mitotic figures were observed on histologic sections, and the diagnosis of malignancy was assigned.
SURGICAL TECHNIQUE The Gamma Knife Unit The first gamma knife unit (Elekta Instrument AB, Stockholm, Sweden) was installed in Stockholm, Sweden in 1968, and it was not until 1987 that the first gamma knife (model U) was installed in the United States at the University of Pittsburgh. The gamma knife model B (1996) is the unit currently most used throughout the United States. The gamma knife model C was introduced three years later and the major upgrade consisted of an automatic positioning system (APS). The unit is otherwise quite similar to the model B and both contain 201 radioactive isotope cobalt 60 (60Co) sources and beam channels. Due to physical restraints these units can only treat lesions intracranially or along the skull base. During 2008, a completely redesigned gamma knife unit, named Perfexion, is being introduced. It uses 192 60Co sources, has a single collimator helmet with variable diameters, and can treat lesions within the entire head and neck, down to the level of the clavicles. The basic principle of gamma knife surgery is to provide focused radiation to the tumor while minimizing radiation delivery to surrounding tissues. As such, a semicircular shield called the collimator helmet is used to generate approximately 200 individual gamma radiation “beams.” In the center of the helmet, where the beams meet, radiation delivery is maximal, but along each individual radiation tract tissue exposure is relatively low. When the collimator helmet is locked into position, the 201 openings of the collimator helmet coincide with the cobalt sources. There is a shielded chamber within which the 60Co sources are contained, and stainless steel shielding doors protect the treatment room from the 60Co sources. There is a treatment couch with an adjustable mattress that slides into the gamma knife unit together
Chapter 65 • Stereotactic Radiosurgery of Skull Base Tumors
787
Automatic positioning system
Helmet with collimators (outside treatment position)
Helmet supports
Cobalt 60 sources
Treatment couch with adjustable mattress Beam channel Shielding Plastic cover
Protective panels
Helmet in treatment position
Shielding doors
Figure 65-1. Gamma knife surgery. Schematic illustration of the Leksell Gamma Knife 4C which utilizes the automatic positioning system. (Published with permission, copyright © 2008, Elekta Instrument AB [Stockholm, Sweden].)
irradiation shot. There are four interchangeable helmets by means of which the size of the collimator (that part of the treatment unit that shapes the beam) can be changed between 4 mm, 8 mm, 14 mm and 18 mm. The combination of four different sized collimators and repositioning the patient in the three-dimensional space defined by the stereotactic headframe are effective to deliver the radiation dose selectively and conformally to radiosurgical targets of any shape.
Frame Attachment
Figure 65-2. Leksell Gamma Knife 4C. (Published with permission, copyright © 2008, Elekta Instrument AB [Stockholm, Sweden].)
with the collimator helmet and the patient. Figure 65-1 schematically shows the orientation of the components of the gamma knife model, Leksell Gamma Knife® 4C and Figure 65-2 shows the overall appearance of the gamma knife model, Leksell Gamma Knife® 4C. When treatment is initiated, the treatment couch is automatically moved from its idle position into the treatment unit together with patient and helmet. Once the couch is docked in its treatment position, the helmet collimator and corresponding collimators in the unit form a beam channel, allowing the radiation that is continuously emitted by the sources to reach the patient. At the end of each irradiation “shot,” the couch is automatically withdrawn, either to its idle position or to a position outside the radiation focus to reposition the patient for the next
The stereotactic head frame is used to coordinate the location of the tumor within the collimator helmet. As such, proper placement is of utmost importance to providing adequate treatment. There are two general principles guiding head frame placement for gamma knife surgery. First, the target should be as close to the center of the frame as possible. This prevents possible collisions of the frame with the sides of the collimator helmet especially when trying to align laterally extended tumors in the center of the unit. Second, the frame attachment should be stable. This prevents movement and ensures accuracy and correlation among the pre-treatment imaging study, workstation treatment plan, and delivery of focused radiation. These principles should be addressed at the time of frame attachment. In lateral targets, such as vestibular schwannomas or glomus tumors, the frame should be shifted toward the tumor side. In skull base tumors the frame should also be positioned lower than for treatment of more superior intracranial lesions. Anteriorposterior alignment should also be accounted for and can be adjusted by varying the lengths of the pins used to
788
OTOLOGIC SURGERY
A
B
C
D
Figure 65-3. A, Stereotactic headframe at the time of assembly. Pins used for fixation prior to imaging, treatment planning, and gamma knife radiosurgery are seen in the foreground. B, With a towel placed on the vertex of the head, the magnetic resonance imaging (MRI) fiducial box is balanced on the head while topical anesthetic is infiltrated at the pin sites. C, While the headframe and attached plastic MRI fiducial box is secured to the skull using the pins, an assistant stabilizes the assembly in place. Note the tightening of pins in opposing vectors. D, After the frame has been secured to the skull, measurements are taken through defined entrance points in a plastic dome representing the size and position of the collimator helmet relative to the frame, head and face. These values are incorporated into the treatment planning software to create a wire grid representing the patient’s head. The post height and pin length are also measured and are inputted into the treatment planning software. These values are used in the calculations used to predict collisions between the collimator helmet and the stereotactic headframe. (Published with permission, copyright © 2008, P. A.Wackym, MD.)
secure the frame. To ensure stability, avoid screw fixation in bone flaps, cranioplasty materials, burr holes, or skull defects. The method of anesthesia used during frame placement is surgeon and patient dependent. In our program, either sedation with versed and fentanyl, or monitored anesthesia with propofol, followed by injection of local anesthetic at the pin sites is used. Figure 65-3A shows the typical array of tools used for the frame attachment. A variety of screw lengths allow the surgeon to choose those ideally suited for the individual location of the posts and tumor. The placement of the frame should begin with an accurate orientation of the location of the target within the patient’s head. Ideally, the target should be located within the fiducial range and placed centrally within the frame thereby avoiding later collisions with the collimator helmet and granting sufficient accuracy for the stereotactic target definition. The stereotactic frame is assembled and preliminarily supported by using external auditory canal support pins, a Velcro band, or a stereotactic fiducial box. When using a fiducial box to facilitate frame placement, it is important to use the MRI fiducial box, rather than the CT or
angiography fiducial box, since this is the smallest of the three plexiglass fiducial boxes (Figs. 65-3B and 65-3C). Asymmetric frame placements are possible and do not impair the accuracy of imaging. The frame can be shifted from side to side or can be moved as far as possible to the front or back to facilitate centering of the tumor. The frame is stabilized against the patient by an assistant and the surgeon should adjust the lengths of the posts to maintain relative tumor position. A low position of the anterior posts can help avoid anterior collisions with the collimator helmet for skull base posterior fossa tumors. In critical positions, collisions can sometimes be avoided by using the curved posts in the anterior position. Once post position is determined the screws can be inserted. The surgeon and assistant should work on diagonally opposing screws to provide the best stability without changing the desired frame position. For asymmetric frame placement apply the longest screws first, thereby defining the desired distance of the target to the frame. Protrusion of the screws from the posts should be kept to a minimum to avoid collisions. Approximately 8 to 10 mm is considered to be sufficient but at our institution we prefer to limit this projection to 4 to 6 mm.
Chapter 65 • Stereotactic Radiosurgery of Skull Base Tumors
789
If a screw extends further it should be exchanged for a shorter screw. Measurements of the frame and placement are then performed to allow the computer to identify any potential collisions after the plan is formulated. These measurements are required for the frame and skull section in Leksell GammaPlan treatment planning software. Measurements include the length of the four posts and the length of the screws that protrude from the posts. Additionally, the volume of the head is measured using the plastic collimator bubble, simulating the relationship of the frame to the treatment collimator helmet (Fig. 65-3D). This concludes frame placement and the patient may proceed to imaging.
Imaging Treatment planning requires imaging of the tumor with respect to the frame as determined by specific fiducial boxes. The MRI fiducial box clips to the frame and care should be taken to ensure that it is flush and square during imaging. The MRI fiducial box has a Z-shaped channel on each side filled with copper sulfate to generate position markers for each axial slice. The box should be checked prior to each use to ensure the channels are filled with solution and no air bubbles are present. The patient, with head frame and fiducial box, is secured into the head holder on the MRI sliding table. For imaging acoustic neuromas and glomus tumors we typically order axial 3D SPGR (spoiled gradient recalled) acquisition with T1 weighting and double dose IV contrast. Before the patient leaves the scanner images are reviewed and the distance between fiducial registration markers is validated for accuracy. Many centers acquire only MRI scans for treatment planning. We prefer to also acquire a non-contrast CT scan through the temporal bone to aid in planning. There is evidence of distortion of MR images and correlation with CT scans at the time of planning can aid in reducing radiation delivery to critical structures such as the cochlea and facial nerve.10 A CT fiducial box is affixed to the frame, the patient secured in the holder attached to the table, and an axial scan through the temporal bone and skull base acquired. Both CT and MR images are imported into the Gamma Knife workstation. Axial scans are defined, and coronal and sagittal reconstructions generated for each.
Treatment Planning Leksell GammaPlan is the dedicated software treatment planning system for Leksell Gamma Knife. Dose planning for gamma knife surgery means precisely conforming the isodose distribution to the target. The isodose distribution is built up by a number of individual shots or isocenters. The Leksell GammaPlan software is designed to help the operator as much as possible to perform this procedure and is quite straightforward to use.
Figure 65-4. Initial treatment planning at the gamma knife workstation involves building a three-dimensional model of the tumor. Determination of the conformation of the treatment plan follows placement of the shots and assignment of the radiation dose delivered to the specified isodose line. (Published with permission, copyright © 2008, P. A.Wackym, MD.)
Currently, for vestibular schwannomas, the routine prescription is 12 to 14 Gy delivered to the 50% isodose line. The 50% isodose line shows where 50% of the prescribed dose lies. In the case of gamma knife treatments the dose is frequently prescribed to the 50% isodose line. This ensures that the periphery of the tumor will receive at least the prescribed dose, that the dose will be higher than the prescribed dose inside the tumor, and that the dose will fall off rapidly outside the tumor thus sparing critical structures. Dose planning using Leksell GammaPlan involves composing shots to develop a conformal isodose. By definition, this includes the whole target but spares the surrounding healthy tissue. Figure 65-4 shows an example of a vestibular schwannoma. The target is well positioned on the screen and magnified for good visibility. When the shot menu is opened, one can select the size of the collimators. The size of the collimator is selected based on the tumor shape and the gaps in coverage of the 50% isodose line displayed over the tumor. Shots are placed sequentially to cover the target as effectively as possible. Changing the position of the shots, adding additional shots, and adjusting the relative weight of shots quickly leads to a conformal dose plan. The dose plan can be checked using Leksell GammaPlan with the three-dimensional (3D) image or the measurement tools, such as dose volume histograms. While the subject of conformity index is beyond the scope of this chapter, an excellent review of available methods has been published.11 Leksell GammaPlan indicates the point in the stereotactic space where a global maximal dose can be found. Leksell GammaPlan also calculates the individual shot times. Once the treatment plan has
790
OTOLOGIC SURGERY
been determined to be appropriate by the gamma knife team (surgeon, radiation oncologist, and radiation physicist), the stereotactic coordinates and irradiation times are printed and used during the gamma knife treatment. An automated approach to initial treatment planning has been developed by Elekta Instruments AB. This software assists the treatment planner in generating a good initial dose plan quickly and is termed “the Wizard.” This is an interactive tool that helps the operator develop the dose plan. The operator first selects the shot size and the degree of density with which the Wizard should fill the target. A mouse click instructs the Wizard to fill the target with shots. If the initial dose distribution is not sufficient, a mouse click on the run button instructs the Wizard to optimize the plan by moving and weighting the shots. This interaction results in a better dose plan, and after a few more changes a satisfactory dose plan can be created. However, in this our experience, manual placement of the shots, particularly for vestibular schwannomas, has always resulted in a better treatment plan. Fine-tuning is made with small adjustments in shot position and weight, allowing optimization of the dose plan. Leksell GammaPlan allows the creation of different plans for the same target. This allows the surgeon and oncologist to follow different strategies and later compare plans and select the best plan for the actual treatment. Treatment plans can utilize as few as one or two shots, such as when treating trigeminal neuralgia, or over 10 shots when treating a large vestibular schwannoma within the cerebellopontine angle and filling the internal auditory canal. With the enhanced capabilities of Leksell Gamma Knife C, plans with 20 shots or more can easily be implemented in a timely manner, since the model C does not require manual adjustments of coordinates in between each shot by the gamma knife treatment team. This allows improved conformity and selectivity of gamma knife surgery, potentially reducing the risk of complications. To shape the dose distribution to avoid critical structures, one or more of the 201 collimators can be replaced with a closed shield called a plug. One can select spherical areas called shields with different diameters and place them over risk centers in the brain, cranial nerves, or cochlea. Once the shields are put in place, the Leksell GammaPlan software closes off all beams that would irradiate through the shielded area. The result is a modified dose plan in the low isodose lines with only little effect on the target peripheral isodose. The beam channels that need to be plugged can be seen in the plug pattern. The plug patterns can be merged for all shots of the same size so that the operator only has to plug the helmets for the treatment once. In the final plan, the peripheral dose is set to a value, which is assessed as optimal for a particular patient. Indication, size, and location of the target are taken into account, as well as clinical experience. The peripheral isodose is usually set to the 50% isodose line. This is exactly
CutBox (MRI_axial)
Sagittal (MRI_axial) Reference Dose for Plan ‘Final_plan’
Absolute Dose Levels Normalize to point Prescription dose
[Gy]
14.00
Prescription dose
[%]
50.00
Maximum dose [Gy] 28.00 OK
Coronal (MRI_axial) Display Isodose
Selected Level Display volume Wire 50 Levels 90 80 45 20 Plot
Set
Exit
Help
Help
OpenBook (CT_axial)
Figure 65-5. Gamma knife surgery. Selecting the Absolute Dose Level and Display Isodose options allows verification that the maximal radiation dose is not delivered near critical structures, such as the facial nerve. In this example, 14 Gy delivered to the 50% isodose line was prescribed. As shown, the maximum dose (28 Gy) is delivered to the center of the tumor (smallest circle). The largest circle represents the 20% isodose line where 6 Gy of radiation is delivered. (Published with permission, copyright © 2008, P. A.Wackym, MD.)
half the maximum dose in the target, referred to as the hot spot (Fig. 65-5). Along the 50% isodose line the dose gradient is usually the steepest ensuring sufficient dose within the target, while the dose level outside falls steeply, sparing the surrounding healthy tissue. Leksell Gamma Plan can also display the absolute dose values if desired. It will show the point in the stereotactic space where the global maximum dose can be found. With vestibular schwannoma it is valuable to complete this exercise, as the maximal dose at the “hot spot” should be positioned well away from the facial nerve and cochlea. In addition, plotting the absolute dose lines will help in determining the actual level of radiation delivered to surrounding structures. When the dose planning is completed, Leksell GammaPlan checks the shots for collisions with the collimator helmet and sorts the plan according to collimator size. The team performs quality assurance steps to check the accuracy of the X, Y, and Z coordinates and to ensure the plan treats the correct side. All relevant data are documented including details of the treatment plan, targets, dose volume histograms, snap shots and images. When the treatment set-up has been finalized the treatment protocol is exported to Leksell Gamma Knife. This is via a special secured direct serial connection. Leksell GammaPlan only accepts valid and verified treatment plans for export. In addition, a protective design limits the transfer of a treatment plan to the Leksell Gamma Knife to one patient at a time. Once the data have been transferred to the operator’s console, it is verified, and the patient can be treated. For the model C unit, the operator does not have to enter the treatment room during a run. However, with the
Chapter 65 • Stereotactic Radiosurgery of Skull Base Tumors
model B the treatment team enters the treatment room after each shot is delivered and manually adjusts the X, Y, and Z coordinates, as well as the gamma angle, i.e., the pitch of the head, if necessary. With both the model B and model C, the team has to change the collimator helmet manually when necessary, as dictated by the treatment plan. Detailed treatment and physics protocols are viewed and printed out.
Treatment Treatment can be performed automatically using the automatic positioning system or manually using trunnions. For the model B, manual setting of the X, Y, and Z coordinates as well as the gamma angle if necessary is accomplished by the treatment team. The Y and Z coordinates are set with the Y, Z slides on the y-bar attached to the coordinate frame, whereas the X coordinate and the gamma angle are set with the trunnions. It is imperative to have a check and balance in place that consists of visual verification of each coordinate by a different team member. Y coordinates need to be verified prior to setting the Z coordinate as the latter will obscure the scale on the Y axis. It is preferable to set the X coordinate of the trunnion on the shorter side first as this will provide more room to manipulate the patient and head frame within the collimator helmet. These coordinates need to be manually changed between each shot on the model B unit. With the automatic positioning system, the treatment is controlled from the operator console. Once the treatment starts, the selected run is carried out automatically. Before repositioning, the couch will move out a short distance to bring the patient out of treatment focus. At this point, the APS will move the patient’s head to the next target position. A run consists of all shots for a specific collimator helmet size. Additional runs are performed after manually changing the collimator helmet. After all runs have been completed the head frame is removed. The anterior fixation sites are dressed with antibiotic ointment and adhesive bandages. The posterior sites are dressed with antibiotic ointment. Often pressure needs to be held to control bleeding and occasionally a staple may need to be used on the posterior sites. Typically patients will experience a transient headache after removal of the frame and some develop nausea and emesis. We typically pre-medicate with decadron and ondansetron prior to frame removal. Patients are observed for several hours post-treatment and discharged home with pain medication and follow-up appointments.
GAMMA KNIFE SURGERY OUTCOMES Just as is the case with other forms of medical and surgical therapy, the techniques and outcomes of gamma knife surgery for vestibular schwannomas have evolved
791
and improved over time. Tumor control and facial nerve motor preservation occurs with virtually all vestibular schwannoma patients treated with current gamma knife protocols. Areas of continued focused investigation include the effects of radiosurgery on hearing and balance, and methods of improving outcomes. The University of Pittsburgh group has the largest clinical experience in treating vestibular schwannomas with gamma knife surgery. Lunsford and colleagues summarized their experience with 829 vestibular schwannomas treated between 1987 and 2002.12 This extensive clinical experience included an average tumor volume of 2.5 cm3 and a median margin dose to tumor of 13 Gy. They reported tumor control in 97% of patients at 10 years, and facial nerve (motor) dysfunction in < 1% of patients. Trigeminal nerve symptoms occurred in < 3% of patients and typically occurred with large tumors reaching the level of the trigeminal nerve. No reporting of balance function was included in their analyses. The reporting of hearing preservation has limited representation in the entire 829 patients. Hearing outcomes data were presented in only 267 patients and “5-year actuarial rates of hearing level preservation and speech preservation” were reported in 103 patients. They reported “unchanged hearing preservation” in 50����� %���� to 77% of these patients, and this method of reporting auditory performance points to the difficulty in interpreting the outcome of most of the studies reporting hearing outcome in patients with vestibular schwannoma who have been treated with gamma knife surgery. They also stated that “for patients with intracanalicular tumors, hearing preservation rates in those treated with 12.5 to 14 Gy at the margin showed 90% preservation of serviceable hearing.”13 Unfortunately, pretreatment and longitudinal data are not available in these reports. In an earlier series of 190 patients the average dose to the tumor margin was reduced to 13 Gy, and excellent tumor control was achieved at 97.1%.14 In this study, issues highlighted earlier with reporting of hearing outcome are equally apparent. They reported “hearing-level preservation” in 71 ± 4.7% of patients. They also reported a “preservation of testable speech discrimination ability” in 91 ± 2.6% of subjects. Obviously, testable speech discrimination ability is far different than useful hearing, and it is unfortunate that these authors did not report the actual auditory thresholds or speech discrimination ability. Most importantly, these were not reported as a function of time post-gamma knife surgery. In addition, they reported that “hearing levels improved” in 10 (7%) of 141 patients who exhibited decreased hearing defined as Gardner-Robertson grades II to V before undergoing gamma knife surgery. Based on our clinical observations and those of other centers, this picture is far more complex over time than is represented in these publications. Prasad and colleagues from the University of Virginia reported their series of 200 vestibular schwannomas treated with gamma knife surgery over a 10-year interval
OTOLOGIC SURGERY
in 2000.15 Of these patients, 153 patients had follow-up data including 96 with primary treatment and 57 with secondary treatment. They reported no hearing pregamma knife in 105 patients, including 53 of 96 primary treatment and 52 of 57 secondary treatment patients. The Gardner-Robertson grading system and subjective assessment of hearing was used; however, no pure-tone average or speech discrimination data were reported. Unfortunately, their data set included audiometric data from only 48 patients, and the intervals of audiometric testing were not reported. Despite these limitations, they found that, except for one patient, no change in hearing was observed in the first two years after gamma knife surgery. Their data also showed that the greatest change in Gardner-Robertson grade occurred between years two and four post-gamma knife; however, without understanding the assessment intervals, the precise onset of the hearing loss is unknown. No outcomes regarding balance function were reported. Kim’s group at the Seoul National University reported the hearing outcomes in 25 patients with vestibular schwannomas with serviceable hearing.16 The median tumor volume was 3.0 cm3 (0.16 to 9.1 cm3), and the dose used was 12 ± 0.7 Gy at the 49.8 ± 1.1% isodose line. They reported the hearing outcomes using the Gardner-Robertson grading system, pure-tone averages, and speech discrimination scores. Pre-gamma knife, interim post-gamma knife, and last post-gamma knife data were reported. Similar to our experience, they found that in 16 patients the hearing deteriorated > 20 dB three to six months post-gamma knife and that this hearing loss continued for 24 months. The only prognostic factor for hearing deterioration that they identified was the maximum dose to the cochlear nucleus. In the Medical College of Wisconsin Acoustic Neuroma and Skull Base Surgery Program, we have established a clinical pathway for all of our patients undergoing gamma knife surgery for primary or secondary treatment of their tumors. Pretreatment they undergo a complete videonystagmography test battery, a complete audiologic assessment, and facial nerve electromyography. At sixmonths intervals post-treatment, each patient undergoes a gadolinium enhanced MRI as well as an audiologic test battery and caloric testing to assess peripheral vestibular function. In addition to other standard reporting methods, we have also presented the data in a longitudinal manner for their objective auditory thresholds (Fig. 65-6), speech discrimination ability (Fig. 65-7), and degree of vestibular paresis (Fig. 65-8). We have recently published an expanded cohort of 54 patients with a median follow-up interval of 54.7 months.17 This report focused on the longitudinal outcomes in vestibular function and changes in the Dizziness Handicap Inventory before and after gamma knife surgery. It is clear that most of the change in hearing and balance function occurs during the first six months after gamma knife surgery; however, continued but less rapid
Patient number:
PTA–3 60 50 PTA–3 change (dB) re: pre-op
792
40 30 20 10 0 −10 −20 −30
0
1
2
3
4
5
6
Postoperative time interval (years)
Figure 65-6. Auditory function over time after gamma knife surgery treatment of unilateral vestibular schwannomas. Three-frequency averages of pure-tone thresholds (PTA-3) in dB HL at 0.5, 1, and 2 kHz were determined for all patients with measures at the preoperative time and at least one postoperative interval. The PTA-3 difference was calculated for each time interval relative to the preoperative PTA-3. The differences are plotted as a function of postoperative time interval, with zero representing the preoperative time. A positive difference value indicates a higher or poorer, postoperative PTA-3. In general, over time, the vast majority of patients were found to have PTA-3s that were poorer or similar to preoperative PTA-3s, although a few individuals showed some initial improvement (e.g., subject 5). The greatest changes in PTA-3 were measured at six months post-treatment although continued changes were observed up to five years post-treatment. (Published with permission, copyright © 2008, P. A.Wackym, MD.)
worsening of function can occur up to 12 months. These objective measurements correspond well to the transient facial nerve dysfunction, trigeminal nerve dysfunction, tinnitus, and dysequilibrium occurring in our patients with vestibular schwannomas undergoing gamma knife surgery.3,10,17 A possible mechanism underlying these changes is that there is an initial increased size of the tumor after radiosurgery. Typically this post-treatment edema persists for six months; however, this may remain for up to one year.3,10,17 The labyrinthine artery, a branch of the anterior inferior cerebellar artery, provides essentially all of the blood supply to the cochlea and vestibule and it is likely that the post-radiation edema compromises this blood supply to the inner ear. The resulting inner ear devascularization could certainly explain the rapid change in hearing and balance function seen at the six-month post-treatment assessment in our patients (see Figs. 65-6 and 65-8). Several of our patients have had tumor control or regression and improvement of hearing and vestibular function. This is clearly divergent from the natural history of vestibular schwannomas. In contrast, worsening of auditory and vestibular function and the development of disequilibrium has occurred in a number of our patients. Continued systematic studies of these patients and expansion of the cohort of patients studied are
793
Chapter 65 • Stereotactic Radiosurgery of Skull Base Tumors
40 20 0 −20 −40 −60 −80
0
1
2
3
4
Vestibular Paresis
100 Vestibular paresis change (%) re: pre-op
Speech recognition change (%) re: pre-op
Patient number: Speech Recognition
5
6
Postoperative time interval (years)
Figure 65-7. Speech recognition testing was performed using the Northwestern University Auditory Test No. 6 (NU-6) monosyllabic words. The stimuli were presented at 40 dB sensation level, i.e., above speech recognition threshold, or if this was too loud, at the patient’s most comfortable listening level. Speech recognition was scored in percent correct. As with PTA, the differences between pre- and postoperative speech recognition were calculated and plotted as a function of postoperative time interval. Positive values are consistent with an improvement in speech recognition. Approximately half of the patients showed improvement in speech recognition at six months post-treatment, while the other half showed a decrease in performance. Of those patients who experienced a reduction in speech discrimination ability, there was a greater range of change than that observed in the patients who enjoyed an improvement in speech discrimination ability. It should be noted that the greatest changes in speech discrimination ability occurred at six months post-gamma knife treatment. However, over time, the patients generally demonstrate speech recognition performance similar to or poorer than pre-treatment performance. (Published with permission, copyright © 2008, P. A.Wackym, MD.)
important to determine the efficacy of gamma knife surgery and to compare to other forms of radiotherapy, as well as microsurgery and expectant management. Recognition of symptoms such as disequilibrium and knowledge regarding the expected time course of vestibular paresis progression are important not only for patient counseling but provide the opportunity to intervene with vestibular rehabilitation or nonspecific vestibular suppression until compensation has been completed, should this be needed clinically.17 One final issue to consider is tumor growth after radiosurgery (Fig. 65-9). It is important to appreciate that there is an increased size of the tumor after radiosurgery. In fact, we observed a statistically significant increase in tumor size for patients whose tumors extended outside of the internal auditory canal six months after gamma knife surgery and a statistically significant decrease at one year post-treatment.3,10 Typically, post-treatment edema persists for six months; however, this may remain for up to one year. Consequently pretreatment counseling should include this information. There have been anecdotal cases discussed and occasionally reported that describe increased tumor size early after radiosurgery.
Patient number:
80 60 40 20 0 −20 −40 −60
0
1
2
3
4
5
6
Postoperative time interval (years)
Figure 65-8. Vestibular paresis was determined with bithermal caloric testing. A positive difference value indicates greater vestibular paresis post-gamma knife surgery. Both degradation and improvement in vestibular paresis are observed across patients. Within a patient, the postoperative degree of vestibular paresis generally tends to remain stable over time after the relatively large initial change observed at the six month post-treatment assessment. In those patients who had continued reduction in their vestibular function, there was continued difficulty with dysequilibrium until vestibular compensation was complete and the vestibular paresis stabilized. (Published with permission, copyright © 2008, P. A.Wackym, MD.)
Figure 65-9. Gamma knife surgery. Example of serial magnetic resonance imaging studies of a small left vestibular schwannoma. Note at six and 12 months post-gamma knife radiosurgery the tumor is larger than pretreatment. By 18 months the tumor is smaller. (Reproduced with permission, copyright © 2008, P. A.Wackym, MD.)
The challenge is in making a decision about whether to resect these tumors and when.2,5,18-22 Pollock and colleagues emphasized the need to demonstrate sustained tumor growth by serial MRI before making the decision to operate and also to review the case with the surgeon who performed the radiosurgery before a surgical decision is made.5 Another related controversy is whether facial nerve dissection and preservation are more difficult during microsurgical resection after radiosurgery. On one end of the spectrum, descriptions of no increased difficulty have
794
OTOLOGIC SURGERY
been reported;5 and, on the other end of the spectrum,19-22 markedly increased difficulty in separating the tumor from the facial nerve and poorer facial nerve function outcome have also been reported. The report of Watanabe and colleagues included a histopathologic analysis of the resected facial nerve.20 They found microvasculitis of the facial nerve, axonal degeneration, loss of axons, and proliferation of Schwann cells. In light of the mechanism of delayed effects following radiosurgery, these findings are not surprising. Moreover, these findings emphasize the need for the neurotologist to be certain that the treatment plan avoids high radiation doses to the facial nerve. Recall as described earlier that a dose of 12 Gy delivered to the 50% isodose line means that the maximum tumor dose is 24 Gy. If the treatment plan delivers this maximal dose to the area of the facial nerve, it should be expected that greater radiation effects will be observed. For this reason, if the neurotologist and the patient have made a decision to resect a tumor previously treated with radiosurgery, it is important to review the treatment plan to determine the amount of radiation delivered to the facial nerve to counsel the patient appropriately preoperatively.
Figure 65-10. Ceiling-mounted diagnostic-energy x-ray sources emit low-dose x-rays through the patient’s tumor treatment area. Amorphous silicon image detectors capture x-ray images from ceilingmounted diagnostic-energy x-ray sources to produce live radiographs. The operating system (typically located adjacent to the treatment room) correlates patient location detected by image guidance system with reconstructed CT scan and directs the robot to adjust position accordingly. The compact linear accelerator mounted on a computercontrolled robotic arm which adjusts position to maintain alignment with the target, compensating for any patient movement and uses X-band technology for mobility. (Published with permission, copyright © 2008, Accuray [Accuray Incorporated, Sunnyvale, CA].)
ALTERNATIVE TECHNIQUES As noted earlier, tumor size and location may dictate that a method other than gamma knife be considered. Indeed, alternative methods of radiosurgery are available for treating a wide variety of skull base neoplasms. These include the Peacock (NOMOS Inc., Cranberry Township, PA), the SmartBeam IMRT (Varian Medical Systems Inc., Palo Alto, CA), the Precise (Elekta, Inc., Stockholm, Sweden), and the CyberKnife (Accuray, Sunnyvale, CA). Among the more common of these modalities is CyberKnife, which will be briefly reviewed here.
Cyberknife Stereotactic Radiology Overview of Treatment Planning The CyberKnife stereotactic radiosurgery system utilizes a compact 6-MeV linear accelerator, a computercontrolled robotic arm with six degrees of freedom, and an image-guidance technology that does not depend on a rigid stereotactic frame and thereby enables treatment of extracranial sites (Fig. 65-10). Potential benefits of this approach include: 1) increased access to and coverage of any target volume including the ability to treat lesions in and around the cranium that are unreachable with other systems, for example, in the lower posterior fossa and foramen magnum; 2) enhanced ability to avoid critical structures; 3) capability to treat lesions in the neck and spine; 4) ability to treat lesions throughout the body; 5) delivery of highly conformal dose distributions; 6) option of fractionating treatment; and 7) potential to target
multiple tumors at different locations during a single treatment, e.g., skull base and neck. The CyberKnife treatment planning system is designed to support the radiosurgery team in determining the optimal plan, including beam weight, targeting positions, dose distributions, and other factors for each patient’s treatment. The CyberKnife stereotactic radiosurgery system permits the following planning and delivery options: 1) inverse planning; 2) nonisocentric delivery; and 3) hypofractionation. In contrast to most gamma knife procedures, CyberKnife is CT based. MR images can be fused with the CT to provide optimal information on soft tissue as well as skeletal anatomy. CT angiography can be used when vascular skull base lesions such as arteriovenous malformations or extensive glomus jugulare tumors are to be treated with this technique. The flexibility of the robotic arm supporting the linear accelerator allows the CyberKnife to implement a wider range of treatment plans than other systems. Furthermore, because the system does not require the use of a stereotactic head frame temporarily attached to the patient’s head, it allows scanning, treatment planning, and quality assurance to take place at any time prior to treatment itself. The CyberKnife system provides a range of treatment options, including the ability to use either forward or inverse treatment planning. With forward treatment planning, the radiation oncologist determines what dose to deliver from a particular targeting position. The total dose within the lesion is then calculated by the
Chapter 65 • Stereotactic Radiosurgery of Skull Base Tumors
system software. With inverse treatment planning, the radiation oncologist specifies total dose to be delivered to the tumor. The surgeon and radiation oncologist are then able to set boundaries to protect adjacent critical structures. The software subsequently determines targeting positions and the dose to be delivered from each targeting position. While other stereotactic radiosurgery systems offer the inverse planning option, the number of possible plans is limited by the constraints of the delivery system.
Dose Distribution The CyberKnife system offers a choice of a nonisocentric or an isocentric treatment approach. With other stereotactic radiosurgery systems, a fixed calculated isocenter is used. Isocentric treatment, or multi-isocentric treatment, involves filling the lesion with a single or multiple, overlapping spherically shaped dose distributions. Isocentric treatment is effective for spherical lesions. However, with irregularly shaped lesions, isocentric delivery can produce significant dose heterogeneity. In this case the surgeon and radiation oncologist must account for the relationship of the maximum dose to critical structures such as the facial nerve or cochlea. Similarly, they must identify regions which may be under-treated by delivery of inadequate doses. Nonisocentric treatment plans are also possible with the CyberKnife system. The delivery of these treatment plans is possible because of the robotic arm which, because of the six degrees of freedom (discussed later) enables the delivery of radiation to complex treatment volumes. The beams originate from arbitrary points in the workspace and are delivered into the lesion. The result is a nonisocentric concentration of beams within the lesion and asymmetric irradiation. Nonisocentric treatment allows the avoidance of critical structures while providing complete coverage of the lesion at the prescribed isodose. With the CyberKnife system, the treatment plan can utilize fractionated or hypofractionated approaches. Fractionated treatment is possible because localization of the lesion is achieved using image guidance technology. Dose delivery over two to five treatment sessions, termed hypofractionation, is another option with the CyberKnife system. Although not directly applicable in managing tumors within the posterior fossa, it has been suggested to be particularly useful in the treatment of large tumors. The argument for fractionation is that lowering the dose for each of a number of treatments, as opposed to a single, larger dose, allows healthy tissue to rejuvenate between treatments. The advantage of fractionated or a single radiation dose remains an active area of investigation and debate. Because of the rigid fixation that occurs with securing the stereotacic headframe in gamma knife surgery, fractionated or hypofractionated delivery of radiation is not possible. Furthermore,
795
it remains to be determined if equal accuracy can be achieved by these two systems or if there is an advantage of fractionation or hypofractionation in the treatment of skull base tumors.
Localization The CyberKnife system’s use of stereotactic principles for tumor localization differs from other stereotactic radiosurgery systems by using an image guidance technology that depends on the skeletal structure of the body as a reference frame. In addition, it continually monitors and tracks patient position during treatment. The CyberKnife’s operating system correlates live radiographic images with preoperative CT scans to determine patient and tumor position repeatedly over the course of treatment. The imaging information is transferred from the computer’s operating system to the robot so that it may compensate for any changes in patient position by repositioning the LINAC.
Treatment Delivery The CyberKnife system’s computer-controlled robotic arm has six degrees of freedom. The robot can position the LINAC to more than 100 specific locations or nodes. Each node has 12 possible approach angles, translating to over 1200 possible beam positions. The treatment planning system determines a set sequence of approach angles, beam weights, and dose distributions. The calculated plan can be incrementally improved by the physicist and physicians. The actual delivery follows a step-and-shoot sequence. The patient is placed in a position approximating that of the CT scan. Image detectors acquire radiographs of the tumor region. The image guidance system software then compares the real time radiographs with the CT information to determine location of the tumor. This information is transmitted to the robot to initialize the pointing of the LINAC beam. The robotic arm then moves the LINAC through the sequence of preset nodes surrounding the patient. At each node, the LINAC stops, and a new pair of images is acquired from which the position is determined again. Corrected position is transmitted to the robot which adapts beam pointing to compensate for any movement. LINAC delivers the preplanned dose of radiation for that position. The entire process is repeated at each node. The total time from imaging to robot compensation is about seven to 10 seconds. The total treatment time depends on the complexity of the plan and delivery paths but is comparable to standard LINAC treatments. Each treatment session ranges from 30 to 90 minutes. Physicians may elect to treat with a single dose, a hypofractionated dose typically of two to five sessions, or a more traditional fractionated regimen. Outcomes following CyberKnife treatment of vestibular schwannomas are emerging at this time.23
796
OTOLOGIC SURGERY
ALTERNATIVE APPLICATIONS Although the majority of stereotactic radiosurgery performed by the neurotologist will be for vestibular schwannoma, other neoplasms and pathologies may be amenable to radiotherapy.24 Several of these were noted earlier in the section on patient selection and this section will focus on a few common pathologies for which the neurotologist may be the primary surgeon. Paragangliomas, more specifically glomus jugulare tumors, are becoming more commonly addressed with primary radiotherapy than with surgical resection. The other chemodectomas such as glomus vagale and carotid body tumors are located too low in the neck for most gamma knife units in use today. Further, these tumors do not typically carry the same morbidity as glomus jugulare tumors and are still commonly addressed surgically. In an effort to reduce the cranial nerve palsies that often accompany glomus jugulare resection, gamma knife surgery has been employed. An evaluation of 42 patients with primary or recurrent/persistent glomus jugulare tumor undergoing gamma knife surgery showed excellent tumor response.25 Approximately 1/3 of tumors shrank and 2/3 showed no size change. A single 3.9 cm tumor was found to have increased 99 months after treatment with 12 Gy at the margin and was re-treated. Progressionfree survival was 100% at seven years and 75% at 10 years. Six patients had complications related to treatment. Five of 26 patients with intact hearing at the time of treatment had subjective decline within the first year. Objective measures of hearing were not performed. One patient had facial parasthesias, one had vocal fold paralysis (the re-treated subject), one had vertigo and imbalance, and one had post-treatment migraine requiring admission. A meta-analysis of glomus jugulare treatment and outcomes compared stereotactic radiosurgery to surgical resection.26 Neurological deficits in those treated with gamma knife, CyberKnife or LINAC showed no change in 58.2%, improved in 39% and permanently worsened in 2.8%. Such deficits included complaints of hearing loss, dizziness, dysphagia, voice change, shoulder dysfunction and headache. Overall, there was an 8.5% incidence of cranial nerve complication with 75% of these being transient. Permanent deficits occurred in three of 141 patients all of which involved facial motor dysfunction; none of which reached House-Brackmann grade VI. Tumor control was achieved in approximately 98% of individuals at 39 months. Conventional surgery for glomus jugulare had a complete resection rate of approximately 92%; some of which represented more than one surgery for resection. The recurrence rate at a mean of 82 months was 3.3%. The mortality rate was 1.3% for conventional surgery compared with 0% for radiotherapy. Cranial nerve deficits varied widely among surgical reports but on
a verage the facial nerve was affected in 4.4% to 11%; the glossopharyngeal in 26% to 42%; the vagus in 13% to 28%; the spinal accessory in 25% to 26% and the hypoglossal in 5% to 21%. Other morbidities included a CSF leakage rate of 8.3%, aspiration in 5.5% and wound infection in 5.5%. Although cranial nerve deficits occur more frequently with conventional surgery, most reports note that the long-term impact of such dysfunction is relatively small. It is important for the surgeon to take into account patient function, age and general health, and tumor size when discussing and weighing treatment options for glomus jugulare tumors. Meningiomas are the second most common benign neoplasm of the cerebellopontine angle and can often present with deficits similar to vestibular schwannomas. Total resection results in excellent tumor control rates and, for all cranial locations, shows a 15-year progression-free survival rate of approximately 68% to 75%.27,28 Experience with partially resected or inoperable meningiomas, however, has shown that radiation therapy can produce excellent tumor control in the majority of cases. The use of stereotactic radiosurgery as a primary treatment to avoid or reduce the incidence of surgical and neurological deficits is increasingly common. Elia and colleagues reviewed stereotactic radiosurgery outcomes for meningioma published since 2001.28 In over 1500 patients the 5-year progression-free control rate was 93.4%. The complication rate ranged from 2.5% to 13% and included neurological and vascular toxicities. Many of these were for tumors around the optic chiasm and carotid arteries and included dosages up to 20 Gy. Kreil and colleagues recently published their series on the treatment of 200 skull base meningiomas with gamma knife surgery.29 There were 21 patients with cerebellopontine angle lesions. Of 20 patients with preoperative hearing loss (not quantified), one improved and 19 remained stable; none showed deterioration. Tinnitus remained stable in seven of seven patients. Vertigo was present in 25 skull base meningiomas and improved in eight and worsened in none. Given the low incidence of complication and the high rate of tumor control, stereotactic radiosurgery should be strongly considered in tumors around sensitive neural structures and in patients medically unsuitable for conventional surgery. In addition to tumors, the neurotologist is often consulted for facial pain syndromes, most notably trigeminal neuralgia. Functional stereotactic radiosurgery using gamma knife has been employed in the treatment of trigeminal neuralgia. In the series of meningiomas reported by Kriel there were 25 patients with preoperative trigeminal neuralgia due to tumor of which 16 improved.29 There were two induced cases of trigeminal neuralgia but these were transient. Such findings indicate that radiation to the trigeminal nerve can induce functional changes. Gorgulho and De Salles reviewed surgical and stereotactic treatments for trigeminal neuralgia.30 Among
Chapter 65 • Stereotactic Radiosurgery of Skull Base Tumors
current treatments, long-term improvement was noted in 70% to 75% of microvascular decompressions, 58% to 77% of radiofrequency rhizotomies, 32% of balloon compressions, 17% to 50% of glycerol rhizotomies, and 45% to 57% of stereotactic radiosurgeries. Immediate improvement was noted in over 90% of patients with stereotactic radiosurgery. Recurrence rates were highest with glycerol rhizotomy and much lower and very similar among the other modalities. Stereotactic radiosurgery was noted to be particularly attractive because it is the least invasive of these methods. Many different treatment protocols for trigeminal neuralgia have been attempted.30 In their review, Gorgulho and De Salles identified several patterns with regards to gamma knife treatment for trigeminal neuralgia that affect outcomes. The root entry zone of the trigeminal nerve, not the nerve proper, should be the preferred target as dosage delivery to this area seems to correlate with pain relief. A minimal dosage of 70 Gy and maximal dosage of 90 Gy should be prescribed. The incidence of post-treatment numbness with this prescription dose ranges from 3% to 55% but bothersome numbness persists in only about 4% to 12%. Treating a longer section of the trigeminal nerve proper does not improve pain control and increases the incidence of post-treatment numbness. Likewise, higher dosage to the nerve does not improve pain control and increases numbness. The overall incidence of complications with stereotactic radiosurgery for trigeminal neuralgia is significantly lower than all other techniques. As with other benign diseases, potential long-term effects of radiation treatment need to be considered in younger individuals.
SUMMARY Stereotactic radiosurgery and radiotherapy are becoming increasingly common in the management of skull base tumors and other disorders. Whether driven by the patient, or the surgeon, the field continues to evolve rapidly. Advances are being made in improving accuracy, effective radiation dose, and parameters necessary to maximize patient outcome. These methods have advantages and disadvantages that must be openly discussed with patients who have vestibular schwannomas or other skull base tumors. It remains the responsibility of the surgeon to provide a balanced view as to the relative risks and benefits of observation, microsurgery, stereotactic radiosurgery or radiotherapy, or a combination of these methods.
FINANCIAL DISCLOSURE None of the authors has a financial interest in any of the companies discussed in this chapter.
797
REFERENCES 1. Kondziolka D, Lunsford L D, McLaughlin M R , Flickinger JC : Long-term outcomes after radiosurgery for acoustic neuroma. N Engl J Med 339:1426-33, 1998. 2. Wackym PA: Stereotactic radiosurgery, microsurgery, and expectant management of acoustic neuroma: basis of informed consent. Otolaryngol Clin North Am 38:653-70, 2005. 3. Wackym PA, Runge-Samuelson C L , Poetker D M, et al: Gamma knife radiosurgery for acoustic neuromas performed by a neurotologist: early experiences and outcomes. Otol Neurol 25:752-61, 2004. 4. Bari M E, Forster D M, Kemeny A A, et al: Malignancy in a vestibular schwannoma. Report of a case with central neurofibromatosis, treated by both stereotactic radiosurgery and surgical excision, with a review of the literature. Br J Neurosurg 16:284-9, 2002. 5. Pollock B E, Lunsford L D, Kondziolka D, et al: Vestibular schwannoma management. Part II. Failed radiosurgery and the role of delayed microsurgery. J Neurosurg 89:949-55, 1998. 6. Rowe J, Grainger A, Walton L , et al: Risk of malignancy after gamma knife stereotactic radiosurgery. Neurosurgery 60:60-5, 2007. 7. Rowe J, Grainger A, Walton L , et al: Safety of radiosurgery applied to conditions with abnormal tumor suppressor genes. Neurosurgery 60:860-4, 2007. 8. Lustig L R , Jackler R K, Lanser M J: Radiation-induced tumors of the temporal bone. Am J Otol 18:230-5, 1997. 9. Hanabusa K, Morikawa A, Murata T, Taki W: Acoustic neuroma with malignant transformation. Case report. J Neurosurg 95:518-21, 2001. 10. Poetker D M, Jursinic PA, Runge-Samuelson C L , Wackym PA : Distortion of magnetic resonance images used in gamma knife radiosurgery treatment planning: implications for acoustic neuroma outcomes. Otol Neurotol 26:1220-8, 2005. 11. Paddick I : A simple scoring ration to index the conformity of radiosurgical treatment plans. Technical note. J Neurosurg 93(Suppl 3):219-22, 2000. 12. Lunsford L D, Niranjan A, Flickinger JC, et al: Radiosurgery of vestibular schwannomas: summary of experience in 829 cases. J Neurosurg 102(Suppl):195-9, 2005. 13. Niranjan A, Lunsford L D, Flickinger JC, et al: Dose reduction improves hearing preservation rates after intracanalicular acoustic tumor radiosurgery. Neurosurgery 45:753-62, 1999. 14. Flickinger JC, Kondziolka D, Niranjan A, Lunsford L D: Results of acoustic neuroma radiosurgery: an analysis of 5 years’ experience using current methods. J Neurosurg 94:1-6, 2001. 15. Prasad D, Steiner M, Steiner L : Gamma surgery for vestibular schwannoma. J Neurosurg 92:745-59, 2000. 16. Paek S H, Chung H -T, Jeong S S, et al: Hearing preservation after gamma knife radiosurgery of vestibular schwannoma. Cancer 104:580-90, 2005. 17. Wackym PA, Hannley MT, Runge-Samuelson C L , et al: Gamma knife surgery of vestibular schwannomas: Longitudinal changes in vestibular function and measurement of the Dizziness Handicap Inventory. J Neurosurg 109(Suppl):137-143.
798
OTOLOGIC SURGERY
18. Pitts L A, Jackler R K : Treatment of acoustic neuromas. N Engl J Med 339:1471-73, 1998. 19. Ho SY, Kveton J F: Rapid growth of acoustic neuromas after stereotactic radiotherapy in type 2 neurofibromatosis. Ear Nose Throat J 81:831-3, 2002. 20. Watanabe T, Saito N, Hirato J, et al: Facial neuropathy due to axonal degeneration and microvasculitis following gamma knife surgery for vestibular schwannoma: a histological analysis. J Neurosurg 99:916-20, 2003. 21. Lee DJ, Westra WH, Staecker H, et al: Clinical and histopathologic features of recurrent vestibular schwannoma (acoustic neuroma) after stereotactic radiosurgery. Otol Neurol 24:650-60, 2003. 22. Friedman R A, Brackmann D E, Hitselberger WE, et al: Surgical salvage after failed irradiation for vestibular schwannoma. Laryngoscope 115:1827-32, 2005. 23. Chang S D, Gibbs IC, Sakamoto GT, et al: Staged stereotactic irradiation for acoustic neuroma. Neurosurgery 56:1254-61, 2005. 24. Knisely J PS, Linskey M E : Less common indications for stereotactic radiosurgery or fractionated radiotherapy for patients with benign brain tumors. Neurosurg Clin N Am 17:149-167, 2006.
25. Pollock B E : Stereotactic radiosurgery in patients with glomus jugulare tumors. Neurosurg Focus 17:63-67, 2004. 26. Gottfried O N, Liu J K, Couldwell WT: Comparison of radiosurgery and conventional surgery for the treatment of glomus jugulare tumors. Neurosurg Focus 17:22-30, 2004. 27. Goldsmith B, McDermott MW: Meningioma. Neurosurg Clin N Am 17:111-120, 2006. 28. Elia A E, Shih H A, Loeffler J S : Stereotactic radiation treatment for benign meningiomas. Neurosurg Focus 23:1-9, 2007. 29. Kreil W, Luggin J, Fuchs I, et al: Long term experience of gamma knife radiosurgery for benign skull base meningiomas. J Neurol Neurosurg Psychiatry 76:1425-1430, 2005. 30. Gorgulho A A, De Salles A A F: Impact of radiosurgery on the surgical treatment of trigeminal neuralgia 66:350356, 2006.
Self-Assessment Questions Q1. Which of the following are true regarding gamma knife surgery for vestibular schwannomas? A. ��������������������������������������������� Tumor control rates are greater than 97% B. Facial nerve motor dysfunction occurs in less than 1% with current dosing C. Trigeminal nerve dysfunction is more common with large tumors D. Malignant transformation or induction is rare E. All of the above Q2. Hearing thresholds as measured by pure-tone averages after gamma knife surgery most commonly: A. Improve immediately B. Behave similar to expectant observation C. Progress rapidly to profound deafness D. Degrade rapidly in the first six months and then slowly worsen E. Do not change Q3. Which of the following tumors would not be amenable to treatment with the more common gamma knife B and C units? A. Intracanalicular acoustic neuroma of 7 mm maximal diameter B: V estibular schwannoma extending into the CPA by 1.5 cm C. 2 cm glomus jugulare tumor extending anterosuperiorly from the jugular bulb D. Glomus vagale tumor extending to the carotid bifurcation E. Petrous apex meningioma of 2.4 cm
Q4. A patient underwent gamma knife surgery of a 2.3 cm CPA vestibular schwannoma six months ago. MRI performed today shows the tumor to be 2.7 cm in maximal diameter. The patient is asymptomatic. The next best course of action is: A. Assure patient this is normal and rescan in six months B. Recommend microsurgical resection for radiation failure C. Counsel patient this may be malignant D. Plan a second round of gamma knife surgery E. Start high-dose steroids with a taper Q5. Which of the following is/are not a component of the gamma knife surgery system? A. Trunnions B. Collimator helmet C. Gamma calipers D. Cobalt 60 (60Co) sources and beam channels E. MRI fiducial box
66
Vascular Considerations in Neurotologic Surgery Robert F. Spetzler, Shaun C. Desai, Vivek R. Deshmukh, and Shervin R. Dashti
The foundation for management of vascular lesions of the petrous bone and cerebellopontine angle (CPA) relies on an understanding of the relevant skull base anatomy and vascular anatomy of this region. Vascular lesions are quite varied and include carotid dissections; carotid and vertebrobasilar aneurysms; arteriovenous fistulas; cavernous malformations; and hypervascular tumors including glomus tumors, hemangiomas, and hemangioblastomas, among others. This chapter presents a concise and clinically relevant review of the anatomy and the diagnostic and therapeutic strategies for these complex lesions.
VASCULAR ANATOMY OF THE PETROUS BONE AND CEREBELLOPONTINE ANGLE Internal Carotid Artery The petrous segment of the internal carotid artery (ICA) begins as the vessel enters the periosteal-lined carotid canal, whereas the cavernous segment originates as it exits the apex of the petrous portion of the temporal bone and traverses the posterior part of the cavernous sinus. The petrous portion of the ICA, which lies immediately posterior to the bony eustachian tube but anterior to the jugular foramen, can be divided into vertical and horizontal segments that connect at the genu. The ICA is lined by cervical sympathetic ganglia throughout its course in the petrous canal, and aneurysms or lesions of the vessel that compress the nerves can cause a classic Horner’s syndrome.1 The ICA is also lined by a venous plexus in the distal part of the canal. The greater superficial petrosal nerve, which originates from the geniculate ganglion of the facial nerve, usually courses above and parallel to the horizontal portion of the petrous ICA. Special caution should be used when exposing the middle fossa floor and the petrous ICA because excessive traction of this nerve can cause postoperative facial nerve palsy.2 Similarly, the trigeminal ganglion lies over the medial portion of the petrous ICA, but it is usually separated from the vessel by dura or a variable layer of bone. The petrous ICA has four main branches of importance to the neurotologic surgeon: (1) caroticotympanic
artery, (2) stapedial artery, (3) artery of the pterygoid canal (vidian), and (4) periosteal artery.1,3 The caroticotympanic artery is often said to be a remnant of the embryonic hyoid artery, and arises from the proximal vertical part of the petrous ICA laterally and supplies the tympanic cavity. It anastomoses most commonly with the inferior tympanic artery. The stapedial artery is the source of the middle meningeal artery during development, but it rarely persists into adulthood. If present, a persistent stapedial artery passes through the floor of the middle ear and runs superiorly in a bony canal, in the obturator foramen of the stapes, and finally through the fal lopian canal into the middle cranial fossa.3 Finally, the artery of the pterygoid canal (vidian) and the periosteal artery are often thought of as small collateral routes most likely branching off from the horizontal segment of the petrous portion or from the internal maxillary artery, and may have relevance in interventional embolization procedures as described later on.1 The intracavernous segment of the ICA begins at the level of the petrolingual ligament near the foramen lacerum, and most consistently gives off two branches known as the dorsal (meningohypophyseal trunk) and lateral (artery of the inferior cavernous sinus) main stem arteries.4 The intracavernous ICA is commonly divided into five segments for anatomic orientation, including a posterior vertical, posterior bend, horizontal, anterior bend, and anterior vertical segments.1 The dorsal main stem artery generally branches from the central one third of the convex outer margin of the posterior bend of the ICA. There is great variability in the origin of the branches of the dorsal main stem, but most investigators agree that three vessels—the tentorial artery (artery of Bernasconi-Cassinari), dorsal meningeal artery, and inferior hypophyseal artery—exist in some form.3 The tentorial arteries travel along the tentorium and may contribute blood supply for tentorial or proximal falcine meningiomas and tentorial dural arteriovenous fistulas, and blood supply to portions of CN III and IV.3,5 The inferior hypophyseal artery supplies the periphery of the anterior pituitary gland and forms a “circulus arteriosus” with the dorsal meningeal arteries around the root of the 799
800
OTOLOGIC SURGERY
dorsum sella. The distal branches of the dorsal meningeal arteries may enter the internal auditory meatus as described in a few patients, and may represent a remnant of the fetal trigeminal artery.3 The second main branch, the artery of the inferior cavernous sinus, generally arises from the central one third of the horizontal segment of the cavernous ICA. Although there is some variability, most studies agree that this vessel supplies various nervous components of the cavernous sinus. Although the dorsal and lateral main stem vessels are the most common branches given off by the intracavernous segment of the ICA, other arteries have been described. McConnell’s capsular artery, when present in about 8% of the population, originates from the horizontal segment of the intracavernous ICA, and supplies any combination of the inferior and peripheral aspect of the anterior lobe of the pituitary gland, the diaphragma sella, and the floor of the sella turcica.6 A persistent fetal trigeminal artery, although rare with an incidence of 0.06% to 0.6%, usually arises from the posterolateral or posteromedial aspect of the intracavernous ICA.3 It is most notable in the literature for producing unique clinical scenarios, including trigeminal neuralgia, oculomotor palsies, hyperprolactinemia, and an increased incidence of intracranial aneurysms.3 Finally, the ophthalmic artery and a superior hypophyseal artery have been shown to originate from the cavernous ICA in 1% to 7.5% and 16% of individuals, respectively.3,7
posterior auricular artery arises above the posterior belly of the digastric and travels between the parotid gland and the styloid process. The artery divides into posterior auricular and occipital branches, and develops a stylomastoid branch, which supplies the structures of the stylomastoid foramen and the facial nerve.2 Although not technically a direct branch of the external carotid artery, but rather a branch of the internal maxillary artery, the middle meningeal artery has clinical relevance in the various neurotologic approaches. As this vessel courses through the foramen spinosum of the sphenoid bone between the two roots of the auriculotemporal nerve, it enters the middle cranial fossa, dividing into two or three main branches. When performing extradural approaches to the floor of the middle fossa, the foramen spinosum is located anteromedial to the geniculate ganglion and anterolateral to the carotid canal. These branches—the anterior branch (frontalis), posterior branch (parietal), and middle branch—supply most of the supratentorial dura and calvaria. On entering the cranium, it also gives off the superficial petrosal artery, which enters the hiatus of the facial canal and supplies the facial nerve. Manipulation of this branch during transpetrosal approaches can result in facial nerve injury.2 Finally, the middle meningeal artery also gives rise to the superior tympanic artery just superior to the foramen spinosum, which runs with the superficial petrosal nerve through the superior tympanic canaliculus and supplies the canal of the tensor tympani muscle.
External Carotid Artery
Vertebrobasilar Arteries
Proximal to its terminal bifurcation, the external carotid artery gives rise to an anterior and posterior group of vessels, the latter of which consists of three arteries that relate to the petrous portion of the temporal bone. The ascending pharyngeal artery usually originates at or near the main bifurcation of the ICA and external carotid artery, and supplies the meninges around the jugular foramen as it passes through the foramen lacerum. As it travels upward with the carotid arteries, it gives off the inferior tympanic artery, which travels through the tympanic cavity with Jacobson’s nerve (inferior tympanic) via the tympanic canaliculus.2 The occipital artery arises from the posterior surface of the external carotid artery and traverses upward between the posterior belly of the digastric muscle and the internal jugular vein, and then medial to the mastoid process. After passing the longissimus capitis muscle, the vessel courses deep to the splenius capitis muscle, finally terminating at the fascia between the attachment of the sternocleidomastoid and the trapezius muscles at the superior nuchal line.2 Its branches supply several muscular and meningeal branches and often anastomose with branches of the external carotid and vertebral arteries.2 Its most notable branch is the mastoid artery, which supplies the posterior portion of the mastoid bone and runs through the mastoid foramen. Finally, the
The vertebrobasilar system has three main branches that pass through the CPA, and any damage to these delicate branches can lead to various brainstem stroke syndromes. The major arterial vessels to consider include the posterior inferior cerebellar artery (PICA), the superior cere bellar artery (SCA), and, most relevant to the CPA, the anterior inferior cerebellar artery (AICA). The AICA typically originates from the lower part of the basilar artery and divides into rostral and caudal trunks as it traverses the central part of the CPA. This main bifurcation often occurs proximal to the facial and vestibulocochlear nerves, and the two trunks formed are usually nerve-related.8 These branches can be divided into a premeatal segment, meatal segment, and subarcuate loop. In one study, a laterally convex curve or “loop” from the meatal segment was found to be medial of the internal acoustic meatus lying in the CPA in 33%, at the entrance of the meatus in 27%, and entering the internal auditory canal (IAC) in 40% of patients.8 A second laterally convex curve, known as a subarcuate loop, can also be seen in the subarcuate fossa and often gives the appearance of an M configuration.8 Although rare, in a few cases the lateral convex loop of the AICA can be embedded in the dura or bone covering the subarcuate fossa, or both.9 In such cases, freeing the artery from the dura and, if
Chapter 66 • Vascular Considerations in Neurotologic Surgery
801
ecessary, drilling the bone around the margin may aid n in freeing this segment in anticipation for exposure of the posterior wall of the internal acoustic meatus.9 In the area of the CPA, the various branches of the AICA give rise to several nerve-related arterial branches from which originate the (1) labyrinthine (IAC), (2) recurrent perforating, (3) subarcuate, and, less commonly, (4) cerebellosubarcuate arteries. The labyrinthine arteries follow the vestibulocochlear nerve into the IAC, supplying the nerves, dura, and bone of the canal, and eventually terminate by giving rise to the anterior vestibular and common cochlear arteries, which supply the inner ear. Although the labyrinthine arteries most commonly arise from the premeatal segment of the AICA, some studies have reported anomalous cases where it originated directly from the basilar artery or from the PICA, or recurrent perforating, subarcuate, and cerebellosubarcuate arteries; however, this difference may be partially explained by the exact anatomic definition of the arteries in the different studies.8 The recurrent perforating arteries often course near the meatus and along the facial and vestibulocochlear nerves before supplying the brainstem. Finally, the subarcuate artery usually travels medial to the meatus, traversing the subarcuate fossa and petromastoid canal, sending branches to the petrous apex, the bone of the semicircular canals, and the posterosuperior portion of the vestibule. The PICA, by definition, originates from the vertebral arteries and usually bifurcates into a medial and lateral trunk at the level of the telovelotonsillar fissure; however, its branches are extremely variable. Because it is located inferiorly, it passes very close to the roots of the lower cranial nerves. The PICA can arise from outside the dura, and at any point from the intradural course of the vertebral artery. The SCA originates most commonly from the basilar artery below, but adjacent to the origin of the posterior cerebral artery. The SCA generally bifurcates into rostral and caudal trunks, and branches from either make up the cortical arteries, which supply the upper two thirds of the petrosal surface, including both lips of the petrosal fissure.8 The first branch of the cortical arteries, when present in about half of the population, is referred to as a marginal branch, and its surface area supplied is inversely related to the supply of the AICA. Several anastomoses exist between the marginal artery and the AICA.
extent of the sigmoid sinus. In addition to draining the petrosal veins that drain the cerebellum and brainstem, it also receives vessels from the inferior surface of the temporal lobe and the cavernous sinus. The transverse sinus begins as a confluence of sinuses at the level of the internal occipital protuberance between the tentorium cerebelli and occipitalis bone and runs laterally to join the sigmoid sinus. The anastomotic vein of Labbé, which drains the temporoparietal region, bridges the inferior surface of the temporal lobe to the transverse sinus. This vein is of special importance in petrosal approaches to the skull base because its injury can lead to speech disturbance and contralateral hemiparesis or hemiplegia.2 This vein is particularly at risk during subtemporal exposures. Excessive retraction of the temporal lobe can result in avulsion of the vein from its insertion into the relatively nonmobile transverse-sigmoid junction. The sigmoid sinus not only drains the superior petrosal sinus, but also receives tributaries from the brainstem, cerebellum, and occipitalis, and vertebral emissary veins. The angle formed between the superior petrosal and sigmoid sinuses and the middle fossa dura represents the sinodural angle, which is an important anatomic landmark. The sinus curves medially and forward, crossing the occipital bone, and finally through the jugular foramen where it enters the superior bulb of the internal jugular vein. Finally, the inferior petrosal sinus connects the cavernous sinus to the medial wall of the bulb of the internal jugular vein as it traverses the petro-occipital fissure and drains the clival area. At its entrance to the jugular foramen, the jugular vein is separated from the inferior petrosal sinus by CN IX, X, and XI. This foramen is located at the lower end of the petro-occipital fissure, and is divided into a larger lateral opening that drains the sigmoid sinus and into a small medial petrosal part that receives the inferior petrosal sinus.2 The jugular bulb is inferior to the posterior floor of the middle ear cavity and generally lies inferior to the ampulla of the posterior semicircular canal; however, its location within the tympanic cavity is highly variable. Often, a high-riding jugular bulb can extend superiorly as high as the lateral semicircular canal and can interfere with visualization in Trautmann’s triangle in transpetrosal approaches as discussed in the next section.2 Finally, the internal jugular vein runs inferiorly from the bulb and is located posterolaterally to the ICA.
Dural Venous Sinuses
VASCULAR CONSIDERATIONS IN NEUROTOLOGIC APPROACHES
One must have an intimate knowledge of venous anatomy in neurotologic surgery to decrease the risk of complications from venous congestion. The petrosal vein (Dandy’s vein) parallels the trigeminal nerve just beneath the tentorium and drains into the superior petrosal sinus. The superior petrosal sinus, which lies at the attachment of the tentorium cerebelli to the superior margin of the petrous ridge, connects the cavernous sinus to the lateral
Transpetrosal Approaches The transpetrosal approach, also known as the transtemporal approach, consists of three craniotomy techniques that open the posterior fossa dura anterior to the sigmoid sinus, through the posterior aspect of the petrous pyramid (Fig. 66-1). This set of approaches requires removal of
802
OTOLOGIC SURGERY
gauze. Intraluminal packing of the sigmoid sinus is not recommended because there is always a risk for embolization of the packing material. Transcochlear Translabyrinthine Retrolabyrinthine
FIGURE 66-1. Surgical corridor afforded by the various transpetrosal approaches. (Courtesy of Barrow Neurological Institute.)
1 to 2 cm of retrosigmoid bone to allow for posterior displacement of the sigmoid sinus. A detailed explanation of the surgical technicalities and indications and exposure to the various approaches is provided elsewhere.
Retrolabyrinthine In this approach, bone is removed up to the bony capsule of the semicircular canals, which provides a limited exposure to the midportion of the CPA. Although this approach was first described as a root entry zone in tic douloureux, it has more recently been used for much less common problems, including vascular decompression of CN VIII for tinnitus and chronic vertigo, and for approaching midbasilar artery aneurysms. More commonly, this approach is used in combination with others, including a subtemporal exposure for transtentorial lesions abutting the brainstem. After the initial incision is made behind the postauricular sulcus, and the mastoid air system is thoroughly exenterated, the posterior limit for drilling is the sigmoid sinus. Using a large diamond burr, the sigmoid sinus must be skeletonized carefully. Some authors have suggested that a thin bone plate be left intact over the anterior aspect of the sinus to protect the vessel from the rotating shaft of the drill as it goes deeper into the temporal bone.2 If this bone plate begins to restrict sigmoid retraction, however, one can always fragment it to facilitate posterior mobilization. Other surgeons believe there is no need for preservation of this bony island, and they remove it and guard the vessel with a retractor. Either way, further exposure must be attained by retracting the sinus superiorly from the jugular bulb. In selected patients, the sigmoid sinus can be ligated to improve exposure; however, this can occur only when angiography shows a patent communication between the two transverse sinuses. Mastoid emissary veins are usually encountered and can be controlled with bipolar cautery; similarly, any lacerations to the sinus can readily be controlled with extraluminal tamponade with hemostatic
Translabyrinthine In the translabyrinthine approach, the semicircular canals and vestibule are completely removed, which allows for excellent access to the IAC and exposure of the CPA. The main indication for this approach is for acoustic neuromas; it also is indicated for meningiomas, epidermoids, and other tumors of the CPA when hearing has been compromised. This approach has also been described for aneurysms of the midportion of the basilar artery. In a translabyrinthine approach, the same precautions must be taken for the sigmoid sinus as discussed earlier for the retrolabyrinthine approach. Sigmoid mobilization must be achieved, however, to have adequate access between the sinus and the external auditory canal. Although there is some anatomic variation to the sigmoid sinus, full exposure can usually be achieved after decompression from the transverse-sigmoid junction superiorly to the jugular bulb inferiorly. A high anterior sigmoid course or a large sinus may inhibit access to the inferior CPA; however, this should not be a problem with acoustic neuroma removal because this neoplasm becomes more accessible in the operative field as it is debulked. Finally, when drilling the mastoid bone toward the semicircular canals, the superior petrosal sinus should be exposed at the sinodural angle. During the labyrinthectomy portion of the procedure, as the lateral and superior semicircular canals are drilled away, bleeding may occur. This is most likely a result of the subarcuate artery because it courses in the bone near the center of the superior semicircular canal. The boundaries of Trautmann’s triangle, the patch of posterior fossa dura in front of the sigmoid sinus, are also important. One limb of the V-shaped incision extends below to the superior petrosal sinus, and the other limb extends above the jugular bulb. The jugular bulb sometimes can lie atypically high behind the posterior semicircular canal and impede the creation of the inferior bony trough, interfering with access to the IAC. This variation can increase the chance of unintentional bleeding and create a potential source for air emboli. If this occurs, better exposure of the superior aspect of the IAC can be achieved by elevating the temporal dura and extending the superior trough to the level of the tentorium. Decompression and inferior displacement of the jugular bulb can always be done, however, if the former option fails to reveal exposure. As the dural flap is reflected posteriorly to expose the meatal structures and the CPA, the subarcuate artery or AICA may be encountered. As noted earlier, the subarcuate artery usually stems from the AICA and passes through the dura on the upper posterior wall of the meatus. This vascular anatomy tends to be variable, however, and in
Chapter 66 • Vascular Considerations in Neurotologic Surgery
some patients the subarcuate artery, along with its origin from the AICA, may be incorporated into the dura on the posterior face of the temporal bone.9
Transcochlear The transcochlear approach is mainly an anteromedial extension of the translabyrinthine approach; the entire inner ear is removed, and the facial nerve is posteriorly rerouted. The surgical extension allows for access to the midclival region, prepontine cistern, and anterior aspect of the CPA. This approach has lost favor in recent years because of substantial morbidity (deafness, facial nerve palsy), although is still indicated in cases of preexisting facial palsy and for life-threatening vascular lesions, such as midbasilar artery aneurysms. The beginning steps of the transcochlear approach represent the same vascular considerations as in the translabyrinthine approach. Similarly, the boundaries in this approach have several important vascular relationships. Medially, the drilling extends to the edge of the clivus, exposing the inferior petrosal sinus from the jugular bulb below to the superior petrosal sinus above. Anteriorly, the ascending portion of the petrous carotid is exposed. After extending dissection medially into the clivus and opening the dura, the initial segment of the AICA and its origin from the basilar artery can be achieved.
Middle Fossa Approaches The middle fossa approaches are a group of procedures that give complete exposure to the IAC and a limited view of the CPA. Several variations exist to this set of approaches, which vary in the exposure of the IAC and CPA. Only Kawase’s approach is discussed here, however.
Kawase’s Approach The middle fossa–transpetrous apex approach, or Kawase’s approach, has mainly been used for lesions that traverse the middle fossa and a small part of the posterior fossa. These include dumbbell-shaped tumors such as petroclival meningiomas and trigeminal schwannomas, and lesions in the petrous apex. This approach can also be used to access aneurysms of the mid-basilar artery and lower basilar artery. Kawase’s approach involves a small posterior fossa craniotomy, after a middle fossa opening, by removing the medial portion of the petrous pyramid and the lateral aspect of the clivus. The ultimate goal is exposure of the anterior CPA and the ventral surface of the pons. As the dura is carefully elevated from the floor of the middle fossa, special attention should be paid to exposing the middle meningeal artery. If necessary, this artery can be sacrificed and divided at the foramen spinosum to avoid excessive bleeding. The boundaries of Kawase’s
803
approach are the IAC posteriorly, petrous carotid artery anteriorly, inferior petrosal sinus inferiorly, and abducens nerve medially. The boundaries of the rhomboid-shaped bony window in the petrous pyramid have several important vascular relationships. The most important is the inferior border of the petrous carotid artery. Often, this vessel can limit posterior fossa exposure. Some authors have advocated that displacement of the intrapetrous carotid artery from its bony canal can be achieved; however, doing so is controversial and probably not justified because of its risk for significant morbidity. More commonly, if more exposure is needed, the horizontal segment of the petrous carotid artery can be skeletonized, leaving a surrounding thin eggshell of bone intact allowing for slightly more mobilization and visualization. A 7 mm length segment of the petrous carotid artery may be exposed without a covering of bone in the floor of the middle fossa deep to the greater petrosal nerve.
Exposure of the Petrous Carotid The petrous ICA may require exposure in particular for revascularization procedures. The petrous ICA is typically located in Glasscock’s triangle, and can be exposed by drilling bone posteromedial to foramen ovale and anterior to the arcuate eminence. The greater superficial petrosal nerve runs superficial to and parallel to the petrous ICA, and can be sacrificed if necessary with resultant disturbance in lacrimation. As stated earlier, excessive traction on the greater superficial petrosal nerve can cause facial nerve palsy.
VASCULAR LESIONS OF THE PETROUS BONE AND CEREBELLOPONTINE ANGLE Cerebral Aneurysms Petrous Internal Carotid Artery Aneurysms Presentation and Pathophysiology Aneurysms of the petrous ICA are rare, although their true incidence is unknown. They are often discovered incidentally on computed tomography (CT) scans for unrelated purposes. They are usually asymptomatic; however, patients can present to an otolaryngologist with a wide variety of signs and symptoms, including cranial nerve palsies, Horner syndrome, epistaxis, vertigo, pulsatile tinnitus, and otorrhagia.1 The triad of otorrhagia, epistaxis, and neurologic deficit is almost pathognomonic for a petrous ICA aneurysm.10 On otoscopic examination, patients with aural symptoms can present with a vascular retrotympanic mass, which is often mistaken for a glomus tympanicum tumor. Biopsy of such a lesion can lead to severe hemorrhage and can be catastrophic. Spontaneous rupture can cause dramatic hemorrhage into the eustachian tube or middle ear or both, and manifest as massive
804
OTOLOGIC SURGERY
epistaxis or otorrhagia or both with the risk of death from exsanguination. These lesions can be classified further as true aneurysms or pseudoaneurysms, which can alter treatment. True aneurysms have walls that are continuous with the unaffected portion of the parent vessel and can develop from a congenitally weakened wall, whereas pseudoaneurysms solely involve the adventitial layer of the vessel and develop when a thrombus and fibrous tissue capsule form in response to a traumatic injury. The pathophysiology of petrous ICA aneurysms is unclear, but traumatic, mycotic (by direct seeding from the middle ear and eustachian tube), and congenital causes have been implicated.11 Trauma is a significant cause of petrous ICA pseudoaneurysms in particular because of the predisposing anatomic arrangement, with a stationary ICA segment connected to a distal mobile cervical segment, predisposing the vessel to stretch forces from forceful trauma. Iatrogenic trauma, especially from temporal bone surgery and myringotomy, has been a well-documented source of petrous ICA aneurysms.1 Management and Treatment After the diagnosis is confirmed by cerebral angiography, the question of management arises. Management of petrous ICA aneurysms remains undefined, with no established treatment paradigm in place. Most authors agree, however, that patients with ruptured petrous ICA aneurysms should undergo urgent treatment to stop active bleeding or to prevent future hemorrhage. Similarly, patients with unruptured petrous ICA aneurysms who experience major chronic symptoms from cranial nerve involvement should undergo intervention. A third group of patients, with incidental findings on CT and who are asymptomatic or experience minimal and minor symptoms, raise the question of whether intervention is necessary. Most surgeons agree that the treatment of asymptomatic patients should be approached on a caseby-case basis. It is reasonable to recommend conservative management, including close observation with serial angiography (magnetic resonance or CT angiography may suffice). Change in lesion size or morphology should prompt further consideration for treatment. Management of petrous ICA dissections should primarily be based on the degree of resultant stenosis of the true lumen. For non–flow-limiting dissections, antiplatelet therapy usually suffices. For dissections with hemodynamically or clinically significant stenoses, stent placement for artery repair should be considered. Various interventional options exist for patients with petrous ICA aneurysms, including endovascular and surgical therapies. Endovascular techniques consisting of coil embolization or stent coiling should be considered the first line of therapy in symptomatic patients or in patients with ruptured aneurysms. For aneurysms that are not amenable to reconstructive procedures, parent vessel sacrifice (deconstructive procedures) should be considered.
A balloon occlusion test with hypotensive challenge is usually performed to evaluate for adequate collateral flow and to assess for any neurologic symptoms or signs of insufficient blood supply. This test produces a high false-negative rate, however, and 2% to 22% of patients with negative findings have been found to have early or delayed ipsilateral cerebral ischemia after permanent occlusion.1 Also, a false-positive result may be obtained if the procedure is complicated by thromboemboli. If the transient occlusion produces no new neuro logic deficits, and if adequate collateral flow exists, occlusion of the ICA can be considered. Detachable balloons were frequently used in the past. More recently, coil occlusion has become the predominant modality for parent vessel sacrifice. Although coils have the advantage of providing a rapid and permanent arterial occlusion, they also have an increased risk of clot formation and distal embolization compared with immediate vessel occlusion with balloon embolization.12 Some reports have shown formation or growth of cerebral aneurysms after balloon or coil embolization.1 Numerous case reports have documented the successful use of stents to treat ruptured petrous ICA aneurysms. Cheng and colleagues12 and Auyeung and associates13 described the successful use of covered stents in three hemodynamically unstable patients who had massive epistaxis as a result of rupture of petrous ICA pseudoaneurysms that were radiation induced. A covered stent may be placed in the petrous ICA because of its lack of major arterial branches, and can be considered for pseudoaneurysms because of their lack of surrounding support. Cheng and colleagues12 concluded that porous stents can alter blood flow pattern, which readily results in stasis and thrombosis of the aneurysm; they cited numerous reports of the successful use of secondary coiling in case of stent failure. Drawbacks to stents include stent-induced intimal hyperplasia, which can cause a hemodynamically significant stenosis and thromboembolic events.14 Longterm studies are still ongoing to determine its overall efficacy and patency rates; however, initial studies show a promising role for this technology. If a patient develops neurologic sequelae after a balloon occlusion test, revascularization should be considered through an extracranial-intracranial bypass. Similarly, a patient with no neurologic deficits on temporary occlusion who exhibits marked asymmetric decrease in hemispheric blood flow (<30 mL/100 g/min) should be a candidate for a revascularization procedure because of the high false-negative rate of the screening test.15 Some groups advocate universal revascularization in patients who undergo carotid artery sacrifice. The rationale for universal revascularization includes the high incidence of false-negative results from balloon test occlusion, the risk of delayed ischemic complications, and the risk for delayed intracranial aneurysm formation because of changes in intracranial flow and increased shear stress at points of bifurcation.
Chapter 66 • Vascular Considerations in Neurotologic Surgery
Surgical options should also be considered if endovascular techniques fail, and include cerebral revascularization procedures with high-flow or low-flow bypass. Umenzu and coworkers16 described a 21-year-old man who had angiographic evidence of a petrous ICA. He failed balloon occlusion twice and eventually was successfully treated via a superficial temporal artery–to–middle cerebral artery bypass.16 Five patients with petrous ICA aneurysms were treated successfully by saphenous vein interposition graft through an extradural cervical carotid artery–to–petrous ICA bypass at another institution.17 Surgical techniques described include a modified infratemporal fossa approach and a combined subtemporal and preauricular infratemporal fossa approach. Liu and colleagues1 describe a less invasive approach in which a submandibular cervical carotid artery–to– supraclinoid ICA interpositional saphenous vein graft precluded the need for direct exposure of the aneurysm or extensive drilling of the temporal bone. From our experience of eight patients with traumatic ICA dissections, we have had great success with saphenous vein ICA bypasses (cervical ICA–to–petrous ICA) with only one patient experiencing early postoperative graft occlusion that was amenable to immediate thrombectomy.18 Relative contraindications to direct surgical repair or bypass include poorly controlled active hemorrhage and lesion extending distal to the carotid canal, which would impede exposure.
High Cervical Internal Carotid Artery Aneurysms Aneurysms of the high cervical ICA are rare and usually occur as a result of trauma, dissection, and atherosclerosis, and iatrogenically from carotid artery surgery.19 Most often, these vascular lesions reach clinical attention after causing symptomatic distal embolization, thrombosis, mass effect, or hemorrhage. Management of an aneurysm in this location is similar to management of aneurysms in the petrous ICA; a balloon occlusion test should be performed to assess for collateral circulation. If the test shows insufficient supply, a revascularization procedure is necessary. Several techniques have been proposed, including a sacrifice of the ICA followed by a standard bypass graft to the middle cerebral artery, exclusion of the aneurysmal ICA by an extra-anatomic cervicopetrous bypass, and a cervicosupraclinoid bypass.19 If the aneurysm is small, and if the ICA tortuosity permits, treatment with a stent (bare or covered) can be considered. Small cervical ICA aneurysms typically can be managed conservatively with the patient placed on aspirin therapy to prevent thromboembolic complications.
Vertebrobasilar Aneurysms Numerous approaches have been described for vertebrobasilar aneurysms, and most often have depended on the precise location of the lesion in the vessel.
805
Aneurysms of the upper third of the basilar trunk are considered more accessible via an orbitozygomatic or subtemporal approach, whereas aneurysms arising from the vertebral artery usually can be accessed via a retrocondylar far-lateral approach.20,21 Surgical access to aneurysms of the lower basilar trunk and vertebrobasilar junction have been controversial, however, and various approaches have been attempted, including the subtemporal with drilling of Kawase’s triangle, suboccipital-retromastoid, far-lateral transcondylar, and trans oral-transclival.20 Exposure is often limited because these vascular lesions are deep within the surgical field and surrounded by vital structures, including brainstem and cranial nerves. More recently, lateral approaches that traverse the petrous bone have gained favor as the approach of choice for such aneurysms. In a series of nine patients with aneurysms of the lower basilar trunk and vertebrobasilar junction, direct clipping via a transpetrosal approach was successfully performed in eight.20 The operative technique consisted of a combined supratentorial/infratentorial craniotomy via burr holes bordering the transverse sinus that were connected using a high-speed drill. After craniotomy, a posterior petrosectomy is performed with careful attention to preserve the labyrinthine structures (particularly the posterior semicircular canal). The temporal dura is excised parallel to the transverse sinus, followed by removal of the presigmoid dura in Trautmann’s triangle to the level of the superior petrosal sinus. This sinus is ligated, the tentorium is transected, and the structures are retracted to expose the basilar artery to the level of the vertebrobasilar junction and to the vertebral artery. The origin of the aneurysm is dissected and clipped. A major advantage over this approach and other skull base approaches compared with standard craniotomies is minimal retraction of brain tissue.
Dural Arteriovenous Malformations Dural arteriovenous malformations are extremely rare, constituting 10% to 15% of all intracranial arteriovenous malformations.5 This vascular anomaly is believed to derive from venous thrombosis of a dural sinus with consequent collateral revascularization that can lead to abnormal fistulous connection between arteries and veins. Dural arteriovenous malformations can be highrisk or low-risk lesions; this differentiation is based primarily on the presence of venous drainage into the cortical veins, which confers high risk for future hemorrhage. Most patients with high-risk cases present with intracranial hemorrhage or progressive neurologic deficit attributed to venous hypertension. Low-risk fistulas are frequently discovered incidentally or manifest with pulsesynchronous bruit. Low-risk fistulas can generally be observed or treated with palliative embolization if the patient is intolerant of the bruit. High-risk fistulas should be treated because
806
OTOLOGIC SURGERY
of the risk for future hemorrhage. Dural arteriovenous malformations are usually amenable to endovascular treatment because the arterial and venous sites of the fistula can usually be accessed. When surgery is necessary, endovascular exploration should still be attempted because it helps guide the surgical approach by simplifying the anatomy and allowing the surgeon to pinpoint the exact location of the fistula. Preoperative arterial embolization may also reduce intraoperative blood loss. Inadequate access to the arterial or draining venous system or multiple sites of fistulization may necessitate surgical management. Endovascular and surgical techniques should complement each other in the management of this group of vascular malformations. For patients with high-risk dural arteriovenous fistulas, treatment with single-mode or multimodal therapy results in a high likelihood of achieving excellent long-term outcomes: 95% have attained Glasgow Outcome Scale scores of 4 or 5 in certain studies.5 The following sections discuss the surgical approaches to dural arteriovenous fistulas traversing the two most common locations for this lesion to develop: the transverse-sigmoid junction and the tentorium.
Transverse and Sigmoid Fistulas The junction of the transverse and sigmoid sinuses is the most common location for dural arteriovenous fistulas. Lesions in this area usually manifest with headache and pulse synchronous bruit. Lesions without cortical venous drainage may be managed conservatively with palliative embolization considered for intractable bruit. Embolization techniques have increasingly become the treatment of choice for dural arteriovenous fistulas located at this sinus junction; however, surgical management may be necessary if transvenous access cannot be attained or in cases of multiples sites of fistulization along the sinus. High-risk lesions are defined by cortical venous drainage. Two types of fistulas have been observed: fistulas with a single leptomeningeal vein along the tentorium in the supratentorial compartment and fistulas in which the entire sinus is involved. Patients with a single fistula focus should have the transverse sinus exposed at the origin of the retrograde draining vein in the region of the tentorium. A lumbar drain should be placed to relax the brain and help with exposure. The arterialized draining vein is identified and coagulated by cauterization or clip-ligation. In the second type of tentorial fistula, the surgeon should expose a significant portion of the sinus for direct vascular access. The sinus can be cannulated intraoperatively and obliterated with coils. Care must be taken, however, to avoid interrupting drainage from the normal venous circulation, particularly from the vein of Labbé. Alternatively, the transverse sinus can be skeletonized, and dural borders above, below, and medial to the sinus can be coagulated. Postoperative angiography
should be performed in all cases to ensure adequate obliteration of the fistulas.
Tentorial Fistulas Preoperative angiograms must be reviewed closely to identify the exact focus of the anomalous fistulous connection. The surgical approach depends on the location of the fistulous connection in the tentorium. If the fistula is situated in the anterior third of the tentorium, an orbitozygomatic approach is most likely the best option; a presigmoid transpetrosal approach is also a reasonable alternative. After achieving exposure by opening the sylvian fissure or retracting the temporal lobe, the tentorium is followed posteriorly until venous drainage is identified most likely through the basal vein of Rosenthal. The vein is coagulated or clip-ligated at the point of dural attachment. Lesions of the middle third of the tentorium, with venous drainage from the petrosal vein or Dandy’s vein, generally require a retrosigmoid-suboccipital approach. Lumbar drainage is recommended to allow for cerebellar relaxation. After achieving cerebellar retraction, the wing of the tentorium is followed medially to the point of the fistula. If the network of normal draining veins complicates identification of the leptomeningeal drainage, microvascular Doppler ultrasonography showing retrograde venous flow or intraoperative angiography can be performed. Venous drainage is clip-ligated or coagulated. Finally, a midline supracerebellar-infratentorial ap proach is recommended for dural arteriovenous malformations traversing the posterior aspect of the tentorium. Cauterization of occipital artery feeders and the meningeal supply during the approach significantly reduces blood flow to the fistula and facilitates surgical revision. Preoperative arterial embolization of external carotid artery branches decreases blood loss and can help mitigate eventual obliteration of the cortical draining veins. When the dura is opened, and the fistula is identified, the vein is cauterized or clip-ligated. Postoperative angiography is recommended to confirm obliteration of the lesion.
Cavernous Malformations and Cavernous Hemangiomas Cavernous malformations and hemangiomas of the petrous bone are rare and often manifest with deficits of CN VII and VIII. Other, more life-threatening presentations are possible, however, including epidural hemorrhage as described by Kessler and associates in 1956.22 There are two subtypes of cavernous malformations, and they are histologically identical. Extra-axial cavernous hemangiomas typically affect the petrous bone, and intraaxial cavernous malformations within the posterior fossa are typically located within the brainstem or cerebellum.
Chapter 66 • Vascular Considerations in Neurotologic Surgery
Cavernous hemangiomas are most commonly located at the fundus of the IAC or in the region of the geniculate ganglion, and can often spread to the CPA.23 These lesions generally produce symptoms by compressing the surrounding cranial nerves causing resultant palsies. They have the capability of perineural invasion, however, which would necessitate resection of the nerve with a cable graft placement.23 These vascular malformations can often resemble an acoustic neuroma in presentation clinically and radiographically, but they can be distinguished from a meningioma based on negative angiography.24 Diagnosis is best made with thin-section, gadolinium-enhanced magnetic resonance imaging (MRI). They are often sharply defined and enhanced compared with cholesteatomas and schwannomas.23 The presence of calcification within the lesion strongly suggests a hemangioma. Sundaresan and colleagues24 described three patients with vascular lesions in the petrous bone, two of which were cavernous hemangiomas, and concluded that these lesions can most likely be distinguished from neuromas by their presentation. The authors observed that small IAC or CPA hemangiomas tend to have early-onset hearing loss associated with marked facial weakness or fasciculation or both. In contrast, neuromas in the same locations, which also manifest with hearing loss, seldom involve such a severe motor palsy of CN VII.24 Madden and Sirimanna25 described a case of a cavernous hemangioma located in the IAC, with the same presentation. Because the patient had permanent sensorineural hearing loss, the group decided on a translabyrinthine approach with a facial nerve graft. The patient did not regain function after 1 year, however.25 Rocco and coworkers26 described a similar presentation of hearing loss and severe facial weakness (House-Brackmann grade IV) except that final pathology revealed a cavernous malformation in the IAC with hematoma formation into the CPA.26 Posterior fossa cavernous malformations are rare and account for only a few cavernous malformations affecting the craniospinal axis. Most of these lesions are approached via standard posterior fossa craniotomies. Cavernous malformations with a ventral and deep location may require a transpetrosal approach, however, to minimize brain retraction during exposure. An anterior transpetrosal approach (Kawase’s approach) was used successfully for a symptomatic cavernous hemangioma that was located in the ventrolateral surface of the pons as described by Saito and colleagues.27 MacDonald and associates28 used an anterior, subtemporal, medial transpetrosal approach for a pontine cavernous hemangioma, which caused intraparenchymal hemorrhage resulting in a moderate right hemiparesis. Most of these lesions can be accessed by a retrosigmoid craniotomy with a generous unroofing of the sigmoid sinus and a few millimeters of the presigmoid bone to allow for retraction of the sinus (see Fig. 66-2).
807
VASCULAR TUMORS OF THE PETROUS BONE AND CEREBELLOPONTINE ANGLE Role of Preoperative Embolization Surgical excision of hypervascular central nervous system tumors in the temporal bone and CPA can be dangerous, with the potential for staggering blood loss. The need to mitigate intraoperative tumoral hemorrhage has fostered the development and refinement of various neuroendo vascular techniques for preoperative embolization of tumors. This concept of interrupting blood supply to a tumor in anticipation for surgical excision has gained popularity for hypervascular tumors such as meningiomas, hemangiopericytomas, hemangioblastomas, paragangliomas, and juvenile nasopharyngeal angiofibromas.29 In the following sections, the principles of embolization, choice of embolic agents, and potential complications associated specifically with glomus tumors, hemangiomas, and hemangioblastomas of the petrous bone and CPA are discussed.
Glomus Tumors (Paragangliomas) Paragangliomas consist of a slow-growing group of benign neoplasms known as glomus tumors that arise from remnants of neural crest paraganglion cells. These tumors are of special interest to neurotologic and head and neck surgeons as they occur in the temporal bone (glomus tympanicum and glomus jugulare), the carotid bifurcation (glomus caroticum), and the upper parapharyngeal space (glomus vagale). Four percent of paragangliomas have documented catecholamine secretion, which can cause a pheochromocytoma-type syndrome.29 If catecholamine secretion is suspected and documented, the patient must be pretreated with α and β blockade before embolization and surgical resection. Similarly, preoperative testing should include angiography to define intracranial and extracranial arterial supply to the tumor and involvement of any dural venous sinus. In such cases, the patency of both transverse-sigmoid systems must be evaluated to determine if the involved sinus can be sacrificed without causing intracranial venous congestion. Temporal glomus tumors often obtain their blood supply from the petrous branches of the ICA, the most common of which is the vidian artery, and from the cavernous-carotid branches, the most common of which is the clival branch of the meningohypophyseal trunk. The most frequent neoplasm arising in the middle ear— glomus tympanicum—seldom requires preoperative embolization, and can usually be resected with conventional tympanoplasty techniques (Fig. 66-3). Alternatively, preoperative embolization may help with removal of glomus jugulare lesions, particularly tumors with extension intracranially. Preoperative embolization of these tumors helps to reduce intraoperative blood loss (which facilitates meticulous neurovascular
808
OTOLOGIC SURGERY
A
B FIGURE 66-2. A, Axial T2-weighted MR image of the brain shows large pontine cavernous malformation in a patient with progressive left hemiparesis. B, Stealth neuronavigation trajectory views during right retrosigmoid craniotomy.
dissection at the skull base), increases the extent of tumor resection, and decreases the overall length of postembolization surgery. These tumors tend to have a multicompartmentalized arterial supply. For embolization to be beneficial, each compartment should be catheterized selectively and embolized. The inferomedial compartment is most often supplied by the ascending pharyngeal artery. The posterolateral compartment is supplied by the stylomastoid branch of the occipital artery or posterior auricular artery or both. The anterior compartment is supplied by the internal maxillary artery and caroticotympanic arteries of the petrous ICA. Finally,
the superior compartment is supplied by various branches of the middle meningeal artery. Balloon test occlusion with preoperative sacrifice is an option for tumors with significant supply from the ICA. Although complications are rare, they do occur. Inadvertent embolization of the vasa vasorum can lead to a palsy of the lower cranial nerves (CN IX-XII). Marangos and Schumacher30 and Herdman and colleagues31 documented two cases of facial palsy after embolization of a glomus jugulare tumor. One patient had recovered completely 1 year after treatment, and the other had improved significantly.30,31 Pandya and coworkers32
Chapter 66 • Vascular Considerations in Neurotologic Surgery
809
days. They also advocated for temporary balloon dilation occlusion with hypotensive challenge if the carotid artery is extensively involved because embolization might be difficult owing to the tumor’s location at the carotid bifurcation.33 Feeders to carotid body tumors tend to have short pedicles, which makes stable microcatheter positioning for embolization challenging, and increases the risk for nontarget embolization.
Hemangioblastomas
C FIGURE 66-2—cont’d C, Postoperative axial T2-weighted MR image of the brain shows complete resection of cavernous malformation. The patient’s strength improved significantly on his left side, and he had House-Brackmann grade II right facial weakness and preserved hearing. (Courtesy of Barrow Neurological Institute.)
described a herniation syndrome that followed direct puncture and absorbable gelatin sponge (Gelfoam) embolization of a glomus jugulare tumor. They theorized that significant tumor infarction exacerbated posterior fossa edema.32 Carotid body tumors usually occur sporadically, whereas bilateral tumors occur in 5% to 10% of cases, with 2% to 6% of lesions showing malignant characteristics such as distant metastatic spread.29 Surgical excision of these tumors is particularly challenging because of their hypervascular nature and propensity to involve cranial nerves (Fig. 66-4). Relevant literature on the use of preoperative embolization techniques for carotid body tumors is scarce, and mostly relegated to a few small single-institution reports. LaMuraglia and associates33 preoperatively embo lized 11 patients with carotid body tumors supplied by the ascending pharyngeal artery, ascending cervical branches of the thyrocervical trunk, and vertebral arteries. They found that the technique significantly decreased intraoperative bleeding compared with nonembolized lesions. One patient had transient aphasia that resolved in 24 hours. The authors concluded that preoperative embolization should be performed for tumors larger than 3 cm, and should be followed by surgical resection within a few
Hemangioblastomas constitute 1.1% to 2% of craniospinal tumors, and most frequently occur within the cerebellar hemispheres and less commonly at the CPA, vermis, or brainstem.29 These benign, highly vascularized tumors most commonly arise sporadically. Twenty percent are associated, however, with the autosomal dominant genetic disorder, von Hippel–Lindau disease. Because these neoplasms are hypervascular, they are exceedingly difficult to resect, particularly in the intricate areas from where they tend to develop. Severe intraoperative hemorrhage is a significant contributor to the morbidity and mortality rates associated with the resection of hemangioblastomas. Based on angiography, blood supply is typically via the PICA and, less commonly, via the AICA or branches of the SCA. Dural branches of the vertebral artery, such as the posterior meningeal artery, may supply superficial lesions. The criteria for preoperative embolization of hemangioblastomas include large tumors (often >3 cm), with well-defined arterial feeders that are inaccessible surgically. Embolization using polyvinyl alcohol (PVA) particles or N-butyl cyanoacrylate (NBCA) glue should be performed with a microcatheter tip that is placed beyond the normal branches.34 The risk of embolization is particularly high with hemangioblastomas because the feeding arteries are often pial vessels. Experience with preoperative embolization of hemangioblastomas in the petrous bone or CPA has been limited mostly to small studies performed at single institutions. Conway and associates35 had mixed success with the use of preoperative embolization in 4 of 40 patients with hemangioblastomas. Although the embolization alone was sufficient to arrest symptom progression in one patient with a sacral hemangioblastoma, another patient had a lateral medullary infarction after attempted embolization of a medullary hemangioblastoma. The authors concluded that embolization should be reserved for tumors with large, surgically inaccessible arterial feeders.35 Tampieri and colleagues36 treated two patients successfully with large hemangioblastomas, one spinal and one involving the posterior fossa, with only minimal blood loss on surgical resection. Eskridge and coworkers37 reported one complication in nine patients with craniospinal hemangioblastomas: A patient developed malignant posterior fossa edema associated with hydrocephalus after treatment with polyvinyl alcohol. The surgeons concluded that embolization
810
OTOLOGIC SURGERY
B
A
C
D
F
G
E
Chapter 66 • Vascular Considerations in Neurotologic Surgery
811
FIGURE 66-3. A and B, Axial (A) and coronal (B) postcontrast MR images of the brain show right-sided glomus jugulare tumor encroaching on the cerebellopontine angle and underlying right cerebellar edema. C and D, Right internal (C) and external (D) carotid artery angiographic injections show prominent vascular blush at right skull base. The tumor was embolized with NBCA glue through ascending pharyngeal, posterior auricular, and occipital branches of right external carotid artery. E and F, Postembolization angiographic injections of right internal (E) and external (F) carotid arteries show significant flow reduction through glomus jugulare tumor. G, Postcontrast coronal MR image of the brain shows successful resection of tumor. (Courtesy of Barrow Neurological Institute.)
B
A
D
C
E
F
FIGURE 66-4. A and B, Postcontrast axial (A) and coronal (B) short tau inversion recovery MR images show large left-sided carotid body tumor. C and D, Angiogram with left common carotid artery injection shows splaying of internal and external carotid arteries by tumor. There is also a large tumor blush. The tumor was embolized with Onyx-15 through multiple left external carotid artery branches. E and F, Repeat left common carotid artery injection shows near-complete occlusion of discernible tumor vessels. One day after embolization, the patient underwent dissection of the left neck with gross total resection of carotid body tumor. (Courtesy of Barrow Neurological Institute.)
facilitated tumor manipulation and surgical resection.37 Finally, from this author’s experience of 35 patients who underwent preoperative embolization with NBCA liquid adhesive, one of the five patients treated for a hemangioblastoma experienced a complication.34 In this case, distal branches of the PICA were inadvertently occluded during glue embolization of a posterior meningeal artery. The
patient had initial mild dysmetria; however, the deficit had resolved by the 12 month follow-up examination.
Hemangiomas The role of preoperative embolization of hemangiomas is unclear because of their low overall incidence.
812
OTOLOGIC SURGERY
A
B
C
D
FIGURE 66-5. Imaging studies obtained in a 66-year-old woman. A and B, Axial T1-weighted MR images without (A) and with contrast material (B) showing recurrent squamous cell carcinoma on the left side. Lesion infiltrates left internal carotid artery (ICA) and cavernous sinus (arrow in B). Note the absence of flow void in left cavernous ICA. These images were obtained after clip occlusion of left ICA in the neck and supraclinoid process. C, Angiogram showing ICA-to-middle cerebral artery bypass with saphenous vein placed before tumor was resected. D, Axial T1-weighted MR image obtained after tumor resection and left orbital exenteration. Skull base was reconstructed with free tissue transfer by using rectus abdominis muscle flap. (From Feiz-Erfan I, Han PP, Spetzler RF, et al: Salvage of advanced squamous cell carcinomas of the head and neck: Internal carotid artery sacrifice and extracranial-intracranial revascularization. Neurosurg Focus 14(3):e6, 2003. Used with permission from Journal of Neurosurgery.)
Role of Cerebral Revascularization and Bonnet Bypass Untreated head and neck cancer that involves the skull base often is associated with a poor prognosis, especially if it involves the ICA. The risk of ICA rupture because of malignant cervical tumor invasion can be 18%, which highlights the need for ICA resection in certain cases.38 ICA sacrifice comes with the inherent risk of stroke, however, and has led many authors to advocate that resection with revascularization should proceed only in patients with unsuccessful results from a balloon test occlusion (radiographically or clinically). The test has significant drawbacks, however, including a high false-negative rate, often leading to a high rate of perioperative stroke despite a successful balloon occlusion test. It is difficult to predict immediate and delayed postoperative ischemic events.
Consequently, we advocate universal revascularization in all patients with cranial base tumors that involve the carotid artery. Other groups have suggested that in severe metastatic disease, aggressive surgery including carotid sacrifice should be performed only if there is a chance of complete cure of the tumor. More prospective studies are needed to determine a specific set of guidelines on the controversy (Fig. 66-5). The concept of a contralateral extracranial–to– ipsilateral intracranial bypass was originally introduced in 1978 by the senior author (R.F.S.) using a saphenous vein as an interposition graft, the so-called bonnet bypass (Fig. 66-6).39 Since then, based on our experience, we have found that the radial artery as an interposition graft can be used as a bonnet bypass and works with a high technical success rate with minimal complications. Bonnet bypass should be considered if the ipsilateral
Chapter 66 • Vascular Considerations in Neurotologic Surgery
813
CONCLUSION Vascular lesions affecting the petrous bone are among the most challenging in neurotologic surgery. Successful management of these lesions requires an understanding of the vascular anatomy and the various surgical approaches. Diagnosis and treatment of vascular lesions have been revolutionized by the advent of contemporary endovascular techniques. A multimodality and interdisciplinary approach to these complex lesions optimizes patient outcomes.
REFERENCES
FIGURE 66-6. Bonnet bypass using saphenous vein graft in a patient with large carotid body tumor that required sacrifice of the carotid artery. (Courtesy of Barrow Neurological Institute.)
carotid artery is unavailable because of previous surgery or radiation, or if a radical neck dissection is planned. The ipsilateral subclavian artery also can be used. In this case, however, the bypass would pass through the surgical field. Experience from this group has shown that the contralateral superficial temporal artery is an excellent donor vessel to perfuse the ipsilateral hemisphere after carotid sacrifice.38 A radial artery interposition graft is positioned between the contralateral superficial temporal artery and the ipsilateral distal middle cerebral artery. The use of a radial artery and a saphenous vein graft has been described as potential interposition grafts. Revascularization using a radial artery graft is often less taxing because the vessel wall is more resilient than vein grafts, and because the caliber of the artery closely matches that of the superficial temporal artery or middle cerebral artery. The patency rates for radial artery grafts are also higher because the arterial endothelium can support sluggish flow without thrombosis, especially because there is a lack of valves. Although the blood flow in radial artery grafts is high at 40 to 70 mL/min, flow rates with saphenous grafts are even higher at 70 to 140 mL/min.38,40 The lower flow makes radial artery grafts more conducive for anastomosis with smaller distal arteries because less turbulence is created. The grafts can be inadequate, however, for some patients who have very poor or no collateral blood flow. In this situation, the surgeon can divide a radial artery graft into two separate branches of the middle cerebral artery, or use one saphenous vein graft to accommodate the need for higher flow.
1. Liu J K, Gottfried O N, Amini A, Couldwell WT: Aneurysms of the petrous internal carotid artery: Anatomy, origins, and treatment. Neurosurg Focus 17:1-9, 2004. 2. Rhoton A L : The temporal bone and transtemporal approaches. Neurosurgery 47:211-265, 2000. 3. Tubbs R S, Hansasuta A, Loukas M, et al: Branches of the petrous and cavernous segments of the internal carotid artery. Clin Anat 20:596-601, 2007. 4. Rhoton A L : The anterior and middle cranial base. Neurosurgery 51:273-302, 2002. 5. Kakarla U K, Deshmukh VR , Zabramski J M, et al: Surgical treatment of high-risk intracranial dural arteriovenous fistulae: Clinical outcomes and avoidance of complications. Neurosurgery 61:447-459, 2007. 6. McConnell E M : The arterial blood supply of the human hypophysis cerebri. Anat Rec 115:175-203, 1953. 7. Hitotsumatsu T, Natori Y, Matushima T, et al: Microanatomical study of the carotid cave. Acta Neurochir 139:869-874, 1997. 8. Rhoton A L : The cerebellar arteries. Neurosurgery 47:2968, 2000. 9. Tanriover N, Rhoton A L : The anteroinferior cerebellar artery embedded in the subarcuate fossa: A rare anomaly and its clinical significance. Neurosurgery 57:314-319, 2005. 10. Constantino PD, Russel E, Reisch D, et al: Ruptured petrous carotid aneurysm presenting with otorrhagia and epistaxis. Am J Otol 12:378-383, 1991. 11. McGrail K M, Heros RC, Debrun G, Beyerl B D: Aneurysm of the ICA petrous segment treated by balloon entrapment after EC-IC bypass. J Neurosurg 65:249-252, 1986. 12. Cheng K M, Chan C M, Cheung YL , et al: Endovascular treatment of radiation-induced petrous internal carotid artery aneurysm presenting with acute haemorrhage: A report of two cases. Acta Neurochir 143:351-356, 2001. 13. Auyeung K M, Lui WM, Chow LC K, Chan FL : Massive epistaxis related to petrous carotid artery pseudoaneurysm after radiation therapy: Emergency treatment with covered stent in two cases. AJNR Am J Neuroradiol 24:1449-1452, 2003. 14. Wakhloo A K, Lanzino G, Lieber B B, et al: Stents for intracranial aneurysms: The beginning of a new endovascular era? Neurosurgery 43:377-379, 1998.
814
OTOLOGIC SURGERY
15. Sekhar L N, Sen C N, Jho H D: Saphenous vein graft bypass of the cavernous internal carotid artery. J Neurosurg 72:35-41, 1990. 16. Umenzu H, Seki Y, Aiba T, Kumakawa K : Aneurysm arising from the petrous portion of the internal carotid artery: Case report. Radiat Med 11:251-255, 1993. 17. Lawton MT, Hamilton MG, Morcos JJ, et al: Revascularization and aneurysm surgery: Current techniques, indications, and outcome. Neurosurgery 38:83-94, 1996. 18. Vishteh AG, Marciano FF, David C A, et al: Long-term graft patency rates and clinical outcomes after revascularization for symptomatic traumatic internal carotid artery dissection. Neurosurgery 43:761-767, 1998. 19. Candon E, Canovas F, Kabbaj J, et al: Anatomic basis for the treatment of aneurysms of the upper cervical segment of the internal carotid artery by extra-intracranial cervico-petrous bypass with inverted “in situ” saphenous vein graft. Surg Radiol Anat 19:1-6, 1998. 20. Seifert V, Stolke D: Posterior transpetrosal approach to aneurysms of the basilar trunk and vertebrobasilar junction. J Neurosurg 85:373-379, 1996. 21. Kawase T, Bertalanffy H, Otani M, et al: Surgical approaches for vertebro-basilar trunk aneurysms located in the midline. Acta Neurochir 138:402-410, 1996. 22. Kessler L A, Lubic LG, Koskoff YD: Epidural hemorrhage secondary to cavernous hemangioma of the petrous portion of the temporal bone. J Neurosurg 14:329-331, 1957. 23. Bottrill I : Imaging quiz case 2: Intraosseous cavernoustype hemangioma of the petrous temporal bone. Arch Otolaryngol Head Neck Surg 121:348-350, 1995. 24. Sundaresan N, Eller T, Ciric I : Hemangiomas of the internal auditory canal. Surg Neurol 6:119-121, 1976. 25. Madden GJ, Sirimanna K S : Cavernous hemangioma of the internal auditory meatus. J Otolaryngol 19:288-291, 1990. 26. Rocco FD, Paterno V, Safavi-Abbasi S, et al: Cavernous malformation of the internal auditory canal. Acta Neurochir 148:695-697, 2006. 27. Saito N, Sasaki T, Chikui E, et al: Anterior transpetrosal approach for pontine cavernous angiomas. Neurol Med Chir 42:272-274, 2002. 28. MacDonald J D, Antonelli P, Day A L : The anterior subtemporal, medial transpetrosal approach to the upper basilar artery and ponto-mesencephalic junction. Neurosurgery 43:84-89, 1998.
29. Deshmukh VR , Fiorella DJ, McDougall CG, et al: Preoperative embolization of central nervous system tumors. Neurosurg Clin N Am 16:411-432, 2005. 30. Marangos N, Schumacher M : Facial palsy after glomus jugulare tumour embolization. J Laryngol Otol 113:268270, 1999. 31. Herdman RC, Gillespie J E, Ramsden RT: Facial palsy after glomus tumour embolization. J Laryngol Otol 107:963-966, 1993. 32. Pandya S K, Nagpal R D, Desai A P, et al: Death following external carotid artery embolization for a functioning glomus jugulare chemdectoma: Case report. J Neurosurg 48:1030-1034, 1978. 33. LaMuraglia G M, Fabian R L , Brewster DC, et al: The current surgical management of carotid body paragangliomas. J Vasc Surg 15:1038-1044, 1992. 34. Kim L J, Albuquerque FC, Aziz-Sultan A, et al: Low morbidity associated with use of n-butyl cyanoacrylate liquid adhesive for preoperative transarterial embolization of central nervous system tumors. Neurosurgery 59:98-104, 2006. 35. Conway J E, Chou D, Clatterbuck R E, et al: Hemangioblastomas of the central nervous system in von HippelLindau syndrome and sporadic disease. Neurosurgery 48:55-62, 2001. 36. Tampieri D, Leblanc R , TerBrugge K : Preoperative embolization of brain and spinal hemangioblastomas. Neurosurgery 33:502-505, 1993. 37. Eskridge J M, McAuliffe W, Harris B, et al: Preoperative endovascular embolization of craniospinal hemangioblastomas. AJNR Am J Neuroradiol 17:525-531, 1996. 38. Deshmukh VR , Porter RW, Spetzler R F: Use of “bonnet” bypass with radial artery interposition graft in a patient with recurrent cranial base carcinoma: Technical report of two cases and review of the literature. Neurosurgery 56:E202, 2005. 39. Spetzler R F, Roski R A, Rhodes R S, Modic MT: The “bonnet bypass”: Case report. J Neurosurg 53:707-709, 1980. 40. Sekhar L N, Duff J M, Kalavakonda C, Olding M : Cerebral revascularization using radial artery grafts for the treatment of complex intracranial aneurysms: Techniques and outcomes for 17 patients. Neurosurgery 49:646-658, 2001.
Index A Abducens nerve monitoring of, 782 rehabilitation of, after neurotologic skull base surgery, 570–571 Abscesses Bezold’s, 186 of brain with acute infections, 186, 187f organisms causing, 183 stages of, 186 treatment of, 191–193, 191f extradural, with acute infections, 186 treatment of, 189–190, 190f Luc’s, 186 subdural with acute infections, 186 treatment of, 193 Absorbable gelatin sponge for patulous eustachian tube, 97 for tympanoplasty, 152 Acoustic reflex testing, 163 Acoustic tumors neuromas as auditory implants and. See Auditory implants. with chronic otitis media, retrosigmoid approach in, 606 combined therapy of, retrosigmoid approach in, 606 resection of, for retrosigmoid approach, 611–614, 612f translabyrinthine approach for. See Acoustic tumors, translabyrinthine approach for. transotic approach for, 621 neurosurgical techniques for, 600–602 for large tumors, 601–602 for small tumors, 601 translabyrinthine approach for, 591–602, 592f, 594f complication(s) of, 597–600 bleeding as, 599 cerebrospinal fluid leak as, 599 facial nerve and, 598–599 meningitis as, 599–600 hearing sacrifice with, 591 patient selection for, 591 postoperative care for, 597 postoperative follow-up for, 600 results with, 600 surgical procedure for, 593–597, 596f, 598f contracted mastoid management and, 597 vestibular nerve resection for, middle fossa approach for, 581 surgical procedure for, 593–597, 597f, 598f
Acoustic tumors (Continued) contracted mastoid management and, 597 vestibular nerve resection for, middle fossa approach for, 581–590 complications of, 588 indications for, 581 patient counseling for, 582 patient selection for, 588 preoperative evaluation for, 581–582 preoperative preparation for, 582–583 results with, 586–588 course of healing and, 586–587 success rate and, 587–588 surgical anatomy and, 583, 583f surgical technique for, 583–586, 584f–587f Acyclovir for Ramsay Hunt syndrome, 339 Adenocarcinoma of temporal bone, 33 Adenoid(s) hypertrophy of, tympanoplasty and, 150 otitis media and, 75 Adenoid cystic carcinoma of temporal bone, 33 Adenoidectomy complications of, 87 draping for, 80 follow-up for, 84 instrumentation for, 81 for otitis media acute, 77 benefits of, 79 chronic, with effusion, 79 limitations of, 79 pitfalls with, 86 postoperative care for, 84 preoperative preparation for, 80 results with, 86 risks of, 80 surgical site preparation and draping for, 80 surgical technique for, 82f, 83 tympanoplasty and, 150 Adhesions after stapes surgery, 317 Aditus ad antrum obstruction, 183, 185f Adson periosteal elevator, 16, 16f Air-bone gaps, small, stapedectomy for, 301–302 Allergic rhinitis, otitis media and, 75 Allergy treatment for, for otitis media with effusion, 87–88 tympanoplasty and, 150 Alloplastic prostheses for ossicular reconstruction, 166 Amikacin, ototoxicity of, 493–494 Aminoglycosides for acute otitis media, 111 ototoxicity of, 493
Amoxicillin for acute otitis media, 110 for otitis media, acute, 77 Amoxicillin/clavulanic acid for acute otitis media, 110 Anaerobic organisms, chronic suppurative otitis media due to, 108 Analgesia. See under specific procedures. Anesthesia. See also under specific procedures. general, complications of, 128 Anspach drill system, 2, 3f Anterior inferior cerebellar artery, anatomy of, 800–801 Anterior inferior cerebellar artery syndrome with retrosigmoid approach to cerebellopontine angle tumors, 616 Anterior sulcus, blunting in, with outer surface grafting tympanoplasty, 125f, 126 Antibiotics. See Antimicrobial therapy; specific agents. Antihistamines for otitis media, chronic, with effusion, 78 Antileukotrienes for otitis media, chronic, with effusion, 78 Antimicrobial therapy antiviral for Bell’s palsy, 338 for Ramsay Hunt syndrome, 339 following canal wall reconstruction tympanomastoidectomy, 177–178, 180–181 for otitis media acute, 77, 77t, 109–115 failed, causes of, 112–115 granulation tissue and, 111–112 preventing bacterial resistance and, 112 chronic, with effusion, 77–78 Argon laser, 281, 282t, 284 Arriaga-University of Pittsburgh tumor lymph node metastasis staging system, 35–36, 35t Arteriovenous malformations, dural, 805–806 tentorial fistulas and, 806 transverse and sigmoid fistulas and, 806 Atresiaplasty, 55–72 bone-anchored hearing appliances in, 58–59 canal wall down approach for, 65, 67f complications of, 68–69 indications for, 58 patient selection for, 58 pitfalls in, 66–67��� postoperative care and, 65 preoperative evaluation for, 60, 60f–61f results with, 67–68 surgical approach for, 60–65, 62f surgical technique for, 62–65, 62f–67f timing of, 58–59 transmastoid approach for, 65, 67f for unilateral atretic ear, 69
Page numbers followed by f indicate figures; t indicate tables.
815
816
Index
Atticotomy, 209, 215–216, 216f Audiology, cochlear implantation and, 374 Audiometry. See under specific procedures and conditions. Auditory brainstem implant, 703. See also Auditory implants. Auditory brainstem implant for vestibular schwannomas in, 698, 698f Auditory canal. See External auditory canal; Internal auditory canal. Auditory evoked brainstem response, monitoring hearing using, 779–780, 781 Auditory implants, 703–714 anatomic considerations for, 704–705, 705f devices for, 704, 705f implantation technique for, 707–709, 707f–709f patient counseling for, 704 patient selection for, 703, 704t postoperative care for, 709 postoperative complications with, 709–710 preoperative evaluation for, 704 research studies on, 711–712, 712f with auditory midbrain implants, 712 with penetrating electrode ABI, 711–712 results with, 710–711, 710f surgical considerations for, 705–707, 706f–707f Auditory midbrain implant, 703. See also Auditory implants. Aural atresia, 55–72 atresiaplasty for. See Atresiaplasty. Baha for, 399 classification systems for, 56–57, 56t–57t computed tomography in, 60, 60f–61f embryology of, 55–56, 56t initial evaluation of, 57–58 unilateral, atresiaplasty for, 69 Aural fullness, differential diagnosis of, 94, 99f Aural toilet for acute otitis media, 109 Auricle, embryology of, 55–56, 60f Auricular nerve, greater, for facial nerve grafting, 236 Austin “reverse elevator” for undersurface graft tympanoplasty, 142–144, 145f Autografts for ossicular reconstruction, 161, 166, 166f Autophony with patulous eustachian tube, 94 in superior canal dehiscence syndrome, 510
B Bacterial resistance, preventing, 112 Bacteroides spp., otitis media and, 74 Baha. See Bone-anchored cochlea stimulator (Baha). Balloon ballottement, stapes surgery and, 312 Balloon microcompression for trigeminal neuralgia, shortcomings of, 525 Barotrauma after stapes surgery, 316 Basal cell carcinoma of temporal bone, 33 “Beginner’s hump” with mastoidectomy, canal wall down, 211–212, 212f Bell’s palsy, 335–338 audiometry in, 336 facial nerve decompression for, 339–344 herpes simplex virus and, 335 management of, 337, 337f natural history of, 337 testing for, 336, 338f Benign paroxysmal positional vertigo, 467–476 pathophysiology of, 468
Benign paroxysmal positional vertigo (Continued) posterior semicircular canal occlusion for patient counseling and, 468–470, 469f patient selection for, 470 preoperative evaluation for, 471 results with, 471–473 surgical technique for, 471, 472f Bezold’s abscess, 186 BICROS for aural atresia, 59 Biofilms failed antibiotic therapy due to, 113 otitis media with effusion and, 88 treatment of, 113 Biscuit footplate, stapes surgery and, 314, 315f Bleeding. See Hemorrhage. Blood within vestibule, stapes surgery and, 316 Bondy modified radical mastoidectomy, 209, 213–214, 214f Bone cement for dural injury, 238 for encephalocele repair, 248–249 for labyrinthine fistula repair, 231 for ossicular prostheses, 162, 166–167 Bone conduction, hearing through, 397–398. See also Hearing loss, conductive. Bone-anchored cochlea stimulator (Baha), 383–384, 397–410 audiometric criteria for, 400–401, 400f–401f for aural atresia, 58 Baha Cordelle II as, 400, 401f Baha Divino as, 400, 400f Baha Intenso as, 400, 400f in children, 408–409 in chronic ear disease, 398 with chronic external otitis, 399 complications of, 407–408 handling of, 408 for conductive loss in only hearing ear, 399 for conductive or mixed hearing loss, 398 for congenital malformations, 399 contraindications to, 401 in Down syndrome, 399 hearing through bone conduction and, 397–398 historical background of, 397 osseointegration and, 398 patient counseling and, 399, 400f patient preparation for, 402, 403f patient selection for, 398–399 pitfalls with, 407–408 postoperative management for, 406–407, 406f for single-sided deafness, 399 stability of, 408, 408f surgical instruments for, 402 surgical technique for, 402–406, 406f Baha coupling placement and, 405–406 for countersink drilling, 405 for guide hole drilling, 405 subcutaneous soft tissue reduction and, 405 for suturing flap, 406 for thin hairless flap preparation, 402–405 Bonnet bypass, internal carotid artery cancer and, 812–813, 812f–813f Boric acid for patulous eustachian tube, 97 Botulinum toxin for hemifacial spasm, 529 for hyperkinesis, with facial reanimation, 770–771 Brain abscesses with acute infections, 186, 187f organisms causing, 183 stages of, 186 treatment of, 191–193, 191f
Brain herniations, 245–252, 246f–247f computed tomography of, 247, 247f diagnosis of, 246–248, 247f–248f magnetic resonance imaging of, 247, 247f pathogenesis of, 245–246, 247f treatment of, 248–250, 248f–249f Brain injury cerebrospinal fluid leaks due to, 246 treatment of, 248, 248f–249f translabyrinthine approach for. See Acoustic tumors, translabyrinthine approach for. Brain prolapse, 245–252, 246f–247f computed tomography of, 247, 247f diagnosis of, 246–248, 247f–248f magnetic resonance imaging of, 247, 247f pathogenesis of, 245–246, 247f treatment of, 248–250, 248f–249f Brow elevation in facial paralysis, 748f, 749–750
C Canal wall defect repair, 206 for defects resulting from disease, 205f, 206 for defects resulting from surgery, 206 Canalithiasis, 468 Canalplasty, 21–32 for collapsing external auditory canal, 28–29 draping for, 22, 23f for external auditory canal exostoses, 21–28 analgesia for, 22, 23f complications of, 28 indications for, 21 pharmacologic preparation for, 22 postoperative care for, 25–28 problems in, 28 psychological preparation for, 22 site preparation for, 22, 23f surgical technique for, 22–25, 23f–24f for anterior exostoses, 25, 26f–27f for posterior exostoses, 24f, 25, 26f for keratosis obturans, 29 for medial third stenosis of external auditory canal, 28, 29f for post-traumatic suture dehiscence, 31 psychological preparation for, 22 for scutum defects, 29–31, 30f for tympanic bone osteonecrosis and osteoradionecrosis, 29 Candida, acute otitis media due to, 109 Canthoplasty for lower lid reapposition in facial paralysis, 742 lateral, 745–747, 746f medial, 744f, 745 Carotid artery external, anatomy of, 800 glomus tumors of, 551 injury of with chronic otitis media surgery, 241 with limited temporal bone resection, 40 with radical temporal bone resection, 50 internal anatomy of, 799–800 aneurysms of high cervical, 805 petrous, 803–805 Cartilage, preparation for ossicular reconstruction, 169 Cartilage tympanoplasty. See Tympanoplasty, cartilage. Cauterization for tympanic membrane closure, 114f, 115–116 Cavernous hemangiomas, 806–807 Cavernous malformations, 806–807
Index Ceftriaxone for acute otitis media, 110 Cefuroxime for acute otitis media, 110 Central nervous system auditory implants. See Auditory implants. Cerebellar artery, superior, anatomy of, 800–801 Cerebellar dysfunction following retrosigmoid approach to cerebellopontine angle tumors, 617 Cerebellopontine angle extended middle cranial fossa approach to, 641–648, 642t advantages of, 647 limitations of, 645–647 surgical anatomy and, 641–642, 642f surgical technique for, 642–645, 643f–646f technique for, 641 vascular anatomy of, 799–801 dural venous sinuses and, 801 external carotid artery and, 800 internal carotid artery and, 799–800 vertebrobasilar arteries and, 800–801 Cerebellopontine angle lesions recurrent, following retrosigmoid approach, 618 retrosigmoid approach to, 603–620 for acoustic neuroma with chronic otitis media, 606 in combined therapy of acoustic neuromas, 606 complications of, 616–618 from patient positioning, 616–617 vascular, 616 contraindications to, 606–607 dressings for, 614 hearing preservation and, 605–606 indications for, 605–606 instruments for, 607–608 patient counseling for, 605 patient positioning for, 607 complications from, 616–617 patient preparation for, 607 patient selection for, 605–607 postoperative care for, 614–615 preoperative evaluation for, 605 results with, 615–616 for revision surgery, 606 surgical anatomy and, 603–605, 604f surgical site preparation for, 607 surgical technique for, 608–614 for acoustic neuroma resection, 611–614, 612f for cerebellopontine angle exposure, 608–609 craniectomy and, 608 for craniotomy closure, 614 hemostasis and, 614 for internal auditory canal closure, 612f, 614 for internal auditory canal exposure, 609–611, 610f for tumors extending into inferior portion of cerebellopontine angle, 606 for tumors with limited extension into Meckel’s cave, 606 transcochlear approach to, 631–640 advantages of, 631 complications of, 639–640 computed tomography and, 632 disadvantages of, 632 magnetic resonance imaging and, 632 patient evaluation for, 632 postoperative care for, 637 preoperative counseling for, 632 results with, 637–639
Cerebellopontine angle lesions (Continued) surgical anatomy and, 632 surgical technique for, 632–637, 633f closure and, 636–637 external auditory canal closure and, 633–634 facial nerve rerouting and, 634–635, 635f incision and, 633 labyrinthectomy and skeletonization of internal auditory canal and, 634, 634f mastoidectomy and, 633, 634f setup and, 633 transcochlear drill-out and, 635, 635f tumor removal and, 635–636, 636f–639f transotic approach for, 621–630 alternatives to, 629–630 draping for, 622 dressings for, 628 indications for, 621 intraoperative monitoring for, 622 patient positioning for, 622 postoperative care for, 628 preoperative evaluation for, 621–622 results with, 629, 629t surgical site preparation for, 622 surgical technique for, 622–628 blind sac closure of external auditory canal and, 622, 623f eustachian tube obliteration and, 622 otic capsule exenteration and, 622–624, 623f skin incision and, 622, 623f subtotal petrosectomy and, 622, 623f tumor removal and, 624–625, 625f–626f unroofing of labyrinthine portion of facial nerve and, 624, 625f wound closure and, 625–628, 627f tips and pitfalls in, 627f, 628–629 vascular, 803–807 cavernous malformations and cavernous hemangiomas as, 806–807, 808f dural arteriovenous malformations as, 805–806 internal carotid aneurysms as, 803–805 high cervical, 805 petrous, 803–805 vertebrobasilar aneurysms as, 805 vascular tumors as, 807–813 cerebral revascularization and bonnet bypass and, 812–813, 812f–813f preoperative embolization and, 807–811 glomus tumors and, 807–809, 810f–811f hemangioblastomas and, 809–811 hemangiomas and, 811 Cerebral blood flow, evaluation of, in temporal bone malignancies, 44 Cerebral revascularization, internal carotid artery cancer and, 812–813, 812f–813f Cerebrospinal fluid leaks with cochlear implantation, 378 diagnosis of, 246–248, 247f–248f with endoscopic endonasal approaches to skull base and paranasal sinuses, 679 following facial nerve tumor surgery, 371 head trauma and, 246 treatment of, 248, 248f–249f management of, 727–732, 731f ear canal closure with eustachian tube and middle ear obliteration for, 731–732, 731f lumbar drain for, 727–729, 729f
817
Cerebrospinal fluid leaks (Continued) middle fossa obliteration of the eustachian tube for, 731f, 732 pressure dressings for, 727, 728f for refractory leaks, 731–732, 731f wound exploration and reclosure for, 731 with retrolabyrinthine vestibular neurectomy, 451 with retrosigmoid approach to cerebellopontine angle tumors, 617 spontaneous encephaloceles and, 246 with surgery for facial nerve trauma, 360 following transcanal labyrinthectomy, 488 with translabyrinthine approach for acoustic tumors, 599 following translabyrinthine vestibular neurectomy, 464 Cerebrospinal fluid otorrhea, 107–118 from fistulas, 107 management of, 109–115 antimicrobial therapies for, 109–115 aural toilet for, 109 failed therapy and, 112–115 granulation tissue and, 111–112 tympanic membrane closure for, office techniques for, 115–116 from otitis media acute, with open tympanic membrane, 109 chronic suppurative otitis, 107–109 spontaneous, 107 from temporal bone fractures, 107 traumatic, 107 Ceruminous carcinoma of temporal bone, 33 Cervical sympathetic chain deficits, rehabilitation after neurotologic skull base surgery for, 578 Chair, hydraulic, for undersurface graft tympanoplasty, 142 Chemotherapy with limited temporal bone resection, 41 Children Baha in, 408–409 ossicular reconstruction in, 163 otosclerosis in, partial stapedectomy for, 279 stapedectomy in, 302 stapedotomy in, 312 trigeminal neuralgia in, microvascular decompression for, 527 Chissone’s classification system for congenital aural atresia, 57 Cholesteatomas canal wall reconstruction tympanomastoidectomy for. See Tympanomastoidectomy, canal wall reconstruction. cartilage tympanoplasty for. See Tympanoplasty, cartilage. diagnosis of, 197 drainage procedures for. See Petrous apex lesions, drainage procedures for. management of, labyrinthine fistulas and, 229–231, 229f–230f mastoidectomy for. See Mastoidectomy. matrix removal and, 229–231, 229f–230f occult, failed antibiotic therapy due to, 113 of pars flaccida, canalplasty for, 29–31, 30f recurrent, 196 residual, tympanoplasty and, 222 Cholesterol granulomas, 76 treatment of, drainage procedures for. See Petrous apex lesions, drainage procedures for. Chorda tympani nerve identification of, 308 injury of, during laser stapedectomy, 271
818
Index
Chorda tympani nerve (Continued) preservation of, in stapes surgery, 308 stapedectomy and, 257, 260f Chronic ear surgery. See Tympanoplasty. Ciprofloxacin for acute otitis media, 110 CO2 laser, 281, 282t, 284 Cochlea, drilling out of, in transcochlear approach, 635, 635f Cochlear blood flow, measurement using laser Doppler flowmetry, 782 Cochlear hypoplasia, 374–375 Cochlear implantation, 373–382 bilateral, 379–380 combined electroacoustic stimulation and, 378f, 380 complications of, 378–379, 379t criteria for, 373, 374t instrumentation for, 15, 15f medical evaluation for, 373–375 audiologic, 374 imaging and, 374–375 physical examination and, 374 promontory evaluation and, 375 patient selection for, 373 revision, 379 selection of ear for, 375 surgical technique for, 375–378, 375f–377f following transcanal labyrinthectomy, 490 Cochlear microphonics in Meniere’s disease, 415 Cochlear nerve, anatomy of, 442–443, 444f Cochleosacculotomy, 477–482 patient selection for, 478 rationale for, 477–478 results with, 481–482 surgical technique for, 478–481, 479f–481f Combined electroacoustic stimulation, 378f, 380 Compound muscle action potential in Bell’s palsy, 336–337 Computed tomography. See under specific conditions and procedures. Conductive hearing loss. See Hearing loss, conductive. Congenital aural atresia. See Aural atresia. Corticosteroids for Bell’s palsy, 337–338 intratympanic, for inner ear conditions, 502–503 clinical studies of, 502–503 experimental studies of, 502 for otitis media, chronic, with effusion, 78 Corynebacterium, otitis media and, 74 Cosmetic deformities, following infratemporal fossa surgery, 664 Cranial base. See Skull base entries. Cranial nerve(s) abducens (VI) monitoring of, 782 rehabilitation after neurotologic skull base surgery for, 570–571 deficits of with limited temporal bone resection, rehabilitation of, 41 after radical temporal bone resection, rehabilitation for, 50 facial (VII). See Facial nerve entries. glossopharyngeal (IX). See Glossopharyngeal �entries. hypoglossal (XII) loss of function of, with limited temporal bone resection, 40 monitoring of, 782 rehabilitation after neurotologic skull base surgery and, 578
Cranial nerve(s) (Continued) hypoglossal/facial anastomosis for facial nerve paralysis and, 755–764 end-to-side, via facial nerve translocation from fallopian canal, 759, 760f nerve growth factors and conduits for, 757–758 newer modifications of, 758–759, 758f–759f patient selection for, 755–756 results with, 759–761, 760t–761t surgical technique for, 756–757, 756f–757f oculomotor (III), monitoring of, 782 spinal accessory (XI) anatomy of, 443, 444f monitoring of, 782 in glomus tumor surgery, 555–556 palsy of, with jugular foramen schwannomas, 552 preservation of, in glomus tumor surgery, 556 rehabilitation after neurotologic skull base surgery and, 577–578 trigeminal (V). See Trigeminal nerve; Trigeminal neuralgia. trochlear (IV), monitoring of, 782 vagus (X). See Vagus nerve. vestibulocochlear (VIII) anatomy of, 445f, 446 compression of, microvascular decompression for, 530 direct recording from, monitoring hearing using, 780–781 monitoring of, during neurologic procedures, 11–12 Cranial rhizopathies, microvascular decompression for. See Microvascular decompression. Craniectomy for retrosigmoid approach, 608 Craniotomy closure of, for retrosigmoid approach, 614 for middle cranial fossa vestibular neurectomy, 431–433, 432f, 434f middle fossa, internal auditory canal decompression without tumor removal with, for vestibular schwannomas, 697 retrosigmoid, with partial removal of vestibular schwannoma, 697, 697f Cross-facial grafting, 755 hypoglossal/facial anastomosis as, 755–764 end-to-side, via facial nerve translocation from fallopian canal, 759, 760f nerve growth factors and conduits for, 757–758 newer modifications of, 758–759, 758f–759f patient selection for, 755–756 results with, 759–761, 760t–761t surgical technique for, 756–757, 756f–757f Crus, division of, 309 Cupulolithiasis, 468 Curettes, stapes, 6, 7f CyberKnife stereotactic radiosurgery, 794–795 dose distribution and, 795 localization and, 795 treatment delivery and, 795 treatment planning and, 794–795, 794f
D De la Cruz classification of congenital aural atresia, 56–57, 56t “Dead ear,” definition of, 457
Deafferentation procedures for vestibular disorders, 457 Deafness. See Hearing loss. Decongestant therapy for otitis media, chronic, with effusion, 78 Delivery technique, failed antibiotic therapy due to, 112–113 Diplacusis, binaural, after stapes surgery, 317 Discharge, chronic, preoperative treatment of, before mastoidectomy, 197–198 Distention theory of Meniere’s disease, 414 Dix-Hallpike maneuver, 468–469, 469f in benign paroxysmal positional vertigo, 467 Dizziness. See also Vertigo. complicating operations for chronic ear infections, 128 following laser stapedectomy, 272 migraine-related, 456, 510 risk of, with mastoidectomy, 210 Dizziness Handicap Inventory in superior canal dehiscence syndrome, 511 Down syndrome, Baha in, 399 Drainage, chronic, preoperative treatment of, before mastoidectomy, 197–198 Drainage theory of Meniere’s disease, 414 Draping. See under specific procedures. Drills, 1–2, 2f–3f “Dry mopping” for acute otitis media, 109 Dry mouth complicating operations for chronic ear infections, 128 Dural defects with limited temporal bone resection, 40 Dural elevation for middle cranial fossa vestibular neurectomy, 433, 434f Dural herniations, 245–252, 246f–247f computed tomography of, 247, 247f diagnosis of, 246–248, 247f–248f magnetic resonance imaging of, 247, 247f pathogenesis of, 245–246, 247f treatment of, 248–250, 248f–249f Dural injury with chronic otitis media surgery, 237–238, 238f–239f Dural venous sinuses anatomy of, 801 mastoidectomy and, 219 Dysgeusia after stapes surgery, 316
E Ear packing after stapes procedures, 7 Ear specula, 5–6, 6f Ectopic tissue, introduced, after stapes surgery, 320 Elderly patients, stapedectomy in, 302 Electrocautery equipment, 1 Electrocochleography in Meniere’s disease, 415 monitoring hearing using, 780–781 perilymphatic fistulas and, 326 Electrodiagnostic testing. See also specific techniques. in Bell’s palsy, 336 Electromyography in Bell’s palsy, 337 for facial nerve monitoring, 773–774, 774f–775f during facial nerve decompression, 339, 342–343 with facial nerve trauma, 352–353 Electroneuronography in Bell’s palsy, 336 with facial nerve trauma, 351 Electronystagmography in Meniere’s disease, 415
Index Electronystagmography (Continued) for translabyrinthine vestibular neurectomy, 458–459 ELITE approach, 715–726, 717f complications of, 724 results with, 724–726 surgical procedure for, 716–724, 718f–722f, 724f Embolization, preoperative, 807–811 glomus tumors and, 807–809, 810f–811f hemangioblastomas and, 809–811 hemangiomas and, 811 Embryology of ear, 55–56, 56t Encephaloceles, temporal bone, 245–252, 246f–247f computed tomography of, 247, 247f diagnosis of, 246–248, 247f–248f pathogenesis of, 245–246, 247f treatment of, 248–250, 248f–249f Endolymphatic hydrops, pathophysiology of, 413–414 Endolymphatic sac surgery, 411–428 complications of, 418 draping for, 10 endolymphatic shunt procedure as, 417–418, 418f–421f instrumentation for, 10–11, 11f, 19 outcomes with, 418–425, 421t hearing and, 424–425 from 1995 to present, 419–423, 422t patients’ perspective on, 423–424 physicians’ perspective on, 424 patient preparation for, 10 patient selection for, 417 physiologic monitoring during, 10 procedure for, 10 sac anatomy and embryology and, 411–412 sac physiology and, 413 Endoscopy, perilymphatic fistulas and, 326 Enterobacteriaceae, chronic suppurative otitis media due to, 108 Entropion of upper lid, correction of, in facial paralysis, 748f, 750 Envoy Esteem, 386t, 391–394, 393f Epithelial cysts following outer surface grafting tympanoplasty, 125f, 126 Estrogen, conjugated, for patulous eustachian tube, 97 Eustachian tube dysfunction of allergic inflammation in, otitis media and, 76 otitis media and, 75 function of, tests of, tympanoplasty and, 151 obliteration of, transotic approach for cerebellopontine angle lesions and, 622 patulous, 93–106 anatomy and physiology of, 93–94, 94f clinical presentation of, 94 diagnosis of, 95–97, 95f, 99f etiology of, 94 transnasal and transoral occlusion of, 98–104, 100f–104f treatment of, 97–98, 99f retracted, aural fullness and, 99f Exostoses of external auditory canal, canalplasty for. See Canalplasty, for external auditory canal exostoses. Extended middle cranial fossa approach, 641–648, 642t advantages of, 647 limitations of, 645–647 surgical anatomy and, 641–642, 642f surgical technique for, 642–645, 643f–646f technique for, 641
External auditory canal aplasia of. See Aural atresia. atresia of, unilateral, Baha for, 399 closure of blind sac, in transotic approach for cerebellopontine angle lesions, 622, 623f in transcochlear approach, 633–634 collapsing, canalplasty for, 28–29 enlargement of, in undersurface graft tympanoplasty, 142, 143f exostoses of, canalplasty for. See Canalplasty, for external auditory canal exostoses. hypoplasia of. See Aural atresia. keratosis obturans of, canalplasty for, 29 medial third stenosis of, canalplasty for, 28, 29f post-traumatic suture dehiscence and, canalplasty for, 31 scutum defects and, canalplasty for, 29–31, 30f stenosis of, atresiaplasty and, 68 trauma to, during tympanostomy tube insertion, 84 tympanic bone osteonecrosis and osteoradionecrosis and, canalplasty for, 29 External otitis, chronic Baha for, 399 Extradural abscesses with acute infections, 186 treatment of, 189–190, 190f Extreme lateral infrajugular transcondylar approach, 715–726, 717f complications of, 724 results with, 724–726 surgical procedure for, 716–724, 718f–722f, 724f Eye(s) in facial paralysis. See Facial nerve paralysis, eye care in. in neurofibromatosis 2, 696, 696t Eye movements in superior canal dehiscence syndrome, 507–508, 508f Eyes-closed-turning test for perilymphatic fistulas, 326
F Facial nerve anatomy of, 443 bone overlying, lowering of, for mastoidectomy, canal wall down, 211–212, 212f congenitally ectopic, stapes surgery and, 314 deficits of with infratemporal fossa surgery, 663 translabyrinthine approach for. See Acoustic tumors, translabyrinthine approach for. with translabyrinthine approach for acoustic tumors, 598–599 dehiscent, stapedectomy and, 257–259, 297 laser, 269–271 embryology of, 56 excessive manipulation of, with limited temporal bone resection, 40 grafting of, 236–237, 236f hypoglossal/facial anastomosis for facial nerve paralysis and, 755–764 end-to-side, via facial nerve translocation from fallopian canal, 759, 760f nerve growth factors and conduits for, 757–758 newer modifications of, 758–759, 758f–759f
819
Facial nerve (Continued) patient selection for, 755–756 results with, 759–761, 760t–761t surgical technique for, 756–757, 756f–757f injury of in atresiaplasty, 60, 61f with cochlear implantation, 378 mastoid segment of, 203, 205f during mastoidectomy, 86, 203 in transcanal labyrinthectomy, 488 tympanic segment of, 203, 205f injury to, audiometry with, 350 labyrinthine portion of, unroofing of, transotic approach for cerebellopontine angle lesions and, 624, 625f mastoid segment of, transection with cutting burr, 235, 235f mastoidectomy and, 219–220, 219f microvascular decompression for hemifacial spasm and, 529–530 overhanging, stapes surgery and, 314, 315f in radical temporal bone resection, 44 rerouting of, in transcochlear approach, 634–635, 635f sacrifice of, with limited temporal bone resection, 40 tumors of, 363–372 computed tomography of, 363, 364f hemangiomas as, 363, 371, 371t magnetic resonance imaging of, 363, 364f neuromas as, 363–365, 369–370, 371t schwannomas as, in neurofibromatosis 2, 694–695 surgery for complications of, 371 graft material for, 367, 368f middle fossa approach for, 365–367, 366f nerve grafting and, 367–369, 368f, 370f patient counseling and, 365, 365f patient selection for, 363–365 pitfalls of, 369 radiation therapy and radiosurgery and, 365 rerouting and, 369, 370f results with, 369–370, 371t surgical techniques for, 365–369 translabyrinthine approach for, 367, 368f transmastoid approach for, 366f, 367 tumor removal and, 367 Facial nerve decompression for Bell’s palsy, 339–344 anesthesia for, 339 draping for, 339 intraoperative electromyography in, 339, 342–343 limitations of, 344 postoperative care for, 343–344 preoperative preparation for, 339, 340f–341f special considerations for, 344 surgical technique for, 340–343, 341f–344f Facial nerve monitoring, 773–778 anesthetic issues and, 775–776 antidromic, 778 during atresiaplasty, 67 electromyography for, 773–774, 774f–775f during endolymphatic sac surgery, 10 historical background of, 773 indications for, 773 during mastoidectomy, 220 during neurologic procedures, 11–12 physiology and, 773
820
Index
Facial nerve monitoring (Continued) practical application of, 776–778 antidromic monitoring and, 778 artifactual responses and, 777 avoiding trauma and, 776 locating facial nerve and, 776 prognosis and, 776–777, 777t troubleshooting facial nerve monitors and, 777–778 stimulation for, 774–775 burst activity and, 775 trains and, 775 during surgery for traumatic facial paralysis, 356–357 during transcanal labyrinthectomy, 485 for tympanoplasty, 152 Facial nerve paralysis with acute infections, 185 treatment of, 189 in Bell’s palsy. See Bell’s palsy. with chronic otitis media surgery, 234–236, 235f complicating operations for chronic ear infections, 128 eye care in, 733–754, 752t anesthesia for, 733–734 brow elevation for, 748f, 749–750 criteria for, 733 draping for, 734–735, 734f lid suture taped to cheek for, 750–752, 751f lower lid reapposition procedures for, 742–749 canthoplasty for, 742 fascia lata suspension of lower lid for, 747, 748f lateral canthoplasty for, 745–747, 746f lid position assessment and, 742–745 lid stents for, 747–749 medial canthoplasty for, 744f, 745 midface support for, 749 patient preparation for, 733 surgical preparation for, 733–734 temporary tarsorrhaphy suture for, 751f, 752 upper lid entropion correction for, 748f, 750 upper lid reanimation procedures for, 735–737 enhanced palpebral spring implantation for, 738–740, 739f, 741f gold weight implantation for, 740–742, 741f, 742t palpebral spring implantation for, 736f, 737–738 silicone rod prosthesis implantation for, 742, 743f hypoglossal/facial anastomosis for, 755–764 end-to-side, via facial nerve translocation from fallopian canal, 759, 760f nerve growth factors and conduits for, 757–758 newer modifications of, 758–759, 758f–759f patient selection for, 755–756 results with, 759–761, 760t–761t surgical technique for, 756–757, 756f–757f mastoidectomy and, 87 after radical temporal bone resection, rehabilitation for, 50 in Ramsay Hunt syndrome, 338–339 reanimation for. See Facial reanimation. risk of, with mastoidectomy, 210 after stapes surgery, 316–318 following transcochlear approach, 639
Facial nerve paralysis (Continued) following translabyrinthine vestibular neurectomy, 464 traumatic, 347–362, 348f extratemporal facial nerve injury and, 349 intracranial facial nerve injury and, 347 intratemporal facial nerve injury and, 347–349, 349t, 350f–353f ossicular damage with, 354 otorrhea with, 354–355 patient evaluation and, 349–354 prognosis based on electric studies in, 351–353 radiologic evaluation of, 353 surgery for complications of, 360 extratemporal nerve segment and, 359 intratemporal nerve segment and, 356–359, 357f–358f patient positioning for, 356 patient preparation for, 356 pitfalls of, 359 postoperative care for, 359 preoperative preparation for, 355–359 results with, 360 surgical technique for, 356–359 timing of, 353–354 Facial reanimation, 765–772 dynamic procedure(s) for, 765–770 free muscle flaps as, 767–770 nerve grafts as, 765–766 nerve substitution as, 766 temporalis muscle transposition as, 766–767, 768f–769f static procedure(s) for, 770–771 for lower lip rehabilitation, 770 static slings as, 770 surgical management of hyperkinesis as, 770–771 Facial recess, opening of, in mastoidectomy, 199–201, 201f, 205f Facial weakness with labyrinthine fistulas, 228 Fascia grafts in tympanoplasty. See Specific techniques under Tympanoplasty. Fascia lata suspension for lower lid reapposition in facial paralysis, 747, 748f Fat graft tympanoplasty, 115f, 116 Fatigability in benign paroxysmal positional vertigo, 467 Fenestra, small, stapedectomy and, 300 Fentanyl citrate for revision stapedectomy, 288 Fisch classification, 715 Fisch operating table, 430, 430f Fistula(s) perilymphatic. See Perilymphatic fistulas. stapedectomy and, 300 Fistula test for perilymphatic fistulas, 326 Flap necrosis with Baha implant surgery, 407, 407f Food allergy, treatment of, for otitis media, 78 Footplate biscuit, stapes surgery and, 314, 315f extraction of, total, instrumentation for, 6–7, 7f fixed foreshortened incus with, ossicular reconstruction for, 165, 165f with mobile malleus and incus, ossicular reconstruction for, 164 floating laser stapedectomy and, 271 in stapedectomy, 259 stapedectomy and, 296f, 297–298, 298t fractured, stapes surgery and, 316 fragments of
Footplate (Continued) left in oval window, revision stapedectomy for, 287 in vestibule, stapes surgery and, 316 mobile, with mobile incus and malleus and absent stapes superstructure, ossicular reconstruction for, 164 obliterated, in stapedectomy, 259, 261f relationship to vestibule, stapes surgery and, 311–312 removal of, 309 Footplate surgery, classification of, 276 Free muscle flaps for facial reanimation, 767–770 Fungi, chronic suppurative otitis media due to, 108
G Gamma knife surgery for skull base tumors. See Skull base tumors, stereotactic radiosurgery of. Gas lasers, 281 Gelfoam for patulous eustachian tube, 97 for tympanoplasty, 152 Gentamicin intratympanic, for inner ear conditions, 497–501 clinical studies of, 497–501, 498t–499t, 499f–500f experimental studies of, 497 results with, 501–502, 501t ototoxicity of, 493–494 Glasscock-Jackson classification, 715 Glomus jugulare tumors, 715–716 extreme lateral infrajugular transcondylar approach for, 715–726, 717f complications of, 724 results with, 724–726 surgical procedure for, 716–724, 718f–722f, 724f radiation therapy for, 566 stereotactic radiosurgery of, 796 surgery for complications of, 565–566 results of, 564–565 Glomus tumors arteriography in, 553 audiometry and, 553 biopsy in, 553 brain perfusion and flow studies in, 553 carotid artery, 551 computed tomography in, 552–553, 555 embolization in, 553 glomus vagale, 552 jugular bulb, 551 jugular foramen schwannomas as, 552 magnetic resonance imaging in, 552–553 surgery for, 551–568 complications of, 564f–567f, 565–566 with glomus jugulare tumors, 565–566 with glomus tympanicum tumors, 565 patient counseling for, 554 patient selection for, 551–552, 552t preoperative evaluation for, 553 results of, 564–565 with glomus jugulare tumors, 564–565 with glomus tympanicum tumors, 564 surgical approaches for, 554–564 complete carotid mobilization as, 562–564 fallopian bridge, 561–562, 563f infratemporal fossa, 557–561, 557f–562f
Index Glomus tumors (Continued) mastoid-extended facial recess, 555, 555f mastoid-neck, 555–557, 556f–557f transcanal, 554–555, 554f transcondylar, 562, 563f transdural, 552 tympanic, 551 tympanomastoid, 551 Glomus tympanicum tumors, surgery for complications of, 565 results of, 564 Glomus vagale tumors, 552 Glossopharyngeal nerve anatomy of, 443, 444f compression of, microvascular decompression for, 531–532 deficits of, rehabilitation after neurotologic skull base surgery for, 571 monitoring of, 782 paresis/palsy of with jugular foramen schwannomas, 552 with limited temporal bone resection, 40 preservation of, in glomus tumor surgery, 556 schwannomas, in neurofibromatosis 2, 695 Glossopharyngeal neuralgia, microvascular decompression for, 531–532 complications of, 532 operative technique for, 532 patient selection for, 531–532 results with, 532 Glycerol rhizotomy, shortcomings of, 525 Gold weight implantation for upper lid reanimation in facial paralysis, 740–742, 741f, 742t Gram-negative organisms, otitis media due to, 74 acute, 109 chronic suppurative, 108 Granulation tissue in chronic suppurative otitis media, 108, 111–112 extradural with acute infections, 186 treatment of, 189–190, 190f Granulomas cholesterol, 76 treatment of, drainage procedures for. See Petrous apex lesions, drainage procedures for. reparative following laser stapedectomy, 272 after stapes surgery, 320 Greater auricular nerve grafts for facial nerve tumors, 367, 368f
H Haemophilus influenzae intracranial complications due to, 183 meningitis due to, with cochlear implantation, 374t, 379 otitis media due to, 74, 77 acute, 109 otogenic complications due to, 183 Head trauma cerebrospinal fluid leaks due to, 246 treatment of, 248, 248f–249f translabyrinthine approach for. See Acoustic tumors, translabyrinthine approach for. Headache migraine-related dizziness and, 456, 510 following retrosigmoid approach to cerebellopontine angle tumors, 605–607 with retrosigmoid vestibular neurectomy, 451
Hearing through bone conduction, 397–398 following canal wall reconstruction tympanomastoidectomy, 181 after endolymphatic sac surgery, 424–425 good, definition of, 457 measurable, definition of, 457 monitoring of, 778–782 auditory evoked brainstem response for, 779–780, 779f direct eighth cranial nerve recording for, 780–781 electrocochleography for, 780–781 indications for, 778–779 laser-Doppler cochlear blood flow for, 782 otoacoustic emissions for, 781 pathophysiology and, 778 preoperative testing and, 778 following ossicular reconstruction, 170 partial stapedectomy in children and, 279 following perilymphatic fistula surgery, 330 preservation of with labyrinthine fistula repair, 231 in neurofibromatosis 2, 696–697 auditory implants for. See Auditory implants. retrosigmoid approach to cerebellopontine angle tumors and, 605–606 serviceable, definition of, 457 stapes prostheses and, stainless steel vs. polytef, 279 following tympanoplasty, 159 tympanostomy tubes and, in otitis media, 79 vestibular schwannomas and, 695 Hearing aids air conduction, implantable hearing aids vs., 384 bone-anchored, for aural atresia, 58–59 implantable, 383–396 Baha. See Bone-anchored cochlea stimulator (Baha). conventional hearing aids versus, 384–385 electromagnetic, 385 Envoy Esteem, 386t, 391–394, 393f historical background of, 385–386 MET, 386t, 394, 395f piezoelectric, 385 requirements for, 384 RION, 386, 386t–387t, 387f Soundtec, 386t, 391, 392f TICA, 386–389, 386t, 388f Vibrant Soundbridge, 386t, 389–391, 390f–391f, 391t reasons patients do not wear, 383 Hearing loss in aural atresia, 57 following cartilage tympanoplasty, 137 with chronic otitis media surgery, 233–234, 234f complicating operations for chronic ear infections, 128 conductive audiometry in, 163 Baha for, 398 differential diagnosis of, 509 inner ear, 272 in only hearing ear, Baha for, 399 stapedectomy and, 298–299, 300t after stapes surgery, 318 following surgery for facial nerve trauma, 360 after endolymphatic sac surgery, 418 with labyrinthine fistula repair, 231
821
Hearing loss (Continued) mastoidectomy and, 87 mixed, Baha for, 398 noise-induced, atresiaplasty and, 68 with perilymphatic fistulas, pathophysiology of, 325 risk of, with mastoidectomy, 210 sensorineural with chronic otitis media surgery, 233–234 with labyrinthine fistulas, 228 following laser stapedectomy, 272 mastoidectomy and, 86 progression of, after stapes surgery, 317 stapedectomy and, 298–299, 298t stapes surgery and, 316 single-sided, Baha for, 399 with translabyrinthine approach for acoustic tumors, 591 Helium-neon laser, 281 Hemangiomas, cavernous, 806–807 Hematomas, complicating operations for chronic ear infections, 128 Hemifacial spasm botulinum toxin for, 529 microvascular decompression for, 529–530 complications of, 530, 531t operative technique for, 528f, 529–530 patient selection for, 529 results with, 530 Hemorrhage adenoidectomy and, 87 from carotid artery, with chronic otitis media surgery, 241 from jugular bulb, with chronic otitis media surgery, 240–241 following limited temporal bone resection, 40 with retrosigmoid approach to cerebellopontine angle tumors, 616 from sigmoid sinus, with chronic otitis media surgery, 238–240, 239f–240f from superior petrosal sinus, with chronic otitis media surgery, 240 following transcochlear approach, 639 with translabyrinthine approach for acoustic tumors, 599 Hemostasis canal injection for, in stapes surgery, 307 for perilymphatic fistula surgery, 330 for retrosigmoid approach, 614 for tympanoplasty, 152 Hemotympanum, idiopathic, 74 diagnosis of, 76 recurrent, mastoidectomy and, 87 tympanostomy tubes for, 80 Herniation of brain, 245–252, 246f–247f computed tomography of, 247, 247f diagnosis of, 246–248, 247f–248f magnetic resonance imaging of, 247, 247f pathogenesis of, 245–246, 247f treatment of, 248–250, 248f–249f Herpes simplex virus, Bell’s palsy and, 335 Herpes zoster cephalicus, 338 Herpes zoster oticus, 338 Histologic evaluation in temporal bone malignancies, 44 Honeycomb bone, 363 Hook for vestibular neurectomy, 13, 14f House-Urban middle fossa retractor, 16, 16f Hydraulic chair for undersurface graft tympanoplasty, 142 Hydrocephalus otitic with acute infections, 186 treatment of, 193
822
Index
Hydrocephalus (Continued) following retrosigmoid approach to cerebellopontine angle tumors, 617 Hydrocortisone for acute otitis media, 110–111 Hydrops endolymphatic, pathophysiology of, 413–414 inner ear, aural fullness and, 99f Hydroset for ossicular prostheses, 162 Hydroxyapatite. See also Bone cement. for ossicular reconstruction, 166 for ossicular replacement, 162 with polyethylene (HAPEX), for ossicular replacement, 162, 166, 167f Hyperacusis after stapes surgery, 317 in superior canal dehiscence syndrome, 507–508 Hyperkinesis with facial reanimation, surgical management of, 770–771 Hypertension, neurogenic, microvascular decompression for, 532–533 complications of, 533 operative technique for, 532 patient selection for, 532 results with, 528f, 532–533 Hypoglossal nerve rehabilitation after neurotologic skull base surgery and, 578 schwannomas, in neurofibromatosis 2, 695 Hypoglossal/facial anastomosis for facial nerve paralysis, 755–764 end-to-side, via facial nerve translocation from fallopian canal, 759, 760f nerve growth factors and conduits for, 757–758 newer modifications of, 758–759, 758f–759f patient selection for, 755–756 results with, 759–761, 760t–761t surgical technique for, 756–757, 756f–757f Hypokinesis with facial reanimation, 770–771
I Immune function, depressed, failed antibiotic therapy due to, 113 Immunotherapy for otitis media, chronic, with effusion, 78 Implantable hearing aids. See Hearing aids, implantable. Incudostapedial joint division of, in stapes surgery, 309 fused, stapedectomy and, 295 identification of, 309 Incudostapedial joint prostheses, 167 Incus absent with mobile stapes and fixed malleus, ossicular reconstruction for, 164 with mobile stapes and mobile malleus, ossicular reconstruction for, 164, 164f dislocation of with facial nerve trauma, 354 during laser stapedectomy, 271 stapes surgery and, 312–313 erosion of revision stapedectomy and, 287 stapedectomy and, 299 fixation of, stapes surgery and, 310f, 313 foreshortened, with fixed footplate or prior stapedectomy, ossicular reconstruction for, 165, 165f
Incus (Continued) mobile with mobile footplate and malleus and absent stapes superstructure, ossicular reconstruction for, 164 with mobile malleus and fixed footplate, ossicular reconstruction for, 164 necrosis of with mobile stapes and malleus, ossicular reconstruction for, 164, 164f after stapes surgery, 318, 319f partial absence of, stapedectomy and, 295–297, 296f Incus/bridge prosthesis, 167 Infection(s). See also specific infections. acute, 183–194 clinical presentation of, 185–186 clinical significance of, 183 complications of aural, 183 intracranial, 183, 184f less obvious, 186, 187f obvious, 185 pathogenesis of, 183, 185f definition of, 183 diagnosis of, 186–188, 188f etiology of, 183 imaging of, 189 treatment of, 188–193 following canal wall reconstruction tympanomastoidectomy, 180–181 chronic, complications of operations for, 128 complicating operations for chronic ear infections, 128 following infratemporal fossa surgery, 663 respiratory upper, after stapes surgery, 317 viral, otitis media and, 75 sequestered nidus of, failed antibiotic therapy due to, 113 wound with cochlear implantation, 378 with retrolabyrinthine vestibular neurectomy, 451 after stapes surgery, 317 with surgery for facial nerve trauma, 360 Infection control for tympanoplasty, 151–152 Infratemporal fossa anterior transfacial (facial translocation) approach for, 660–661, 661f–662f anesthesia for, 652 diagnostic and staging work-up for, 649–650, 650f, 650t patient selection for, 649 pitfalls and complications of, 663–664 postoperative care for, 662–663 reconstructive considerations for, 651 rehabilitation considerations for, 650–651 lateral Fisch infratemporal approaches for anesthesia for, 652 diagnostic and staging work-up for, 649–650, 650f, 650t patient selection for, 649 pitfalls and complications of, 663–664 postoperative care for, 662–663 reconstructive considerations for, 651 rehabilitation considerations for, 650–651 lateral Fisch infratemporal fossa approaches for, 658–660, 660f type B, 658–659 type C, 659–660 type D, 660 postauricular (transtemporal) approach to, 657–658, 658f–659f
Infratemporal fossa (Continued) anesthesia for, 652 diagnostic and staging work-up for, 649–650, 650f, 650t patient selection for, 649 pitfalls and complications of, 663–664 postoperative care for, 662–663 reconstructive considerations for, 651 rehabilitation considerations for, 650–651 preauricular (subtemporal) approach to, 652–657, 653f, 655f–657f anesthesia for, 652 diagnostic and staging work-up for, 649–650, 650f, 650t patient selection for, 649 pitfalls and complications of, 663–664 postoperative care for, 662–663 reconstructive considerations for, 651 rehabilitation considerations for, 650–651 transorbital approach to, 661–662 anesthesia for, 652 diagnostic and staging work-up for, 649–650, 650f, 650t patient selection for, 649 pitfalls and complications of, 663–664 postoperative care for, 662–663 reconstructive considerations for, 651 rehabilitation considerations for, 650–651 Inner ear conductive hearing loss, 272 Instrumentation, 1–20 for adenoidectomy, 81 for Baha implant surgery, 402 for cochlear implant surgery, 15, 15f for endolymphatic sac surgery, 10–11, 11f, 19 for mastoidectomy, 8–9, 81 for middle cranial fossa surgery, 16, 16f–17f, 19–20 for middle cranial fossa vestibular neurectomy, 430, 431f for neurologic procedures, 11–12, 12f–15f for neurologic surgery, 19 for retrosigmoid approach, 607–608 for stapes surgery, 5, 6f–7f, 18 for tympanoplasty, 7–8, 9f–11f, 18 for tympanostomy tubes, 80–81, 82f for vestibular neurectomy, 13, 14f Internal auditory canal anatomy of, 442, 443f closure of, for retrosigmoid approach, 612f, 614 decompression of, without tumor removal, middle fossa craniotomy with, for vestibular schwannomas, 697 exposure of, 433–435, 436f for retrosigmoid approach, 609–611, 610f extended middle cranial fossa approach to, 641–648, 642t advantages of, 647 limitations of, 645–647 surgical anatomy and, 641–642, 642f surgical technique for, 642–645, 643f–646f technique for, 641 skeletonization of, in transcochlear approach, 634, 634f tumors with extension into, as contraindication to retrosigmoid approach, 606 Internal carotid artery anatomy of, 799–800 aneurysms of high cervical, 805 petrous, 803–805 management and treatment of, 804–805 presentation and pathophysiology of, 803–804 Irrigation for acute otitis media, 109
Index
J
Kanamycin, ototoxicity of, 494 Keratosis obturans, canalplasty for, 29 Klebsiella pneumoniae, otitis media and, 74
Labyrinthine fistulas (Continued) secondary to chronic otitis media, 227–228, 228t Labyrinthitis with acute infections, 185 serous, after stapes surgery, 317 treatment of, 189 Laser(s) comparison of, 284–285, 285f limitations of, 287–288 physics of, 281–282, 282f–284f safety and use of, for stapes surgery, 307 types of, 281, 282t use of, during 1990s to 2007, 285, 286t Laser myringotomy, 81–83 Laser stapedectomy. See Stapedectomy, laser. Lateralization, atresiaplasty and, 68 Lermoyez’s syndrome, 414 Liberatory maneuver, 468 Lid position, assessment of, for lower lid reapposition, in facial paralysis, 742–745 Lid stents for lower lid reapposition in facial paralysis, 747–749 Lid suture taped to cheek in facial paralysis, 750–752, 751f Lower lid reapposition procedures in facial paralysis, 742–749 brow elevation for, 749–750 canthoplasty for, 742 fascia lata suspension of lower lid for, 747, 748f lateral canthoplasty for, 745–747, 746f lid position assessment and, 742–745 lid stents for, 747–749 medial canthoplasty for, 744f, 745 midface support for, 749 Lower lip rehabilitation for facial reanimation, 770 Luc’s abscess, 186
L
M
Labyrinth, chemical treatment of, 493–506 intramuscular streptomycin for, 494–496 clinical studies of, 494 results with, 496 subtotal vestibulectomy using, 494–496 intratympanic corticosteroids for, 502–503 clinical studies of, 502–503 experimental studies of, 502 intratympanic gentamicin for, 497–501 clinical studies of, 497–501, 498t–499t, 499f–500f experimental studies of, 497 results with, 501–502, 501t streptomycin application to lateral semicircular canal for, 496–497 clinical studies of, 496 experimental studies of, 496 results with, 496–497 surgical method for, 496 Labyrinthectomy incomplete, with transcanal labyrinthectomy, 488–489 partial, 473–474 transcanal. See Transcanal labyrinthectomy. in transcochlear approach, 634, 634f transmastoid, translabyrinthine vestibular nerve section vs., 458 Labyrinthine fistulas iatrogenic, 231–233, 232f–233f intraoperative management of, 229–231, 229f–230f mastoidectomy for, 197, 205–206 patient counseling about, 198
Magnetic resonance imaging. See under specific conditions and procedures. Malleus fixation of, stapes surgery and, 310f, 313 fixed with fixed stapes, but absent incus, ossicular reconstruction for, 165, 165f with mobile stapes and absent incus, ossicular reconstruction for, 164 stapedectomy and, 257, 260f, 295, 297t, 300 fracture of, with facial nerve trauma, 354 mobile with fixed stapes, but absent incus, ossicular reconstruction for, 165, 165f with mobile footplate and incus and absent stapes superstructure, ossicular reconstruction for, 164 with mobile incus and fixed footplate, ossicular reconstruction for, 164 with mobile stapes and absent incus, ossicular reconstruction for, 164, 164f with mobile stapes and incus necrosis, ossicular reconstruction for, 164, 164f Malleus/oval window prosthesis, 167 Mastoid exenteration of, in mastoidectomy, 199, 201f obliteration of. See Tympanomastoidectomy, canal wall reconstruction, with mastoid obliteration. Mastoid cavity draining, management of, 113 exteriorized, indications for, 197
Jahrsdoerfer’s grading system for congenital aural atresia, 57, 57t “Joker,”, 16, 16f Jugular bulb injury with chronic otitis media surgery, 240–241 Jugular bulb tumors, 551 Jugular foramen tumors arteriography in, 553 audiometry and, 553 brain perfusion and flow studies in, 553 computed tomography in, 552–553, 555 embolization in, 553 magnetic resonance imaging in, 552–553 schwannomas as, 552 surgery for, 551–568 complications of, 564f–567f, 565–566 patient counseling for, 554 patient selection for, 551–552, 552t preoperative evaluation for, 553 results of, 564–565 surgical approaches for, 554–564 complete carotid mobilization as, 562–564 fallopian bridge, 561–562, 563f infratemporal fossa, 557–561, 557f–562f mastoid-extended facial recess, 555, 555f mastoid-neck, 555–557, 556f–557f transcanal, 554–555, 554f transcondylar, 562, 563f
K
823
Mastoid surgery. See also Mastoidectomy. complications of, 128 obliteration procedure as, 209, 216–217 reconstructive, 209, 217–218, 218f risks of, 128 Mastoid tip, mastoidectomy and, canal wall down, 212–213, 213f Mastoidectomy. See also Tympanomastoi dectomy. canal wall down procedure for, 209–220 atticotomy and, 209, 215–216, 216f decision making for, 210 definitions for, 209 dural venous sinuses and, 219 facial nerve and, 219–220, 219f indications for, 209–210 mastoid obliteration procedure and, 209, 216–217 mastoid reconstruction procedure and, 209 modified radical (Bondy), 209, 213–214, 214f patient counseling for, 210–211 patient positioning for, 211 patient preparation for, 211 pitfalls in, 219 postoperative care for, 218–219 preoperative evaluation for, 210 radical, 209, 216, 217f radical cavity reconstruction and, 217–218, 218f revision, 218 surgical techniques for, 211–213 facial ridge lowering adequacy and, 211–212, 212f mastoid tip management and, 212–213, 213f meatoplasty adequacy and, 213, 214f saucerization adequacy and, 211, 211f with tympanoplasty, 214–215, 215f tympanoplasty with, 209 canal wall up technique for. See Mastoidectomy, intact canal wall procedure for. complications of, 87 draping for, 80 for dural venous thrombophlebitis, 190–191, 191f follow-up for, 84 instrumentation for, 8–9, 81 intact canal wall procedure for, 195–208 canal wall defect repair and, 206 for defects resulting from disease, 205f, 206 for defects resulting from surgery, 206 complete canal wall up, for mastoiditis, 188 controversy over, 196 definitions for, 195 evolution of technique and, 195–196 exteriorized mastoid cavity and, indications for, 197 facial nerve and, 203 mastoid segment of, 203, 204f–205f tympanic segment of, 203, 204f–205f indications for, 196–197 labyrinthine fistula management and, 205–206 modified radical, 195 patient preparation for, 198 postoperative care for, 203 preoperative counseling for, 198 preoperative evaluation for, 197–198 preoperative preparation for, 197–198 radical, 195 surgical technique for, 198–203 completion of operation and, 201–202 elimination of disease and, 201, 204f–205f
824
Index
Mastoidectomy (Continued) mastoid exenteration and, 199, 200f–201f opening facial recess and, 199–201, 200f–201f������������������ , 204f–205f plastic sheeting and, 201, 204f–205f for removal of middle ear disease, 198 tympanoplasty with, 195 for labyrinthine fistulas, 197 laser, 81–83 modified radical, 214–215, 215f for otitis media, chronic, with effusion, 79 for petrositis, 189, 189f–190f pitfalls with, 86 postoperative care for, 84 preoperative audiometry for, 210 preoperative preparation for, 80 results with, 86 surgical site preparation and draping for, 80 surgical technique for, 174f–177f, 175–176 laser, 83–84, 85f in transcochlear approach, 633, 634f tympanoplasty with, 152, 195 Mastoiditis with acute infections, 185–186 masked, with acute infections, 186 treatment of, 188–189 Meatoplasty, mastoidectomy and, canal wall down, 213, 214f Meckel’s cave, tumors with extension into, retrosigmoid approach for, 606 Meniere’s disease, 414–417 audiometry in, 415 aural fullness and, 96–97 cochleosacculotomy for, 477–482 patient selection for, 478 rationale for, 477–478 results with, 481–482 surgical technique for, 478–481, 479f–481f definition of, 411 diagnosis of, 415 endolymphatic sac surgery for. See Endolymphatic sac surgery. epidemiology of, 414 superior canal dehiscence syndrome vs., 510 theories explaining symptoms of, 414 treatment of, 415–416, 457 chemical. See Labyrinth, chemical treatment of. intratympanic, 416–417 retrolabyrinthine vestibular neurectomy for. See Vestibular neurectomy, retrolabyrinthine. surgical. See Endolymphatic sac surgery; Transcanal labyrinthectomy; Vestibular neurectomy, middle cranial fossa; Vestibular neurectomy, retrosigmoid (suboccipital). vertigo in, 456 Meningiomas in neurofibromatosis 2, 695, 696f, 696t removal of, in transcochlear approach, 635–636, 636f–639f stereotactic radiosurgery of, 796 Meningitis with acute infections, 185 aseptic, following retrosigmoid approach to cerebellopontine angle tumors, 617 bacterial organisms causing, 183 following retrosigmoid approach to cerebellopontine angle tumors, 617 with cochlear implantation, 379 with facial nerve trauma, 355
Meningitis (Continued) with retrolabyrinthine vestibular neurectomy, 451 following transcochlear approach, �������� 640 following translabyrinthine approach for acoustic tumors, 599–600 treatment of, 192f, 193 Meningoencephaloceles, 245–252, 246f–247f computed tomography of, 247, 247f diagnosis of, 246–248, 247f–248f magnetic resonance imaging of, 247, 248f pathogenesis of, 245–246, 247f treatment of, 248–250, 248f–249f Meperidine (Demerol), setup of, 264 MET, 386t, 394, 395f Microraspatory, angled, 430, 431f Microscope, operating, 3 Microtia, 56 repair of, before atresiaplasty, 60 Microvascular decompression, 523–536 audiometric tests for, 523 brainstem auditory evoked potentials for, 523, 524f–526f for glossopharyngeal neuralgia, 531–532 complications of, 532 operative technique for, 532 patient selection for, 531–532 results with, 532 for hemifacial spasm, 529–530 complications of, 530, 531t operative technique for, 528f, 529–530 patient selection for, 529 results with, 530 for neurogenic hypertension, 532–533 complications of, 533 operative technique for, 532 patient selection for, 532 results with, 528f, 532–533 patient positioning for, 526–527, 528f for positional vertigo and tinnitus, 530–531 complications of, 531 operative technique for, 531 patient selection for, 528f, 530–531 results with, 531 preoperative evaluation for, 523, 524f–526f for trigeminal neuralgia, 523–529 in children, 527 complications of, 529 operative technique for, 526–527, 528f patient selection for, 523–526 results with, 527–529, 528f Midazolam (Versed) for revision stapedectomy, 288 setup of, 264 Middle cranial fossa approach, extended, 641–648, 642t advantages of, 647 limitations of, 645–647 surgical anatomy and, 641–642, 642f surgical technique for, 642–645, 643f–646f technique for, 641 Middle cranial fossa surgery draping for, 15 instrumentation for, 16, 16f–17f, 19–20 operating room setup for, 3, 5f patient preparation for, 15, 16f vestibular neurectomy as. See Vestibular neurectomy, middle cranial fossa. Middle cranial fossa vestibular neurectomy. See Vestibular neurectomy, middle cranial fossa. Middle ear allergic inflammation in, otitis media and, 76 exposure of, for tympanoplasty, 154, 155f–156f
Middle ear (Continued) preparation of, for undersurface graft tympanoplasty, 144, 147f Middle ear effusion, asymptomatic, 74 Middle fossa craniotomy, internal auditory canal decompression without tumor removal with, for vestibular schwannomas, 697 Middle fossa retractor articulated, 430, 431f introduction of, 433, 436f Midface support for lower lid reapposition, in facial paralysis, 749 Migraine-related dizziness, 456, 510 Mimmix for ossicular prostheses, 162 Minor’s syndrome. See Superior semicircular canal dehiscence syndrome. Mixed hearing loss, Baha for, 398 Mondini malformation,192f Moraxella catarrhalis otitis media and, 77 otitis media due to, 74 acute, 109 Mucosal disease. See Otitis media. Muscle plication for facial reanimation, 770 Musculoskeletal disorders, aural fullness and, 96–97 Myectomy, regional, for hyperkinesis, with facial reanimation, 770–771 Myringoplasty complications of, 128 risks of, 128 Myringotomy for facial paralysis, 189 for labyrinthitis, 189 for mastoiditis, 188 for meningitis, 192f for patulous eustachian tube, 97 for tube placement. See Tympanostomy tubes.
N Nasal obstruction, otitis media and, ������� 75 Nasal septum, tympanoplasty and, 150 Nasopharynx, posterior wall of, deep removal of, during adenoidectomy, 86 Neodymium:yttrium-aluminum-garnet (Nd:YAG) laser, 281 Neomycin for acute otitis media, 110–111 Nerve grafts for facial reanimation, 765–766 Nerve growth factors for hypoglossal/facial anastomosis, 758 Nerve substitution for facial reanimation, 766 Neurectomy peripheral, for tic relief, 525 vestibular. See Vestibular neurectomy. Neurofibromatosis 2, 691–702 definition of, 691–692, 692t eye findings in, 696 family history of, 693 Gardner form of, 692 genetic testing for, 699 imaging studies in, 692 magnetic resonance imaging in, 692 management of, 698–699 meningiomas in, 695, 696f, 696t molecular genetics of, 692–693 neurofibromatosis 1 differentiated from, 691 prevalence and incidence of, 692, 692f screening for, 693 spinal tumors in, 695–696
Index Neurofibromatosis (Continued) symptoms of, 696t tumor types in, 693–694, 694f vestibular schwannomas in, 694–695 hearing loss and, 695 treatment option(s) for, 696–698 auditory brainstem implant as, 698, 698f hearing preservation and, 696–697. See also Auditory implants. middle fossa craniotomy and internal auditory canal decompression without tumor removal as, 697 non-hearing preservation, translabyrinthine/suboccipital approach and total tumor removal as, 697–698 observation without surgery as, 697 retrosigmoid craniotomy with partial removal as, 697, 697f stereotactic irradiation as, 698 Wishart form of, 691–692 Neurofibromatosis 1, neurofibromatosis 2 differentiated from, 691 Neurogenic hypertension, microvascular decompression for, 532–533 complications of, 533 operative technique for, 532 patient selection for, 532 results with, 528f, 532–533 Neurolysis for hyperkinesis with facial reanimation, 770–771 Neuromas, acoustic auditory implants and. See Auditory implants. with chronic otitis media, retrosigmoid approach in, 606 combined therapy of, retrosigmoid approach in, 606 resection of, for retrosigmoid approach, 611–614, 612f translabyrinthine approach for. See Acoustic tumors, translabyrinthine approach for. transotic approach for, 621 Neurosurgery for brain abscess, 191–193, 191f draping for, 11, 11f–12f instrumentation for, 11–12, 12f–15f, 19 for subdural abscesses, 193 Neurotologic surgery skull base cranial nerve deficit rehabilitation after, 569–580, 570f for abducens nerve, 570–571 for cervical sympathetic chain, 578 for glossopharyngeal nerve, 571 for hypoglossal nerve, 578 patient evaluation for, 569 for spinal accessory nerve, 577–578 for trigeminal nerve, 569–570, 570f, 573f for vagus nerve, 571–577, 573f–574f, 576f transtemporal, 519–522 collaboration in, 520–521 nomenclature for, 520, 520f organizational framework for, 519, 520f vascular considerations in, 801–803 with middle fossa approaches, 803 of Kawase, 803 with petrous carotid exposure, 803 with transpetrosal approaches, 801–803, 802f retrolabyrinthine, 802 transcochlear, 803 translabyrinthine, 802–803
Neurovascular complications following infratemporal fossa surgery, 663–664 Nitinol-polytef piston for laser stapedectomy, 269 Noncompliance, failed antibiotic therapy due to, 112 Nose, allergic inflammation in, otitis media and, 76 Nystagmus in benign paroxysmal positional vertigo, 467 perilymphatic fistulas and, 325–326
O Oculomotor nerve monitoring, 782 Operating microscope, 3 setup of, for stapes surgery, 307, 308f Operating room, 1–3 drills in, 1–2, 2f–3f electrocautery equipment in, 1 general equipment for, 18 for middle cranial fossa procedures, 3, 5f observers in, 1, 2f operating microscope in, 3 setup of for laser stapedectomy, 263–264, 264f for neurologic surgery, 3 for routine otologic surgery, 3, 4f for tympanoplasty, outer surface grafting techniques, 120, 121f suction in, 1, 2f Operating table, 1, 2f positioning at, 3, 4f Osseointegration, 398 Ossicular chain. See also Incus; Malleus; Stapes enties. damage to, with facial nerve trauma, 354 dislocation of, during stapedectomy, 295 total fixation of, ossicular reconstruction for, 165, 165f tympanoplasty and, 222 Ossicular prostheses extrusion of, 237 materials for, 161 for partial ossicular replacement, 161–162 selection of, 165–167, 166f–167f for total ossicular replacement, 161–162 Ossicular reconstruction, 161–172 factors influencing outcome with, 170 historical background of, 161–162 patient evaluation for, 162–163 patient selection for, 162–163 postoperative care for, 170 prosthesis selection for, 165–167, 166f–167f surgical considerations in, 163–165, 164f–165f surgical technique for, 167–170 cartilage preparation and, 169 exposure and assessment and, ��������� 167, 168f prothesis placement and, 169–170, 169f tympanoplasty and, 152 Ossiculoplasty, complications of, 237 Osteogenesis imperfecta, stapes surgery and, 313 Osteonecrosis of tympanic bone, canalplasty for, 29 Osteoradionecrosis of tympanic bone, canalplasty for, 29 Otic capsule exenteration, transotic approach for cerebellopontine angle lesions and, 622–624, 623f Otitis externa, chronic Baha for, 399
825
Otitis media acute chronic suppurative otitis media arising from, 108 with effusion, 73 epidemiology of, 74 guidelines for management of, 88 with intact tympanic membrane, 109 with open tympanic membrane, 109 pathophysiology of, 75 recurrent, 73 treatment of, 77, 77t after stapes surgery, 316 treatment of, 77, 77t without effusion, 73 chronic, 227–244 acoustic neuroma with, retrosigmoid approach in, 606 with effusion pathophysiology of, 75–76 treatment of, 77–79 labyrinthine fistulas and iatrogenic, 231–233, 232f–233f secondary to otitis media, 227–231, 228t, 229f–230f ossicular reconstruction for, 162 preoperative counseling in, 227 second-stage tympanoplasty for, 222–223 stapedectomy in, 302 suppurative, 73–74, 107–109 bacteriology of, 108 definition of, 107–108 etiology of, 108 granulation tissue in, 108 pathophysiology of, 108–109 surgery for, complications of dural injury as, 237–238, 238f–239f facial nerve grafting and, 236–237, 236f facial nerve injury as, ��������� 234, 235f labyrinthine fistulas as, 231–233, 232f–233f sensorineural hearing loss as, 233–234, 234f vascular injury as, 238–241, 239f–240f definition of, 73–74 diagnosis of, 76 with effusion, 73 allergy treatment for, 87–88 biofilms and, 88 guidelines for management of, 88 risk factors for, 74 microbiology of, 74 tympanoplasty and, 221–223, 224f tympanostomy tubes for. See Tympanostomy tubes. Otoacoustic emissions, monitoring hearing using, 781 Otologic drills, 1–2, 2f–3f Otorrhea cerebrospinal fluid. See Cerebrospinal fluid otorrhea. in chronic suppurative otitis media, 73–74 with facial nerve trauma, 354–355 tympanoplasty and, 150–151 tympanostomy tubes and, 86–87 biofilm formation and, 88 Otosclerosis far advanced, stapedectomy for, 301 juvenile, partial stapedectomy for, 279 obliterative laser stapedectomy and, 271 stapedectomy and, 297 stapes surgery and, 314, 315f
826
Index
Otosclerosis (Continued) stapedectomy for, total. See Stapedectomy, total. surgery for. See Stapedectomy; Stapes surgery. Otosclerotic inner ear syndrome after stapes surgery, 317 Otosclerotic regrowth after stapes surgery, 318, 319f Oval window deep, stapedectomy and, 261 narrow niche and, stapes surgery and, 314 narrow oval window niche and promontory drilling and, stapedectomy and, 297 protection of, with labyrinthine fistula repair, 232, 232f–233f tissue seal at, 312 Oval window drill-out, atresiaplasty and, 69 Oval window seal, stapes surgery and, 312
P Pain following transcochlear approach, 639–640 Palisade technique for cartilage tympanoplasty, 133–135, 134f–136f Palpebral spring implantation enhanced, for upper lid reanimation, in facial paralysis, 738–740, 739f, 741f for upper lid reanimation, in facial paralysis, 736f, 737–738 Paper patching for tympanic membrane closure, 114f, 115–116 Paragangliomas stereotactic radiosurgery of, 796 of temporal bone, radiation therapy for, 715 Paranasal sinuses, endoscopic endonasal approaches to, 667–680 complications of, 679 coronal plane, 668, 670f anterior fossa, 676–677 infratemporal fossa, 677–679, 678f lower infratemporal fossa/parapharyngeal space, 679 posterior fossa, 679 supraorbital, 676 transcondylar, 679 transorbital, 676 transpterygoid, 676, 677f–678f historical background of, 667 learning curve for, 679 reconstruction and, 679–680 sagittal plane, 668, 668f foramen magnum, 675 middle clivus, 674–675, 674f superior clivus, 673–674, 674f transclival, 673–675 transcribriform, 669–671, 671f transfrontal, 668–669 transodontoid, 675, 675f transplanum/transtuberculum, 671–672, 672f transsellar, 672–673, 673f Particle repositioning technique, 468–469, 469f Patient positioning. See under specific procedures. Patulous eustachian tube reconstruction procedure, 98–101 Pediatric patients Baha in, 408–409 ossicular reconstruction in, 163 otosclerosis in, partial stapedectomy for, 279 stapedectomy in, 302 stapedotomy in, 312 trigeminal neuralgia in, microvascular decompression for, 527
Penetrating auditory brainstem implant, 703. See also Auditory implants. Peptococcus spp., otitis media and, 74 Peptostreptococcus spp., otitis media and, 74 Perichondrial grafts harvesting of, 158f, 159 for undersurface graft tympanoplasty, 144, 146f Perichondrium/cartilage island flap for tympanoplasty, 132–133, 132f–133f placement problems with, 136–137 poor fit of, 135–136 Perilymph, excessive flow of, stapedectomy and, 259 Perilymphatic fistulas, 323–334, 324f acquired, 323 congenital, 323 hearing loss and, pathophysiology of, 325 idiopathic, 323 implosive and explosive, 323 recurrent, 331 after stapes surgery, 320 superior canal dehiscence syndrome vs., 510 surgery for anesthesia for, 327, 330 dressings and, 330 instruments for, 329 patient counseling and, 326–327 patient selection for, 325 pitfalls with, 330 postoperative care for, 330 preoperative evaluation for, 325–326, 327f preoperative preparation for, 327–329 results with, 330–331 surgical technique for, 328f, 329–330 Perilymphatic gusher, stapes surgery and, 314–316 Petroclival area, petrosal approach to, 681–690 closure and, 687–688 combined petrosal variations of, 681–682, 682f preoperative evaluation for, 682–683 preoperative preparation for, 683 results with, 688–689, 688f surgical techniques for, 683–687 middle fossa exposure and, 685–687 posterior exposure and, 683–685, 684f–687f Petrosectomy, subtotal, 544t, 546, 546f transotic approach for cerebellopontine angle lesions and, 622, 623f Petrositis with acute infections, 186 treatment of, 189, 189f–190f Petrous apex lesions drainage procedures for, 537–550, 538t complications of, 544–545, 544t computed tomography and, 537, 538t infralabyrinthine, 539f, 540–541, 544t pitfalls of, 541 middle fossa approach for, 544t, 546, 549f patient counseling for, 540 patient selection for, 538, 539f preoperative evaluation for, 538–540 results with, 544 subtotal petrosectomy for, 544t, 546, 546f transcanal infracochlear approach for, 541–544, 542f, 544t translabyrinthine, 544t transsphenoidal approach for, 544t, 545–546 magnetic resonance imaging and, 537, 538t
Petrous bone vascular anatomy of, 799–801 dural venous sinuses and, 801 external carotid artery and, 800 internal carotid artery and, 799–800 vertebrobasilar arteries and, 800–801 vascular lesions of, 803–807 cavernous malformations and cavernous hemangiomas as, 806–807, 808f dural arteriovenous malformations as, 805–806 internal carotid aneurysms as, 803–805 high cervical, 805 petrous, 803–805 tentorial fistulas as, 806 transverse and sigmoid fistulas as, 806 vertebrobasilar aneurysms as, 805 vascular tumors of, 807–813 cerebral revascularization and bonnet bypass and, 812–813, 812f–813f preoperative embolization and, 807–811 glomus tumors and, 807–809, 810f–811f hemangioblastomas and, 809–811 hemangiomas and, 811 Physiologic monitoring during endolymphatic sac surgery, 10 Pilots, stapedectomy in, 302 Plastic sheeting for tympanoplasty, 223 Platform posturography for translabyrinthine vestibular neurectomy, 458–459 Plester ossicular prosthesis, 167 Pneumatization, lack of, atresiaplasty and, 60 Pneumocephalus with facial nerve trauma, 355 Polyethylene, thermal fused (Polycel), for ossicular reconstruction, 161–162 Polyethylene sponge (Plastipore) for ossicular reconstruction, 161–162, 166 Polymyxin B for acute otitis media, 110–111 Positional vertigo microvascular decompression for, 530–531 complications of, 531 operative technique for, 531 patient selection for, 528f, 530–531 results with, 531 paroxysmal, benign, 467–476 pathophysiology of, 468 posterior semicircular canal occlusion for patient counseling and, 468–470, 469f patient selection for, 470 preoperative evaluation for, 471 results with, 471–473 surgical technique for, 471, 472f Posterior canal dehiscence, posterior semicircular canal occlusion for, 474 Posterior inferior cerebellar artery, anatomy of, 800–801 Posterior semicircular canal occlusion for benign paroxysmal positional vertigo patient counseling and, 468–470, 469f patient selection for, 470 preoperative evaluation for, 471 results with, 471–473 surgical technique for, 471, 472f for posterior canal dehiscence, 474 for superior semicircular dehiscence, 473 Posturography, platform, for translabyrinthine vestibular neurectomy, 458–459 Potassium iodide for patulous eustachian tube, 97 Potassium titanyl phosphate crystal laser, 281, 282t, 284 Prednisone for Bell’s palsy, 337–338 for Ramsay Hunt syndrome, 339
Index Premarin for patulous eustachian tube, 97 Prepontine cistern, extended middle cranial fossa approach to, 641–648, 642t advantages of, 647 limitations of, 645–647 surgical anatomy and, 641–642, 642f surgical technique for, 642–645, 643f–646f technique for, 641 Promontory, evaluation of, for cochlear implantation, 375 Propionibacterium acnes, otitis media and, 74 Proteus mirabilis, intracranial complications due to, 183 Pseudomonas, acute otitis media due to, 109 Pseudomonas aeruginosa chronic suppurative otitis media due to, 108 intracranial complications due to, 183 otitis media and, 74 otogenic complications due to, 183
Q Quix test for perilymphatic fistulas, 326
R Radiation therapy for facial nerve tumors, 365 for glomus jugulare tumors, 566 following limited temporal bone resection, 41 with radical temporal bone resection, 50 stereotactic, for vestibular schwannomas, 698 Radiofrequency rhizotomy, shortcomings of, 525 Radiosurgery for facial nerve tumors, 365 stereotactic, for skull base tumors. See Skull base tumors, stereotactic radiosurgery of. Ramsay Hunt syndrome, 338–339 Reanimation for facial paralysis. See Facial reanimation. of upper lid in facial paralysis, 735–737 enhanced palpebral spring implantation for, 738–740, 739f, 741f gold weight implantation for, 740–742, 741f, 742t palpebral spring implantation for, 736f, 737–738 silicone rod prosthesis implantation for, 742, 743f Respiratory infections upper, after stapes surgery, 317 viral, otitis media and, 75 Retrolabyrinthine vestibular neurectomy. See Vestibular neurectomy, retrolabyrinthine. Retrosigmoid craniotomy with partial removal of vestibular schwannoma, 697, 697f Retrosigmoid vestibular neurectomy. See Vestibular neurectomy, retrosigmoid (suboccipital). Revision stapedectomy, laser. See Stapedectomy, laser, revision. Rhinitis, allergic, otitis media and, 75 Rhinorrhea following infratemporal fossa surgery, 664 Rhizotomy glycerol, shortcomings of, 525 radiofrequency, shortcomings of, 525 RION, 386, 386t–387t, 387f Round window closure of, stapedectomy and, 259 obliteration of, stapes surgery and, 314 Rupture theory of Meniere’s disease, 414
S Sacculotomy, perilymphatic fistulas and, 325 Salicylic acid for patulous eustachian tube, 97 Salivary gland tumors, involving temporal bone, 33 limited temporal bone resection for. See Temporal bone resection, limited. Saucerization for mastoidectomy, canal wall down, 211, 211f Scalp flap, necrosis of, following infratemporal fossa surgery, 663 Schuknecht’s classification system for congenital aural atresia, 57 Schwannomas facial nerve, in neurofibromatosis 2, 694–695 glossopharyngeal, in neurofibromatosis 2, 695 hypoglossal, in neurofibromatosis 2, 695 jugular foramen, 552 trigeminal nerve, in neurofibromatosis 2, 694 vagal nerve, in neurofibromatosis 2, 695 vestibular in neurofibromatosis 2. See Neurofibromatosis 2, vestibular schwannomas in. stereotactic irradiation for, 698 Scutum defects, canalplasty for, 29–31, 30f Sedation See under specific procedures. Selector ultrasonic aspirator for vestibular neurectomy, 13, 14f Semicircular canal, lateral, streptomycin application to, 496–497 clinical studies of, 496 experimental studies of, 496 results with, 496–497 surgical method for, 496 Sensorineural hearing loss. See Hearing loss, sensorineural. Sigmoid sinus injury with chronic otitis media surgery, 238–240, 239f–240f Silicone rod prosthesis implantation for upper lid reanimation in facial paralysis, 742, 743f Silicone sheeting for tympanoplasty, 223 extrusion of, 225 Sinusitis, tympanoplasty and, 150 Skin grafts following limited temporal bone resection, 38f, 39–40 Skin reactions with Baha implant surgery, 407 Skull base, tumors with extension into, as contraindication to retrosigmoid approach, 606 Skull base tumors endoscopic endonasal approaches to surgery for, 667–680 complications of, 679 coronal plane, 668, 670f anterior fossa, 676–677 infratemporal fossa, 677–679, 678f lower infratemporal fossa/parapharyngeal space, 679 posterior fossa, 679 supraorbital, 676 transcondylar, 679 transorbital, 676 transpterygoid, 676, 677f–678f historical background of, 667 learning curve for, 679 reconstruction and, 679–680 sagittal plane, 668, 668f foramen magnum, 675 middle clivus, 674–675, 674f superior clivus, 673–674, 674f
827
Skull base tumors (Continued) transclival, 673–675 transcribriform, 669–671, 671f transfrontal, 668–669 transodontoid, 675, 675f transplanum/transtuberculum, 671–672, 672f transsellar, 672–673, 673f extreme lateral infrajugular transcondylar approach for resection of, 715–726 complications of, 724 results with, 724–726 surgical procedure for, 716–724, 718f–722f, 724f neurotological surgery for. See Neurotologic skull base, surgery���� �����������. stereotactic radiosurgery of, 785–798 CyberKnife, 794–795 dose distribution and, 795 localization and, 795 treatment delivery and, 795 treatment planning and, 794–795, 794f outcomes with, 791–794, 792f–793f patient selection for, 785–786 preoperative counseling for, 786 surgical technique for, 786–787 anesthesia and, 788 frame attachment and, 787–791, 788f gamma knife unit and, 786–787, 787f imaging and, 789 treatment and, 791 treatment planning and, 789–791, 789f–790f Solid state lasers, 281 Sonotubometry in patulous tympanic membrane, 95 Soundtec Direct Drive Hearing System, 386t, 391, 392f Spinal accessory nerve anatomy of, 443, 444f monitoring of, 782 in glomus tumor surgery, 555–556 palsy of, with jugular foramen schwannomas, 552 preservation of, in glomus tumor surgery, 556 rehabilitation after neurotologic skull base surgery and, 577–578 Spinal tumors in neurofibromatosis 2, 695–696 Squamous cell carcinoma of external auditory canal, staging of, 35–36, 35t of temporal bone, 33 Stapedectomy, 293–304 atrophic tympanic membrane and, 295 bilateral, 302 in children, 302 in chronic otitis media, 302 contraindications to, 254 dehiscent facial nerve and, 297 drill, laser stapedectomy vs., 311 in elderly patients, 302 in far advanced otosclerosis, 301 fixed malleus and, 295, 297t, 300 floating footplate and, 296f, 297–298, 298t fused incudostapedial joint and, 295 indications for, 254 instrumentation for, 6 intraoperative audiometry and, 293–294, 294f intraoperative audiometry for, 293–294, 294f laser, 263–274 complications of intraoperative, 271 postoperative, 272
828
Index
Stapedectomy (Continued) draping for, 264, 264f drill stapedectomy vs., 311 with floating footplate, 271 nitinol-polytef piston for, 269 in obliterative otosclerosis, 271 operating room setup for, 263–264 with overhanging facial nerve, 269–271 patient counseling about, 263 patient selection for, pitfalls in, 272–273 postoperative care for, 271–272 preoperative evaluation for, 263 preparation for, 263–264 results with, 272–273 revision, 281–292 advantages of, 287 anesthesia for, 288 comparison of lasers for, 284–285, 285f failed stapes surgery analysis and, 285–287, 287t historical background of, 282–288 laser development and, 283–284 laser physics and principles for, 281–282, 282f–284f, 282t laser use during 1990s to 2007 and, 285, 286t limitations of lasers and, 287–288 technique for, 288–290, 289f sedation for, 264 surgical technique for, 264–269, 264f–270f narrow oval window niche and promontory drilling and, 297 obliterative otosclerosis and, 297 ossicular dislocation during, 295 partial, 275–280 cup/piston stapes prostheses for indications for, 278 modification of, 278 historical background of, 275–276 in juvenile otosclerosis, 279 piston concept and, 275 results with, 278 stainless steel versus polytef prostheses for, 279 surgical technique for, 276–278, 277f wire/stapes prostheses for, 278–279 partial absence of incus and, 295–297, 296f patient counseling for, 253 patient selection for, 253–254 perilymphatic fistulas following, 323 in pilots, 302 pitfall(s) in, 257–261 chorda tympani nerve and, 257, 260f deep oval window as, 261 eardrum perforation as, 257 facial nerve abnormalities as, 257–259 floating footplate as, 259 intraoperative vertigo as, 259–261 malleus fixation as, 257, 260f obliterated footplate as, 259, 261f perilymph flow as, 259 round window closure as, 259 postoperative care for, 257 prior, foreshortened incus with, ossicular reconstruction for, 165, 165f revision, 298–301 for conductive hearing loss, 298–299, 300t laser. See Stapedectomy, laser, revision. recommendations for, 300–301 for sensorineural hearing loss, 298–299, 298t surgical findings in, 299–300, 300t surgical technique for, ������������������� 299, 296f
Stapedectomy (Continued) routine, 294–295, 296f for small air-bone gaps, 301–302 small fenestra technique for, 300 stapedotomy vs., 311 surgical technique for, 254–257, 256f, 258f total, 253–262 anesthesia for, 254 tympanic membrane perforation during, 295 tympanomeatal flap tears during, 295 Stapedial artery, persistent, stapes surgery and, 313 Stapedial footplate. See Footplate. Stapedotomy anesthesia for, 312 in children, 312 laser, laser type for, 312 prosthesis displacement at, 312 site for, 311 stapedectomy vs., 311 Stapes absent superstructure of, with mobile footplate, incus, and malleus, ossicular reconstruction for, 164, 164f displacement of, 315f, 318 fixation of, congenital, stapes surgery and, 313 fixed, with fixed or mobile malleus, but absent incus, ossicular reconstruction for, 165, 165f fracture of crura of, with facial nerve trauma, 354 mobile with fixed malleus and absent incus, ossicular reconstruction for, 164 with mobile malleus and absent incus, ossicular reconstruction for, 164, 164f mobile with mobile malleus and incus necrosis, ossicular reconstruction for, 164, 164f Stapes curettes, 6, 7f Stapes prosthesis(es) availability of, 311 ballottement of, 312 cup/piston indications for, 278 modifications of, 278 wire prostheses compared with, 278–279 displacement of, at stapedotomy, 312 extrusion of, 318 incus erosion and, 295 long, overinsertion of, 318, 319f loose coupling of, 318–320 nitinol, 288 placement of, 309 polytef, 299 preparation and loading of, 309 removal of, 309 Robinson fistulas with, 300 malfunction of, 299 sizes of, 311, 311t sizing of, 309, 310f stainless steel, Robinson, 294 stainless steel vs. polytef, 279 types of, 311 wire, incus erosion with, 295–297, 296f, 299 wire loop, 318, 319f Stapes superstructure, removal of, 309 Stapes surgery, 305–322. See also Stapedectomy; Stapedotomy. balloon ballottement and, 312 biscuit footplate and, 310f, 314
Stapes surgery (Continued) blood within vestibule and, 316 congenital stapes fixation and, 313 congenitally ectopic facial nerve and, 314 draping for, 4, 5f failed, analysis of, 285–287, 287t footplate fragments in vestibule and, 316 footplate/vestibular relationships and, 311–312 fractured footplate and, 316 incus dislocation and, 312–313 incus fixation and, 310f, 313 instrumentation for, 5, 6f–7f, 18 malleus fixation and, 310f, 313 narrow oval window niche and, 314 obliterative otosclerosis and, 314, 315f operative setup for, 3–7, 7f–8f osteogenesis imperfecta and, 313 oval window seal and, 312 overhanging facial nerve and, 314, 315f patient positioning for, 4–5 patient preparation for, 3, 5f perilymphatic gusher and, 314–316 persistent stapedial artery and, 313 preoperative evaluation for, 305–306 informed consent and, 306 medical conditions and, 305–306 physical examination in, 306 primary, complications after, 316–320 medical management of, 316–317 surgical management of, 317–320 prostheses for. See Stapes prosthesis(es). revision, surgical risk reduction in, 320 round window obliteration and, 314 sensorineural hearing loss and, 316 stapedectomy as. See Stapedectomy. stapedotomy as. See Stapedotomy. surgical technique prerequisites for residents and, 306–311 tympanic membrane perforation and, 312 tympanomeatal flap tears and, 312 tympanosclerosis and, 313 vascular anomalies and, 313 Staphylococcus, acute otitis media due to, 109 Staphylococcus aureus intracranial complications due to, 183 otitis media and, 74 otogenic complications due to, 183 Static slings for facial reanimation, 770 Stereotactic irradiation for vestibular schwannomas, 698 Stereotactic surgery for glomus jugulare tumors, 566–567 radiosurgery as of skull base tumors. See Skull base tumors, stereotactic radiosurgery of. for trigeminal neuralgia, shortcomings of, 525–526 Steroids for Bell’s palsy, 337–338 intratympanic, for inner ear conditions, 502–503 clinical studies of, 502–503 experimental studies of, 502 for otitis media, chronic, with effusion, 78 Streptococcus pneumoniae intracranial complications due to, 183 meningitis due to, with cochlear implantation, 374t, 379 otitis media due to, 74, 77 acute, 109 otogenic complications due to, 183 Streptococcus pyogenes intracranial complications due to, 183 otogenic complications due to, 183
Index Streptomycin application to lateral semicircular canal, for inner ear conditions, 496–497 clinical studies of, 496 experimental studies of, 496 results with, 496–497 surgical method for, 496 intramuscular, for inner ear conditions, 494–496 clinical studies of, 494 results with, 496 subtotal vestibulectomy using, 494–496 ototoxicity of, 493–494 Subdural abscesses with acute infections, 186 treatment of, 193 Suboccipital vestibular neurectomy. See Vestibular neurectomy, retrosigmoid (suboccipital). Suction in operating room, 1, 2f Suction tubes for stapes procedures, 7, 7f Superior petrosal sinus injury with chronic otitis media surgery, 240 Superior semicircular canal dehiscence syndrome, 456, 507–518 audiography in, 508, 508f aural fullness and, 96, 99f bilateral, 512 computed tomography in, 509, 510f diagnosis of, 96 diagnostic evaluation for, 507–509, 508f–510f differential diagnosis of, 509–510 long-term results with, 516 operative technique for, 512–515, 513f–516f posterior semicircular canal occlusion for, 473 postoperative care for, 515 preoperative decision making for, 510–512 symptoms of, 507 Suppurative otitis media, chronic. See Otitis media, Chronic suppurative. Supramid Extra for tympanoplasty, 223 Surgeons, tremor in, 307 Surgical drills, 1–2, 2f–3f Suture dehiscence, post-traumatic, canalplasty for, 31 Synkinesis with facial reanimation, 770–771
T Tarsorrhaphy suture, temporary, in facial paralysis, 751f, 752 Tast disturbance, complicating operations for chronic ear infections, 128 Teflon for patulous eustachian tube, 97 Temporal bone drainage procedures for tumors of. See Petrous apex lesions, drainage procedures for. encephaloceles of, 245–252, 246f–247f computed tomography of, 247, 247f diagnosis of, 246–248, 247f–248f pathogenesis of, 245–246, 247f treatment of, 248–250, 248f–249f fractures of cerebrospinal fluid otorrhea due to, 107 facial nerve injury due to, 347–349, 349t longitudinal, 347, 349–354, 350f–351f transverse, 348, 352f–353f glomus tumors of, surgery for. See Glomus tumors, surgery for. gunshot wounds to, 349 malignancies of, 43–54. See also specific tumors.
Temporal bone (Continued) angiography in, 44 biopsy of, 34 diagnostic evaluation of, 43–44 diagnostic tests for, 34–35 history and physical examination in, 34 limited temporal bone resection for. See Temporal bone resection, limited. recurrence of, 51 staging of, 35–36, 35t temporal bone resection for. See Temporal bone resection, limited; Temporal bone resection, radical. ossification centers forming, 245 paragangliomas of, radiation therapy for, 715 post-traumatic suture dehiscence and, canalplasty for, 31 Temporal bone resection lateral, 43. See also Temporal bone resection, radical. limited, 33–42 chemotherapy and, 41 complications of, 40 postoperative reconstruction and, 38f, 39–40 radiation following, 41 rehabilitation of lower cranial nerve defects and, 41 surgical technique for, 36–39, 37f–38f radical, 43–54 adjuvant treatment with, 50 complications of, 50 diagnostic evaluation for, 43–44 follow-up for, 51 postoperative care and, 50 rehabilitation following, 50 results with, 51–52, 51f surgical procedure for, 44–50, 45f, 47f–49f subtotal, 43. See also Temporal bone resection, radical. total, 43. See also Temporal bone resection, radical. Temporalis fascia in undersurface graft tympanoplasty, 142 Temporalis muscle flaps following limited temporal bone resection, 40 for middle cranial fossa vestibular neurectomy, 430–431, 432f Temporalis muscle retractor, 430, 431f Temporalis muscle transposition for facial reanimation, 766–767 complications of, 767 patient evaluation for, 766 patient selection for, 766 results with, 767 surgical technique for, 766–767, 768f–769f Temporomandibular joint dysfunction of, aural fullness and, 96–97, 99f with infratemporal fossa surgery, 663 Tensor veli palatini, 93 release of, for patulous eustachian tube, 98 Thrombophlebitis, dural venous sinus with acute infections, 186 treatment of, 190–191, 191f Tinnitus complicating operations for chronic ear infections, 128 microvascular decompression for, 530–531 complications of, 531 operative technique for, 531 patient selection for, 528f, 530–531 results with, 531 perilymphatic fistulas and, 325 Titanium for ossicular prostheses, 162, 166 Topognostic testing in Bell’s palsy, 337
829
Torus tubarius, trauma to, during adenoidectomy, 86 Totally Implantable Communication Assistance device, 386–389, 386t, 388f Transcanal labyrinthectomy, 483–492 anesthesia for, 484–485 cochlear implantation and, �������� 490 complications of intraoperative, 488 postoperative, 488–489 histopathology of, 489, 489f informed consent for, 484 patient counseling for, 484 patient selection for, 483–484 postoperative care for, 485–488 preoperative evaluation for, 484 results with, 490 surgical technique for, 484–485 incision and exposure and, 485, 486f preparation for vestibule opening and, 485, 486f vestibular end organ removal and, 485, 486f–488f Transdural glomus tumors, 552 Translabyrinthine vestibular nerve section, transmastoid labyrinthectomy vs., 458 Translabyrinthine vestibular neurectomy. See Vestibular neurectomy, translabyrinthine. Transmastoid labyrinthectomy, translabyrinthine vestibular nerve section vs., 458 Transtemporal skull base surgery, 519–522 collaboration in, 520–521 nomenclature for, 520, 520f organizational framework for, 519, 520f Transtemporal-supralabyrinthine approach. See Vestibular neurectomy, middle cranial fossa. Tremor of surgeon’s hand, 307 Triad of Gradenigo, 186 Trichloroacetic acid for tympanic membrane closure, 116 Trigeminal nerve anatomy of, 443, 444f deficits of with infratemporal fossa surgery, 663 rehabilitation after neurotologic skull base surgery for, 569–570, 570f, 573f schwannomas of, in neurofibromatosis 2, 694 Trigeminal neuralgia microvascular decompression for, 523–529 in children, 527 complications of, 529 operative technique for, 526–527, 528f patient selection for, 523–526 results with, 527–529, 528f stereotactic radiosurgery of, 796 Trismus following infratemporal fossa surgery, 663 Trochlear nerve monitoring, 782 T-tubes with cartilage tympanoplasty, 137, 137f Tullio phenomenon, 507 Tumarkin’s otolithic catastrophe, 478 Tympanic bone osteonecrosis of, canalplasty for, 29 osteoradionecrosis of, canalplasty for, 29 Tympanic glomus tumors, 551 Tympanic membrane atrophic management of, in cartilage tympanoplasty, 137 stapedectomy and, 295 collapse of, following tympanoplasty, 221 in idiopathic hemotympanum, 74 lateralization of, with outer surface grafting tympanoplasty, 125f, 126
830
Index
Tympanic membrane (Continued) perforation of during laser stapedectomy, 271 in stapedectomy, 257 stapedectomy and, 295 stapes surgery and, 312 perforations of closure of cauterization and paper patching for, 114f, 115–116 fat graft tympanoplasty for, 115f, 116 persistent, tympanostomy tubes and, 87 reconstruction of. See Tympanoplasty. remnant of, preparation for tympanoplasty, 154 re-retraction of, following canal wall reconstruction tympanomastoidectomy, 181 Tympanomastoid glomus tumors, 551 Tympanomastoidectomy, canal wall reconstruction completeness of cholesteatoma removal and, 181 with mastoid obliteration, 173–182 advantages and disadvantages of canal wall up and canal wall down techniques and, 173–174 draping for, 174 hearing results with, 181 mastoidectomy considerations and, 174f–177f, 175–176 patient positioning for, 174 patient preparation for, 174 patient selection for, 174 pitfalls in, 180–181 postoperative care for, 177–178 rationale for, 174 reconstruction and, 176–177, 178f–180f surgical technique for, 175–177 Tympanomeatal flaps design and elevation of, 307–308, 308f placement of, for undersurface graft tympanoplasty, 144, 147f tears of during stapedectomy, 295 stapes surgery and, 312 Tympanometry, impedance, in patulous eustachian tube, 95, 95f Tympanoplasty, 221–226 cartilage, 131–140, 157–159 complications of intraoperative, 135–137 postoperative, 137–138, 137f patient selection for, 131–132 postoperative care for, 135 results with, 138 revision surgery and, 138 surgical technique for, 132–135, 158f, 159 palisade technique and, 133–135, 134f–136f perichondrium/cartilage island flap and, 132–133, 132f–133f postauricular approach and, 132 complications of, 128 draping for, 7, 8f fat graft, 115f, 116 instrumentation for, 7–8, 9f–11f, 18 with mastoidectomy, 149 canal wall down, 209, 214–215, 215f intact canal wall, 195 mastoidectomy with, 152 outer surface grafting technique for, 119–130 advantages and disadvantages of, 126 anesthesia for, 120 blunting in anterior sulcus and, 125f, 126 epithelial cysts and, 125f, 126
Tympanoplasty (Continued) healing problems with, 126 historical background of, 119 operating room arrangement for, 120, 121f patient counseling for, 120 patient evaluation of, 119–120 patient positioning for, 120 patient selection for, 119–120 postoperative care for, 124–126 preoperative preparation for, 120 preparation in operating room for, 120 surgical technique for, 122–124 canal skin replacement and, 124, 125f closure and, 124 dead skin removal and, 122, 123f ear canal enlargement and, 122–124, 123f fascia placement and, 123f, 124 fascia removal and, 122 packing preparation for, 124 postauricular exposure and, 122 transmeatal incisions and, 121f, 122 tympanic membrane remnant deepithelialization and, 124 vascular strip replacement and, 124, 125f tympanic membrane lateralization and, 125f, 126 packing after, 8–9, 10f patient positioning for, 7 for petrositis, 189, 189f plastic extrusion and, 225 plastic sheeting for, 223 preoperative counseling for, 223 preoperative evaluation for, 223 risks of, 128 second look, following canal wall reconstruction tympanomastoidectomy, 178 staging of, 221–223 of canal wall down procedures, 223–225 controversy regarding, 225 mucosal disease factors and, 221–222 mucous membrane indications for, 223, 224f ossicular chain factors and, 222 residual cholesteatoma and, 222 timing of second stage and, 222–223 tympanic membrane collapse following, 221 undersurface graft technique for, post auricular approach for, 149–160 adenoidal hypertrophy and adenoidectomy and, 150 allergy and, 150 anesthesia for, 153, 153f cartilage graft tympanoplasty and, 157–159, 158f complications of, 151 disease eradication and, 154 eustachian tubal tests before, 151 facial nerve monitoring and, 152 fascia harvest for, 154, 155f fundamental preoperative principles of, 150 graft placement and, 154–156, 156f–157f grafting techniques and exposure for, 152 hemostasis and, 152 historical background of, 149–150 imaging before, 151 incisions for, 154, 155f infection control and, 151–152 informed consent for, 151 mastoidectomy and, 152 middle ear exposure for, 154, 155f–156f nasal or sinus condition and, 150 objectives of, 150 ossicular reconstruction and, 152 otorrhea and, 150–151 postoperative care for, 156–157
Tympanoplasty (Continued) predisposing conditions and, 150 preoperative preparation for, 150–151 rare disorders and, 150 results with, 157 surgical preparation for, 153, 153f surgical technique for, 153–156 tympanic membrane remnant preparation and, 154 undersurface graft technique for, transcanal approach for, 141–148 graft selection for, 141–142 patient selection for, 141 postoperative care for, 145–147 preoperative evaluation for, 141 preparation for, 142, 143f rationale for, 141 requirements for, 141 surgical evaluation for, 142 surgical technique for, 142–144 canal enlargement and speculum placement and, 142–144, 143f, 145f canal incisions and middle ear exposure and, 144, 145f external canal packing and, 144, 147f graft placement stabilization and, 144, 147f injection and, 142 middle ear preparation and, 144, 147f perforation placement and, 143f, 144, 145f perichondrial graft and, 144, 146f wound closure following, 9–10 Tympanosclerosis, stapes surgery and, 313 Tympanostomy tubes, 73–92 biofilms and, 88 complications of, 79–80, 86–87 dislodgement of, 86 draping for, 80 follow-up for, 84 for idiopathic hemotympanum, 80 insertion of, 81–84 instrumentation for, 80–81 for otitis media acute, 77 chronic, with effusion, 78–79 limitations of, 79 for patulous eustachian tube, 97 pitfalls with, 84–86 postoperative care for, 84 preoperative patient counseling for, 79–80 preoperative preparation for, 80 results with, 86 risks of, 79–80 surgical site preparation and draping for, 80 Tympanotomy, posterior, for mastoiditis, 188
U Upper lid entropion correction in facial paralysis, 748f, 750 Upper lid reanimation procedures in facial paralysis, 735–737 enhanced palpebral spring implantation for, 738–740, 739f, 741f gold weight implantation for, 740–742, 741f, 742t palpebral spring implantation for, 736f, 737–738 silicone rod prosthesis implantation for, 742, 743f Urban rotary suction-dissector for vestibular neurectomy, 13, 14f Utricle, failure to find, in transcanal labyrinthectomy, 488
Index
V Vagus nerve anatomy of, 443, 444f compression of, microvascular decompression for, 531–532 deficits of, rehabilitation after neurotologic skull base surgery for, 571–577, 573f–574f, 576f jugular foramen schwannomas and, 552 monitoring of, 782 in glomus tumor surgery, 555–556 paresis/palsy of with jugular foramen schwannomas, 552 with limited temporal bone resection, 40 preservation of, in glomus tumor surgery, 556 schwannomas of, in neurofibromatosis 2, 695 Valacyclovir for Ramsay Hunt syndrome, 339 Vascular anomalies, stapes surgery and, 313 Vascular clips for vestibular neurectomy, 13–14, 14f Vascular compressive syndromes, microvascular decompression for. See Microvascular decompression. Vascular injury with chronic otitis media surgery, 238–241 Vascular lesions, neurotologic surgery for, 801–803 middle fossa approaches for, 803 of Kawase, 803 with petrous carotid exposure, 803 transpetrosal approaches for, 801–803, 802f retrolabyrinthine, 802 transcochlear, 803 translabyrinthine, 802–803 Vascular tumors of petrous bone and cerebellopontine angle, 807–813 cerebral revascularization and bonnet bypass and, 812–813, 812f–813f preoperative embolization and, 807–811 glomus tumors and, 807–809, 810f–811f hemangioblastomas and, 809–811 hemangiomas and, 811 Velopharyngeal insufficiency, adenoidectomy and, 87 Ventilation tubes. See Tympanostomy tubes. Vertebrobasilar arteries anatomy of, 800–801 aneurysms of, 805 Vertigo. See also Dizziness. differential diagnosis of, 456, 467–468, 510 after endolymphatic sac surgery, 418 intraoperative, stapedectomy and, 259–261 positional, microvascular decompression for, 530–531 complications of, 531 operative technique for, 531 patient selection for, 528f, 530–531 results with, 531 positional, benign paroxysmal, 467–476 pathophysiology of, 468 posterior semicircular canal occlusion for patient counseling and, 468–470, 469f patient selection for, 470 preoperative evaluation for, 471 results with, 471–473 surgical technique for, 471, 472f after stapes surgery, 317 transcanal labyrinthectomy to control, 483 in vestibular system disorders, 455–456 Vestibular disorders, treatment of, 457–458. See also Vestibular neurectomy, translabyrinthine. decision making and, 457–458
Vestibular disorders, treatment of (Continued) transmastoid labyrinthectomy vs. translabyrinthine vestibular nerve section for, 458 Vestibular evoked myogenic potential(s) in Meniere’s disease, 415 for translabyrinthine vestibular neurectomy, 458–459 Vestibular evoked myogenic potential thresholds in superior canal dehiscence syndrome, 508, 509f Vestibular nerve resection, middle fossa approach for, for acoustic tumors. See Acoustic tumors, vestibular nerve resection for. Vestibular nerve resection, translabyrinthine, transmastoid labyrinthectomy vs., 458 Vestibular neurectomy endoscopic assisted, 448 instrumentation for, 13, 14f middle cranial fossa, 429–440 alternatives to, 439–440 anesthesia for, 430 complications of, 439 draping for, 430 dressing for, 435 instrumentation for, 430, 431f intraoperative monitoring for, 430 patient positioning for, 430, 430f patient selection for, 429 pitfalls in, 437–439 postoperative care for, 435 preoperative counseling and, 430 preoperative evaluation for, 429–430 preoperative medication for, 430 results with, 439 surgical site preparation for, 430 surgical technique for, 430–435, 431f bony exenteration and blue lining of superior semicircular canal and, 433, 436f craniotomy and, 431–433, 432f, 434f dural elevation and, 433, 434f exposure of meatal plane and arcuate eminence and, 433, 434f internal auditory canal exposure and, 433–435, 436f medial cranial fossa retractor introduction and, 433, 436f middle cranial fossa floor repair and, 435, 438f skin incision and, 430, 432f temporal muscle flap and, 430–431, 432f vestibular neurectomy and, 435, 437f–438f wound closure and, 435, 438f tips for, 437–439 retrolabyrinthine, 439, 441–454, 445f combined with retrosigmoid approach, 448, 448f complications of of, 451–452, 451t efficacy of, 449–451, 449t–450t historical background of, 441–442 neuroanatomy and, 442–443, 443f–444f postoperative course and, 448–449 preoperative evaluation for, 442 retrosigmoid (suboccipital), 439, 441–454, 447f combined with retrolabyrinthine approach, 448, 448f complications of, 451–452, 451t efficacy of, 449–451, 449t–450t historical background of, 441–442 neuroanatomy and, 442–443, 443f–444f postoperative course and, 448–449 preoperative evaluation for, 442
831
Vestibular neurectomy (Continued) translabyrinthine, 455–466 anesthesia for, 459 complications of, 464 diagnostic considerations for, 455–456 draping for, 459 electronystagmography and, 458 follow-up for, 461–464 functional auditory system assessment and, 459 high-resolution computed tomography and, 459 historical background of, 455 literature overview for, 464 magnetic resonance imaging and, 459 patient positioning for, 459 platform posturography and, 458–459 postoperative care for, 461–464 results with, 464 surgical procedure for, 459–461, 459f–460f, 462f–463f transmastoid labyrinthectomy vs., 458 vestibular evoked magnetic potentials and, 458 Vestibular neuronitis, 456 Vestibular schwannomas in neurofibromatosis 2. See Neurofibromatosis 2, vestibular schwannomas in. stereotactic irradiation for, 698 Vestibular symptoms. See also Dizziness; Vertigo. following perilymphatic fistula surgery, 330 Vestibular-evoked myogenic potential, 456 Vestibule, footplate relationship with, 311–312 Vestibulectomy, subtotal, with intramuscular streptomycin, 494–496 indications for, 494–495 pretreatment evaluation and patient counseling for, 495 treatment technique for, 495–496 Vestibulocochlear nerve anatomy of, 445f, 446 compression of, microvascular decompression for, 530 direct recording from, monitoring hearing using, 780–781 monitoring of, during neurologic procedures, 11–12 Vibrant Soundbridge, 167, 386t, 389–391, 390f–391f, 391t Vibratory transducer in ossicular reconstruction, 167 Vibroplasty, 170 Visible-spectrum lasers, 281, 282t, 284, 285f
W Wound infection with cochlear implantation, 378 with retrolabyrinthine vestibular neurectomy, 451 after stapes surgery, 317 with surgery for facial nerve trauma, 360
Y Yeasts, acute otitis media due to, 109
Z Zygomas, 186
Self-Assessment Questions Chapter 1 NONE
Chapter 2 1. E xostoses of the external auditory canal would be expected most frequently in the following: a. Patient with chronic otitis externa b. College swimmer from a warm climate c. 23-year-old thrice-weekly surfer from Florida d. 43-year-old thrice-weekly surfer from Oregon Answer: d.
2. R apid and complete healing of the skin of the external auditory canal is facilitated by a. Performance of surgery prior to severe skin attenuation from chronic otitis externa b. Preservation of skin of the external auditory canal c. Local anesthesia d. a and b e. b and c Answer: d.
3. C onductive hearing impairment occurs with large exostoses of the external auditory canal when a. Debris prevents normal tympanic membrane vibration b. Narrowing of the external auditory canal reaches 5 mm c. Narrowing of the external auditory canal reaches 2 mm d. All of the above e. a and c Answer: e.
4.
edial third stenosis may result in M a. Weeping and chronic otitis externa b. Conductive hearing impairment c. Formation of dense scar lateral and medial to the tympanic membrane d. a and b e. All of the above
Answer: d.
5. O steonecrosis of the tympanic bone presents with bone exposure in the external auditory canal and may be associated with a. Radiation history to the temporal bone b. Lupus vasculitis c. Diabetes d. a and b e. All of the above Answer: e.
Chapter 3 1.
he most common malignancy of the temporal bone is T a. Adenoid cystic carcinoma b. Pleomorphic adenoma c. Warthin’s tumor d. Squamous carcinoma e. Ceruminoma
Answer: d.� Most temporal bone carcinoma originates in the
skin of the external auditory canal. 2. T he mechanism for salivary gland tumor involvement of the temporal bone is a. Direct extension b. Hematogenous metastases c. Ectopic salivary rests within the temporal bone d. Squamous metaplasia e. Absence of the fissures of Santorini e
e
SELF-ASSESSMENT QUESTIONS
Answer: a.� The proximity of the partoid and fissures of San-
torini make direct extension the most common mode for salivary gland tumors to involve the temporal bone. 3. A fungating mass of the external auditory canal is biopsied in the office. The pathologist reports acute and chronic inflammation. The next step in management is a. Consult infectious disease b. Image and rebiopsy c. Lateral temporal bone resection d. Subtotal temporal bone resection e. Radiation therapy Answer: b.� Necrotic carcinomas frequently are confused with
inflammatory changes. The otologist should be suspicious and plan deeper biopsies once imaging proves that vital structures are not jeopardized by the planned biopsy. 4. T he best study to distinguish postoperative scar from tumor recurrence is a. PET-CT imaging b. PET imaging c. CT d. MRI with contrast e. MRI-CT fusion imaging Answer: a.� PET-CT has proven increasingly useful at dem-
onstrating the expected increased metabolic activity of carcinoma in relation to the osseous anatomy. 5. T he most significant prediction of postoperative survival in temporal bone carcinoma is a. Preoperative chemotherapy and radiation b. Clear surgical margins c. Postoperative chemotherapy and radiation d. Patient age e. Extensiveness of resection Answer: b.� Clear surgical margins reflect complete oncologi-
cally sound removal of the lesion. While adjuvant treatments definitely have a role, a clear surgical margin is the best predictor of survival.
Chapter 4 1. T he overall prevalence of primary temporal bone malignancy is about a. Six cases per thousand b. Six cases per hundred thousand c. Six cases per million d. Six cases per hundred million Answer: c.
dvanced malignancies of the temporal bone are best 2. A treated with a. Surgery only b. Radiation only c. Surgery and radiation d. Chemotherapy Answer: c.
3. R adiologic assessment of temporal bone malignancies may include a. CAT scan b. MRI c. Cerebral angiography d. All of the above Answer: d.
ong (>8 cm) facial nerve defects should be grafted 4. L with a. The greater auricular nerve b. The cervical cutaneous nerves c. The cranial nerve XI d. The sural nerve Answer: d.
5. T he most common site for temporal bone treatment failure is a. Local b. Locoregional c. Regional d. Distant site Answer: a.
Chapter 5 1. What are the goals of atresia surgery? a. Perform ossicular reconstruction with the patient’s own ossicular chain instead of using a prosthesis b. Create a patent, skin-lined external auditory canal and achieve a postoperative air-bone gap within 20 to 30 dB c. Perform surgery before age 6 in both unilateral and bilateral cases d. Perform hearing restoration surgery and microtia repair in a one-stage procedure Answer: b.
2. W hat are the four anatomic parameters, as seen radiographically, crucial for atresia surgical operability? a. The status of the inner ear, the degree of pneumatization of the mastoid, the course of the facial
SELF-ASSESSMENT QUESTIONS
nerve, and the presence of the oval window and stapes footplate b. Thickness and form of the atretic bone, soft tissue contribution to the atresia, size and status of the middle ear cavity, and the presence or absence of congenital cholesteatoma c. The status of the inner ear, the degree of pneumatization of the mastoid, thickness and form of the atretic bone, and the presence or absence of congenital cholesteatoma d. Soft tissue contribution to the atresia, size and status of the middle ear cavity, the course of the facial nerve, and the presence of the oval window and stapes footplate
e
c. Noise-induced sensorineural hearing loss. Minimize drilling on the ossicular chain when dissecting it away from the atretic bone. d. Ossicular reconstruction prosthesis extrusion. Cover the prosthesis with cartilage prior to grafting the new drum.
Answer: b.
Chapter 6 None
Answer: a.
Chapter 7
3. W hat is the most common cause of inoperability in congenital aural atresia? a. Absence of the oval window b. A facial nerve overlying the oval window c. Poor mastoid pneumatization d. Unilateral atresia
1. W hich of the following does not contribute to the closing of the lumen of the Eustachian tube? a. Coiling properties of the cartilaginous skeleton b. Active contraction of the associated muscles c. Pressure of tissues neighboring the lumen d. All of the above
Answer: c.
Answer: b.
4. W hat are the main postoperative complications of atresiaplasty? a. Lateralization of the tympanic membrane, large meatus, sensorineural hearing loss, and facial nerve palsy b. Ossicular reconstruction prosthesis extrusion, otorrhea, stenosis of the meatus, and facial nerve palsy c. Sensorineural hearing loss, tympanic membrane perforation, and ingrowth of grafted skin into external auditory canal Merocel wicks d. Lateralization of the tympanic membrane, stenosis of the meatus, sensorineural hearing loss, and facial nerve palsy
2. W hich are the most common symptoms of the patulous Eustachian tube? a. Autophony made worse in recumbent position b. Autophony and pressure-induced dizziness c. Aural fullness and autophony d. Symptoms associated with retraction and atelectasis of the tympanic membrane
Answer: d.
5. W hat is the most common delayed cause of a poor hearing outcome in atresia surgery and what can be done to prevent it? a. Meatal stenosis. Use one-piece split-thickness skin grafts to cover all exposed bone, and Silastic sheets and Merocel wicks in the external auditory canal for 2 weeks. b. Tympanic membrane lateralization. Use Silastic sheets to line the external auditory canal and Merocel wicks for 2 weeks, Gelfim disks, and tabs in the fascia graft to maintain the tympanic membrane graft in position.
Answer: c.
3. W hich of the following is not considered to participate in the pathogenesis of the patulous Eustachian tube? a. Cigarette smoking b. Weight loss c. Changes in estrogen levels d. Scarring from previous nasopharyngeal procedures Answer: d.
4.
iagnosis of patulous Eustachian tube is certain with D a. Appropriate findings on axial CT b. Appropriate findings on sonotubometry c. Presence of typical symptoms d. Observation of lateral and medical excursions of the drum
Answer: d.
e 5.
SELF-ASSESSMENT QUESTIONS
reatment options do not currently include: T a. Discontinuing topical antidecongestants b. Instillation of local irritants c. Augmentation of the nasopharyngeal Eustachian tube orifice d. Injection of Botox to the tensor veli palatine
Answer: d.
Chapter 8 1. W hich of the following is not normally found as part of the pathophysiology of chronic suppurative otitis media? a. Capillary proliferation b. Rarefying osteitis c. Multinucleated giant cells d. New bone formation e. Granulation tissue Answer: c.� All of the other four items are specifically men-
tioned as routinely present in individuals with chronic suppurative otitis media. Large multinucleated giant cells are classic for tuberculosis and/or the presence of foreign bodies but do not appear regularly in patients with chronic suppurative otitis media. 2. W hich of the following statements is true of acute otitis media through a tympanostomy tube or perforation, but not true of acute otitis media with an intact tympanic membrane? a. It occurs frequently without fever. b. Drainage from the area is commonly the only sign. c. Significant pain is uncommon. d. Pseudomonas is the causative organism in a significant number of infections. e. S treptococcus pneumoniae is the causative organism in a number of infections. Answer: d.� Pseudomonas is only very rarely recovered from
children with otitis media and an intact tympanic membrane. However, it makes up a significant percentage of primary isolates in children with acute otitis media through a tympanostomy tube, especially in older children. Answers a–c and e are true of acute otitis media through a tympanostomy tube but are not true of acute otitis media with an intact tympanic membrane. 3. W hich of the following statements is true about commercially available antibiotic ear drops? a. The antibiotic concentration in commercially available ear drops is hundreds of times higher than the concentration that can be achieved in middle ear fluid following systemic administration.
b. Antibiotic–steroid suspensions are too viscous to pass through a tympanostomy tube. c. The use of topical quinolone drops is directly responsible for the increase in quinolone-resistant isolates in children with ear disease. d. The currently available evidence suggests that the use of a steroid compromises healing of the tympanic membrane. e. There is no evidence that resolution of acute otitis media through a tympanostomy tube or tympanic membrane perforation is enhanced when a steroid is added to a topical antibiotic drop.
Answer: a.� The others are all false. Although concerns have
been raised that a steroid might delay wound healing, there is no evidence to support the assertion that outcomes are worse when steroids are used. There is now good evidence that a steroid improves both clinical cure and facilitates the eradication of bacteria when it is included in an antibiotic drop. 4. W hich of the following statements is true of fat graft tympanoplasty? a. Donor fat is usually harvested from the abdomen. b. Fat graft tympanoplasties do not require rimming of the perforation. c. The literature justifies performing this procedure bilaterally and simultaneously. d. It is an important technique for perforations involving 75% of the tympanic membrane or less. e. It is appropriate for any patient with a tympanic membrane perforation and a hearing loss of 50 dB or less. Answer: c.� Mitchell, et al assert that the risks and complica-
tions from fat graft tympanic myringoplasty are sufficiently low so that they can be performed bilaterally and simultaneously. The procedure is not likely to be successful in tympanic membrane perforations that involve more than 25% of the drum. The procedure should probably not be performed in individuals with hearing losses over 25–30 dB because the middle ear ossicles cannot be explored or repaired. Fat grafts are usually taken from earlobes. The margin of the perforation needs to be de-epithilialized as usual. 5. W ith respect to paper patching of tympanic membrane perforations, which of the following statements is true? a. The procedure usually needs to be performed in an outpatient surgery setting. b. It is an excellent technique for large perforations. c. Closure may require repeated applications of the paper patch. d. If successful, the perforation closes in a few days. e. “Rimming” of the perforation is not necessary. Answer: c.� Multiple repetitions of the procedure are almost
always required to get complete closure. Paper patching is often
SELF-ASSESSMENT QUESTIONS
done in the office. The perforation must be de-epithelialized. A solution, usually trichloroacetic acid, is used to de-epitheliaze the perforation. Several months are usually required for successful closure.
e
Chapter 10
1. O ne of these complications is not seen with underlay tympanoplasty grafting: a. Reperforation b. Blunting c. Infection d. Myringitis of the graft
1. A ll of the following are true concerning cartilage tympanoplasty except a. Hearing results are worse than with conventional materials. b. Toothed forceps should be avoided during manipulation. c. Although softening occurs, long-term graft survival is the norm. d. Cartilage becomes more brittle with age and more subject to fracture. e. The postoperative tympanogram is often type B despite no effusion present.
Answer: b.
Answer: a.� Hearing results are worse with cartilage. Several
2. O ne of these complications is not seen with overlay grafting: a. Reperforation b. Blunting c. Lateralization d. Short healing time
studies comparing cartilage, fascia, and perichondrium have been published that show no significant difference in hearing results between the materials. Cartilage can be somewhat brittle, especially in the older patient, so toothed forceps should not be used to manipulate the graft. Softening can occur, and the postoperative tympanogram is frequently a small volume B, due to the noncompliant nature of the graft, even with normal hearing and no effusion.
Chapter 9
Answer: d.
3. Which one of the following statements is correct? a. Blunting is a complication of underlay tympanoplasty. b. The rate of take for overlay tympanoplasty is greater than for underlay tympanoplasty. c. Lateralization is a complication of underlay tympanoplasty. d. Myringitis is not seen after underlay tympanoplasty. Answer: b.
4. A dvantages of the outer surface technique include all the following except a. It provides excellent exposure. b. It allows for removal of as much remnant as necessary. c. The take rate is lower than overlay tympanoplasty. d. This technique can be used in all cases
2. C artilage should be considered as a graft material in the following situations: a. The atelectatic ear b. Cholesteatoma c. A perforation anterior to the annulus d. A draining perforation e. All of the above Answer: e.� Cartilage can be used as a graft material in any
ear considered to be at high risk for failure with traditional techniques using temporalis fascia or perichondrium. Included in this would be the high-risk perforation, the atelectatic ear, and cholesteatoma.The high-risk perforation comprises a revision surgery, a perforation anterior to the annulus, a perforation draining at the time of surgery, a perforation larger than 50%, or a bilateral perforation.
5. D isadvantages of the outer surface technique include all of the following except a. Healing problems b. Lateralization of the graft c. Blunting in the anterior sulcus d. Absence of epithelial cysts
3. T he following are true statements concerning the perichondrium/cartilage island flap except a. A strip of cartilage 1–2 mm in width is removed to facilitate the malleus. b. The graft is placed as an underlay graft, medial to the malleus. c. Tragal cartilage is most suitable for total tympanic membrane reconstruction. d. One should expect a cosmetic defect after harvest in the tragal area. e. The graft is placed so that the perichondrium side is out, toward the canal.
Answer: d.
Answer: d.� One should expect a cosmetic defect after harvest
Answer: c.
in the tragal area. If the cartilage is harvested correctly, leaving
e
SELF-ASSESSMENT QUESTIONS
a strip of cartilage in the dome of the tragus, no cosmetic deformity is seen. The rest of the statements are true concerning the island flap technique described here.
Chapter 11
4. W hen using the palisade technique for cartilage tympanoplasty, the following statements are considered true, except for a. Cartilage can be harvested from the tragus or cymba. b. The cartilage is cut into rectangular strips. c. The technique is favored when ossiculoplasty is performed in a malleus-present situation. d. The technique does not require one large piece of cartilage. e. The technique is particularly suited for cholesteatoma surgery.
1. T he authors identified four advantages associated with transcanal medial graft placement tympanoplasty. Which one of the following was not considered an advantage? a. It is quicker and more direct. b. It results in less surgical trauma and reduces the likelihood of healing problems such as adhesions, narrowing, and stenosis of the ear canal. c. It is technically easier to perform. d. It results in less postoperative discomfort for the patient. e. There is reduced risk of tympanic membrane blunting and burying of squamous epithelium beneath the graft.
Answer: b.� The cartilage is cut into rectangular strips. One
Answer: c.� Although transcanal medial graft placement may
major difference between the palisade technique described here and that described by Heermann, et al, is that, instead of placing rectangular strips of cartilage side to side, an attempt is made to cut one major piece of cartilage in a semilunar fashion. A second semilunar piece is placed between this first piece and the canal wall to reconstruct the scutum precisely, and any spaces that result between this cartilage and the canal wall or scutum are filled in with small slivers of cartilage.The remaining statements are all true with regard to the palisade technique.
be technically difficult in some circumstances, the advantages gained with this approach are felt by the authors to outweigh the difficulties associated with the procedure.
5. W hich of the following statements apply to the postoperative period following cartilage tympanoplasty? a. Eardrum intubation, if necessary, will likely require a trip to the operating room. b. A tympanogram may be unreliable. c. A second-look surgery may be required in cases of cholesteatoma removal. d. Persistent effusion is the most significant complication. e. All of the above Answer: e.� While the tympanic membrane remains relatively
insensate after cartilage reconstruction, it is often necessary to take the patient to the operating room for eardrum intubation as tube placement can be difficult. Impedance tympanometry is unreliable after cartilage tympanoplasty and will generally yield a low-volume, type B tympanogram, despite normal hearing, due to the noncompliant nature of the cartilage. One serious disadvantage of using cartilage for reconstruction in cholesteatoma surgery is that it creates an opaque tympanic membrane posteriorly, which could potentially hide residual disease. If disruption of the cholesteatoma sac occurs, consider the advisability of performing a second-look surgery. The most significant pitfall seen in the postoperative period is persistent effusion with conductive hearing loss, requiring intubation of the reconstructed eardrum. This is seen in about 7% to 10% of cases and can be problematic in cases in which the entire tympanic membrane is reconstructed with the cartilage/perichondrium island flap.
2. I dentify which one of the following is not a contraindication for transcanal tympanoplasty surgery. a. A narrow ear canal that will only accommodate up to a 4.5-mm speculum b. An elderly and debilitated patient c. Involvement of only hearing ear and patient not a satisfactory surgical risk d. A perforation with persistent and active drainage e. Very questionable Eustachian tubal function Answer: d.� Drainage at the time of surgery is not believed
by the authors to be an absolute contraindication to surgery. If drainage cannot be controlled preoperatively, this constitutes an indication for surgery. 3. W hich one of the following techniques is not advocated by the authors? a. The head is secured with tape. b. Preoperative antibiotics are usually administered. c. The ear, auricle, and surrounding skin are cleansed with povidone-iodine solution. d. The hair is secured with liquid spray adhesive and usually is not shaved. e. The head is extended with the chin up. Answer: b.� Studies have shown that no significant benefits
result from the use of preoperative antibiotics. 4. W hich one of the following tympanoplasty techniques is advocated by the authors? a. The entire bony ear canal skin is removed to allow satisfactory exposure. b. If there is a prominent anterior bony canal wall “hump,” a postauricular incision should be used.
SELF-ASSESSMENT QUESTIONS
c. The graft is placed lateral to the malleus handle. d. The ear speculum should be supported and manipulated by the surgeon’s fingers during the procedure. e. Following graft placement, the ear canal is filled with moist Gelfoam.
Answer: c.� The authors believe that graft placement lateral
to the malleus handle avoids interference with ossicular chain reconstruction and lessens the likelihood of postoperative middle ear adhesions. 5. W hich one of the following postoperative methods is advocated by the authors? a. Cotton is placed in the ear canal and usually left in place for 3 weeks. b. A mastoid dressing is usually placed at the end of the procedure. c. If drainage occurs postoperatively, the patient is advised to come into the office immediately and the ear canal is cleaned using the microscope. d. The patient is advised to keep the operated ear dry for 3 weeks, at which time the ear is examined in the office with a microscope. e. Active autoinflation of the ear is begun 3 days postoperatively. Answer: d.� It is important to keep the ear dry and to avoid
inflating the ear for 3 weeks. Cotton should be placed in the meatus, changed as necessary for drainage, and discontinued after drainage has ceased. If drainage worsens, the patient is started on antibiotic ear drops.
Chapter 12 1. W hich of the following conditions is not associated with a predisposition for chronic ear disease? a. Chronic tonsillitis b. Adenoid hypertrophy c. Allergic rhinitis d. Chronic sinusitis Answer: a.
2. F or which of the following are temporal bone CT scans without contrast used for preoperative assessment in chronic ear surgery? a. Sinonasal disease b. Epidural abscess c. Brain hernia d. Inner ear fistula Answer: d.
3. A ntimicrobial prophylaxis is not indicated in chronic ear surgery in which of the following situations?
e
a. Violation of the dura with or without cerebrospinal fluid leak b. Simple perforation c. Labyrinthine fistula d. Presence of indwelling devices such as a cochlear implant
Answer: b.
4. T he undersurface grafting technique greatly reduces which of the following postoperative complications? a. Infection b. Anterior sulcus blunting c. Graft failure d. Recurrent cholesteatoma Answer: b.
5. T he vascular strip is defined as the canal skin between a. The annulus and bony cartilaginous junction b. The 12 o’clock and 6 o’clock positions in the ear canal c. The malleus and the anulus d. The tympanomastoid and tympanosquamous sutures Answer: d.
Chapter 13 1. W hich of the following factors may increase the risk of extrusion following ossicular reconstruction with a titanium prosthesis? a. Age of patient b. Inadequate length of prosthesis c. Absence of cartilage between prosthesis and tympanic membrane d. Hydroxyapatite platform on prosthesis e. Absence of malleus handle Answer: c.� Many allograft prostheses, including titanium
and Plastipore, require a disk of cartilage to be interposed between the platform and tympanic membrane to prevent extrusion. Hydroxyapatite has demonstrated excellent biocompatibility and the use of cartilage is not always required with this allograft material; however, it is probably prudent to use cartilage even in these cases. 2. W hich of the following reconstruction techniques can be used in the setting of a mobile malleus and stapes superstructure and a foreshortened incus? a. Total ossicular reconstruction prosthesis b. Malleus-to-oval-window prosthesis c. Stapes piston
e
SELF-ASSESSMENT QUESTIONS
d. Bone cement e. Removal of incus and placement of total ossicular replacement prosthesis
Answer: d.� When the gap between a foreshortened incus and
a mobile stapes superstructure is small, bone cement can be used to reconstruct the ossicular chain and restore continuity between the incus and stapes. In situations where the gap is large, an incus bridge prosthesis can be used in conjunction with bone cement.
3. S uccessful ossicular reconstruction is most likely to be achieved when a. The Eustachian tube function is normal. b. The canal wall remains intact. c. Adequate tension is placed on the prosthesis. d. A mobile stapes superstructure exists. e. All of the above are correct. Answer: e.� Each of the above factors contributes to improved
a. Graft the tympanic membrane and use a partial ossicular replacement prosthesis to reconstruct the ossicular chain. b. Pack the ear with antibiotic-soaked packing and stage the operation. c. Place a ventilation tube in the newly reconstructed tympanic membrane and then use a partial ossicular replacement prosthesis. d. Place Gelfilm or Sialastic sheeting over the promontory and reconstruct at a later time. e. Stage the operation and once the tympanic membrane has healed reconstruct the ossicular chain with a malleus-to-footplate prosthesis.
Answer: d.� A totally denuded promontory may result in
severe scarring between it and the tympanic membrane. In cases where the mucosa is missing or severely diseased, it is best to cover the promontory with Gelfilm or Sialastic sheeting, and reconstruct the ossicular chain during a second-look procedure.
success in ossicular reconstruction. 4. I n your preoperative discussions with your patients undergoing ossicular chain reconstruction, what are the anticipated hearing results for partial ossicular replacement prostheses (PORPs) and total ossicular replacement prostheses (TORPs)? a. Complete closure of air-bone gap with PORPS and 5 dB air-bone gap with TORPs b. PORPs with a closure of air-bone gap to within 5 dB in 90% of cases and to within 10 dB in 90% of TORPs c. Closure of air-bone gap to within 15 dB with PORPs and to within 25 dB with TORPs in majority of patients d. PORPs with closure of air-bone gap to within 5 dB and TORPs to within 10 dB in the majority of patients e. An equivalent air-bone gap if the stapes footplate is mobile
Chapter 14
Answer: c.� Although we all would like to completely close the air-bone gap in our patients, it is important to have realistic expectations for yourself and your patients. The results with PORPs are generally better than with TORPs. In patients undergoing ossicular chain reconstruction with PORPs, two thirds should close the air-bone gap to within 15 dB, and two thirds of patients with TORPs should be within 25 dB.
2. I n the canal-wall-reconstruction tympanomastoidectomy technique, bone pate collection should be performed a. After tympanic membrane grafting and replacement of the posterior canal wall b. After a complete mastoidectomy has been performed but prior to making bony canal cuts with a microsagittal saw c. Prior to any entry into mastoid cells d. Following removal of the bony posterior canal wall
5. I n a patient undergoing cholesteatoma removal using an intact canal wall approach, the surgeon had to remove the mucosa off the promontory and also remove the involved incus. The malleus was mobile and the stapes was intact and mobile. What are the surgeon’s best options for the management of this ear?
1. A ll of the following are thought to contribute to the higher rate of recidivism seen with canal-wall-up tympanomastoidectomy (vs. canal-wall-down technique) except a. Tympanic membrane re-retraction with recurrent antral or epitympanic cholesteatoma formation b. Suboptimal exposure to the attic, antrum, and middle ear due to the presence of the posterior canal wall leading to persistent disease c. The presence of nitrogen-resorbing mucosa lining the mastoid cavity postoperatively d. Compromised exposure of the sinus tympani due to the presence of the posterior canal wall Answer: d.
Answer: c.
3. I n the canal-wall-reconstruction tympanomastoidectomy technique, slices of calvarial bone are placed
SELF-ASSESSMENT QUESTIONS
a. In the attic and sinus tympani b. In the sinus tympani and facial recess c. In the facial recess and attic d. Only in the attic
Answer: c.
4. T he bony posterior canal wall is cut in the following fashion during the canal-wall-reconstruction tympanomastoidectomy technique: a. One beveled cut superiorly and one beveled cut inferiorly b. Two cuts superiorly making a right angle and one beveled cut inferiorly c. Two cuts inferiorly making a right angle and one beveled cut superiorly d. One beveled cut superiorly and one straight cut inferiorly
e
with a right-sided unilateral profound sensorineural hearing loss. You suspect she has what? a. A right-sided meningoencephalocele through the mastoid tegmen b. A right-sided Mondini malformation c. A right-sided enlarged vestibular aqueduct syndrome d. A patent Hyrtl’s fissure
Answer: b.� She had a severe Mondini malformation with
dehiscence of the medial wall of the vestibule with a large cerebrospinal fluid (CSF) connection between the internal auditory canal and the vestibule. She also had an incomplete stapes footplate allowing the CSF to bulge a CSF containing thin mucosal cyst filling the oval window niche.
5. C ontraindications to canal wall reconstruction tympanomastoidectomy include a. Mastoid cholesteatosis b. Sinus tympani involvement c. Facial paralysis d. Tegmen defect with meningoencephalic herniation
3. A 40-year-old woman presents with a seizure, from which she fully recovered. She had never had seizures before. Her complete neurological examination was normal. The only abnormality found on full examination was an infected cholesteatoma in her right ear, which had drained intermittently for several years. You are concerned about what? a. Brain abscess b. Subdural abscess c. Meningitis d. Otitic hydrocephalus
Answer: a.
Answer: a.� She had a large right temporal lobe brain
Answer: b.
abscess.
Chapter 15 1. A 35-year-old man with no previous ear disease presents with persistent left ear conductive hearing loss for 1 month following an upper respiratory infection concomitant with left ear pain, all of which resolved quickly with oral antibiotics. He did ultimately mention that the left ear was still very mildly painful. On examination, the left ear had seromucinous middle ear effusion; otherwise, the history and physical examination was normal. What would you do? a. Follow him for another month expecting full recovery. b. Give him another round of oral antibiotic treatment. c. Have him self-inflate the ear to speed aeration of the middle ear. d. Order a temporal bone CT because you are concerned he might have masked mastoiditis. Answer: d.� He had mastoiditis with fairly extensive bone
destruction. 2. A 13-year-old girl presents with a history of three episodes of meningitis associated with episodes of rightsided acute suppurative otitis media. She was born
4. A 60-year-old woman presents with a spontaneous conductive hearing loss in the left ear that had been present for 3 months. She had been treated with antibiotics with no resolution. She had no previous ear problems. On examination, the left middle ear is filled with amber serous-appearing fluid. After a myringotomy and tube in the office, her ear developed a copious thin discharge continuously so that she had to, at times, place a cotton ball in the meatus and change it several times a day. You investigate for what? a. Chronic suppurative otitis media b. Cerebrospinal fluid leak c. Chronic serous otitis media d. Fungal external otitis Answer: b.� She had a cerebrospinal fluid leak from a menin-
goencephalocele through the tegmen of the mastoid. 5. A 20-year-old man presents with a draining right ear and pain. When questioned about the pain, he tells you that the pain is in his right ear and behind his right eye. On examination, he has a perforation in the right drum and mucopurulent discharge. The rest of the history and examination is normal. You suspect what?
e10
SELF-ASSESSMENT QUESTIONS
a. Mastoiditis b. Otitis external associated with the drainage c. Brain abscess d. Petrositis
a. 3–6 months b. 6–9 months c. 9–18 months d. 18–24 months e. 24–36 months
Answer: d.� He had petrositis that was discovered by CT. Answer: c.� These are guidelines established by Sheehy.
Chapter 16
Chapter 17
5.
NONE
Answer: c.� This thickness is commonly used for staging. The
NONE
Chapter 18 1. T he indications for staging tympanoplasty include which of the following? a. Mucosal disease b. Stapes fixation c. Concern of residual cholesteatoma d. All of the above Answer: d.� All are well-recognized indications.
2. F actors that have been blamed for postoperative collapse of the tympanic membrane include which of the following? a. Multiple surgeries b. Eustachian tube dysfunction c. Recurrent infections d. Middle ear mucosal disease e. b and d
Thick” Silastic sheeting has which dimension? “ a. .005 inch b. .010 inch c. .040 inch d. .1 inch e. .2 inch
.005-inch thickness is too easily displaced by scar tissue for use with severe mucosal disease.
Chapter 19 1. H ow is a cholesteatoma from a pars flaccida retraction pocket most likely to spread? a. Anteriorly, in the lateral mallear space, into the anterior epitympanum b. Posteriorly, lateral to the body of the incus, to the aditus ad antrum c. Inferiorly via the posterior pouch of von Tröltsch, lateral to the incus, and into the mesotympanum. d. Posteriorly, medial to the body of the incus, to the aditus ad antrum. e. Inferiorly via the posterior pouch of von Tröltsch, medial to the incus, and into the mesotympanum. Answer: b.� Anterior spread is relatively rare. Inferior spread
is lateral to the incus but is less common than the posterior route.
Answer: e.� Classically, Eustachian tube dysfunction has
3. F or patients that have middle ear cholesteatoma found at the first stage, what percentage will have residual disease in the middle ear at the second stage? a. 5% b. 15% c. 33% d. 66% e. 90%
2. F or which situation is it appropriate to attempt repair of a cholesteatoma fistula? a. A 3-mm lateral semicircular canal fistula during a second-stage surgery, with normal hearing in the contralateral ear b. A 1-mm lateral semicircular canal fistula in an only hearing ear. c. A 1.5-mm lateral semicircular canal fistula, with normal hearing in the contralateral ear. d. A 0.5-mm cochlear fistula with normal hearing in the contralateral ear. e. A 1.5-mm fistula involving the lateral semicircular canal and vestibule.
Answer: c.� This is an empiric observation.
Answer: c.� Repair should not be attempted with fistulas
4. W hat is the time interval between stages when the indication is possible residual cholesteatoma?
greater than 2 mm, in an only hearing ear, or when it involves any labyrinthine structure other than the lateral semicircular canal.
been blamed for retraction and collapse of the grafted tympanic membrane. With the increased use of staging, many surgeons have recognized the role of middle ear mucosal disease in the pathogenesis of tympanic membrane collapse.
SELF-ASSESSMENT QUESTIONS
e11
3. W hich would be the best approach for an acquired cholesteatoma that involves the geniculate ganglion, but without a sensorineural hearing loss? a. Mastoidectomy combined with a middle fossa craniotomy b. Radical mastoidectomy c. Bondy modified radical mastoidectomy d. Canal-wall-up mastoidectomy with malleus head removal to expose the anterior epitympanum e. Translabyrinthine
Chapter 20
Answer: a.� A middle fossa craniotomy will give the expo-
bone and encephalocele, and 59% occurred as a complication of previous mastoid surgery.
sure needed for cholesteatoma removal. Mastoidectomy alone is unlikely to give adequate exposure of the geniculate ganglion. A radical mastoidectomy would not improve the exposure or facilitate postoperative care compared with other canal-wall-down mastoidectomy techniques. A translabyrinthine approach should be considered if there is no salvageable hearing. 4. A pproximately 50% of the mastoid segment of the facial nerve is disrupted during removal of a large cholesteatoma. What should be done? a. Remove all cholesteatoma and plan nerve repair at a second stage. b. Remove all cholesteatoma and evaluate facial function following surgery. c. Decompress the nerve, remove the injured section, and place an interposition graft. d. Reroute the facial nerve and anastomose it. e. Decompress the nerve and put a small nerve graft into the defect. Answer: c.� Repair is best done at the time of injury. The
injured segment should be removed if it is greater than 30% of the total diameter of the nerve. Rerouting requires much more extensive drilling, as well as transection of the greater superficial petrosal nerve. 5. T here is profuse bleeding from a 2-mm tear in the jugular bulb. How should you proceed? a. Firm intraluminal packing with Surgicel b. A muscle plug reinforced with fibrin glue c. Ligation of the sigmoid sinus and internal jugular vein d. Cover the injury with gelatin foam e. Surgicel packing between the bulb and the overlaying bone Answer: e.� Packing between the bulb and the overlaying
bone usually works with a defect this size. Firm intraluminal packing may injure the nerves in the jugular foramen. The muscle plug and gelatin foam are inadequate. Ligation is rarely necessary.
1. E ncephaloceles occur most commonly with what other condition? a. Chronic otitis media b. Head trauma c. Idiopathic d. Previous mastoid surgery e. Arachnoid granulations Answer: d.� In 1989, Iurato reviewed 139 cases of temporal
2. F or an encephalocele to develop which event or events must occur? a. Dural injury b. Bone dehiscence c. Dural injury and bone dehiscence d. Chronic otitis media e. Head trauma Answer: c.� For an encephalocele to develop, two preexisting
conditions are necessary and must occur: bone dehiscence and dural injury. 3. T he most common presentation of cerebrospinal fluid (CSF) leak in the temporal bone is a. Middle ear effusion b. Otorrhea c. Meningitis d. Seizure e. Mass in the ear canal Answer: a.� Middle ear effusion and hearing loss are the most
common presentation of CSF leak and encephalocele in the temporal bone. Subsequent myringotomy may result in otorrhea. Meningitis, seizures, and mass in the ear canal are all uncommon presenting conditions. 4. C onfirmation of spinal fluid in the ear is best made with which test? a. Glucose content b. Protein content c. β-2 transferrin d. CT e. MRI Answer: c.� Identification of β-2 transferrin in a specimen is
over 90% sensitive for identifying spinal fluid. CT and MRI are more valuable for locating possible sites of spinal fluid leakage. 5. The most common site of encephalocele formation is: a. Middle fossa b. Posterior fossa
e12
SELF-ASSESSMENT QUESTIONS
c. Ear canal d. Middle ear e. Oval window
Answer: a.
Chapter 21 1. P atients with far advanced otosclerosis are not candidates for a stapedectomy. a. True b. False Answer: b.
2. T he advantages of stapes surgery using local anesthesia includes all of the following except a. More bleeding during surgery b. Immediate feedback in terms of vertigo when manipulating the footplate c. Typically shorter recovery time Answer: a.
3. I f too much posterosuperior canal wall bone is removed, adhesions to the incus or a retraction pocket can form. a. True b. False Answer: a.
a. Tympanic membrane perforation b. Acute otitis externa c. Active Ménière’s disease d. Conductive hearing loss of at least 15 dB confirmed by tuning fork e. Only hearing ear
Answer: d.� Perforation, infection, or active Ménière’s disease
present at stapedectomy can result in a postoperative hearing loss. 2. P ossible risks and complications of stapedectomy surgery include a. Worsened hearing b. Tinnitus c. Dizziness d. Taste disturbance e. All of the above Answer: e.� All are well-recognized postoperative complica-
tions. 3. T he tympanomeatal flap should be elevated so that which two landmarks can be visualized? a. Neck of the malleus and round window b. Half of the diameter of the fallopian canal and stapedial tendon c. Cochlear form process and Jacobson’s nerve d. Eustachian tube orifice and a long process of incus e. Subiculum and ponticulum Answer: a.� Exposure of these two landmarks will allow the
4. T he limits of exposure in stapes surgery include all of the following except a. Round window niche b. Pyramidal eminence c. Upper edge of tympanic fallopian canal d. Malleus handle Answer: c.
5. A surgeon might consider stopping the procedure if he or she encounters a: a. Fixed malleus b. Tear in the tympanic membrane c. Solid or obliterated footplate d. Prolapsed facial nerve covering the footplate Answer: d.
tympanomeatal flap to be folded forward, providing adequate access to the middle ear and facilitating crimping of the prosthesis. 4. F or a right-handed surgeon operating on a right ear, what two landmarks should be visualized at the conclusion of curetting? a. Neck of the malleus and round window b. Half of the diameter of the tympanic fallopian canal and stapedial tendon c. Cochlear forum process and Jacobson’s nerve d. Eustachian tube orifice and a long process of incus e. Subiculum and ponticulum Answer: b.� This exposure will allow visualization of the foot-
Chapter 22
plate and introduction of instruments. On a left ear, a righthanded surgeon will require more exposure for instrument access and the entire diameter of the fallopian canal should be exposed.
1. R elative contraindications for stapedectomy include all of the following except
5. W hich of the following statements are true about reparative granuloma?
SELF-ASSESSMENT QUESTIONS
a. A reparative granuloma is a controversial entity whose modern existence is in doubt. b. It may be due to retained ethylene oxide in gelfoam wire prostheses. c. The attributed symptomatology may be explained by serous labyrinthitis. d. The presentation can be explained by expected postoperative findings. e. All of the above
Answer: e.
Chapter 23 1. F or the partial stapedectomy technique, how much of the footplate should be removed? a. All of it b. 25% c. Only that part which comes out easily d. 75% Answer: c.
2. T o properly size a piston-cup prosthesis, how is the correct length determined? a. The measurement from the surface of the footplate to the lateral surface of the incus is used. b. The measurement from the surface of the footplate to the medial surface of the incus is used. c. A prosthesis 4 mm in length is used universally, as it protrudes 0.2 to .03 mm into the vestibule. d. One half of a mm is added to the measurement from the surface of the footplate to the lateral surface of the incus. Answer: c.
e13
5. F or the treatment of juvenile otosclerosis, the prosthesis length should be sized a. The same as adult b. One half mm shorter than adult c. One half mm longer than adult d. None of the above Answer: a.
Chapter 24 1. L aser energy power (watts) per area of spot size (cm2) × exposure time (sec) is called a. Power density b. Fluence c. Intensity d. Photon Answer: b.
2. T he most common indication of failure of primary stapedectomy requiring revision surgery is a. Dizziness and vertigo b. Sensorineural hearing loss c. Recurrence of conductive hearing loss d. Facial nerve palsy Answer: c.
3. T he most common intraoperative finding associated with revision stapedectomy is a. Perilymph fistula b. Prolapsed facial nerve c. Displaced prosthesis d. Fibrosis of oval window tissue Answer: c.
3. W hat is the minimum diameter of the opening into the vestibule required for the piston-cup prosthesis? a. Larger than 0.8 mm b. Larger than 2.0 mm c. Half of the footplate d. The entire footplate Answer: a.
4. The vein graft size and oval window position should be a. 15 × 15 mm and adventitial side toward the vestibule b. 5 × 5 mm and endothelial side toward the vestibule c. 4 × 8 mm and adventitial side toward the vestibule d. 15 × 15 mm and endothelial side toward the vestibule Answer: c.
4. T he major advantage of lasers in revision stapes surgery is a. The ability to define the margins of the oval window b. Less bleeding and better visualization of oval window c. Thinning of existing membrane to more thoroughly identify the status of the oval window (footplate fragments, bone regrowth, etc.) d. All of the above Answer: d.
5. T he cardinal rule of laser surgery of the oval window, regardless of the type of laser, is a. To never fire the laser directly into an open vestibule b. Only perform revision surgery once
e14
SELF-ASSESSMENT QUESTIONS
c. That the inner ear is immune to damage from laser energy d. That there is nothing more to learn about laser revision stapes surgery
c. There is no difference in audiometric results in stapedotomy as compared to partial stapedectomy, except in overclosure. d. The vein graft should be placed with adventitia up. e. Place the prosthesis with two-handed technique.
Answer: a. Answer: d.� The vein graft should be placed adventitia down,
Chapter 25 1. W hich of the following statements regarding intraoperative audiometry is correct? a. It is not useful in cases done under local anesthesia. b. It requires testing by an audiologist. c. The frequency with the greatest air-bone gap is usually tested. d. It does not assist in placement of the prosthesis in revision stapedectomy. e. It will not prevent opening of the oval window membrane in revisions. Answer: c.� Intraoperative audiometry is done under local
anesthesia, does not require an audiologist, uses the greatest air-bone gap frequency, and assists in decisions regarding the footplate and prosthesis placement. 2. W hich of the following statements is correct regarding promontory drilling? a. It cannot be done with patient under local anesthesia. b. It will cause sensorineural hearing loss. c. It will yield poorer hearing results in stapedectomy. d. It should be done in a superior to inferior direction. e. It should be done in a medial to lateral direction. Answer: e.� Promontory drilling begins at the footplate prom-
ontory junction and sweeps laterally. All other statements are false. 3. A vein tissue graft is not useful in which of the following intraoperative situations? a. Sealing of the oval window after footplate removal b. Centering of the stapes prosthesis over the central area of footplate removal c. Preventing perilymph leak d. Recreating a mobile oval window membrane e. Repair of a necrosed distal long process Answer: e.� Vein tissue graft will not repair a necrosed incus.
All other statements are correct. 4. In routine stapedectomy, which is not correct? a. Always use a speculum holder. b. Check both the malleus and incus visually and by palpation.
intima up. All other statements are correct. 5. T he most appropriate candidate for revision stapedectomy is a patient with a history of: a. Sensorineural hearing loss without an oval window tissue graft b. Sensorineural hearing loss with an oval window tissue graft c. Malleus/incus fixation that is total d. Stapedectomy in which the hearing initially improved then declined e. Significant erosion of the incus Answer: d.� The most likely candidate is one whose hearing
initially improved, then declined and who presents with a conductive hearing loss.
Chapter 26 1. P resence of acoustic reflexes on a preoperative audiogram should prompt the operating surgeon to a. Cancel the surgery. b. Order a CT scan. c. Consider superior semicircular canal dehiscence as a cause of the conductive hearing impairment. d. Do all of the above. Answer: d.
2. W ith gentle pressure on a stapedotomy prosthesis after placement, patient dizziness should prompt the surgeon to consider that a. The prosthesis may be too long. b. The otosclerotic inner ear syndrome may be operative and may require treatment with calcium fluoride. c. A fistula exists in the round window. d. Placing a vein seal would be helpful. Answer: a.
3. The law of additive inadequacy describes a. Reduction in surgeon ability with age b. Unavoidable deterioration in surgical equipment over time c. Surgical inadequacies that add up to poor results d. Additive reduction in outcome with multiple surgeons Answer: c.
SELF-ASSESSMENT QUESTIONS
4. A dequate surgical results can be achieved in all of the following except a. Superior semicircular canal dehiscence b. X-linked progressive mixed deafness c. Revision stapedotomy with obliterative otosclerosis d. a and b e. a, b, and c Answer: d.
5. S tudies have shown the following may produce sensorineural hearing loss following stapedotomy: a. Viral infection b. Blood in the inner ear c. Gusher d. Otosclerosis e. All of the above Answer: e.
Chapter 27 1. E xamination findings suggestive of perilymphatic fistula (PLF) include a. Rotatory nystagmus with fatigue and direction reversal b. Nystagmus induced by loud noise or pressure to the eardrum c. Abnormal electrocochleography (ECoG) with elevated summating potential/action potential (SP/AP) ratio (>0.5) d. Positional nystagmus and a positive turning test Answer: d.� Nystagmus induced by noise or positive or nega-
tive pressure with intact tympanic membrane is indicative of a superior semicircular canal fistula. Rotary nystagmus with fatigue and direction reversal is classic for benign positional paroxysmal vertigo. Abnormally elevated SP/AP ratios have been noted in patients with PLF; however, they cannot be distinguished from Ménière’s patients who also have abnormal ECoG findings. 2. M iddle ear exploration for PLF is recommended in which of the below cases: a. Mild hearing loss and/or disequilibrium following recent trauma b. Sudden onset of hearing loss with no disequilibrium c. Fluctuating hearing loss, episodic vertigo d. Disequilibrium, worsening with coughing or straining, conductive hearing loss Answer: a.� Conservative, nonsurgical treatment is advo-
cated in patients seen in the first 7 days after development of symptoms. However, if the patient’s hearing fails to recover
e15
in 10 days of bed rest, surgical intervention should be carried out in the next 4 days. Sudden sensorineural hearing loss without history of prior antecedent traumatic event should be treated medically. Fluctuating hearing loss and episodic vertigo is likely Ménière’s disease. Conductive hearing loss with disequilibrium worsened by Valsalva-type maneuvers is likely superior semicircular canal dehiscence. 3. C lear fluid is encountered in the middle ear space during exploration of PLF. The most appropriate next step would be to a. Send for β-2 transferrin assay b. Suction the fluid c. Have the patient perform a Valsalva maneuver d. Patch both the oval and round windows Answer: b.� It is important to remove any fluid present on
initial exposure of the middle air space as this is likely diffusion of local anesthetic into the middle ear space. Once the fluid is suctioned, having the patient perform a Valsalva maneuver and putting pressure over the incudostapedial joint can help determine the presence of a leak. β-2 transferrin assay has not been shown to be helpful for intraoperative determination of perilymph. There is lack of general consensus as to whether both the oval and round windows should be patched without visual confirmation of PLF. 4. A natomical features of the round window that increase the likelihood of PLF are a. Deep round window niche b. Large overhanging promontory c. The inability to visualize round window transtympanically d. A 45-degree angle of the round window membrane to the promontory Answer: d.� Tears of the round window membrane occur most
frequently when its position is 45 degrees to the promontory and there is little or no overhanging promontory, allowing direct visualization of the round window membrane transtympanically. 5.
ecurrent PLF should be suspected if R a. Disequilibrium persists 2 weeks after surgery. b. Whirling episodic vertigo persists or develops. c. Hearing levels fail to improve. d. A fat graft was used to patch the fistula and 6 weeks of healing have taken place.
Answer: d.� Multiple authors have recommended avoiding use of fat as a graft material as it has been noted to have an increased failure rate. Six weeks should be allowed for the wound to completely heal before contemplating reexploration.Whirling episodic vertigo is typical for endolymphatic hydrops and not PLF. Hearing improvement has been reported in less than 50% of patients who have undergone PLF repair.
e16
SELF-ASSESSMENT QUESTIONS
Chapter 28 1. B ell’s palsy is characterized by all of the following pathophysiologic findings except a. HSV-1 DNA in facial perineural fluid at the time of nerve decompression in some patients b. Lymphocytic infiltration, neural edema, and myelin degeneration most prominent at the labyrinthine segment c. Constriction of the facial nerve at the meatal foramen with distal ischemia d. Degeneration of the cisternal segment of the facial nerve e. A spectrum of injury to the facial nerve from neuropraxia to neurontomesis Answer: d.
2. W hich of the following clinical findings is most consistent with the diagnosis of Bell’s palsy? a. Progression to complete paralysis over 1–7 days b. A palpable parotid mass c. Recent hearing loss associated with the facial paralysis d. Other cranial neuropathies associated with the onset of the facial paralysis e. Continued complete facial paralysis 3–6 months after the initial onset Answer: a.
3. W hen is electrodiagnostic testing of the facial nerve indicated in the setting of Bell’s palsy? a. Immediately after the onset of complete paralysis b. Two weeks after the onset of complete paralysis c. Worsening paralysis after appropriate medical therapy d. Seventy-two hours after the onset of complete facial paralysis e. Immediately once eye closure becomes incomplete Answer: d.
4. A ppropriate surgical management of Bell’s palsy involves all of the following except a. Complete paralysis of less than 14 days’ duration b. A transmastoid approach c. Greater than 90% duration on electroneurography performed at least 72 hours after the onset of complete paralysis d. Lack of voluntary motor unit potentials on facial electromyogram e. Bony decompression of the labyrinthine, geniculate, and proximal tympanic segments Answer: b.
5. A ll of the following findings are consistent with the diagnosis of Ramsay-Hunt syndrome in a patient with acute facial paralysis except a. Periauricular vesicular eruptions b. Otalgia c. Predominant involvement of the frontal and orbital branches of the facial nerve d. Other cranial neuropathies e. Skip regions of facial nerve involvement Answer: c.
Chapter 29 1. W hat segment of the facial nerve is most commonly damaged as a result of temporal bone trauma? a. Intracranial b. Meatal c. Labyrinthine/perigeniculate d. Tympanic e. Mastoid Answer: c.� The labyrinthine/perigeniculate portion of the
facial nerve is involved in up to 90% of cases. 2. A ll of the following are most commonly associated with longitudinal temporal bone fractures except a. Frontal/occipital impact b. Conductive hearing loss c. Ossicular damage d. Bloody otorrhea e. Fracture through the foramen ovale Answer: a.� Frontal/occipital impact is most commonly asso-
ciated with transverse temporal bone fractures. 3. W hich of the following findings would favor surgical exploration rather than observation in a patient with traumatic facial nerve paralysis? a. Immediate onset of complete facial nerve paralysis after penetrating injury b. Delayed onset of facial nerve paralysis with greater than 95% degeneration on electroneuronography (ENoG) c. Bony fragment impingement of the facial nerve on CT scan d. None of the above e. All of the above Answer: e.� All of the findings would favor surgical explora-
tion. In a and c, the chance of a complete transection of the facial nerve is high. Progressive deterioration of facial nerve function with a greater than 95% degeneration on ENoG are also associated with poor outcomes.
SELF-ASSESSMENT QUESTIONS
4. W hat is the best approach for surgical decompression of the facial nerve in a patient who has complete hearing loss on the paralyzed side? a. Transmastoid b. Translabyrinthine c. Middle cranial fossa d. Suboccipital e. None of the above
e17
a. Transcanal b. Transmastoid/extended facial recess c. Translabyrinthine d. Suboccipital e. Transcondylar
Answer: c.� The translabyrinthine approach provides expo-
sure of the facial nerve from the brain stem to the stylomastoid foramen.
Answer: b.� In a patient without residual hearing, the pre-
ferred approach is translabyrinthine because it provides adequate exposure of the perigeniculate area without many of the risks of serious complications associated with the middle cranial fossa approach. 5. W hat is the best result (House-Brackmann grade) in terms of facial nerve function after exploration and primary anastamosis? a. I b. II c. III d. IV e. V Answer: c.� The best surgical result that can be attained after
primary anastamosis is House-Brackmann III.
Chapter 30 1. T he most common presenting symptom for a facial nerve tumor patient is a. Facial paralysis b. Facial twitching c. Hearing loss d. Pulsatile tinnitus e. Otorrhea Answer: c.� Facial nerve tumors can cause a sensorineural
hearing loss if they develop in the internal auditory canal or a conductive hearing loss if they are present in the middle ear or mastoid. 2. P roven approaches to the management of facial nerve neuromas include a. Observation b. Surgical resection and nerve repair c. Decompression d. Radiation therapy e. a, b, and c Answer: e.� There are few reports of radiation treatment for
facial nerve tumors and no assessments of long-term efficacy. 3. T he entire length of the intracranial/intratemporal nerve can be accessed through which approach?
4. T he expected facial nerve result after facial nerve repair is a. I–II b. II–III c. III–IV d. IV–V e. V–VI Answer: c.� The intracranial and intratemporal facial nerve
is monofasicular. After facial nerve anasatomosis, synkenesis is an expected result. 5.
acial nerve repair is best accomplished F a. With as many sutures as possible b. Without a nerve graft c. In a tensionless fashion d. In a second-stage procedure e. With a laser welding technique
Answer: c.� It has been shown that a tensionless anastomosis
is important for success.
Chapter 31 1. W hich of the following is considered to be an absolute contraindication to cochlear implantation? a. Duration of deafness greater than 30 years b. Auditory neuropathy c. Enlarged vestibular aqueduct d. Michel aplasia e. All of the above Answer: d.
2. P atients with cochlear implants are at higher risk for developing meningitis. The CDC has recommended which of the following vaccinations to reduce the incidence of meningitis in cochlear implant recipients? a. Pneumovax for all patients greater than 2 years of age b. Prevnar vaccine for all patients less than 5 years of age c. Hib vaccine for all patients less than 5 years of age
e18
SELF-ASSESSMENT QUESTIONS
d. Meningococcal vaccination for all patients greater than 5 years of age e. a, b, and c f. All of the above
a. Patient age b. Handling of soft tissue c. Cooling during drilling d. Gender e. Patient hygiene
Answer: e. Answer: c.� Excessive heat during drilling will cause trauma
3. B ilateral cochlear implantation is experimental and should only be performed as part of a clinical trial. a. True b. False Answer: b.
4. A ssuming audiometric criteria for cochlear implantation are met, which of the following patients would be the poorest candidate for cochlear implantation? a. A 14-year-old male with congenital, profound sensorineural hearing loss who uses sign language as his primary mode of communication b. An 8-year-old female with progressive hearing loss following meningitis c. An 82-year-old female with gradually progressive hearing loss d. A 14-month-old male with congenital sensorineural hearing loss e. A 15-year-old female with congenital, progressive sensorineural hearing loss with normal speech and language Answer: a.
to the osteocytes and could result in soft tissue healing instead of osseointegration. 2. T o establish a reaction-free skin penetration, it is important to a. Remove all periosteum b. Be careful not to remove any of the periosteum c. Use an extra long abutment d. Make the skin at the penetration site thin and hairless e. Use firm packing under healing cap Answer: d.� Thin skin at the implant site will reduce the rela-
tive mobility between implant and skin, which is the key to a lasting, reaction-free skin penetration. The important daily cleaning is facilitated if no hair follicles are present. 3. I mportant for success with an ear-level, bone-anchored hearing aid is when the a. Air-bone gap is less than 20 dB b. Air-bone gap is larger than 20 dB c. Cochlea reserve is better than 35 dB d. Cochlea reserve is worse than 60 dB e. Bilateral cochlea deafness
5. W hich of the following techniques should be used during surgery for cochlear implantation? a. Skin and periosteal incisions should overlap by at least 1 cm b. The internal device should be placed as close as possible to the postauricular crease c. The cochleostomy should not be packed with soft tissue, as this increases the risk of developing meningitis d. The cochleostomy should be placed between the round window and oval window to ensure insertion occurs in the scala vestibuli
Answer: c.� Air-bone gap is of no importance at any level.
Answer: a.
Answer: a.� Even in large defects, healing will take place with
At 60 dB cochlea reserve, a body-worn aid is probably needed. 4. T he recommended way to handle postoperative skin necroses is a. Conservative handling with mild ointment b. Revision surgery as soon as possible with a free graft c. Removal of skin-penetrating abutment d. Removal of coupling and bone implant e. Long-term intravenous antibiotics a conservative attitude, even if it could take some time.
Chapter 32 NONE
Chapter 33 1. W hat is the most important factor to establish osseointegration?
5. W hich of the following is the best method to use to handle damage to the sigmoid sinus during boneanchored hearing aid surgery? a. Plug with bone wax, close wound, and wait 6 months for next trial b. Plug with periosteum and find a new implant site c. Perform a mastoidectomy to identify the damaged area
SELF-ASSESSMENT QUESTIONS
d. Use a p-PTFE membrane to stop bleeding e. Enlarge the defect, fill with muscle tissue, and use fibrin glue
Answer: b.� Sigmoid sinus is a low-pressure system and dam-
age of the wall is of minor importance. It is easily stopped with some soft tissue, and a new implant site close by can often be found.
Chapter 34 1. W hat is the pathophysiologic correlate of Ménière’s disease? a. Scarring and fibrosis of the periductal and saccular tissues within the vestibular aqueduct and opercular regions b. Contraction of the mastoid cavity with narrowing of Trautman’s triangle by anteromedial displacement of the sigmoid sinus c. Dilation of the membranous endolymphatic spaces including the scala media, saccule, and endolymphatic system d. Shortening and narrowing of the vestibular aqueduct e. Episodic vertigo lasting more than 30 minutes associated with nausea and vomiting, fluctuating or deteriorating hearing often in the low frequencies initially, and tinnitus and/or aural pressure. Answer: c.� Endolymphatic hydrops, as documented by tem-
poral bone histopathologic analysis of Ménière’s patients, shows dilatation of the membranous endolymphatic spaces, particularly the scala media and saccule. Answers a, b, and d are common findings within temporal bones of Ménière’s patients that may contribute to the symptomatology and development of the pathophysiologic state of hydrops. Answer e is the 1995 Committee on Hearing and Equilibrium symptom diagnosis for possible Ménière’s disease. 2. The functional role of the endolymphatic sac is to a. Maintain homeostasis of the endolymph with a graded concentration of sodium and potassium b. Provide immunologic support for the inner ear c. Remove debris and infectious waste by phagocytes d. Provide hormonal maintenance of inner ear fluid balance by secretion of Saccin and other cell-signaling effectors e. All of the above Answer: e.� Physiologic evidence that the endolymphatic
sac participates in inner ear immunity and control of fluid dynamics by humoral and local cellular activity are helping to identify the direct role that the endolymphatic system plays on hearing and balance function.
e19
3. A ccording to the 1995 Committee on Hearing and Equilibrium guidelines, the appropriate minimal follow-up time prior to reporting data in Ménière’s disease is a. 6 months b. 12 months c. 18 months d. 24 months e. 36 months Answer: d.� 1995 guidelines of the American Academy of
Otolaryngology—Head and Neck Surgery standardize the diagnosis and reporting of Ménière’s disease to minimize confounding effects of the natural history of the disease. They also promote standard benchmarks with which to compare data between competing treatment arms and studies. According to these guidelines, observations over the 6 months preceding intervention should be compared with observations within the 18 to 24 months following treatment. For comparison of audiometric data, the worst audiogram in the 6 months prior to treatment should be compared to the poorest audiogram following treatment. 4. E ndolymphatic sac procedures accumulatively offer class A and B results in what percentage of patients? a. Less than 50% b. 51%–65% c. 66%–80% d. 81%–90% e. Greater than 90% Answer: d.� According to the 1995 Committee on Hearing
and Equilibrium guidelines, class A responses to treatment of Ménière’s disease completely eliminate vertigo when comparing the 6 months prior to treatment with the 6-month period 18 to 24 months following intervention. Class B results occur when an intervention reduces the frequency of definite vertigo spells to less than 40% of the pretreatment levels. Looking at endolymphatic sac interventions between 1985 and 1995 with adequate follow-up, Grant andWelling showed a weighted effect of 86% class A/B results. A similar analysis between 1995 and 1997 shows endolymphatic surgery to be 84% effective. 5. W hat is the most common intervention for medically refractory Ménière’s disease in patients with serviceable hearing? a. Intratympanic gentamicin b. Tube insertion with application of micropressure, Meniette system c. Labyrinthectomy d. Vestibular nerve section e. Endolymphatic sac procedure Answer: e.� Endolymphatic sac surgery is the preferred
primary surgical treatment of Ménière’s disease by American Otologic and Neurotologic Society members in patients refractory to medical management with serviceable hearing,
e20
SELF-ASSESSMENT QUESTIONS
b ilateral disease, and in an only hearing ear. The total numbers of endolymphatic procedures is greater than numbers for other interventions. Intratympanic gentamicin therapy evolved throughout the 1990s and is used as first-line surgical treatment in patients with hearing loss and unilateral disease. Surgical labyrinthectomy numbers are falling, and numbers of vestibular nerve sections remain at a steady level.
Chapter 35
Chapter 36
1. The meatal plane is defined by a. The blue line of the posterior canal and a 60-degree angle b. The blue line of the superior canal and a 60-degree angle c. The blue line of the lateral canal and a 60-degree angle d. Bisecting the angle defined by the superior canal and the greater superficial petrosal nerve e. The flat bone 1 cm medial to the geniculate ganglion
1. W hich of the following is the most important component of the preoperative evaluation when considering vestibular neurectomy? a. Clinical history and physical examination b. CT scan of the temporal bone c. Caloric testing results d. MRI e. Audiometric findings
Answer: b.
2. A n absolute contraindication to middle fossa vestibular nerve section is a. An only hearing ear b. Bilateral Ménière’s disease c. Incapacitating vertigo d. Diuretic allergy e. Age less than 60 years Answer: a.
3. E xposure of the tegmen tympani facilitates orientation by a. Identification of the malleus and incus b. Identification of the lateral canal c. Identification of the superior canal d. Identification of the posterior canal e. Identification of the stapedius muscle
a. Less than 5% b. Greater than 20% c. Greater than 30% d. Greater than 40% e. Greater than 50%
Answer: a.
Answer: a.� All of the other components of the preopera-
tive workup may be helpful, but clinical history and physical examination are the most important components when identifying the cause of dizziness prior to considering any surgical treatment for relief of vertigo. 2. W hich patient is best suited for treatment of vertigo with a vestibular neurectomy? a. A patient with vestibular neuritis b. A patient with benign paroxysmal positional vertigo c. A patient with bilateral Ménière’s disease d. A patient with migrainous vertigo e. A patient with unilateral Ménière’s disease Answer: e.� Patients with unilateral Ménière’s disease have
the best response to selective vestibular neurectomy. There is no surgical treatment of migrainous vertigo and, if surgical treatment is undertaken for benign paroxysmal positional vertigo, it is most likely to be canal occlusion or singular neurectomy. Vestibular neuritis is much more unlikely to be treated successfully with vestibular neurectomy. Bilateral Ménière’s disease is a relative contraindication to vestibular neurectomy.
Answer: a.
4. C utting rather than avulsing the vestibular nerve is preferred because of possible a. Injury of the cochlear nerve b. Cerebrospinal fluid leak c. Increased dural injury d. Increased facial nerve injury e. Less effective vertigo control Answer: a.
5. Middle fossa vestibular nerve section hearing loss risk is
3. W hich of the following statements best describes the course of the facial nerve from the brain stem to the internal auditory canal? a. The facial nerve leaves the brain stem inferior and slightly anterior to the eighth cranial nerve and rotates in an anterior and superior direction so that it resides anterior to the superior vestibular nerve and superior to the cochlear nerve in the internal auditory canal. b. The facial nerve leaves the brain stem anterosuperior to the eighth cranial nerve and maintains that position into the internal auditory canal where
SELF-ASSESSMENT QUESTIONS
it is situated anterior to the cochlear nerve and inferior to the superior vestibular nerve. c. The facial nerve leaves the brain stem anterosuperior to the eighth cranial nerve and rotates until it is located anterior to the cochlear nerve and inferior to the superior vestibular nerve. d. The facial nerve leaves the brain stem posteroinferior to the eighth cranial nerve and rotates until it is anterior to the cochlear nerve and inferior to the superior vestibular nerve in the internal auditory canal. e. The facial nerve leaves the brain stem directly inferior to the eighth cranial nerve and rotates posteriorly such that it is located posterior to the cochlear nerve and inferior to the superior vestibular nerve in the internal auditory canal.
e21
Chapter 37 NONE
Chapter 38 1. A ll of the following are characteristic features of the nystagmus of a positive Dix-Hallpike maneuver except a. Predominantly torsional, side specific b. Latency of 3–5 seconds c. Fatigability with repeat testing d. Limited duration of 60–90 seconds e. Reversal when resuming the sitting position Answer: d.
Answer: a.
4. W hich of the following is the most commonly cited complication associated with the retrosigmoid vestibular neurectomy? a. Cerebrospinal fluid leak b. Wound infection c. Aseptic meningitis d. Subdural hematoma e. Chronic headache
2. B enign paroxysmal positional vertigo (BPPV) can result from, or has been correlated to, all of the following except a. Ménière’s disease b. Autoimmune inner ear disease c. Migraine d. Stapedectomy e. Labyrinthitis Answer: b.
Answer: e.� The retrosigmoid approach is associated with an
approximately 10% chance of headache that can be long-lasting and disabling. 5. W hich of the following statements best describes the success rate of posterior fossa vestibular neurectomy at eliminating vertigo and maintaining hearing? a. Posterior fossa vestibular neurectomy is greatly inferior to labyrinthectomy in both controlling vertigo and preserving hearing. b. Posterior fossa vestibular neurectomy is slightly inferior to labyrinthectomy in controlling vertigo and preserves hearing in approximately two thirds of patients. c. Posterior fossa vestibular neurectomy is slightly inferior to labyrinthectomy in preserving hearing while controlling vertigo in approximately two thirds of patients. d. Posterior fossa vestibular neurectomy is superior to labyrinthectomy at controlling vertigo, but not at preserving hearing. e. Posterior fossa vestibular neurectomy is superior to labyrinthectomy in both controlling vertigo and preserving hearing. Answer: b.� Labyrinthectomy is the gold standard for control of
vertigo but results in complete loss of hearing. In contrast, posterior fossa vestibular neurectomy controls vertigo in greater than 90% of patients and preserves hearing in 60% to 70% of patients.
3.
he most common form of BPPV results from T a. Anterior canal canalithiasis b. Lateral canal cupulolithiasis c. Lateral canal canalithiasis d. Posterior canal cupulolithiasis e. Posterior canal canalithiasis
Answer: e.
4. T he following statements about posterior canal occlusion surgery for BPPV are all correct except a. Free-floating particles can be seen in about 30% of cases. b. The risk of hearing loss is about 20%. c. Transient disequilibrium is seen in virtually all patients. d. Bilateral occlusions can be safely done in a sequential manner. e. Intraoperative auditory monitoring is not required. Answer: b.
5. T ransmastoid canal occlusion surgery has been used for all of the following except a. Otosclerosis b. Horizontal canal BPPV c. Acoustic neuroma excision
e22
SELF-ASSESSMENT QUESTIONS
d. Superior canal dehiscence e. Trigeminal schwannoma
Answer: a.
Chapter 39 1. Cochleosacculotomy creates a permanent defect in a. Osseous spiral lamina and cochlear duct and saccule b. Vestibular aqueduct and endolymphatic sac and saccule c. Cochlear aqueduct and endolymphatic sac and saccule d. Singular nerve e. Round window membrane Answer: a.
2. T he appropriate pick length for cochleosacculotomy without drilling the round window niche is a. 1 mm b. 2 mm c. 3 mm d. 4 mm e. Any pick length which can pierce the round window Answer: c.
3.
earing loss after cochleosacculotomy is H a. Almost zero b. Progressive for weeks c. Immediate d. Reversible with mannitol e. Only in the high frequencies
Answer: b.
4.
ertigo control with cochleosacculotomy is V a. 99% b. 50% c. 65% d. Improved with time e. Predictable by the presence of immediate nystagmus
Chapter 40 1. T he most important element of a transcanal labyrinthectomy to achieve mechanical destruction of the five vestibular end organs is a. Identification and removal of the saccule b. Identification and removal of the utricle c. Mechanical probing of the three semicircular canals d. Surgical widening of the oval window Answer: b.
2. T he most common cause of cerebrospinal fluid leakage during the course of a transcanal labyrinthectomy is a. Fracture of the crisbose bone at the medical aspect of the vestibule b. Widely patent cochlear aqueduct c. Anomalous spinal fluid communication along course of the facial nerve d. Widely patent communication of the spinal fluid with the perilymphatic space of the posterior semicircular canal Answer: a.
3. F acial nerve injury may complicate a transcanal labryinthectomy. The most common site of the injury is at the a. Descending segment b. Horizontal segment c. Geniculate ganglion d. Internal auditory canal Answer: b.
4. T he most difficult vestibular neuroephithelium to destroy by a transcanal labyrinthectomy is a. Maculae utriculi b. Macular sacculi c. Crista ampullaris of lateral semicircular canal d. Crista ampullaris of posterior semicircular canal Answer: d.
5. T he vertigo control rate in Ménière’s disease is affected by a strong placebo effect. a. True b. False
5. F ailure to obliterate the vestibule with a tissue seal following transcanal labyrinthectomy may a. Result in incomplete destruction of the vestibular neuroepithelium b. Result in delayed facial paresis c. Increase the chance of postoperative meningitis d. Result in tramautic neuroma of the superior or inferior vestibular nerve
Answer: a.
Answer: c.
Answer: c.
SELF-ASSESSMENT QUESTIONS
Chapter 41
NONE
Chapter 42 1. W hat is the most common serious complication of superior canal dehiscence (SCD) plugging surgery in the initial 24 hours after surgery? a. Aphasia b. Hematoma c. Seizure d. Meningitis
e23
c. Reassure the patient that the symptoms will improve with time d. Hold high-dose steroids and reassess in 4 hours
Answer: a.
Chapter 43 NONE
Chapter 44 NONE
Answer: b.
2. W hich factor puts patients at considerably greater risk of hearing loss after SCD plugging surgery? a. SCD greater than or equal to 4 mm in size b. Preoperative air-bone gap greater than 40 dB c. Prior stapes surgery d. Pulsatile tinnitus Answer: c.
3. W hich of the following is not a relative contraindication to SCD plugging? a. Prior middle ear surgery b. Prior successful SCD plugging on the contralateral side c. Age greater than 70 years d. Preoperative oscillopsia symptoms Answer: d.
4. W hich of the following symptoms are not likely to be improved after SCD plugging? a. Disorientation when rapidly rotating the head to look at something on the floor b. Disorientation and oscillopsia in response to loud noise c. Disturbing sound of one’s own voice d. Pulsatile tinnitus Answer: a.
5. T wo hours after SCD plugging surgery the patient has severe head pain on the side of the surgery and difficultly naming some common items. What is the appropriate next step in management? a. CT scan of the head without intravenous contrast b. Order a patient-controlled anesthesia machine so he or she does not have to ask the nurse for narcotics
Chapter 45 1. G radenigo’s syndrome usually includes all of the following except a. Retro-orbital pain b. Fourth cranial nerve palsy c. Otorrhea d. Hearing loss e. Sixth nerve palsy Answer: b.
2. T he most serious complication of the infracochlear approach to the petrous apex is a. Sensorineural hearing loss b. Facial nerve injury c. Tearing of the tympanic membrane d. Carotid artery injury e. Rupture of the jugular bulb Answer: d.
3. C haracteristic imaging of cholesterol granuloma includes a. Lack of pneumatization of contralateral petrous apex on CCT b. Hyperintense T1W and T2W images on MRI c. Hypointense T1W and hyperintense T2W images on MRI d. Hyperintense T1W and hypointense T2W images on MRI e. None of the above Answer: b.
4. I n the infralabyrinthine approach to the petrous apex, the structure at greatest risk is a. The carotid artery b. The jugular bulb
e24
SELF-ASSESSMENT QUESTIONS
c. The facial nerve d. The posterior semicircular canal e. The endolymphatic sac
Answer: b.
5. T he infracochlear approach to the petrous apex includes all of the following except a. Transection of the external auditory b. Elevation of the tympanomeatal flap c. Disarticulation of the incudostapedial joint d. Removal of tympanic bone e. Skeletonization of the carotid artery
c. Ascending pharyngeal artery d. Occipital artery e. Infratemporal fossa artery
Answer: c.� The ascending pharyngeal branch of the external
carotid artery is the most common arterial supply of glomus tumors. 5. P reoperative biopsy of glomus tumor is routinely used for surgical planning and patient counseling. a. True b. False Answer: b.� Biopsy risks uncontrollable hemorrhage or major
Answer: c.
arterial/venous injury.
Chapter 46
Chapter 47
1. T he distinction between a tympanomastoid glomus tympanicum versus a tympanomastoid glomus jugulare is a. Visible mass in a tympanicum b. Erosion of jugular bulb c. Erosion of carotid canal d. Middle ear and mastoid involvement e. Secretion of vasoactive chemicals
1. D uring resection of an anterior skull base malignancy, extensive removal of bone at the lateral aspect of the sphenoid is required. Abrupt bleeding is encountered that obliterates the entire surgical field. Vigorous packing with absorbable hemostatic agents successfully controls the bleeding. Postoperatively the patient complains of ipsilateral double vision on lateral gaze. No other focal neurologic deficit can be detected. The most likely cause of this deficit is a. Acute cerebral hemorrhage b. Edema in Dorello’s canal c. Overzealous packing of the cavernous sinus d. Carotid artery aneurysm e. Injury to cranial nerve V
Answer: b.� Jugulare tumors arise from the jugular bulb.
2.
ocal cord paralysis occurs soonest with V a. Glomus tympanicum b. Glomus jugulare c. Glomus vagale d. Jugular foramen schwannoma e. Jugular foramen meningioma
Answer: c.� Early onset of vocal cord paralysis is characteris-
tic of glomus vagale. 3.
reoperative embolization in glomus jugulare surgery P a. Reduces intraoperative blood loss b. Eliminates the need for jugular bulb resection c. Improves cranial nerve outcomes d. Alters periauricular incisions due to vascular compromise e. Has no proven intraoperative or postoperative effect
Answer: a.� Preoperative embolization significantly reduces
intraoperative blood loss. 4. T he most common arterial supply of glomus jugulare tumors is the a. Internal carotid artery b. External carotid artery
Answer: c.
2. A ppropriate studies for detection of aspiration in patients with lower cranial nerve deficits after skull base surgery include (choose all that apply) a. Bedside flexible endoscopic evaluation of swallowing (FEES) b. Subjective report from the patient c. Chest x-ray d. Modified barium swallow (MBS) e. Flexible fiberoptic laryngoscopy Answer: a, c, d, e.
3. V elopalatal insufficiency associated with lateral skull base surgery may be caused by all of the following except a. Tensor veli palatini paralysis b. Loss of cranial nerve X nodose ganglion fibers c. Nerve of Hering injury d. Palatopharyngeus muscle dysfunction Answer: c.
SELF-ASSESSMENT QUESTIONS
4. A patient undergoes lateral skull base tumor resection that includes sacrifice of cranial nerve VII and removal of tumor at foramen ovale. Acceptable rehabilitation of the cranial deficits would include (choose all that apply) a. Facial nerve cable graft b. Gore-tex or alloderm orbicularis oris sling c. Temporalis muscle sling d. Masseter muscle sling e. Temporal fossa implant Answer: a, b, e.
5. A 26-year-old sales professional undergoes routine resection of a glomus vagale tumor. She has previously undergone Silastic medialization at an outside institution but complains of a breathy voice and nasal speech. Swallowing evaluation and airway evaluations show decreased pharyngeal squeeze without aspiration, poor glottal closure with vocal cord atrophy, and paralysis of the left hemipalate. The best surgical option(s) for rehabilitation include (choose all that apply) a. Vocal cord augmentation with collagen injection b. Arytenoid adduction only c. Revision silastic medialization with arytenoid adduction d. Palatal adhesion e. Cricopharyngeal myotomy Answer: c, d.
Chapter 48 1. T he indications for a middle fossa approach for removing an acoustic tumor include a. Small tumor b. Good hearing c. Intracanalicular location d. Location mainly in the cerebellopontine angle e. a, b, and c Answer: e.� The exposure of the cerebellopontine angle is lim-
ited through the middle fossa approach. 2. A coustic tumors arising from which nerve have a higher incidence of hearing preservation? a. Superior vestibular nerve b. Inferior vestibular nerve c. Cochlear nerve d. Facial nerve Answer: a.� Tumors arising in the superior compartment of
the internal auditory canal tend to have less involvement of the cochlear nerve and cochlear blood supply.
e25
3. S earch of the landmarks that aid in the identification of the internal auditory canal include a. Arcuate eminence b. Greater superficial petrosal nerve c. Geniculate ganglion d. Superior semicircular canal e. All of the above Answer: e.� The arcuate eminence overlies the superior semi-
circular canal. Bisecting the angle between the superior canal and the greater superficial petrosal nerve locates the course of the internal auditory canal. The geniculate ganglion is at the lateral end of the internal auditory canal. 4.
he direction of tumor dissection should be T a. Lateral to medial b. Medial to lateral c. Anterior to posterior d. Inferior to superior e. All of the above
Answer: b.� Medial to lateral dissection prevents traction
injury to the facial and cochlear nerves at their exit points in the lateral internal auditory canal. 5.
he first middle fossa approach was reported by T a. William House b. Howard House c. John House d. Walter Dandy e. R. H. Parry
Answer: e.� Parry reported the use of the middle fossa
approach in the early part of the 20th century. It was refined and popularized by William House in the 1960s.
Chapter 49 1. S ome advantages the translabyrinthine approach has over the retrosigmoid approach for removal of acoustic neuromas include a. Extradural drilling that decreases the seeding of bone dust into the subarachnoid space b. Exposure of the entire facial nerve c. Less cerebellar retraction during tumor removal d. a and c e. All of the above Answer: e.� All drilling is complete prior to opening the dura.
In a retrosigmoid craniotomy, any internal auditory canal drilling must be performed after the cistern has been opened. By completing a mastoidectomy and opening the entire length of the internal auditory canal, the facial nerve can be exposed from brain stem to stylomastoid foramen (and possibly beyond if parotidectomy is performed), which allows for easier grafting if necessary.The trajectory of the translabyrinthine craniotomy
e26
SELF-ASSESSMENT QUESTIONS
enters the cerebellopontine angle from a more anterior origin than the retrosigmoid craniotomy; this allows for less cerebellar retraction.
2. A 45-year-old male with a 3.6-cm enhancing right cerebellopontine angle (CPA) lesion, which fills the internal auditory canal and does not have a detectable “dural tail” on MRI, is a good candidate for a translabyrinthine craniotomy because a. No other surgical approach will allow removal of a tumor this large. b. There is an 80% chance that the facial nerve function will be House-Brackmann grade I or II 1 year after surgery. c. There is a poor chance that hearing will be spared if total tumor removal is attempted, and the entire internal auditory canal and CPA component of the tumor can be exposed. d. This is an urgent case given the tumor size, and the translabyrinthine approach is a faster method of exposure. e. None of the above.
Answer: c.� Because of the tumor’s size, this patient has a
poor chance of hearing preservation with tumor removal, regardless of approach. Its size does not prohibit its removal through other approaches, and the translabyrinthine craniotomy is not necessarily faster than other craniotomies. The reported percentage of patients with House-Brackmann grade I or II 1 year following surgery for removal of unilateral, sporadic vestibular schwannomas greater than 3.5 cm in a single stage is about 50%. 3. W hich of the following is not routinely employed during a translabyrinthine craniotomy? a. Facial nerve monitoring b. Facial electromyography c. Short-acting paralytic agents for anesthesia d. Auditory brain stem response e. Monopolar electrocautery Answer: d.� Auditory brain stem response is not routinely
used during any case in which hearing is expected to be lost. Short-term paralytics, facial nerve monitoring, and facial electromyography are routinely employed as they allow the surgeon feedback regarding facial nerve irritation/trauma during tumor dissection. There are no contraindications to monopolar electrocautery during the approach. 4. Y ou are planning a translabyrinthine craniotomy for removal of an acoustic neuroma in a 59-year-old male in whom you had previously performed a tympanoplasty with mastoidectomy 10 years ago for chronic otitis media. His tympanic membrane appears healed, but your old operative report indicates that he had a small mastoid. For this patient you should
a. Perform the case by taking down the bony canal wall, closing off the ear canal laterally, and converting to a transcochlear or transotic approach to improve exposure. b. Perform the case as a routine translabyrinthine craniotomy, decompressing a large amount of middle fossa dura and dura posterior to the sigmoid sinus. c. Stage the case because of risk of infection by first closing the ear canal skin and taking down the bony canal wall followed by a translabyrinthine craniotomy 6 months later if no evidence of infection is noted. d. Abort the procedure as the risk of infection is high and the exposure will be poor. e. None of the above.
Answer: b.� A contracted mastoid is not a contraindication
to the translabyrinthine craniotomy.Wide dural decompression will be necessary to improve exposure. 5. T he potential benefits of performing a cranioplasty after a translabyrinthine craniotomy include a. Decreasing the rate of cerebrospinal fluid leak b. Prevention of a noticeable postauricular defect c. Protects the bony ear canal d. a and b e. a and c Answer: d.� Performing a cranioplasty helps bolster the fat
graft into the dural defect and potentially drops the cerebrospinal fluid leak rate.The nature of the cranioplasty is prevention of a noticeable defect in the skull; it does nothing to protect the bony ear canal.
Chapter 50 NONE
Chapter 51 1. T he transotic exposure adds which dimension of exposure of the internal auditory canal beyond what the translabyrinthine exposure offers? a. Anterior b. Posterior c. Inferior d. Superior e. Lateral Answer: a.
SELF-ASSESSMENT QUESTIONS
2.
ar canal management in transotic exposure requires E a. Sterile preparation with antibacterial solutions b. Two-layer closure c. Temporary transection and stenting to prevent stenosis d. Skin graft lining e. Wide meatoplasty
e27
c. Sinonasal tumors invading the infratemporal fossa, the masticator space, or the pterygomaxillary fossa, and tumors of the nasopharynx extending into the infratemporal fossa d. Trigeminal nerve sections, vestibular nerve sections, and CPA biopsies
Answer: a.
Answer: b.
3. T he facial nerve near the root-entry zone in relation to the anterior inferior cerebellar artery (AICA) is always: a. Posterior b. Superior c. Anterior d. Inferior e. Lateral Answer: c.
4. T he distinction between transcochlear and transotic exposure is that a. The facial nerve is mobilized posteriorly in transotic exposure. b. The facial nerve is mobilized anteriorly in transotic exposure. c. The facial nerve is mobilized anteriorly in transcochlear exposure. d. The facial nerve is mobilized posteriorly in transcochlear exposure. e. The facial nerve is better visualized in transotic exposure. Answer: d.
5. T he meatal foramen of the fallopian canal is situated at the a. Internal auditory canal porous b. Internal auditory canal fundus c. Geniculate ganglion d. Stylomastoid foramen e. Second genu Answer: b.
Chapter 52 1. W hat are the indications for the transcochlear approach? a. Lesions arising anterior to the internal auditory canal (IAC), petrous apex lesions, skull base and midline intradural lesions from the clivus b. Small acoustic tumors, with moderate extension into the cerebellopontine angle (CPA), and good preoperative hearing
2. W hat are the advantages of the transcochlear approach? a. The added exposure of removing the external auditory canal (EAC) and the cochlea, which increases access medially and anteriorly to the facial nerve with the safety of not requiring transposition of this nerve b. Hearing preservation and provision of an anterior plane of dissection on IAC in cases of acoustic neuroma c. Excellent exposure of the midline, allowing complete removal of the tumor, its base of implantation, and its blood supply with no cerebellar or temporal lobe retraction d. Extensive exposure inferiorly in the area of the jugular foramen and foramen magnum Answer: c.
3. W hat are the disadvantages of the transcochlear approach? a. Does not allow exposure of the lateral aspect of the pons and upper medulla, cranial nerves V through XI, and the midbasilar artery b. Sacrifice of residual hearing in the operated ear and risk of temporary facial palsy c. Limited by the cerebellum and the brain stem d. Requires the use of brain retractors Answer: b.
4. W hat are the intracranial structures that can be exposed by the transcochlear approach? a. Arcuate eminence, middle meningeal artery, greater and lesser superficial petrosal nerves, V3, V2, and VI entering into the superior orbital fissure b. The jugular foramen and foramen magnum c. Cerebellum and middle cranial fossa d. Entire lateral aspect of the pons and upper medulla, cranial nerves V through XI, as well as the midbasilar artery Answer: d.
5. W hat is the most common complication after transcochlear surgery and what steps should be taken to treat it?
e28
SELF-ASSESSMENT QUESTIONS
a. Temporary facial nerve paresis. Prompt eye care including lubrication with drops, nighttime ointments, and a moisture shield to prevent corneal complications should be given. Soft lenses, spring and gold weights, canthoplasty should be administered, and “watchful waiting” in cases of complete paralysis, if the facial nerve is anatomically intact, should be done. b. Temporary facial nerve paresis. Prompt eye care including lubrication with drops, nighttime ointments, and a moisture shield to prevent corneal complications should be given. Surgical intervention for facial reanimation in cases of complete paralysis, with a facial nerve that is anatomically intact, should be done. c. Delayed postoperative intracranial hemorrhage. Immediate reopening of the surgical wound and removal of the fat in the intensive care unit, and operative evacuation of the hematoma and control of the bleeding site or sites, should be done. d. Meningitis. Aggressive antibiotic therapy, if it is infectious, and treatment with dexamethasone if it is chemical aseptic meningitis, should be administered.
Answer: a.
Chapter 53 1. T he electromagnetic field (EMF) approach is a useful one for a. Acoustic neuroma greater than 2.5 cm b. Anterior petrous meningioma c. Large pituitary adenoma d. Meningioma extending into the jugular foramen Answer: b.
2. T he superior vestibular nerve can be safely followed laterally for a distance equaling a. 3 mm b. Three quarters the distance of the labyrinthine facial nerve c. One half the distance of the labyrinthine facial nerve d. 2.5 mm Answer: c.
4. T he EMF approach can be used for all of the following lesions except a. Petroclival meningioma b. Acoustic neuroma extending into the posterior fossa c. Lower clival lesions d. Infraclinoidal basilar tip aneurysms Answer: c.
5. A s a rule, the internal auditory canal (IAC) can be identified topographically by a. Bisecting the angle between the greater superficial petrosal nerve (GSPN) and the arcuate eminence b. Bisecting the angle between the arcuate eminence and the middle meningeal artery c. Following the course of the GSPN d. Following the course of the IAC Answer: a.
Chapter 54 1. H igh risk of stroke with carotid sacrifice on xenon blood flow studies with test occlusion of the carotid are predicted by flows a. Greater than 35 mL/min/100 gm of tissue b. 21–35 mL/min/100 gm of tissue c. Less than 20 mL/min/100 gm of tissue d. Greater than 50 mL/min/100 gm of tissue e. Greater than 100 mL/min/100 gm of tissue Answer: c.
2. T he frontalis portion of the facial nerve is best protected in the lateral subtemporal skull base approach by dissecting a. Deep to the deep temporal fascia off the zygomatic arch b. Superficial to the deep temporal fascia off the zygomatic arch c. Superficial to the fat pad over the temporal muscle d. Along each branch of the frontalis portion of the facial nerve e. Along at least two branches of the frontalis portion of the facial nerve Answer: a.
3.
he inferior extent of the EMF approach includes T a. The foramen magnum b. The inferior petrosal sinus c. The midclivus d. The superior petrosal sinus
Answer: b.
3. T he most common morbidity associated with surgery of the infratemporal fossa is a. Trigeminal nerve dysfunction b. Facial nerve dysfunction c. Dysphagia
SELF-ASSESSMENT QUESTIONS
d. Dysphonia e. Ataxia
Answer: a.
4.
he best treatment of postoperative trismus is T a. Temporomandibular joint (TMJ) resection b. TMJ prosthesis c. Stretching therapy d. Laser scar ablation e. Steroid injection into the pterygoid muscles
Answer: c.
5. U nilateral rhinorrhea not related to postoperative cerebrospinal fluid leak following skull base surgery occurs from a. Loss of sympathetic fibers along the internal carotid artery b. Loss of parasympathetic fibers along the internal carotid artery c. Pterygopalatine nerve ablation d. Sinusitis e. Hematoma liquefaction Answer: a.
e29
3. W hat are the four anatomical structures that form the boundaries of the “quadrangular space”? a. Cavernous internal carotid artery (ICA), optic nerve, first branch of trigeminal nerve, and vidian canal b. Vertical/paraclival ICA, petrous/horizontal ICA, cranial nerve (CN) VI, and maxillary branch of trigeminal nerve c. Vertical/paraclival ICA, petrous/horizontal ICA, CN IV, and cavernous sinus d. Cavernous ICA, optic nerve, vidian canal, and gasserian ganglion e. Vertical/paraclival ICA, petrous/horizontal ICA, CN VI, and gasserian ganglion Answer: b.� These four structures are the key anatomic struc-
tures that limit the endonasal approach to Meckel’s cave. 4. W hich of the following are options for vascularized tissue repair of endonasal skull base defects? a. Temporoparietal fascial flap b. Nasal septal mucosal flap c. Turbinate flap d. All of the above e. None of the above Answer: d.� All are potential candidates for a vascularized
Chapter 55 1. T he lateral extent of access via endoscopic endonasal approach at the level of the superior orbit is a. Lamina papyracea b. Periorbita c. Midorbital line d. Medial rectus muscle e. Anterior ethmoidal artery Answer: c.� Following removal of the lamina papyracea, the
periorbita can be gently displaced laterally to allow access to lesions above the orbit as far lateral as midorbit. 2. I nferior extent of access via endonasal approach can be determined preoperatively by drawing a line (Kassam line) between which two structures extended into the depth on sagittal CT? a. Nasal tip and odontoid process b. Tip of bony nasal bridge and hard palate c. Tip of bony nasal bridge and odontoid process d. Middle turbinate and hard palate e. Middle turbinate and C1 Answer: b.� These two bony structures create a fulcrum that
limits the inferior extent of exposure. By drawing a line between them on preoperative sagittal imaging, one can get a rough idea of the extent of lowest access.
flap. The nasal septal flap is the most versatile given its size and location. However, it usually must be harvested during the approach. 5. W hich of the following is the most common complication of endoscopic endonasal surgery prior to the introduction of the nasal septal flap? a. Cerebrospinal fluid leak b. Arterial injury c. Infection d. Stroke e. Cranial nerve palsy Answer: a.� Cerebrospinal fluid leak was the most common
complication of endoscopic endonasal surgery for skull base lesions prior to the introduction of the septal flap. Despite this, infection rates were very low.
Chapter 56 1. A combined petrosal approach is appropriate for petroclival tumors that a. Are superior to the internal auditory meatus b. Are inferior to the internal auditory meatus c. Extend into the infratemporal fossa d. a, b, and c e. a and b
e30
SELF-ASSESSMENT QUESTIONS
Answer: e.� The combined petrosal approach is best suited for
petroclival tumors that are both superior and inferior to the internal auditory canal. Otherwise, the middle fossa approach alone or a posterior approach alone can be used. The infratemporal fossa is not accessible with the combined petrosal approach.
a. Cerebrospinal fluid leak b. Cranial nerve injury c. Cerebrovascular accident d. Pituitary hypofunction e. Subdural hematoma
Answer: b.� Cranial nerve injury is the most common com-
2. M edial access to petroclival tumors with the combined petrosal approach is most commonly limited by a. The jugular bulb b. The angle between the clivus and brain stem c. The distance between the sigmoid sinus and facial nerve d. The tentorium and superior petrosal vein e. The angle between the cerebellum and brain stem
plication. Cranial nerves V and VII are the most commonly compromised, but III–XI are at risk with this approach. Some authors advocate subtotal excision with stereotactic irradiation as a result, but the long-term results are unknown.
Chapter 57 NONE
Answer: b.� The two structures that cannot be retracted,
removed, or moved are the brain stem and clivus. It is therefore the angle between them that limits access to the medial aspect of these tumors. More anterior exposure helps to open this angle. 3. P reoperative evaluation should include examination of the venous drainage system to exclude a. A dominant sigmoid sinus system b. A high jugular bulb c. Failure of the transverse sinus to communicate with the confluence (torcular herophili) d. Insertion of the vein of Labbé into the superior petrosal sinus e. a, c, and e Answer: e.� All of these vascular variations place the patient
at risk for venous infarction if not recognized preoperatively. 4. S ignificant posterior compression of the brain stem by the tumor a. Often requires more aggressive removal of the otic capsule for adequate brain stem exposure and tumor excision b. Requires no change in the amount of otic capsule removal because brain stem position does not affect tumor exposure c. Often requires less aggressive removal of the otic capsule for adequate brain stem exposure and tumor excision d. Requires more aggressive cerebellar retraction to adequately expose the brain stem e. Often makes sigmoid sinus transection necessary Answer: c.� The angle between the brain stem and clivus can
often be opened by posterior brain stem compression, allowing better access to the clivus than with a tumor that does not compress the brain stem. 5. T he most common complication from the combined petrosal approach is
Chapter 58 1. T he auditory brain stem implant (ABI) and the other CNS auditory implants described in this chapter a. Are direct replacements for cochlear implants b. Provide equivalent speech perception performance to cochlear implants c. Were developed to provide hearing sensations to deaf individuals with nonviable peripheral auditory neural systems d. All of the above Answer: c.
2. T he primary benefit of first-tumor side ABI implantation includes a. Better outcomes because implantation can occur when tumors are smaller b. Better outcomes because implantation can occur when patients are younger c. The opportunity to gain experience with the device before becoming completely deaf on the second-tumor side d. An opportunity to implant a more advanced device when the second-side tumor is removed Answer: c.
3. A primary contributor to a satisfactory outcome with an ABI is a. Implantation when vestibular schwannomas are as small as possible b. Implantation as soon as possible after onset of complete deafness c. A thorough and frank appraisal preoperatively of the potential benefits and limitations of the device d. Learning sign language and lipreading preoperatively Answer: c.
SELF-ASSESSMENT QUESTIONS
4. T he primary benefit of the penetrating auditory brain stem implant (PABI) was found to be a. Improved speech recognition performance b. Fewer nonauditory sensations c. Lower electrical auditory thresholds and a wide range of pitch sensations d. Greater ease of placement in the target neurons than the surface ABI Answer: c.
5. W hich of the following statements is true about auditory midbrain implantation? a. Speech perception is much better than the PABI. b. Speech perception is much worse than the regular surface ABI. c. Initial patients were not able to understand speech without lipreading cues. d. It clearly avoids any issues related to performance limitations with regular ABIs because of possible neural damage due to vestibular schwannomas or their removal. Answer: c.
Chapter 59 1. T he major advantage of the ELITE procedure for resection of large glomus jugulare tumors is a. Easier reconstruction of the skull base b. Better results in hearing preservation c. Improved exposure of intradural tumor d. Exposure of the tympanic portion of the facial nerve Answer: c.
2. I n the classic ELITE approach, management of the facial nerve typically includes a. Classic transposition b. Skeletonization of intratympanic portions of nerve c. Resection for improved access, with later grafting d. Skeletonization of vertical portion with limited anterior translocation
e31
Answer: Craniovertebral instability can occur secondary
to over-resection of the condyle. At least 50% of the condyle should be preserved. 5. D iscuss options for cranial base reconstruction and for prevention of cerebrospinal fluid leak. Answer: Options for reconstruction include fascia for
smaller defects, a pericranial flap, or a microvascular free flap for reconstruction of larger defects. Use of a vascularized myofascial flap for reconstruction is thought to reduce the incidence of postoperative cerebrospinal fluid leak. Abdominal fat is used to fill the mastoid defect. Lumbar drainage is used in most cases.
Chapter 60 1. A ll of the following are advantages of using a pumpregulated drainage system for lumbar drainage of a cerebrospinal fluid leak EXCEPT: a. Patient can move about more freely b. Decreased risk of tension pneumocephalus c. Strictly regulated flow of CSF d. Lower risk of meningitis Answer: d. There is no study showing that pump-regulated
drainage is associated with a lower risk of meningitis, but pump-regulated drainage does allow the patient to move about more freely, decreases the risk of accidental tension pneumocephalus due to unregulated CSF drainage, and does provide for a strict regulation of CSF flow. 2. T RUE or FALSE: Titanium mesh cranioplasty and hydroxyapatite cranioplasty reduce the risk of cerebrospinal fluid leak after translabyrinthine acoustic neuroma surgery. a. True b. False Answer: a. This is true (cf. references by Fayad and
Arriaga).
Answer: Dissection of the suboccipital triangle, identification
3. W hich of the following techniques would NOT be appropriate for closure of persistent CSF rhinorrhea after translabyrinthine acoustic neuroma surgery? a. Ear canal closure with Eustachian tube and middle ear obliteration (blind sac closure) b. Wound exploration and reclosure with new abdominal fat graft c. Middle fossa obliteration of the Eustachian tube d. Lumbar drainage
of the vertebral sulcus (“J” groove of the C1 lamina), or use of a Doppler probe.
Answer: c. Since the patient no longer has hearing, middle
Answer: d.
3. N ame three methods of identifying the V3 segment of the vertebral artery
4. N ame one uncommon complication of the ELITE surgical procedure that is unique to this approach.
fossa surgery to obliterate the Eustachian tube is not indicated after translabyrinthine surgery.
e32
SELF-ASSESSMENT QUESTIONS
4. M ethods of sealing the CSF space after posterior fossa surgery include: a. Abdominal fat b. Hydroxyapatite cement c. Fibrin glue d. Collagen matrix e. All of the above Answer: e. All of the answers represent methods used to seal
the CSF space after skull base surgery.
4. T he best method of assessing lower lid position on the paralyzed side of the face is a. Ask the patient how much the lid position has changed. b. Assess the position with straight ahead gaze. c. Assess the position with up gaze. d. Assess the position with down gaze. e. The gaze of the patient should be directed such that the limbus of the contralateral lid is placed just tangential to the inferior limbus. Answer: e.
Chapter 61 1. P atients who are to be considered for eyelid surgery include a. Those who are either symptomatic or who show signs of conjunctival or corneal injury, or both, despite maximum tolerated medical therapy b. Those who require rapid ocular rehabilitation to resume their usual occupation and responsibilities c. Those whose ocular status is currently stable but who are at high risk of corneal complications d. All of the above e. None of the above
5. A lid suture taped to the cheek and temporary tarsorrhaphy suture are a. Old methods that are only of historical interest b. Ways to close the eye on a windy day c. Complex procedures requiring special skills d. Only useful for a day or two e. Useful temporizing measures until a definitive solution to the eyelid closure problem can be implemented Answer: e.
Chapter 62
2. U nipolar cautery should not be used in eyelid surgery if a. The patient has an auditory brain stem implant b. The patient has a cochlear implant c. The patient has a prior spring d. All of the above e. Only a and b
1. W hat is the best method, when possible, for repairing a facial nerve injury where the nerve ends are visible during surgery for removal of a cerebellopontine angle neoplasm? a. Cable grafting b. Primary tension-free anastomosis c. Hypoglossal-facial anastomosis d. Cross-facial grafting e. Temporalis muscle dynamic reanimation
Answer: e.
Answer: b. Primary anastomosis of nerve ends, without
Answer: d.
3. A dvantages of the palpebral spring compared to the gold weight include the following: a. The spring closes the eye during sleep, when the patient is supine, whereas the weight does not. b. Lids that are hard to close may require large bulky weights, compared to a fine wire spring. c. Patients who are exposed to extremes of hot or cold may experience discomfort in the lid due to the heating or cooling of the weight, which is much less of a concern with the spring due to less mass to heat or cool. d. To adjust the closing tension with a gold weight requires replacement, whereas the spring can be adjusted without removal. e. All of the above. Answer: e.
tension, is the preferred method for repairing discontinuities of the facial nerve. Cable grafting requires more than one suture line, but should be used in cases where a tensionless anastomosis is impossible. The other options are appropriate when direct facial nerve repair has failed to reanimate the facial palsy. 2. W hich of the following procedures generally results in the most tongue dysfunction? a. Jump graft interposition between the hypoglossal nerve and facial nerve b. Facial nerve transposition from the fallopian canal, with end-to-side anastomosis to the hypoglossal nerve c. Direct hypoglossal-facial anastomosis using the distal end of the hypoglossal nerve
SELF-ASSESSMENT QUESTIONS
e33
Answer: c. The conventional direct anastomosis of the distal
Answer: d. A single epineurial suture is the standard method
hypoglossal nerve to the main trunk of the facial nerve provides the most amount of viable neurons to the facial nerve, but results in hemiglossal atrophy. Partial anastomoses such jump grafting and end-to-side anastomosis may largely spare ipsilateral hypoglossal function.
of securing a nerve to the end of a collagen tubule. Multiple sutures are used when nerve ends are directly anastomosed.
3. W hat clinical situation is the most appropriate for use of the hypoglossal-facial anastomosis? a. Extirpation of malignant parotid neoplasm with sacrifice of the pes anserinus and upper and lower divisions of the intraparotid facial nerve b. Extirpation of cerebellopontine angle tumor with loss of the facial nerve at the brainstem with no visible proximal stump c. Bell’s Palsy with no visible facial function but with active motor unit potentials on voluntary EMG d. Extirpation of cerebellopontine angle tumor with laceration of the facial nerve at the porus acousticus, with present proximal and distal nerve ends Answer: b. The most appropriate use of hypoglossal-facial
nerve anastomosis is when there is no potential for recovery of nerve function from the facial nerve itself, either through tensionless anastomosis or grafting. In the case of the parotid neoplasm, direct facial nerve anastomosis or muscular facial reanimation may be attempted, but hypoglossal-facial nerve anastomosis is impossible due to loss of the main trunk and upper and lower divisions of the facial nerve. 4. W hich electrophysiological test results are most compatible with the use of the hypoglossal-facial anastomosis? a. Fibrillation potentials on voluntary EMG, with 100% degeneration on electroneuronography (ENoG) b. Motor unit potentials on voluntary EMG, but with >90% degeneration on ENoG c. No degeneration on ENoG and active motor unit potentials on EMG Answer: a. Option 1 is the only test result that signifies com-
plete loss of facial nerve function, and is hence compatible with hypoglossal-facial nerve anastomosis. 5. W hat is the most appropriate way to secure a collagen tubule to the nerve end to be anastomosed? a. Six 9-0 nylon interrupted epineurial sutures b. Fibrin glue c. Cyanoacrylate d. One 8-0 interrupted polypropylene epineurial suture
6. T /F Recent studies suggest that partial hypoglossal nerve anastomoses may provide similar facial nerve outcomes while sparing patients morbidity from hemiglossal atrophy. a. True b. False Answer: a. This is true. References may be found in the
chapter text.
Chapter 63 1. T he ideal procedure for any patient with a disruption of the facial nerve is a. Primary repair b. Grafting to reestablish continuity between the facial nerve nucleus and facial musculature c. Temporalis muscle transposition d. None of the above e. a and b Answer: e.
2. B ased on the author’s experience, what percentage of patients have fair to superb nerve graft results? a. 5% b. 20% c. 60% d. 80% e. 95% Answer: d.
3. T he goal of the temporalis muscle transposition procedure is to restore spontaneous mimetic expression to a patient with facial paralysis. a. True b. False Answer: b.
4. T ime post-onset of the facial paralysis is an important consideration in which of the following procedures? a. Temporalis muscle transposition b. Greater auricular nerve graft c. Free muscle grafting d. All of the above e. None of the above Answer: b.
e34 5.
SELF-ASSESSMENT QUESTIONS
he preferred nerve to use in a grafting procedure is T a. Sural cutaneous b. Medial branchial cutaneous c. Greater auricular d. None of the above e. a and b
Answer: c.
6. W hich of the following are important factors in considering a temporalis muscle transposition procedure? a. Airway structure b. Appearance of nasolabial structures c. Type of smile on unaffected side d. b and c e. All of the above Answer: e.
Chapter 64 1. T he rationale for using intraoperative monitoring is to reduce the risk of permanent postoperative neurologic deficits. The way in which monitors do this is to a. Identify facial nerve when not visible in field b. Preserve useful hearing in very small acoustic tumors c. Assist in microvascular decompression for hemifacial spasm, trigeminal neuralgia, and vestibular nerve section d. Direct intracranial dissection when anatomy is unknown e. All of the above Answer: e.� There are a variety of scenarios in which intra-
operative monitoring can be a useful adjunct to reduce the risk of permanent postoperative neurological deficits. 2. A pitfall of intraoperative monitoring includes a. Operating room is noisy and electrical interference is abundant. b. Result of intraoperative monitoring is easily obtainable. c. The computer equipment can identify clinically significant changes in waveform without neurophysiologic personnel. d. Facial nerve monitoring is required in all otologic procedures. e. Maintenance of equipment is rarely necessary. Answer: a.� Obtaining a reproducible signal in the operat-
ing room is difficult due to electrical noise and interference. Neurophysiologic personnel should be available for setup and troubleshooting equipment that is regularly maintained. Neurophysiologic personnel help to identify changes in the amplitude and latencies and can help distinguish true changes from
rtifact. Facial nerve monitoring is required in otologic proa cedures when the facial nerve is particularly at risk, that is, unusual anatomy, previously operated ear, or extensive disease. 3. M ethods to monitor neural conductivity of the eighth cranial nerve include recording a. Brain stem–evoked auditory potentials b. Evoked potentials from exposed intracranial portion of cranial nerve VIII c. Surface electrode on cochlear nucleus d. Laser Doppler blood flow analysis e. Visual inspection Answer: a, b, c.� These are three methods to check neural
conductivity of cranial nerve VIII. 4. T o reduce the time to obtain interpretable evoked potentials a. Minimize electrical interference reaching recording electrodes b. Optimize filtering of recorded potential to accentuate background noise c. Decrease stimulus repetition rate and strength d. Increase electrode impedance e. Ignore methods for quality control that do not require record replication Answer: a.� Evoked potentials are most readily obtained in
environments with minimal electrical interference, background noise, low electrode impedance, and optimal stimulus repetition rate and strength. Low electrode impedance and increased stimulation rate and strength improve the quality of electrical signal information to be interpreted. 5. C hanges involving the following reflect clinically important changes in hearing a. Increase peak V latency b. Decreased peak V amplitude c. Change in peak V without change in peak III d. Change in interpeak latency e. Loss of reproducible signal Answer: e.� Whereas all of these changes could indicate
potential hearing compromise, the loss of reproducible signal is the strongest indication of a clinically important change in hearing.
Chapter 65 1. W hich of the following is true regarding gamma knife surgery for vestibular schwannomas? a. Tumor control rates are greater than 97%. b. Facial nerve motor dysfunction occurs in less than 1% with current dosing. c. Trigeminal nerve dysfunction is more common with large tumors.
SELF-ASSESSMENT QUESTIONS
d. Malignant transformation or induction is rare. e. All of the above.
Answer: e.
2. H earing thresholds as measured by pure-tone averages after gamma knife surgery most commonly a. Improve immediately b. Behave similar to expectant observation c. Progress rapidly to profound deafness d. Degrade rapidly in the first 6 months and then slowly worsen e. Do not change
e35
4. A patient underwent gamma knife surgery of a 2.3-cm CPA vestibular schwannoma 6 months ago. MRI performed today shows the tumor to be 2.7 cm in maximal diameter. The patient is asymptomatic. The next best course of action is a. Assure patient this is normal and rescan in 6 months b. Recommend microsurgical resection for radiation failure c. Counsel patient this may be malignant d. Plan a second round of gamma knife surgery e. Start high-dose steroids with a taper Answer: a.
Answer: d.
3. W hich of the following tumors would not be amenable to treatment with the more common gamma knife B and C units? a. Intracanalicular acoustic neuroma of 7 mm maximal diameter b. Vestibular schwannoma extending into the cerebellopontine angle (CPA) by 1.5 cm c. 2-cm glomus jugulare tumor extending anterosuperiorly from the jugular bulb d. Glomus vagale tumor extending to the carotid bifurcation e. Petrous apex meningioma of 2.4 cm maximal diameter Answer: d.
5. W hich of the following is not a component of the gamma knife surgery system? a. Trunnions b. Collimator helmet c. Gamma calipers d. Cobalt 60 (60Co) sources and beam channels e. MRI fiducial box Answer: c.
Chapter 66 NONE