Surgical Disorders of the Peripheral Nerves Second Edition
Rolfe Birch
Surgical Disorders of the Peripheral Nerves S...
39 downloads
1403 Views
64MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Surgical Disorders of the Peripheral Nerves Second Edition
Rolfe Birch
Surgical Disorders of the Peripheral Nerves Second Edition
Author Rolfe Birch M. Chir, FRCP&S (Glas), FRCS (Edin), FRCS (Eng) by election Professor in Neurological Orthopaedic Surgery, University College, London and Visiting Professor, Department of Academic Neurology, Imperial College, London and Honorary Orthopaedic Consultant, Hospital for Sick Children Great Ormond Street, London and The National Hospital for Nervous Diseases, Queen Square, London and Raigmore Hospital, Inverness and Honorary Orthopaedic Surgeon to the Royal Navy and Consultant in Charge, War Nerve Injuries Clinic at the Defence Medical Rehabilitation Centre, Headley Court, Leatherhead, Surrey
ISBN 978-1-84882-107-1 e-ISBN 978-1-84882-108-8 DOI 10.1007/978-1-84882-108-8 Springer London Dordrecht Heidelberg New York British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Control Number: 2010933604 © Springer-Verlag London Limited 2011 1st edition by R. Birch, G. Bonney and C.B. Wynn Parry published in 1998 by Churchill Livingstone, ISBN 978-0-443-04443-4 Apart from any fair dealing for the purposes of research or private study, or criticism or review, as permitted under the Copyright, Designs and Patents Act 1988, this publication may only be reproduced, stored or transmitted, in any form or by any means, with the prior permission in writing of the publishers, or in the case of reprographic reproduction in accordance with the terms of licenses issued by the Copyright Licensing Agency. Enquiries concerning reproduction outside those terms should be sent to the publishers. The use of registered names, trademarks, etc., in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant laws and regulations and therefore free for general use. Product liability: The publisher can give no guarantee for information about drug dosage and application thereof contained in this book. In every individual case the respective user must check its accuracy by consulting other pharmaceutical literature. Cover design: xxxxxx Printed on acid-free paper Springer is part of Springer Science+Business Media (www.springer.com)
In memoriam George Bonney
v
Preface to the First Edition
“Opportunity comes to the mind prepared.” Opportunity came in full measure to H.J. Seddon with his appointment as surgeon in charge of the Medical Research Coucil’s (MRC) Peripheral Nerve Injury Unit in Oxford during the Second World War (1939–1945). It was after all in the Oxford of those days that J.Z. Young was doing the work on the nervous system that was to be the beginning of so many of the later developments in neuroanatomy and neurophysiology. The rewards of the MRC’s far sighted planning were abundant: to Seddon in particular was due that happy result. He was, however, the first to acknowledge the fortunate circumstances that enabled him to undertake the work in the company of so many doctors and scientists distinguished in the fields of anatomy, physiology and pathology of the peripheral nerves. He acknowledged too his debt to two great colleagues in clinical medicine: Hugh Cairns and George Riddoch. It was also fortunate that the Oxford unit was not allowed to die away at the end of the War, and that Seddon was able to continue at the Royal National Orthopaedic Hospital and the Institute of Orthopaedics the work that had begun and flourished at Oxford. After Seddon’s retirement his close colleague Donal Brooks and others developed and extended his work in London, while in other centres those who had worked with these men made their own contributions. Sadly, the bright hope that with introduction of the National Health Service the planning that contributed so much to the earlier success would be continued, has withered and died. Seddon’s belief that “the necessity for this segregation, this concentration of cases” would be recognised has been proved wrong. The hope may finally have been extinguished in this country by the introduction through the National Health Service Act of 1990 of an artificial internal market in health care, with competition between “providers of health care” and by the necessary corollary of the forced Gleichschaltung of the medical profession. Seddon’s firmness of purpose, clarity of thought, immense capacity for sustained hard work and powers of organisation were shown in “Peripheral Nerve Injuries” (1954) presented by the Nerve Injuries Committee of the Medical Research Council under his chairmanship. These characteristics were complemented by the qualities of those who collaborated in the work at Oxford. In the preface to the first edition of Surgical Disorders, Seddon paid generous tribute to J.Z. Young, P.B. Medawar, Graham Weddell and others. A special prominence was accorded to the contribution made by Donal Brooks. One who knew and worked with both has recorded the view that even that recognition was inadequate. He has made the disrespectful comparison of Brooks and Seddon with Jeeves and Wooster, and others who shared that experience may recognise the origins of the impious thought. Seddon’s character was of course the antithesis of that of Wooster, but those who saw these men at work cannot doubt that the calm guidance from the Irish Jeeves greatly influenced Seddon’s work and actions. Both editions of Surgical Disorders bear the mark of Seddon’s personality: the ordered thought; the meticulous observation and recording; the awareness of the ambient scientific field; the occasional dogmatic assertion; the love of tabulation. No one can read the book and vii
viii
Preface to the First Edition
not admire its depth and scope; no one can read it without astonishment at the comprehensive manner in which the subject is treated. One finds in it flashes of insight which, thought by the reader to have originated with him or her, turn out to be the subliminal origins of that thought. Some may consider it too bold an undertaking to attempt the revision of a work which is a classic of British clinical science and an abiding monument to Seddon’s work and leadership. However, things have moved on since 1975. Advances have been made in this country. In the field of disorders of peripheral nerves by Eames, Fullerton, Gamble, Gilliatt, Thomas, Urich and others. In continental Europe the work of Brunelli, Carlstedt, Gilbert, Hagbarth, Landi, Lundborg, Millesi, Morelli, Narakas, Slooff, Torebjörk, Wallin and others has opened new possibilities and destroyed old certainties. In the USA, Gelberman, Kline, Leffert, Omer, Spinner, Terzis, Wilbourn and many others have made massive contributions. In Canada, Hudson and Mackinnon have made great clinical and experimental contributions. In Australia, the doctrine of primary repair of injuries in the upper limb and hand was developed by Rank, O’Brien and others. In China, where the feasibility of “replantation” was first demonstrated, Professor Gu of Shanghai has made and continues to make advances in the field. The development of neurotisation was largely due to work in Japan, where Nagano and Sugioka and many others continue that and other work. In particular, ideas about all types of lesion of the brachial plexus have changed; conceptions of pain mechanisms have developed, and much enlightenment has come to the understanding of the pathology of tumours. Lastly, and most sadly the incidence of “iatrogenic” lesions has greatly increased, though in this connection great advances have been made in the treatment of birth injuries of the brachial plexus. The authors of the present work hope that they may have succeeded in restoring Surgical Disorders to its place as the British text on the present state of affairs in the field. George Bonney had the privilege and pleasure and occasional pain, of working with Seddon; Christopher Wynn Parry came to the Royal National Orthopaedic Hospital at the time of its Renaissance under the leadership of Lipmann Kessel when new fields in neurophysiology, in the treatment of pain and in the surgery of the brachial plexus were being explored; Rolfe Birch came to the field at St Mary’s and the Royal National Orthopaedic Hospitals armed with experience in microsurgery, and in histological and electron-microscopic techniques. Most of the work on the results of which we have drawn was done at St Mary’s and the Royal National Orthoapaedic Hospitals. The original layout of the book has largely been retained, but the text has been entirely rewritten. Chapters on “iatrogenic” lesions, on birth injuries of the brachial plexus and on recovery of sensibility after repair have been added. The subjects of pain and of tumours are considered in more depth than formerly. The subject of electrophysiological examination is considered by an expert in the field, Dr. Shelagh Smith. Rather more attention is given to anatomical considerations than was formerly the case. As Last (1949) remarked with some asperity “restatement of the facts appears to be warranted by the misconceptions shown by many postgraduate students.” Not just by students. Evidently, we have tried to keep abreast of continuing advances in this developing field, but we shall inevitably be overtaken by the march of events. One does what one can. As was shown by his magisterial reorganization of a then famous London medical library, Seddon saw a clear separation between “medicine” and “surgery.” As the title of this book suggests, we have aimed to deal mainly with disorders which are generally amenable to treatment by operation, and with the appropriate techniques of operation. However, we maintain a belief in the unity of medicine. The book is aimed at surgeons in training and in practice, at physicians in general and at neurologists in particular. We even hope that undergraduates will come to no great harm through reading it. The aim has been determined by observation over the years in the clinic and in the courts of a general lack of knowledge and want of interest in conditions of peripheral nerves. These defects in medical education have generally been unhelpful to patients: injuries of nerves have been “missed” in accident departments; the delay so
Preface to the First Edition
ix
produced is compounded by a lack of appreciation of the urgency of the situation and of the possibilities of repair; tumours of nerves too often go unrecognised until at operation a junior practitioner is confronted by a tumour occupying, or in close proiximity to, a major nerve. We hope that this book may help in stimulating interest in conditions affecting peripheral nerves, that it may aid in enhancing the quality of treatment of such lesions, and that it may bring the field of study to the notice of clinicians highly placed in surviving teaching institutions. Lesions of peripheral nerves are indeed very useful in teaching the interpretation of physical signs, which is after all the main business of the clinician. It has not been possible to treat exhaustively every aspect of the now large subject, but we hope that sufficient and sufficiently well chosen references have been given to open avenues for further reading. We had hoped to include a chapter on disorders of selected cranial nerves, and in particular of the facial nerve. That hope was, alas, born to die: we have in the event restricted the cranial nerve study to one which is really a spinal nerve – the spinal accessory – and to some aspects of damage to the fifth, seventh, tenth and twelfth nerves. Rashly, perhaps, we have proposed one or two new terms for varieties of nerve injury and for pain arising from the nerve injury. We have also revived an old suggestion for a term to replace that commonly used to designate injury inflicted by doctors. We do not hope to escape criticism for this presumption, but we hope that, at least, our derivations will be found by classicists to be correct. We hope that those derivations do no discredit to the august institution where their principles were imparted. All who work in this field owe a debt of gratitude to the late Sir Herbert Seddon, and to the late Sir Sydney Sunderland. The extent of their contribution is overwhelming: no list of references, however long, can indicate its magnitude. Any book on this subject must draw heavily on Peripheral Neuropathy, and in particular on the third edition (1993). We are glad to acknowledge our debt to the editors and authors contributing to that majestic work. The late Professor Roger Gilliatt was foremost amongst those who after the Second World War made advances in the field and stimulated the interest of his colleagues and juniors at Queen Square and in other centres. No-one contributed more to the study and treatment of lesions of the brachial plexus than did the late Professor Algimàntas Narakas of Lausanne. We gratefully acknowledge the contribution of these two men and the lasting influence of their work.
Rolfe Birch George Bonney† Christopher Wynn Parry
Preface to the Second Edition
This work is dedicated to George Bonney who began preparation of the second edition of Surgical Disorders of the Peripheral Nerves shortly after the publication of the first. The structure of the book was well advanced by the time of his death. One main reason for undertaking this task was the rising tide of iatrogenous injuries. It became clear that it was no longer reasonable to assume that modern medical education provides graduates with a sound grasp of the anatomical and physiological principles of the peripheral nervous system. The work has been almost entirely rewritten with much greater emphasis upon the causes and manifestations of injuries to nerves, particularly iatrogenous injuries and the effects of ischaemia. Shelagh Smith and Ravi Knight have rewritten the chapter on electrodiagnosis which now takes its proper place in the central part of the book rather than at the end. Tara Renton has provided a welcome addition about the risks to the branches of the trigeminal nerve during facio-maxillary and dental work. The field of entrapment neuropathy has been reduced to a discussion about how to avoid error in diagnosis and in execution. With considerable reluctance the field of tumours of peripheral nerves has been approached in the same way. Information about cause, course and outcome in more than 6000 nerve injuries are summarised. The immense task of collation of data was undertaken by the staff of the Peripheral Nerve Injury Unit under the direction of the research coordinator, Margaret Taggart. Sanjay Patel provided exceptional skills in the development of different data bases and he analysed the extensive material about the birth lesion of the brachial plexus. Dirk de Camp, photographer of the Institute of Orthopaedics undertook all of the photographic work and developed an archive of several thousand images with particular precision. This involved the retrieval of much earlier material and he showed extraordinary patience in preparing the final order of the photographs for the different chapters. All of the drawings in this edition were done by Philip Wilson, who took on the constant revision and alteration without demur. We have indeed been fortunate in the close collaboration with distinguished colleagues in two other Institutions. Praveen Anand, now at the Hammersmith Hospital (Imperial College), has, with his team, provided extensive information from investigations of tissues obtained from patients with nerve injuries or suffering from neuropathic pain and these findings have been matched with those drawn from clinical examination and quantitative sensory testing performed in the Joint Clinics held with him, and with Peter Misra, at the Hammersmith Hospital. Uma Anand provided beautiful illustrations from the successful culture of neurones from human dorsal root ganglia. Nicholas Murray, Shelagh Smith, Peter Misra, and Carla Cordivari of the Department of Neurophysiology at the National Hospital for Nervous Diseases, Queen Square examined more than 1000 patients seen in our Joint Diagnostic Clinics. Susan Standring provided a great deal of her own original material and, through her good offices, Martyn Cooke curator and John Carr photographer, went to great lengths to provide photographs of specimens held in the Wellcome Museum of the Royal College of Surgeons of England. xi
xii
Preface to the Second Edition
A number of colleagues undertook the laborious task of reading the manuscripts. The first four chapters were read by Susan Standring, Praveen Anand and Peter Misra who made many cogent suggestions and important corrections. Frank Horan, Editor Emeritus of the Journal of Bone and Joint Surgery, and Michael Laurence of the Editorial Board of that Journal undertook the arduous task of reading the entire work. Their contribution went far beyond the detection of innumerable blemishes. Their highly informed and meticulous criticism strengthened the structure and concept of the work. The responsibility for errors remains with the author. Margaret Taggart indexed and catalogued all references and transcribed the entire manuscript as well as developing and implementing methods within the Peripheral Nerve Injury Unit for the prospective collection of data for all patients. Thomas Carlstedt joined us in 1995, transferring his work from the Karolinska Institute, Stockholm and he shared his thoughts and work in the fields of injuries to the lumbo sacral plexus, reconnection between the central and the peripheral nervous systems in cases of avulsion of spinal nerves, entrapment of the pudendal nerve and in aspects of nerve tumours. Marco Sinisi, who joined us five years ago generously provided material from his experience with peripheral nerve tumours, neuropathic pain and prolonged conduction block. For many years our house surgeons and registrars have made important contributions by reviewing both case notes and patients and going to great lengths to retrieve classical references. We have been extremely fortunate in our many visiting colleagues, who brought with them new ideas and who undertook much original work. Without their contributions the work could not have gone forward and wherever possible those contributions have been acknowledged in the text. Particular thanks are due to the many hundreds of colleagues, for the most part orthopaedic surgeons, by whose acumen and professional authority so many patients have been sent to us with a clear and accurate diagnosis at the best possible time for their treatment. Particular thanks too go to our editors, Melissa Morton and Denise Roland of Springer. They provided so much encouragement and interest throughout but also offered highly critical reviews of earlier drafts which ensured some clarity of purpose. Rolfe Birch
Contents
1 The Peripheral Nervous System: Gross Anatomy . . . . . . . . . . . . . . . . . . . . . . . . . 1.1 The Cranial Nerves. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2 The Spinal Nerves. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2.1 The Anterior Primary Rami. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2.2 Thoracic Anterior Primary Rami. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2.3 Lumbar and Sacral Anterior Primary Rami. . . . . . . . . . . . . . . . . . . . . . . . 1.2.4 The Posterior Primary Rami. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3 The Autonomic Nervous System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3.1 The Sympathetic System. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3.2 The Parasympathetic Nervous System. . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3.3 Afferent Autonomic Pathways. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.4 Nerves at Risk from Musculo Skeletal Injury. . . . . . . . . . . . . . . . . . . . . . . . . . . . References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1 4 7 7 18 18 32 33 34 36 37 37 41
2 The Microscopic Structure of the Nervous System: Its Function . . . . . . . . . . . . . 2.1 The Neurotrophins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.1 Nerve Growth Factor. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 The Peripheral Nerve Fibres. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.1 Conduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.2 The Basis of the Action Potential: Ion Channels. . . . . . . . . . . . . . . . . . . . 2.2.3 Axonal Transport. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.4 The Blood Supply of Nerves. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.5 The Nervi Nervorum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.6 Changes in Nerves with Ageing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 The Somatic Motor System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4 The Somatic Sensory System. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.1 Cutaneous Sensibility. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.2 The Skin. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.3 Cutaneous Sensory Receptors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.4 Deep Sensibility. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.5 The Muscle Spindles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.6 The Golgi Tendon Organs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.7 Central Connections. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.8 Visceral Afferents. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5 Cortical Maps. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.6 Synaptic Activity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
43 44 44 46 54 56 57 58 59 59 61 62 62 64 65 66 66 68 69 70 71 72 72
xiii
xiv
Contents
3 Reactions to Injury . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Axonotmesis – Neurotmesis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 The Cell Body and Proximal Stump. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.1 Wounds of the Perineurium. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.2 Contralateral Effects. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 The Distal Stump . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4 Types of Lesion Produced by Different Physical Agents. . . . . . . . . . . . . . . . . . . 3.4.1 Acute Ischaemia. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.2 Ischaemia from Tamponade. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.3 Ischaemia and Acute Compression Within Neurovascular Fascial Compartments. . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.4 Ischaemia by Acute Compression from Swollen Muscle . . . . . . . . . . . . . 3.4.5 Ischaemia Caused by Traction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.6 Reperfusion Injury. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.7 Chronic Ischaemia. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.8 Crush . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.9 Compression – Acute. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.10 Chronic Nerve Compression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.11 Traction or Stretch Injury. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.12 Thermal Injury. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.13 Electric Shock . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.14 Percussion Injury. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.15 Injection Injury. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.16 Vibration Injury. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5 The Perineurium and Neoplasm or Infiltration. . . . . . . . . . . . . . . . . . . . . . . . . . . 3.6 Radiation and Peripheral Nerves. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.7 Envenomation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.8 The Peripheral Effects of Denervation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.9 Changes at the Higher Levels: The Phantom Limb . . . . . . . . . . . . . . . . . . . . . . . References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
90 92 95 96 97 97 98 99 100 103 104 106 106 107 107 108 108 109 110 111
4 Regeneration and Recovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 The Response of the Nerve and Axon to Transection. . . . . . . . . . . . . . . . . . . . . . 4.2 The Repair of Large Gaps. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.1 Other, Non Neural, Material for Grafts: Entubation . . . . . . . . . . . . . . . . . 4.3 Nerve Transfer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.1 Recovery of Cutaneous Sensation after Nerve Transfer . . . . . . . . . . . . . . 4.3.2 Recovery of the Deep Afferent Pathways after Nerve Transfer. . . . . . . . . 4.3.3 Complications of Regeneration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4 Regeneration after Intradural Injury. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.5 Recovery of Function after Nerve Repair. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.5.1 Methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.6 Factors in Prognosis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.7 Severity of Injury . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.8 Delay. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
115 116 123 126 128 130 130 132 135 137 138 139 140 140 140
5 Clinical Aspects of Nerve Injury . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1 Associated Symptoms and Signs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2 Recognition of the Level and the Depth of Injury . . . . . . . . . . . . . . . . . . . . . . . . 5.3 Signs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4 Tinel’s Sign. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.5 Eliciting the Tinel Sign in Closed Lesions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
145 148 151 154 156 159
77 81 82 84 85 85 87 87 89
Contents
xv
5.6 5.7 5.8 5.9
Tinel’s Sign and Recovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Examination of Sensibility. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Quantitative Sensory Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Examination of Muscles. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.9.1 Some Pitfalls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.9.2 Our Methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.9.3 Clinical Examination. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.9.4 The Lower Limb. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.9.5 Late Signs of Nerve Injury. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.9.6 Signs of Reinnervation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.10 Records. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.11 Aids to Diagnosis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
161 161 163 166 166 168 168 174 177 180 182 185 188
6 Clinical Neurophysiology in Peripheral Nerve Injuries . . . . . . . . . . . . . . . . . . . . Shelagh Smith and Ravi Knight 6.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2 Electrodiagnostic Techniques. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2.1 Glossary of Electrodiagnostic Procedures. . . . . . . . . . . . . . . . . . . . . . . . . 6.2.2 Intra-operative Neurophysiological Procedures. . . . . . . . . . . . . . . . . . . . . 6.2.3 Electrodiagnostic Techniques and Localisation. . . . . . . . . . . . . . . . . . . . . 6.3 Limitations and Pitfalls of Electrodiagnostic Investigation . . . . . . . . . . . . . . . . . 6.4 Safety Aspects. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.5 Pathophysiological Correlates. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.5.1 Types of Nerve Lesion: the Electrophysiological Consequences . . . . . . . 6.5.2 Regeneration and Reinnervation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.6 Clinical Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.6.1 Upper Limb Neuropathies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.6.2 Lower Limb Neuropathies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.6.3 Diffuse Problems. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
191
7 Operating on Peripheral Nerves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.1 Indications and Objects of Intervention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.1.1 Special Units: Their Role. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2 General Principles of Operation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2.1 Control of Bleeding. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3 Apparatus and Instruments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.4 Methods of Repair. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.4.1 The Vascular Repair. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.4.2 The Nerve Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.5 The Nerve Repair . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.5.1 Methods of Suture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.5.2 Grafting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.5.3 Indications For and Methods of Nerve Transfer . . . . . . . . . . . . . . . . . . . . 7.5.4 Immobilisation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.6 Approaches to Individual Nerves. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.6.1 The Transverse Supraclavicular Approach: (Anterior, or Anterolateral). . 7.6.2 The Transclavicular Exposure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.6.3 The Postero-Lateral Route. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.7 Repair of the Roots of the Spinal Nerves in an Avulsion Lesion. . . . . . . . . . . . . 7.7.1 The Spinal Accessory Nerve. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.7.2 The Suprascapular Nerve. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.7.3 The Infraclavicular Part of the Brachial Plexus. . . . . . . . . . . . . . . . . . . . .
231 231 232 233 233 234 239 239 241 244 245 247 249 252 254 254 256 259 262 264 265 265
191 191 192 197 199 199 201 201 202 205 206 206 219 223 225
xvi
Contents
7.7.4 7.7.5 7.7.6 7.7.7 7.7.8 7.7.9 7.7.10 7.7.11 7.7.12
The Circumflex Nerve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Median and Ulnar Nerves in the Arm and the Axilla. . . . . . . . . . . . . . . . The Radial Nerve. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Posterior Interosseous Nerve. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Lower Part of the Median Nerve. . . . . . . . . . . . . . . . . . . . . . . . . . . . The Lower Part of the Ulnar Nerve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Nerves in the Abdomen and Pelvis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Sciatic Nerve. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Tibial and Common Peroneal Nerves in the Popliteal Fossa and Below. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.7.13 The Lower Tibial Nerve and the Plantar Nerves . . . . . . . . . . . . . . . . . . . 7.8 Entrapment Neuropathy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.8.1 The Thoracic Outlet Syndromes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.8.2 The Suprapleural Membrane (Sibson’s Fascia). . . . . . . . . . . . . . . . . . . . 7.8.3 The First Rib. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.8.4 The Seventh Cervical Rib. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.8.5 Considerations About Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.8.6 Carpal Tunnel Syndrome. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.8.7 Technique of Operation (Open Method) . . . . . . . . . . . . . . . . . . . . . . . . . 7.8.8 Operations for Ulnar Neuropathy at the Elbow. . . . . . . . . . . . . . . . . . . . 7.8.9 The Less Common Entrapment Syndromes – Upper Limb. . . . . . . . . . . 7.8.10 Some Entrapment Neuropathies in the Lower Limb . . . . . . . . . . . . . . . . 7.8.11 Meralgia paraesthetica. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.8.12 Entrapment of the Pudendal Nerve. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.8.13 The Piriformis Syndrome. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.8.14 The Tarsal Tunnel Syndrome. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.8.15 “Morton’s Metatarsalgia”. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.9 Pitfalls in Operating on Tumours of Peripheral Nerves. . . . . . . . . . . . . . . . . . . . 7.9.1 The Solitary Benign Schwannoma (Neurolemmoma, Neurinoma). . . . . 7.9.2 The Intraneural Ganglion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.9.3 The Solitary Neurofibroma. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.9.4 Malignant Peripheral Nerve Sheath Tumours (MPNST). . . . . . . . . . . . . References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
266 266 266 267 267 269 270 271
8 Compound Nerve Injury . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.1 The Wound. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.1.1 War Wounds: Current Practice. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.2 The Vascular Lesion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.2.1 The Shoulder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.2.2 The Elbow. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.2.3 The Knee . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.2.4 The False Aneurysm and Arteriovenous Fistulae. . . . . . . . . . . . . . . . . . . 8.2.5 Ischaemia and the Nerve. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.2.6 Iatrogenous Ischaemic Injury. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.3 Skin. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.4 Penetrating Missile Injuries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.4.1 Suture and Grafting During World War Two: The MRCR Evidence. . . . 8.4.2 Grafts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.4.3 Experience from St Mary’s and the Royal National Orthopaedic Hospitals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.4.4 The Brachial Plexus. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.4.5 Pain. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.4.6 The Peripheral Nerves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.4.7 Recent Experience . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
303 304 305 306 309 312 312 313 316 319 320 322 323 324
272 273 273 275 276 276 276 280 282 282 284 284 285 285 285 286 286 287 287 287 289 290 293 298
326 326 328 328 330
Contents
xvii
8.5 Neurovascular Injuries: Amputation Revascularisation. . . . . . . . . . . . . . . . . . . . 332 8.5.1 The Brachial Plexus. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333 8.5.2 The Closed Infraclavicular Lesion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 334 8.6 Nerves and Bone and Joint Injuries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339 8.6.1 The Nerve and the Pattern of Fracture. . . . . . . . . . . . . . . . . . . . . . . . . . . 341 8.6.2 The Shoulder Girdle and Gleno-Humeral Joints. . . . . . . . . . . . . . . . . . . 343 8.6.3 The Clavicle. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 344 8.6.4 Dislocation of the Gleno-Humeral Joint . . . . . . . . . . . . . . . . . . . . . . . . . 345 8.6.5 The Radial Nerve and Fractures of the Humerus. . . . . . . . . . . . . . . . . . . 351 8.6.6 Incidence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351 8.6.7 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355 8.6.8 The Musculocutaneous Nerve. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355 8.6.9 Elbow. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 356 8.6.10 Iatrogenous Injuries in the Adult. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 359 8.6.11 The Forearm. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 359 8.7 The Lower Limb. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 360 8.7.1 The Femoral Nerve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 360 8.7.2 The Lumbo Sacral Plexus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 361 8.7.3 The Hip. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 363 8.7.4 The Common Peroneal and Tibial Nerves. . . . . . . . . . . . . . . . . . . . . . . . 365 8.7.5 More Recent Experience at the Royal National Orthopaedic Hospital. . 367 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 369 9 The Closed Supraclavicular Lesion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.1 Mechanisms of Injury. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.2 Anatomy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.2.1 Course in the Neck. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.2.2 Micro Anatomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.2.3 Functional Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.3 Lesions of the Spinal Cord. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.4 The Evolution of Our Policy of Treatment of Closed Traction Lesions of the Brachial Plexus. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.5 The Definition of Pre and Postganglionic Lesion: Prognosis for Recovery. . . . . 9.6 Epidemiology. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.7 Diagnosis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.7.1 Conduction studies: The Current Situation . . . . . . . . . . . . . . . . . . . . . . . 9.7.2 The Operation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.8 Some of the Techniques Used for Repair. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.9 Strategies of Repair. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.9.1 The Upper Lesion: Rupture or Avulsion of C5, C6 (C7) with Intact (C7) C8, T1. . . . . . . . . . . . . . . . . . . . . . . . . . 9.9.2 The Lower Lesion: Intact C5, C6 (C7), Rupture or Avulsion C8, T1. . . 9.9.3 The Middle Group. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.9.4 The Complete Lesion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.9.5 Free Functioning Muscle Transfer (FFMT). . . . . . . . . . . . . . . . . . . . . . . 9.9.6 The Bilateral Lesion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.10 Results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.10.1 Methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.10.2 Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.10.3 Conventional Nerve Transfers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.10.4 Repair of Avulsed Ventral Roots. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.11 Recovery of Function by Patients. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.11.1 Age. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
375 376 378 378 379 379 380 384 385 388 389 392 398 398 399 399 404 404 404 406 407 409 409 409 410 410 411 415
xviii
Contents
9.12 Relief of Pain by Repair. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 416 9.12.1 Return to Work. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 420 9.13 Reimplantation of Avulsed Spinal Nerves. . . . . . . . . . . . . . . . . . . . . . . . . . . . 421 9.13.1 The First Clinical Case – George Bonney, 1977 (from the first edition of this work) . . . . . . . . . . . . . . . . . . . . . . . . . . 421 9.13.2 Subsequent Work by Thomas Carlstedt. . . . . . . . . . . . . . . . . . . . . . . 423 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 425 10 Birth Lesions of the Brachial Plexus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 429 10.1 The Lesion of the Spinal Nerve in Birth Lesion of the Brachial Plexus (BLBP)����������������������������������������������������������������������������������������430 10.1.1 The Central Affect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 431 10.2 Methods of Study. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 432 10.2.1 Diagnosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 432 10.3 Epidemiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 436 10.3.1 Incidence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 436 10.3.2 Risk Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 437 10.4 Recovery in the Complete Lesion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 444 10.4.1 Group 3. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 445 10.4.2 Group 4. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 445 10.5 Treatment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 447 10.6 Supplementary Investigations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 448 10.6.1 Neurophysiological Investigations (NPI). . . . . . . . . . . . . . . . . . . . . . 448 10.6.2 Imaging. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 453 10.7 Nerve Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 453 10.8 The Indications for Operation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 453 10.8.1 The Incomplete Lesion: Groups 1 and 2 . . . . . . . . . . . . . . . . . . . . . . 454 10.9 The Operation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 454 10.9.1 Methods of Repair . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 456 10.9.2 Post-operative Care. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 456 10.10 Results of Nerve Operations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 457 10.10.1 Experience at St Mary’s – Royal National Orthopaedic Hospital . . . 458 10.10.2 Late Reinnervation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 462 10.10.3 Cocontraction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 463 10.11 Deformities. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 463 10.12 Posterior Subluxation (PS) and Posterior Dislocation (PD) of the Gleno-humeral Joint with Related Contractures . . . . . . . . . . . . . . . . . . 466 10.12.1 Onset and Progression of the Secondary Deformities. . . . . . . . . . . . 467 10.12.2 Diagnosis and Classification. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 469 10.12.3 Medial Rotation Contracture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 470 10.12.4 The Development of Our Preferred Operation for Posterior Subluxation or Dislocation of the Gleno-Humeral Joint. . . . . . . . . . 471 10.12.5 The Operation for Reduction of the Posterior Subluxation and Dislocation of the Gleno-Humeral Joint. . . . . . . . . . 473 10.12.6 Deformities at the Elbow and Forearm. . . . . . . . . . . . . . . . . . . . . . . . 478 10.13 Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 479 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 479 11 Iatrogenous Injuries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Part 1: General Considerations Rolfe Birch 11.1 Incidence and Audit. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2 Causes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2.1 Generalised Disorders. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2.2 Alcohol. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
483
485 486 486 487
Contents
xix
11.2.3 Diabetes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2.4 Connective Tissue Disease. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2.5 Warning and consent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2.6 Teaching and Training. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2.7 Specialisation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2.8 Continuity of Care: Timing of Operation. . . . . . . . . . . . . . . . . . . . . . . 11.3 Nerve Lesions in Total Hip Arthroplasty. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.3.1 The Nerve Lesion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.3.2 Findings in the 110 Patients Seen Between 2001 and 2007. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.3.3 Indications for Urgent Reexploration . . . . . . . . . . . . . . . . . . . . . . . . . 11.4 Radiation Neuropathy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.4.1 The Place of Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.5 Prevention of Iatrogenous Lesions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.5.1 Teaching and Training. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.5.2 Audit and Consent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.5.3 Conduct of Affairs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.5.4 Recognition and Action. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
487 488 490 492 493 493 494 494
Iatrogenous Injuries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Part 2: Minimising and Managing Iatrogenous Trigeminal Nerve Injuries in Relation to Dental Procedures Tara Renton 11.6 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.6.1 Signs and Symptoms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.7 Mechanisms of Nerve Injuries. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.7.1 Local Analgesic Related Trigeminal Nerve Injuries . . . . . . . . . . . . . . 11.7.2 Implant Related Nerve Injuries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.7.3 Management of Implant Related Nerve Injuries . . . . . . . . . . . . . . . . . 11.7.4 Endodontic Related Nerve Injury. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.7.5 Third Molar Surgery. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.7.6 Dental Extraction of Other Teeth Proximal to IAN Canal. . . . . . . . . . 11.7.7 Socket Medications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.7.8 Post Operative Infection Related Nerve Injuries. . . . . . . . . . . . . . . . . 11.7.9 Management of Trigeminal Nerve Injuries . . . . . . . . . . . . . . . . . . . . . 11.7.10 Evaluation of Trigeminal Nerve Injuries. . . . . . . . . . . . . . . . . . . . . . . 11.7.11 Possible Interventions and Timing. . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.7.12 Possible Management Tools. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.7.13 Reassurance Counselling/Cognitive Behavioural Therapy. . . . . . . . . 11.7.14 Medical Intervention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.7.15 Surgical Intervention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.7.16 Reasoning for Early Nerve Repair. . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.7.17 Surgical Technique. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.7.18 Medico Legal Issues. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.7.19 Improved Consent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.7.20 Improved Management of These Injuries . . . . . . . . . . . . . . . . . . . . . . 11.7.21 Can We Prevent These Injuries? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
501
12 Pain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.1 Nocicipient. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.1.1 The Gate Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.1.2 Events After Nerve Injury . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.1.3 Central Pain: The Preganglionic Injury of the Brachial Plexus. . . . . . 12.1.4 The Sympathetic Nervous System and Pain . . . . . . . . . . . . . . . . . . . .
527 528 530 530 533 534
494 496 497 498 499 499 499 499 500
501 503 503 503 504 505 507 509 513 513 513 513 513 515 515 515 516 516 517 517 518 521 521 521 521
xx
Contents
12.2 Complex Regional Pain Syndrome (CRPS) Type 1 and (CRPS) Type 2. . . . . . 534 12.2.1 CRPS Type 2 (Causalgia). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 535 12.2.2 The Painful, Stiff, Swollen Part After Fracture or Soft Tissue Injury (CRPS Type I, Reflex Sympathetic Dystrophy). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 536 12.3 Diagnosis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 538 12.3.1 Drugs and Other Measures Short of Operation. . . . . . . . . . . . . . . . . . 538 12.3.2 Indications for Operation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 542 12.3.3 Causalgia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 542 12.3.4 Neurostenalgia. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 549 12.3.5 Post Traumatic Neuralgia. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 551 12.4 Interventions upon the Central Nervous System. . . . . . . . . . . . . . . . . . . . . . . . 555 12.4.1 Interventions on the Central Nervous System: Stimulation. . . . . . . . . 556 12.4.2 Interventions upon the Central Nervous System: Ablation. . . . . . . . . 557 12.5 Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 558 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 558 13 Reconstruction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.1 Requirements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.1.1 The Tension of Transfer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.1.2 Fixed Deformity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.1.3 Correction of Fixed Deformity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.2 Principles of Operation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.2.1 The Equino-varus Deformity of Ankle and Foot. . . . . . . . . . . . . . . . . 13.2.2 Flexor Muscle Slide, by the Technique of Books . . . . . . . . . . . . . . . . 13.2.3 Fixed Extension at the Metacarpo-phalangeal Joints. . . . . . . . . . . . . . 13.2.4 Release of Contracted Small Muscles of the Hand. . . . . . . . . . . . . . . 13.3 Methods of Reconstruction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.3.1 The Shoulder Girdle. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.3.2 Extension of the Elbow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.3.3 Flexion of Elbow. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.3.4 The Operations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.4 Paralysis of the Extensors to the Wrist and Fingers, and of the Abductors and Extensors to the Thumb. . . . . . . . . . . . . . . . . . . . . . . 13.4.1 Pre and Post-operative Care. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.4.2 The Operations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.5 The High Median Palsy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.5.1 Abduction and Opposition of the Thumb. . . . . . . . . . . . . . . . . . . . . . . 13.6 The High Ulnar Palsy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.7 Combined Nerve Lesions and the Hand. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.8 The Lower Limb. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.8.1 The Insensate Foot. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.8.2 Abduction and Flexion at the Hip . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.8.3 The Knee . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.8.4 Foot and Ankle: The Drop Foot from Common Peroneal Palsy. . . . . 13.9 Vascularized Bone and Muscle Transfer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.10 Amputation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.11 Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
563 565 568 569 571 572 573 573 574 575 576 576 580 580 582 584 586 587 589 589 591 593 594 596 596 597 598 601 601 602 603
14 Rehabilitation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 607 14.1 Communication. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 608 14.2 History. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 609
Contents
xxi
14.3 The Rehabilitation Team. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.4 The Method of Work. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.5 Some Technical Aspects. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.6 Motivation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.7 Rehabilitation in Progressive Neurological Disease. . . . . . . . . . . . . . . . . . . . . 14.8 The Choice of Intervention. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.9 Patients with Neuropathic Pain. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.10 Paralysis not Physically Determined. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.10.1 Conversion Paralysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.10.2 The Munchausen Syndrome. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.10.3 Malingering. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.11 Rehabilitation in the War Wounded . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.11.1 Outcomes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.12 The Future. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
610 611 612 616 624 624 625 626 626 627 628 628 629 630 630
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 631
Contributors
Ravi Knight, MB BS MD MRCPCH Department of Clinical Neurophysiology, John Radcliffe Hospital, Oxford, UK Tara Renton, PhD MDSc BDS FRACDS(OMS) FDSRCS MHEA Oral Surgery Department, King’s College London, London, UK Shelagh Smith, BSc MBChB FRCP Department of Clinical Neurophysiology, National Hospital for Neurology and Neurosurgery, London, UK
xxiii
1
The Peripheral Nervous System: Gross Anatomy
Restatement of the facts appears to be warranted by the misconceptions shown by many post graduate students (Last 1949). Definition of the nervous system; separation of the central and peripheral nervous system; gross anatomy of the peripheral nervous system, including cranial and spinal nerves and the autonomic nervous system; nerves particularly at risk from skeletal injury and bleeding. The nervous system is the mechanism through which the organism is kept in touch with its internal structures and external environments and reacts to changes in them. The central nervous system – the brain and its caudal prolongation the spinal cord – is connected to the periphery by the peripheral nervous system. The latter includes the cranial nerves, the spinal nerves with their roots and rami, the peripheral nerves and the peripheral components of the autonomic nervous system, the sympathetic, parasympathetic and enteric divisions (Gardner and Bunge 2005). The peripheral nerves contain motor fibres (to end plates in skeletal muscle), sensory fibres (from organs and endings in skin, muscle, tendon, periosteum, and bone and joint), efferent autonomic fibres (to blood vessels, sweat glands and arrectores pilarum muscle), and visceral afferent fibres. In no other system is so much functional and relay capacity concentrated in so small a volume of tissue. The cervical spinal cord, with a width of about 2 cm and a depth of about 1.5 cm, contains all the apparatus transmitting control of somatic function from the neck down, together with that of control of much visceral function. Because of their greater content of connective tissue, the peripheral nerves have proportionately a lesser functional content, yet severance in an adult’s arm of the median nerve of 5 mm diameter effectively ruins the function of the hand and forearm (Fig. 1.1).
Phrenic n.
Dorsal root ganglion C5 Dorsal root ganglion C6
Fig. 1.1 The fifth, sixth cervical nerves avulsed from the spinal cord. The ventral root is easily distinguishable from the dorsal rootlets. Note the dorsal root ganglion, the dural sleeve merging into the epineurium and the spinal nerve itself. The small pieces of tissue on the proximal ends of the dorsal rootlets (below) are probably portions of the spinal cord.
Ventral root C5 Ventral root C6 Dorsal roots C5
R. Birch, Surgical Disorders of the Peripheral Nerves, DOI: 10.1007/978-1-84882-108-8_1, © Springer-Verlag London Limited 2011
Dorsal roots C6
1
2
Surgical Disorders of the Peripheral Nerves
The cranial and spinal nerves: Twelve pairs of cranial nerves arise from the brain and brain stem. The second of these, the optic nerves, are in fact prolongations of the central nervous system. Thirty-one pairs of spinal nerves – 8 cervical, 12 thoracic, 5 lumbar, 5 sacral and 1 coccygeal – arise from the spinal cord. Each spinal nerve leaves or enters the cord by a ventral, largely motor, root, and a dorsal sensory root (Figs. 1.2 and 1.3). Each sensory root splits into several rootlets as it approaches the spinal cord; these enter the cord along the line of the postero lateral sulcus. The division of the anterior roots into rootlets is less obvious and takes place nearer the cord. Because in the adult the spinal cord extends caudally only so far as the first lumbar vertebral level, the obliquity of the emerging and entering roots in the theca increases from above downwards (Figs. 1.4 and 1.5). The theca below the first lumbar level is occupied by the lumbar, sacral and coccygeal roots forming a leash whose appearance has been likened to that of a horse’s tail (cauda equina). The cell bodies of the fibres forming the anterior roots are mostly situated in the anterior horn of the grey matter of the spinal cord; those of the fibres of the dorsal root are in the dorsal root ganglion, situated in or near the intervertebral foramen. As they approach the foramen, the two roots join to form the spinal nerve, which outside the foramen divides into anterior and posterior primary rami (Fig. 1.6). The autonomic nervous system: Three divisions of the autonomic nervous system – the sympathetic, the parasympathetic and the enteric – are usually described. In each, pre-ganglionic fibres arise from cells in the brain stem or spinal cord. These relay in ganglia to post-ganglionic fibres innervating cardiac muscle, smooth muscle and glands. Most viscera are supplied by both sympathetic and parasympathetic divisions; the cell bodies of the enteric system are confined to the wall of the bowel (Gardner and Bunge 2005). Dorsal root
Dura mater Arachnoid Pia mater
Dorsal root ganglion
Fig. 1.2 The origin of the roots from the cord, their junction just distal to the dorsal root ganglion, and the emergence of the nerve from the spinal cord.
Ventral root
Posterior primary ramus
Fig. 1.3 The division of the spinal nerve (cervical region) into anterior and posterior primary rami.
Anterior primary ramus
Dorsal root ganglion
Spinal cord Dura mater
The Peripheral Nervous System: Gross Anatomy Fig. 1.4 The spine and spinal cord, seen from the front (a) and behind (b). Some of the dorsal bone has been excised. There is increasing obliquity of the roots in relation to the spinal cord from above down. Below the first lumbar level the spinal canal is occupied by the cauda equina. The dorsal root ganglia lie within the intervertebral foramina. The posterior primary rami are generally smaller than the anterior primary rami. The thoracic sympathetic chain is seen (a). The formation of the sacral plexus and the origin of the sciatic nerve is shown (b).
a
Fig. 1.5 The left side of the spinal cord exposed by hemilaminectomy. The cord has temporarily been rotated by stays passed through the denticulate ligaments. The upper most roots (left) are those of the fifth cervical nerve: they emerge at an angle of about 45° to the cord. The lowest roots (right) are those of the seventh cervical nerve, emerging at a very oblique angle to the cord.
3
b
4
Surgical Disorders of the Peripheral Nerves
1.1 The Cranial Nerves (Figs. 1.6–1.9, 1.12 and 1.14–1.16) The first, olfactory, mediates the sense of smell; the second, optic, mediates that of sight. The latter nerve is a prolongation of the central nervous system. The third, fourth and sixth nerves control the muscles concerned with movement of the eye. The fifth, (trigeminal) nerve has an extensive motor and sensory function, controlling the muscles of the jaw and conveying sensation from the skin of the face and the mucosa of the mouth and nose, and probably from the superficial muscles of the face. The lingual branch which conveys sensation from the tongue and buccal mucosa is, with the inferior alveolar nerve, particularly at risk during operations upon the mouth and jaws. Compression of the trigeminal sensory root by branches from the superior cerebellar artery is a common cause of trigeminal neuralgia (Hughes 2005). The seventh (facial) nerve innervates the superficial muscles of the face and neck. It is remarkable for its vulnerability to damage in each of the three parts of its course – intra-cranial, intra-osseous (in the petrous part of the temporal bone) and extra-temporal. The eighth (auditory) nerve mediates the senses of hearing and of balance. The ninth (glosso-pharyngeal) nerve conveys sensibility from the pharynx and from the back of the tongue and has a small motor function. The tenth (vagus) nerve has, as its name suggests, wide ranging branches and functions, most of the latter being parasympathetic. Motor branches innervate the muscles of the larynx, and sensory branches convey sensation from it. Its recurrent laryngeal branch is in the ascending part of its course in close relationship with the trachea and oesophagus and with the thyroid and parathyroid glands (Fig. 1.16a and b).
Spinal accessory n.
Vertebral a.
Dorsal roots C3, C4 and C5 Superior cervical sympathetic ganglion Vagus n. Dorsal root ganglia C5 and C6
Posterior primary ramus of C7
Fig. 1.6 The brainstem and cervical cord exposed by laminectomy. The spinal accessory nerve passes anterior to the dorsal roots, and emerges through the jugular foramen accompanied by the vagus and glosso-pharyngeal nerves. The vertebral artery courses anteriorly to the spinal nerves.
The Peripheral Nervous System: Gross Anatomy Fig. 1.7 The formation of the 11th cranial (cranial and spinal accessory) nerve. Note the temporary “cranialisation” of the spinal component of the nerve.
5
Pons
Vagus n.
Abducens n.
Jugular foramen
Cranial root
Spinal root C1
Accessory n. Foramen magnum
C2
C3 Spinal cord C4
C5
Trigeminal (V) n. Facial (VII) and vestibulocochlear (VIII) n.n. Glossopharyngeal (IX), vagus (X) and accessory (XI) n.n. Spinal part of accessory n.
Vertebral a. Posterior primary ramus C2 Anterior primary ramus C2 Dura Dorsal root ganglion C3
Fig. 1.8 The junction between the spinal cord and the brain stem shown by excision of a posterior bony element. The first cervical nerve passes away from the spinal cord at almost a right angle.
6 Fig. 1.9 (a) The terminal branches of the facial nerve emerging from deep to the parotid gland. The greater auricular nerve courses obliquely across the sternocleidomastoid. (b) The deeper course of the facial nerve is shown after excision of parotid, the spinal accessory nerve is seen crossing the floor of the posterior triangle.
Surgical Disorders of the Peripheral Nerves
a
Greater occipital n. Lesser occipital n. Parotid gland and facial n. branches at anterior border
Great auricular n. External jugular v. Transverse cervical n.
Marginal mandibular branch of facial n. Cervical branch of facial n.
b
Spinal accessory n.
The Peripheral Nervous System: Gross Anatomy
7
The spinal part of the eleventh (accessory) nerve arises from cells in the accessory nucleus, “a column of cells of the somatic efferent type, extending from the second to the fifth and sixth cervical segments of the cord” (Brodal 1981). These cells are in the dorsolateral part of the anterior horn of the grey matter. The fibres emerge segmentally from each side of the cord, to unite to form on each side a nerve which passes rostrally, posterior to the denticulate ligament, into the cranial cavity through the foramen magnum. In the cranial cavity the nerve unites briefly with its cranial part, derived mainly from the cells in nucleus ambiguus, before passing out of the skull with it through the jugular foramen. Outside the skull the two parts separate, the cranial portion going to join the vagus nerve and the spinal part passing obliquely down the neck to innervate the sternocleidomastoid and trapezius muscles. The spinal accessory nerve is particularly at risk to the activities of surgeons in the posterior triangle of the neck yet its course here is consistent. It emerges from beneath the sternocleidomastoid muscle at about 5 mm cephalad to the point where the greater auricular nerve begins its upward course over the anterior face of the muscle. The course of the nerve distal to the posterior triangle has been described by Pereira and Williams (1999). The innervation of the trapezius has been described by Brown (2002) and by Kierner et al. (2001) who showed that the middle and lower parts of the muscle may be partially supplied by branches from the cervical plexus whereas the upper one-third is innervated solely by the spinal accessory nerve. Although Bremner-Smith et al. (1999) thought that 60% of the individual nerve fibres were unmyelinated or finely myelinated, and that many of these were afferent fibres, this observation was not confirmed by Bunting and Standring (2000), who did not detect unmyelinated axons in immunofluorescence light microscopic studies. The 12th (hypoglossal) nerve leaves the skull through the hypoglossal canal in the occipital bone to supply the intrinsic and all but one of the extrinsic muscles of the tongue. Although there are receptor organs in the muscles of the human tongue (Cooper 1953), it is likely that most of the impulses from them travel in the lingual nerve (Weddell et al. 1940). The sensation of taste is mediated by fibres in the facial nerve (anterior two-thirds of the tongue) and by fibres in the glossopharyngeal nerve (posterior one-third of the tongue). In the upper part of the neck the hypoglossal nerve is joined by fibres from the anterior rami of the uppermost two cervical nerves. These soon leave the nerve to form the ansa hypoglossi from which the infrahyoid muscles are supplied.
1.2 The Spinal Nerves 1.2.1 The Anterior Primary Rami The anterior primary rami of the uppermost four cervical nerves unite and branch to form the cervical plexus, through which the skin of the neck and part of the face and some of the muscles of the neck are innervated. A branch of the fourth cervical anterior ramus, with contributions from the third and fifth rami, passes caudally into the thorax as the phrenic nerve, to supply motor fibres to the diaphragm and sensory fibres to the related pleura, fibrous pericardium and peritoneum (Figs. 1.10–1.12). The anterior primary rami of the lowest four cervical nerves and most of that of the first thoracic nerve unite and branch to form the brachial plexus in the lower part of the neck and behind the clavicle (Figs. 1.13–1.16). The upper limb receives its innervation through the branches of this important plexus. The most proximal muscles are supplied by branches from the rami; the intermediate muscles by branches from the trunks and cords; the muscles of the limb itself by branches from the main terminal nerves – the median, ulnar, musculo-cutaneous, radial and circumflex (axillary). There is a segmental pattern to this innervation: the most proximal muscles are supplied by branches of the uppermost rami; the most distal muscles are supplied by branches derived from the eighth cervical and first thoracic nerves. The segmental pattern of innervation is shown more clearly in the cutaneous supply (Figs. 1.17 and 1.18). The cervical root supply has been, as it were, extruded from the supply to the trunk. Thus, in the transition of innervation from the skin of the neck to that of the trunk there is anteriorly a change from the fourth cervical to the second thoracic segment; posteriorly, from the fifth cervical to the first thoracic segment. An important anatomical and functional differentiation of the plexus takes place with the division of the trunks into anterior and posterior divisions. From the anterior divisions the lateral and medial cords are formed; from the posterior divisions the posterior cord is formed. The lateral and medial cords innervate pre-axial (flexor) musculature; the posterior cord innervates post-axial (extensor) musculature. The plexus and the distribution of its nerves vary considerably from one individual to another: the contributions made by the component nerves vary; the origin and method of formation of the main nerves vary; in some cases the contribution of the fifth nerve is large and that of the first thoracic nerve is small, while in others the reverse is the case. The truly autonomous area of cutaneous supply of each main component nerve is small and variable in extent and location. The contributions made
8
Surgical Disorders of the Peripheral Nerves
Fig. 1.10 The right cervical plexus and its (mainly sensory) branches.
C1 Hypoglossal (XII n.) Lesser occipital
C2
Great auricular Transverse cervical
C3
Superior root of ansa cervicalis Inferior root of ansa cervicalis
C4
C5
Supraclavicular n.n.
Phrenic
Ramus to n. to serratus anterior
by the fourth cervical and second thoracic nerves vary: usually their contributions are small, but occasionally the fourth cervical nerve makes an important contribution to the innervation of scapulo-humeral muscles (Figs. 1.19–1.29). The supply of the skin and of the hand is divided between the median, ulnar and radial nerves. The first two supply the palmar aspect; all supply the dorsal aspect. The median nerve supplies the skin of the radial side of the palm, the palmar aspects of the thumb, index and middle fingers and of the radial part of the ring finger, and the terminal parts of the dorsal aspect of these digits. The ulnar nerve supplies the skin of the ulnar side of the palm, the palmar aspects of the little finger and the ulnar part of the ring finger, the dorsal aspect of the ulnar half of the hand, the little and ring fingers and the ulnar side of the proximal part of the middle finger. The radial nerve supplies the radial side of the dorsum of the hand, of the proximal parts of the thumb and index fingers and of the radial side of the middle finger. Damage to the terminal branches of these nerves of cutaneous sensation, which is usually caused by a needle or scalpel, often leads to pain which is quite out of proportion to the functional importance of the nerve. Mok et al. (2006) studied the cutaneous innervation of the dorsum of the wrist and of the hand; Tindall et al (2006) describe a “safe zone” which avoids damage to the dorsal cutaneous branch of the ulnar nerve; Beldner et al. (2005) studied the relations between the lateral cutaneous nerve of the forearm and the superficial radial nerve and MacAvoy et al. (2006) offer practical advice about avoiding damage to the posterior cutaneous nerve of the forearm (Fig. 1.29). Gruber (1870) having dissected the forearm of 125 cadavers, reported an incidence of 15.2% of branches between the median and ulnar nerves in the forearm . Most commonly, a branch runs from the anterior interosseous branch of the median nerve to the ulnar nerve, but Srinivasan and Rhodes (1981) recognize six varieties of this anomaly, the “Martin-Gruber” anastomosis. Bhadra et al. (1999) displayed, by dissection, important variations in the innervation of the flexor digitorum profundus, whilst Branovovacki et al. (1998) and Mahakkanukrauh and Somsarp (2002) reveal different patterns of distribution of the radial nerve. Whilst the variations of the distribution of the peripheral nerves are significant we have found that the variations in the distribution of the spinal nerves forming the brachial plexus are much more important. At least one-third of patients with complete lesions of C5, C6 and C7 retain powerful extension of the digits and this is seen in some patients in whom only the first thoracic nerve has survived.
The Peripheral Nervous System: Gross Anatomy
9
Fig. 1.11 The course of the phrenic nerve. Note the small branch to the pericardium.
Great auricular n. Transverse cervical n. Spinal accessory n. Cervical n. passing to accessory Supraclavicular n.n.
Omohyoid m. Internal jugular v. Nerve to serratus anterior Phrenic n. Lateral pectoral n.
Fig. 1.12 The cervical and brachial plexus. The right supraclavicular nerves are shown. The left brachial plexus is exposed after excision of sternocleidomastoid and the clavicular head of pectoralis major.
10
Surgical Disorders of the Peripheral Nerves
Fig. 1.13 The right brachial plexus. Note the sequence: the anterior primary rami; trunk; divisions; cord; nerves. Note that the trunks are upper, middle and lower, and that the cords are lateral, medial and posterior from their position in relation to the axillary artery which is, in fact, variable.
C3 C4 C5
Supraclavicular n.
C6
Upper trunk
Lateral cord Posterior cord Medial cord Lateral pectoral n. Circumflex n.
C7
Dorsal scapular n. Middle trunk Suprascapular n. Lower trunk
C8 T1 Phrenic n. Nerve to serratus anterior Medial pectoral n.
Axillary a. Thoracodorsal n. Medial cutaneous n. of forearm Radial n. Musculocutaneous n. Median n. Ulnar n.
Vagus n. Transverse cervical a. Suprascapular a. Thoraco-acromial a.
Phrenic n. Suprascapular n.
Fig. 1.14 The brachial plexus revealed by excision of both clavicles. The right subclavian vein has been removed, both left subclavian vessels have been excised.
The Peripheral Nervous System: Gross Anatomy
11
a
b Stellate ganglion Vertebral a.
Suprascapular n. Phrenic n. Vagus n. Recurrent laryngeal n.
Suprascapular a. Subclavian a.
Subclavian a. Common carotid a.
Subclavian v.
Brachiocephalic v.
Fig. 1.15 The right brachial plexus and the great vessels. (a) Left: showing the relations of the brachiocephalic vein and its branches. (b) Right: the subclavian vein has been excised, revealing the subclavian artery passing deep to scalenus anterior.
a
b
Internal carotid a. Vagus n.
Phrenic n. Thyrocervical a.
Superior cervical sympathetic ganglion
Spinal accessory n. Internal jugular v.
Subclavian a. Vagus n. Recurrent laryngeal n.
Thoracic duct
Common carotid a.
Common carotid a.
Thoracic duct
Subclavian a. Recurrent laryngeal n.
Fig. 1.16 The left brachial plexus. (a) Left: sternocleidomastoid has been excised. (b) Right: the clavicle and part of pectoralis major has been removed.
12
Surgical Disorders of the Peripheral Nerves
C3
C3
C4
C4 T2 T3 T4 T5 T6 T7
T3 T4 T2
T2
T10 T11
T5 T6
T8 T9
T2
C5
C5
T1
C6
C6
T7 T1
T8 T9
T12
C6
L1
T10
C6
T11 T12 L1
C8
C7
Fig. 1.17 Approximate distribution of dermatomes on the posterior aspect of the right upper limb (From Aids to the Examination of the Peripheral Nervous System, 4th edition. By kind permission of Dr. Michael O’Brian and Elsevier Ltd).
C7
C8
Fig. 1.18 Approximate distribution of dermatomes on the anterior aspect of the right upper limb (From Aids to the Examination of the Peripheral Nervous System, 4th edition. By kind permission of Dr. Michael O’Brian and Elsevier Ltd).
Upper subscapular n.
Musculocutaneous n.
Circumflex n. Lower subscapular n. Thoracodorsal n.
Latissimus dorsi tendon Median n.
Nerves to triceps
Ulnar n. Nerve to brachialis
Lateral cutaneous n. forearm
Fig. 1.19 The main nerves in the right axilla and arm.
Nerve to medial head triceps
The Peripheral Nervous System: Gross Anatomy Fig. 1.20 The right circumflex (axillary) and suprascapular nerves.
13
Suprascapular n.
Superior transverse scapular ligament
Supraspinatus
Suprascapular n. Joint capsule Infraspinatus
Deltoid Circumflex n.
Teres minor Branches of radial n. to triceps Teres major
Radial n.
Triceps brachii
Suprascapular a. Suprascapular n.
Acromioclavicular joint
Medial head Long head Lateral head
CIRCUMFLEX (AXILLARY) NERVE
Deltoid RADIAL NERVE UPPER CUTANEOUS NERVE OF THE ARM
Posterior circumflex humeral a.
Teres minor
Circumflex n.
Fig. 1.21 The posterior aspect of the right shoulder after excision of trapezius and most of deltoid showing the course of the suprascapular and the circumflex nerves and their vessels.
Fig. 1.22 The right (axillary) circumflex nerve (From Aids to the Examination of the Peripheral Nervous System, 4th edition. By kind permission of Dr. Michael O’Brian and Elsevier Ltd).
14
Coracobrachialis MUSCULOCUTANEOUS NERVE
Surgical Disorders of the Peripheral Nerves
CIRCUMFLEX (AXILLARY) NERVE Triceps, long head Triceps, lateral head
Biceps Triceps, medial head Brachialis
Brachioradialis
RADIAL NERVE
Extensor carpi radialis longus Extensor carpi radialis brevis Supinator
POSTERIOR INTEROSSEOUS NERVE (deep branch)
Extensor carpi ulnaris Extensor digitorum Lateral cutaneous nerve of the forearm
Extensor digiti minimi Abductor pollicis longus Extensor pollicis longus Extensor pollicis brevis Extensor indicis
Fig. 1.23 The right musculocutaneous nerve (From Aids to the Examination of the Peripheral Nervous System, 4th edition. By kind permission of Dr. Michael O’Brian and Elsevier Ltd).
SUPERFICIAL RADIAL NERVE
Fig. 1.24 The right radial nerve (From Aids to the Examination of the Peripheral Nervous System, 4th edition. By kind permission of Dr. Michael O’Brian and Elsevier Ltd).
The Peripheral Nervous System: Gross Anatomy
15
Fig. 1.25 The anterior aspect of the right elbow. Lateral cutaneous n. of the forearm
Brachioradialis m. Posterior interosseous n.
Superficial radial n. Flexor digitorum superficialis m. (radial head cut) Anterior interosseous n. Nerve to flexor pollicis longus
Ulnar n. Brachial a. Median n. Nerve to pronator teres Nerve to palmaris longus, flexor dig. sublimis and flexor carpi radialis Nerve to flexor carpi ulnaris Nerve to flexor digitorum profundus Nerves to flexor digitorum sublimis
Nerve to flexor digitorum profundus Ulnar n. and a.
Flexor pollicis longus
Flexor carpi ulnaris Flexor digitorum profundus
16
Surgical Disorders of the Peripheral Nerves
Sensory Dorsal cutaneous branch
MEDIAN NERVE
ANTERIOR INTEROSSEOUS NERVE
Flexor carpi radialis Palmaris longus Flexor digitorum superficialis
Deep motor branch Superficial terminal branches
Flexor digitorum profundus I & II Flexor pollicis longus
Flexor pollicis brevis Opponens pollicis
Flexor carpi ulnaris Flexor digitorum profundus III & IV MEDIAL CUTANEOUS NERVE OF THE FOREARM
Motor Abductor pollicis brevis
MEDIAL CUTANEOUS NERVE OF THE ARM
Palmar cutaneous branch
Pronator teres
Pronator quadratus
ULNAR NERVE
Palmar branch Motor
Flexor retinaculum
Adductor pollicis Flexor pollicis brevis
Abductor Opponens digiti Flexor minimi
1st Dorsal interosseous
First lumbrical
1st Palmar interosseous
Second lumbrical
Third lumbrical
Fourth lumbrical
Sensory
Fig. 1.26 The right median nerve (From Aids to the Examination of the Peripheral Nervous System, 4th edition. By kind permission of Dr. Michael O’Brian and Elsevier Ltd).
Fig. 1.27 The right ulnar nerve (From Aids to the Examination of the Peripheral Nervous System, 4th edition. By kind permission of Dr. Michael O’Brian and Elsevier Ltd).
The Peripheral Nervous System: Gross Anatomy
17
Fig. 1.28 The median and ulnar nerves in the left hand. Inset shows the normal course of median nerve at the wrist and also the palmar cutaneous nerve.
Common palmar digital n. Median n. dividing into common palmar digital branches Recurrent branch of median n. Flexor carpi radialis m. Radial a.
Superficial palmar arch Nerve to hypothenar m.m. Deep branch of ulnar n. Superficial branch of ulnar n. Ulnar n. Ulnar a. Flexor carpi ulnaris m.
Median n.
Recurrent branch of median n.
Flexor retinaculum
Palmar cutaneous branch of median n. Median n. Flexor carpi radialis m. Flexor digitorum superficialis m.
18 Fig. 1.29 The terminal branches of the superficial radial and lateral cutaneous nerves of forearm seen during operation at the right wrist.
Surgical Disorders of the Peripheral Nerves Superficial radial n.
Extensor carpi radialis longus
Lateral cutaneous n. forearm
Extensor carpi radialis brevis
1.2.2 Thoracic Anterior Primary Rami The second to the sixth thoracic anterior primary rami innervate the intercostal muscles and the skin of the anterior and lateral chest wall. Most of the first nerve goes to join the brachial plexus; most of the second goes as the intercosto-brachial nerve to innervate the skin of the axilla and of the medial side of the arm. The lower six thoracic anterior rami continue from the intercostal spaces to the anterior wall of the abdomen, innervating its skin and muscles. The lowest nerves supply sensory fibres to the lateral part of the diaphragm. The lowest (12th) thoracic ventral ramus, sometimes called the subcostal nerve, is larger than the others and connects with the ilio-hypogastric branch of the first lumbar nerve.
1.2.3 Lumbar and Sacral Anterior Primary Rami (Fig. 1.30) The first lumbar anterior primary ramus gives rise to two mainly cutaneous nerves and part of a third. The iliohypogastric, iloinguinal and genitofemoral nerves supply respectively the skin of part of the buttock, of the groin and the greater part of the external genitalia. The second, third and fourth lumbar anterior rami unite and branch to form the lumbar plexus from which arise the nerves innervating the skin of the thigh and its anterior and medial muscles. The plexus is formed in the anterior part of the psoas major muscle, in the posterior wall of the abdomen. Its terminal branches lie under the parietal peritoneum. Some of these emerge lateral and some medial to the psoas major. The most important terminal branch is the femoral nerve, which passes, lateral to the psoas major and femoral vessels, under the inguinal ligament to reach the upper part of the thigh (Figs. 1.31 1.32 and 1.39). Through its anterior and posterior divisions it supplies the skin of the anterior surface of the thigh and the quadriceps and sartorius muscles. The saphenous branch of the posterior division descends with the femoral artery to emerge from the femoral canal above the knee and supply the skin of the medial side of the leg and foot. The obturator nerve emerges medial to the psoas major and, passing along the lateral wall of the pelvis, emerges into the thigh through the obturator foramen. Through anterior and posterior branches the adductor muscles and the skin of the medial side of the thigh are supplied. Part of the fourth lumbar ramus and all the fifth ventral ramus join to form the lumbo-sacral trunk, which emerges medial to the psoas major to enter the pelvis and join the first, second and third sacral nerves to form the sacral plexus on the posterolateral wall of the pelvis (Figs. 1.33 and 1.34).
The Peripheral Nervous System: Gross Anatomy
19
Ganglionated chain
L1
L1
L2
L2
Iliohypogastric n. L3
L3
Ilioinguinal n. L4
L4
Genitofemoral n. Psoas muscle
L5
L5
Femoral n. Lumbosacral trunk
S1
Lateral cutaneous n. thigh Femoral n.
S2
Obturator n.
S3 S4
Pudendal n.
Fig. 1.30 The femoral and sacral plexuses and the ganglionated sympathetic chain.
Sciatic n.
20
Surgical Disorders of the Peripheral Nerves
Fig. 1.31 The left femoral nerve. Lateral cutaneous n. of thigh
Femoral n. Femoral a. Medial circumflex femoral a. Lateral circumflex femoral a. Anterior branch of obturator n. Medial cutaneous nerve of thigh 2nd perforating a. 3rd perforating a. Saphenous n. Femoral a. (cut)
Femoral a. (cut) Ascending branch of lateral circumflex femoral a. Profunda femoris 1st perforating a. Intermediate cutaneous nerve of thigh Descending branch of lateral circumflex femoral a.
The Peripheral Nervous System: Gross Anatomy
21
Fig. 1.32 The left femoral nerve.
Femoral n.
Nerves to iliacus
External iliac a. External iliac v. Obturator n.
a
1st sacral n.
Lumbosacral trunk Genitofemoral n.
Sympathetic trunk 2nd sacral n. 3rd sacral n.
Femoral n.
4th sacral n. Pudendal n.
Fig. 1.33 The relations of the left sacral plexus. (a) The female pelvis. (b) The male pelvis.
Obturator n.
22 Fig. 1.33 (continued)
Surgical Disorders of the Peripheral Nerves
b
1st sacral n.
Lumbosacral trunk Genitofemoral n.
Sympathetic trunk 2nd sacral n. Femoral n.
3rd sacral n. 4th sacral n.
Obturator n.
Pudendal n.
Union of ventral rami of S2 and S3
Ventral ramus of S1
Ventral ramus of S4
Fig. 1.34 The left sacral plexus.
Sacrospinous ligament
Lumbosacral trunk
The Peripheral Nervous System: Gross Anatomy
23
The innervation of the perineum and most of the lower limb is derived from the branches of this plexus (Figs. 1.37 and 1.38). The sciatic nerve, the largest in the body, leaves the pelvis through the greater sciatic foramen and passes behind the hip joint into the back of the thigh. This great trunk has two main components, which are functionally and often anatomically quite distinct (Figs. 1.35, 1.36, 1.39 and 1.40). The tibial nerve innervates the medial hamstrings, it descends in the midline through the popliteal fossa into the back of the leg, to supply its superficial and deep muscles. It has an important and frequently useful branch, the sural
Gluteus maximus (reflected) Superior gluteal n. Inferior gluteal n. Pudendal n. Sciatic n. Posterior cutaneous n. of thigh
Tibial n. Common peroneal n. Nerves to lateral and medial gastrocnemius Nerve to popliteus Nerves to soleus Sural n.
Sural communicating branch
Tibial n.
Medial calcaneal n.
Fig. 1.35 The right sciatic nerve and its major components in the lower limb.
24
Surgical Disorders of the Peripheral Nerves
Superior gluteal n.
T10
Common peroneal division of sciatic trunk
L1 T11
L2 T12
Inferior gluteal n.
L1
S5
S4
S3
S3
Pudendal n.
S4
Posterior cutaneous n. of thigh
L2
L2 S2
L3
L3
Tibial division of sciatic trunk
S2 L4
Fig. 1.36 The right sciatic nerve and its two main components, the common peroneal and tibial nerves. The CPN is quite distinct and in this specimen it passes separately through piriformis muscle, a not uncommon variation.
L5 L4 L5
S1
L3
L3
S2
S1 L5
S2
S1 L5
S4 S3
S3 S5
L2
L2
L2
L2
Fig. 1.38 Approximate distribution of dermatomes on the perineum (From Aids to the Examination of the Peripheral Nervous System, 4th edition. By kind permission of Dr. Michael O’Brian and Elsevier Ltd).
Fig. 1.37 Approximate distribution of dermatomes on the right lower limb (From Aids to the Examination of the Peripheral Nervous System, 4th edition. By kind permission of Dr. Michael O’Brian and Elsevier Ltd).
The Peripheral Nervous System: Gross Anatomy Fig. 1.39 The nerves on the anterior aspect of the right lower limb (From Aids to the Examination of the Peripheral Nervous System, 4th edition. By kind permission of Dr. Michael O’Brian and Elsevier Ltd).
25
Iliacus FEMORAL NERVE
LATERAL CUTANEOUS NERVE OF THE THIGH
OBTURATOR NERVE
Cutaneous branch Adductor brevis MEDIAL CUTANEOUS NERVE OF THE THIGH
Rectus femoris Quadriceps Vastus lateralis femoris Vastus intermedius Vastus medialis INTERMEDIATE CUTANEOUS NERVE OF THE THIGH
Adductor longus Gracilis
Adductor magnus
COMMON PERONEAL NERVE SUPERFICIAL PERONEAL NERVE Peroneus longus Peroneus brevis LATERAL CUTANEOUS NERVE OF THE CALF Peroneus tertius
Extensor digitorum brevis
DEEP PERONEAL NERVE Tibialis anterior Extensor digitorum longus Extensor hallucis longus SAPHENOUS NERVE
26 Fig. 1.40 The nerves on the posterior aspect of the right limb (From Aids to the Examination of the Peripheral Nervous System, 4th edition. By kind permission of Dr. Michael O’Brian and Elsevier Ltd).
Surgical Disorders of the Peripheral Nerves
Gluteus medius Gluteus minimus SUPERIOR GLUTEAL NERVE Piriformis
SCIATIC NERVE
Tensor fasciae latae INFERIOR GLUTEAL NERVE Gluteus maximus
Semitendinosus Semimembranosus Adductor magnus
TIBIAL NERVE
POSTERIOR CUTANEOUS NERVE OF THE THIGH Biceps, long head Biceps, short head
COMMON PERONEAL NERVE
Gastrocnemius, medial head Soleus
Gastrocnemius, lateral head
Tibialis posterior Flexor digitorum longus
Flexor hallucis longus
TIBIAL NERVE SURAL NERVE CALCANEAL BRANCH
MEDIAL PLANTAR NERVE to: Abductor hallucis Flexor digitorum brevis Flexor hallucis brevis Cutaneous branches
LATERAL PLANTAR NERVE to: Abductor digiti minimi Flexor digiti minimi Adductor hallucis Interossei Cutaneous branches
The Peripheral Nervous System: Gross Anatomy
27
nerve, which arises in the upper part of the popliteal fossa, descends between the two heads of gastrocnemius and pierces the deep fascia in the proximal part of the leg (Figs. 1.41 and 1.42). Usually, a branch from the common peroneal (or fibular) nerve joins the sural nerve; at times, it is larger than the contribution from the tibial nerve. Rarely, the sural nerve arises wholly from the common peroneal division. The nerve then descends to pass lateral to the tendo Achilles to supply the skin on the lateral side of the foot. The tibial nerve continues into the foot behind the medial, tibial, mallelous and through its terminal medial and lateral plantar branches supplies the intrinsic muscles of the foot and the skin of the sole (Fig. 1.44a). The common peroneal nerve innervates the lateral hamstring muscles in the thigh. It diverges laterally from the mid line to pass behind the head of the fibula and lateral to its neck. Here it divides into deep and superficial peroneal nerves. The former passes into the anterior compartment of the leg to innervate the anterior muscles and finally to supply the extensor digitorum brevis and the skin of the dorsum of the first interdigital space (Figs. 1.43 and 1.44b). The superficial peroneal (musculocutaneous) nerve passes deep to the upper part of the peroneus longus muscle to supply both peronei. Its continuation pierces the deep fascia in the distal part of the leg to supply the skin of the dorsum of the foot and anterolateral part of the ankle.
Popliteal a.
Sciatic n.
Tibial n. Common peroneal n. Superior lateral genicular a. Superior medial genicular a. Saphenous n. Inferior medial genicular a.
Sural communicating branch Inferior lateral genicular a. Nerve to popliteus
Tibial n.
Sural n.
Fig. 1.41 The right popliteal fossa.
28
Surgical Disorders of the Peripheral Nerves
Fig. 1.42 The right popliteal fossa.
Tibial n. Sartorius m. Gracilis m. Semitendinosus m. Nerve to medial gastrocnemius m.
Common peroneal n. Plantaris m. Lateral head of gastrocnemius m.
Semimembranosus m. Popliteal a.
Medial head of gastrocnemius m.
Sural n.
The more proximal branches of the sacral plexus supply the gluteal muscles and the skin and muscles of the perineum (Figs. 1.35, 1.36 and 1.38). The superior gluteal nerve, emerging above the piriformis muscle supplies the short gluteus medius and minimus and the tensor fasciae latae. The inferior gluteal nerve, emerging below the piriformis muscle, supplies the gluteus maximus muscle. Kampa et al. (2007) have described an internervous “safe zone” for exposure of the capsule of the hip. The pudendal nerve leaves the pelvis through the greater sciatic foramen and, entering the pudendal canal through the lesser sciatic foramen, passes into the perineum to innervate its skin and muscles. As in the case of the upper limb, there is a segmental innervation of the muscles and, more easily seen, of the skin. Again, the segments innervating the limb have been extruded from the innervation of the trunk and perineum, so that in the transition from trunk to perineum there is posteriorly a segmental change from the third lumbar to the third sacral dermatome. The skin of the foot is supplied by all the main nerves of the lower limb save the obturator (Figs. 1.44–1.46). The plantar surface is supplied by the tibial nerve through its plantar branches; the medial side by the saphenous branch of the femoral nerve; the lateral side by the sural branch of the tibial nerve, and the dorsum by the superficial and deep divisions of the common peroneal nerve. Apparently trivial injuries to the terminal branches of the nerves of cutaneous sensation sometimes cause even more trouble in the lower than in the upper limbs. Tennent et al. (1998) describe painful neuromas of the infrapatellar branch of the saphenous nerve whilst MacNicol and Kelly (2002) provide an ingenious method for preventing this by transillumination during arthroscopic work. Solan et al. (2001) advise against the conventional dorsal incision over the hallux because of the constant dorso medial branch which arises from the superficial peroneal nerve.
The Peripheral Nervous System: Gross Anatomy
29
Fig. 1.43 The course relations and branches of the right common peroneal nerve.
Lateral sural cutaneous n. Common peroneal n. Lateral sural cutaneous n.
Deep peroneal n. Superficial peroneal n.
Sural n. Deep peroneal n. Superficial peroneal n.
Fascia of the leg
Lateral calcaneal branches
Dorsal branch of plantar digital n. of great toe
Deep peroneal n., cutaneous branch
30 Fig. 1.44 (a) Above: the medial aspect of the right ankle. (b) Below: the lateral aspect of the left ankle and heel.
Surgical Disorders of the Peripheral Nerves
a Posterior tibial a. Medial calcaneal n.n.
Tibialis posterior m. Flexor digitorum longus m. Flexor hallucis longus m. Medial plantar n.
Lateral plantar n.
b Short saphenous v. Sural n.
Superficial peroneal n.
Peroneus longus m. Anterior tibial a. and deep peroneal n.
Peroneus brevis m.
The Peripheral Nervous System: Gross Anatomy
Dorsal branch of plantar digital n. of great toe
31
Proper plantar digital n.n.
Common plantar digital n.n. Flexor hallucis longus Flexor digitorum longus
Medial plantar n.
Lateral plantar n., superficial branch Lateral plantar n., deep branch
Lateral plantar n.
Dorsal branches of plantar digital n. great toe
Flexor hallucis longus Flexor digitorum longus
Medial plantar n. Lateral plantar n. Flexor digitorum brevis Tibial n.
Fig. 1.45 The sole of the left foot.
Plantar aponeurosis
Fig. 1.46 The sole of the left foot.
32
Surgical Disorders of the Peripheral Nerves
1.2.4 The Posterior Primary Rami The posterior primary rami are usually smaller than the anterior. Most divide into medial and lateral branches to supply the muscles and skin of the posterior part of the neck and trunk. They do not enter the limbs. The posterior primary rami of the uppermost three spinal nerves differ from those of the lowest five in extending their supply to the back of the head (Fig. 1.47). The posterior ramus of the first nerve, actually larger than the anterior ramus, chiefly supplies muscles between the atlas and occiput. The posterior ramus of the second cervical nerve is the largest of the cervical posterior rami and larger than its anterior ramus. Emerging between the posterior arch of the atlas and the lamina of the axis, it divides into a large medial and a smaller lateral branch. The former goes on as the greater occipital nerve to innervate the skin of the back of the scalp. At its beginning, it is in close relationship with the back of the atlanto-axial joint. The third cervical posterior ramus provides a third occipital branch. The posterior rami of the lowest five cervical nerves innervate the posterior vertebral muscles; the medial branches of the fourth and fifth rami also innervate the skin. The thoracic posterior primary rami similarly pass posteriorly close to the posterolateral (zygapophyseal) intervertebral joints to supply posterior vertebral muscles and the skin of the back of the chest. The lumbar posterior rami are similarly disposed, but only the uppermost three reach the skin. The sacral posterior rami are small, having a small distribution to muscle and to the skin over the sacrum.
Vertebral a.
Atlantoaxial facetal joint
Greater occipital n.
Fig. 1.47 The uppermost right three cervical vertebrae and the greater occipital nerve.
Lower articular process of C3
Vertebral a.
The Peripheral Nervous System: Gross Anatomy
33
1.3 The Autonomic Nervous System Conventionally, the autonomic nervous system is divided into two parts – the sympathetic and the parasympathetic nervous systems. Langley (1903, 1926) proposed the enteric nervous system as a third component. Indeed, the complexity and extent of the innervation of the viscera are such as to make this further division attractive (Furness and Costa 1980). Both systems are characterized by having relays between their cells of origin and their terminations: in the case of the sympathetic system the relays are in paravertebral or axial ganglia; in the case of the parasympathetic system they are in or near the organs innervated (Fig. 1.48).
Afferent Dorsal root Sympathetic chain Dorsal horn (sensory neurone) ganglion and ganglion
Blood vessel
Ventral horn Fig. 1.48 Efferent (red) and afferent (green) autonomic paths in the spinal cord and ganglionic chain.
Ventral Efferent (motor neurone) root
Grey ramus
Skin
Intestine
Splanchnic ganglion (coeliac) White ramus
34
Surgical Disorders of the Peripheral Nerves
1.3.1 The Sympathetic System The pre-ganglionic cells of the efferent fibres of the sympathetic system are in the lateral horn of the grey matter of the spinal cord from the first thoracic to the second lumbar level. Most of the ganglia are in the paravertebral sympathetic chains extending from the top to the bottom of the spinal column; others lie in autonomic plexi in the abdominal cavity. Usually there are on each side 2 cervical, 1 cervico-thoracic (stellate), 11 thoracic, 4 lumbar and 5 sacral or pelvic ganglia. Preganglionic myelinated fibres enter the cervico-thoracic, thoracic and upper two lumbar ganglia in white rami from the first thoracic to the second lumbar spinal nerve. These fibres relay in the corresponding ganglia or proceed up or down the chain to relay in other ganglia of the chain or in one of the ganglia of the autonomic plexi. The distribution to the spinal nerves is by way of grey rami, which contain unmyelinated fibres, from the corresponding paravertebral ganglia. Fibres pass directly from the autonomic plexi to their destinations. Afferent fibres have their cells in the posterior root ganglia; their sites of relay are not clearly identified. The sympathetic supply to the head and neck arises mainly from the uppermost three thoracic segments, passes cranially, and relays in the cervical ganglia to be distributed to vessels and sweat glands and in particular to the dilator of the pupil, and the smooth muscle fibres in orbitales and levator palpebrae superioris muscles. Most of the supply to these muscles of the eye arises from the first thoracic segment. The sympathetic supply to the upper limb arises principally from the second to the sixth thoracic segments. Fibres pass up the chain to the middle cervical and cervicothoracic stellate ganglia, where they relay to be distributed by grey rami to the nerves of origin of the brachial plexus (Figs. 1.49 and 1.50). The sympathetic supply to the lower limbs arises from the lowest three thoracic and uppermost two lumbar segments. Fibres enter the first and second lumbar ganglia by white rami, descend in the chain, relay in the lumbar and sacral ganglia and are distributed by grey rami to the lumbar and sacral nerves (Figs. 1.51 and 1.52).
Stellate ganglion
8th cervical n. Lower trunk of plexus
C7
1st intercostal n. 1st posterior intercostal vessels
T1
Kuntz n. Rib 1 T2
T3
Ventral ramus T1
Rib 2
Ventral ramus C8
Rib 3
Lower trunk
Sympathetic chain and ganglion
T4 Rib 4 Superior intercostal v.
Superior costotransverse ligament
Fig. 1.49 The relations of the cervico thoracic (stellate) ganglion.
Unusually large communication between T2 and T1
Fig. 1.50 The relations of the cervico thoracic (stellate) ganglion.
The Peripheral Nervous System: Gross Anatomy Fig. 1.51 The autonomic nerves in the chest and abdomen.
35
Common carotid a. Internal jugular v. Vagus n. Thoracic duct Subclavian v. Intercostal n.n.
Arch of aorta
Azygos v. Sympathetic trunk and ganglia Branches from sympathetic trunk to the greater splanchnic n.
Thoracic duct
Aorta
Greater splanchnic n.
Coeliac plexus Superior mesenteric plexus
Inferior mesenteric plexus
Superior hypogastric plexus
36
Surgical Disorders of the Peripheral Nerves Coeliac plexus
Superior mesenteric plexus Subcostal n. Iliohypogastric n. Inferior mesenteric plexus
Ilioinguinal n. Genitofemoral n.
Superior hypogastric plexus
Inferior hypogastric plexus and pelvic splanchnic n.
Fig. 1.52 The autonomic nerves in the abdomen and pelvis.
1.3.2 The Parasympathetic Nervous System The efferent outflow of the parasympathetic nervous system arises from nuclei in the mid-brain and part of the hind brain and the sacral part of the spinal cord. The pre-ganglionic fibres are distributed by the third, seventh, ninth and tenth cranial nerves and by the second to the fourth sacral spinal nerves. From the last arise the pelvic splanchnic nerves (nervi evigentes) which supply the ganglia in which the pre-ganglionic fibres relay are in or near the organs supplied. The effect of parasympathetic activity is inhibitory in the heart, motor to the muscle of the bladder and bowel and dilator in small vessels. The central control of both sympathetic and parasympathetic function is exercised from nuclei in the hypothalamus which themselves receive input from higher centres. The fibres from the hypothalamus almost certainly descend in a column in the lateral part of the white matter of the spinal cord (Nathan and Smith 1987). Parasympathetic fibres to the papillary and ciliary muscles pass with the oculomotor (III) cranial nerve via the ciliary ganglion, those to the lacrimal, submandibular and sublingual salivary glands travel with the facial (VII) cranial nerve via the submandibular ganglion. Those to the parotid gland are conveyed by the glossopharyngeal (IX) cranial nerve. The main visceral plexi are the cardiac, pulmonary oesophageal, coeliac, mesenteric and hypogastric (Fig. 1.52). They are fed from the cervical and cervico-thoracic ganglia, from the middle and lower thoracic ganglia (the thoracic splanchnic nerves), from the lumbar ganglia (the lumbar splanchnic nerves) and from the sacral ganglia. Both sympathetic and parasympathetic systems contribute to these plexi, the vagus (tenth cranial) nerve being the principal source of parasympathetic fibres to the chest and abdomen, and the second, third and fourth sacral nerves to the pelvis. The effect of efferent sympathetic activity is to cause sweating, to constrict small vessels and to cause contraction of the arrectores pilarum muscle. The visceral actions are to stimulate the action of the heart and to cause sphincteric contraction.
The Peripheral Nervous System: Gross Anatomy
37
1.3.3 Afferent Autonomic Pathways Autonomic afferent fibres have their cells of origin in the dorsal root ganglia and in some cranial ganglia. Their peripheral processes run with efferent fibres, terminating in receptor endings in the walls of viscera and vessels. Impulses from these endings do not necessarily obtrude on consciousness, but evidently mediate sensations such as hunger, distension of the bladder and perhaps pain. Most, presumably, are concerned with the initiation of visceral reflexes. Schott (1994) re-examined the role of afferent autonomic pathways in the transmission of painful impulses, and has proposed a return to Langley’s (1903) concept of the unitary nature of the sensory system. In particular, he proposes the inclusion of autonomic afferents from the peripheral organs such as blood vessels with those from organs in the head, neck, thorax and abdomen under the general head “visceral afferents.” Edgar (2007) reviewed the considerable evidence showing that there is dual innervation to the lumbar intervertebral disc. The somatic afferent pathway enters the adjacent dorsal root segmentally via the sinu-vertebral nerves. The other pathway is non segmental and sends fibres through the paravertebral sympathetic chain which re enter through the thoraco lumbar white rami communicantes (Suseki et al. 1998). The nociceptor terminals from the spine have been mapped by Bucknill et al. (2002) by detecting sodium channels Nav 1.8 and Nav 1.9 which are associated with pain fibres. The afferent component of the enteric division is described by Furness (2006).
1.4 Nerves at Risk from Musculo Skeletal Injury The anatomical arrangements of some of the peripheral nerves make them particularly vulnerable to damage from skeletal injury. The sacral nerves are particularly at risk in fractures involving the sacral foramina. The proximity to bone of the main nerves at the elbow render all three vulnerable to skeletal injury (Fig. 1.53). The sciatic trunk is damaged by the posterior displaced head of the femur. Fleming et al. (2004) showed, by “video extensometry” that strain on the sciatic nerve was increased by 26% with the hip flexed to 45° and the knee extended. The course of the circumflex nerve is described by Coene (1985) “in the axillary region the nerve has a free course in loose fatty tissue, but when it turns around the subscapularis muscle, it is captured in a tunnel formed by the fascia of the subscapularis muscle cranially, the teres major muscle caudally and the coraco-brachialis muscle laterally. The fascii of these muscles join at the entrance to the quadrilateral tunnel, where they firmly surround the hitherto independent axillary nerve and posterior circumflex vessels which meet the nerve at this point. This neurovascular complex turns around the inferior border of the subscapular tendon, it passes cranially over the superior border of the teres major tendon, and enters a horizontal ‘tube’.” This arrangement puts the nerve at risk by forward displacement of the head of humerus and bleeding from the posterior circumflex vessels strangles the nerve. The radial nerve is at risk from fractures of the shaft of the humerus between the two relatively fixed points of the nerves to the lateral head of triceps and the tunnel through the inter muscular septum (Lambert 2005). The common peroneal nerve, which passes above or through the piriformis in as many as 30% of cases, is tethered above in relationship to the piriformis and below at the neck of the fibula. The fascia surrounding the biceps femoris muscle and tendon sweeps around to embrace the nerve. The deep peroneal nerve passes rather acutely forward to enter the anterior compartment of the leg. Sleeves of fascia surround main nerves and main vessels in some regions, an arrangement which predisposes the nerve to injury from ischaemia or compression or both, from bleeding. The anterior primary rami of C7, C8, and T1 are enclosed in quite a rigid space after they enter the posterior triangle of the neck. This is bounded, posteriorly, by the dorsal part of the first rib, the transverse processes of the cervical vertebrae and by the fascia of the levator scapulae muscle. The nerves are embraced by the scalenus anterior and scalenus medius muscles, both of which are invested in an unyielding fascia. This is one envelope of the prevertebral fascia which also serves to bind the phrenic nerve down to the anterior face of scalenus anterior. The prevertebral fascia is particularly well developed in front of the vertebral column and also at the base of the posterior triangle where it envelops C7, C8, and T1, the phrenic nerve, the cervical sympathetic chain, and the subclavian and vertebral arteries. Infusion of relatively large volumes of fluid, from 10 to 20 mL, deep to the prevertebral fascia for the
38
Surgical Disorders of the Peripheral Nerves
purpose of inducing regional block may cause tamponade of the radicular vessels which enter the spinal canal and contribute to the anterior spinal artery. Wilbourn (2005) has described the medial brachial fascial compartment which extends from the axilla to the elbow, and is bounded by the tough medial intermuscular septum and the axillary sheath. Wilbourn suggests that bleeding into this compartment is responsible for the majority of infraclavicular plexopathies following regional block and possibly for many of the neurovascular injuries which result from closed or penetrating missile injuries into this region. The anterior interosseous nerve and its accompanying artery may be damaged by compression because of swelling in the deepest part of the flexor compartment of the forearm. The ulnar nerve is accompanied by the ulnar artery, in a discrete fascial compartment in the distal two-thirds of the forearm. Donaghy (2005) suggests that the femoral nerve is more seriously damaged by haematoma where it passes deep to the thick fascia over the iliacus muscle. The nerve is especially at risk from bleeding into the femoral triangle (Fig. 1.54). The deep peroneal nerve is accompanied by the anterior tibial artery, an end artery, throughout most of the anterior compartment of the leg. The tibial nerve is accompanied by the posterior tibial artery in the distal one-half of the leg in a sheath of fascia similar to the arrangements for the ulnar vessels and nerve. The collateral circulation at the knee is particularly poor (Figs. 1.55 and 1.56).
Superior ulnar collateral a.
Brachial a.
Radial recurrent a.
Common interosseous a.
Anterior ulnar recurrent a.
Radial a.
Ulnar a.
Fig. 1.53 The collateral vessels at the elbow accompany the nerves. They must be compromised if the nerves are injured. The specimen is the left elbow.
Anterior interosseous a.
The Peripheral Nervous System: Gross Anatomy
39
Fig. 1.54 The femoral nerve and vessels deep to the inguinal ligament. The drawing shows the left inguinal region.
Lateral femoral cutaneous nerve Inguinal ligament Femoral nerve Femoral branch of genitofemoral nerve Iliacus m. Femoral artery Femoral vein Femoral canal Pectineus m. Psoas major m.
40
Surgical Disorders of the Peripheral Nerves
Fig. 1.55 The collateral circulation about the knee is poor. The drawing shows the left knee.
Descending genicular a. Articular branch of descending genicular a. Saphenous branch of descending genicular a. Medial superior genicular a.
Descending branch of lateral circumflex femoral a.
Lateral superior genicular a.
Common peroneal n.
Patellar tendon
Lateral inferior genicular n.
Medial inferior genicular a.
Circumflex fibular a.
Anterior tibial a.
Anterior tibial recurrent a. Deep peroneal n. Superficial peroneal n.
a
Superior lateral genicular a.
b
Superior medial genicular a. Popliteal a. Popliteal a.
Superior lateral genicular a. Inferior lateral genicular a.
Anterior tibial a. Anterior tibial a.
Inferior lateral genicular a.
Tibiofibular trunk
Inferior medial genicular a.
Fig. 1.56 The popliteal and the anterior tibial arteries. Lateral view (a) (left) posterior view (b) (right). In this specimen the anterior tibial artery passes above the popliteus muscle. Other variations are common. The specimen is the right knee.
The Peripheral Nervous System: Gross Anatomy
41
References Beldner S, Zlotolow DA, Melone CP, Agnes AM, Jones MH (2005) Anatomy of the lateral antebrachial cutaneous and superficial radial nerves in the forearm: a cadaveric and clinical study. J Hand Surg 30A:1226–1230 Bhadra N, Keith MW, Peckham PH (1999) Variation of innervation of flexor digitorum profundus muscle. J Hand Surg 24A:700–703 Branovovacki G, Hanson M, Cash R, Gonzalez M (1998) The innervation pattern of the radial nerve at the elbow and in the forearm. J Hand Surg 23B:167–169 Bremner-Smith AT, Unwin AJ, Williams WW (1999) Sensory pathways in the spinal accessory nerve. J Bone Joint Surg 81B:226–228 Brodal A (1981) The spinal accessory nerve. In: Neurological anatomy. Oxford University Press, New York/Oxford, p 458 Brown H (2002) Anatomy of spinal accessory nerve plexus: relevance to head and neck cancer and atherosclerosis. Exp Biol Med 227:570–578 Bucknill AT, Coward K, Plumpton C, Tate S, Bountra C, Birch R, Sandison A, Hughes SP, Anand P (2002) Nerve fibres in lumbar spine structures and injured spinal roots express the sensory neurones specific sodium channels SNS/PN3 NaN/SNS2. Spine 27:135–140 Coene LNJEM (1985) Axillary nerve lesions and associated injuries. Thesis, University of Leiden Cooper S (1953) Muscle spindles in the intrinsic muscles of the human tongue. J Physiol 67:1–13 Donaghy M (2005) Lumbosacral plexus lesions. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy. 4th edn. Philadelphia, Elsevier, Chapter 56. pp 1375–1390 Edgar MA (2007) The nerve supply of the lumbar intervertebral disc. J Bone Joint Surg 89B:1135–1139 Fleming P, Lenehan B, Sankar J, Folan-Curran P, Curtin W (2004) One third, two thirds: relationship of the radial nerve to the lateral intermuscular septum in the arm. Clin Anat 17:26–29 Furness JB (2006) Novel gut afferents: intrinsic afferent neurones and intestinofugal neurones. Auton Neurosci 125:81–85 Furness JB, Costa M (1980) Types of nerves in the enteric nervous system. Neuroscience 5:1–20 Gardner E, Bunge RP (2005) Gross anatomy of the peripheral nervous system. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Saunders Elsevier, Philadelphia, pp 11–34 Gruber W (1870) Über die Verbindung des Nervus medianus mit dem Nervus ulnaris am Unterarme der Menschen und der Säugethiere. Archive Anatomie Physiologie Medizin (Leipzig) 37:501–522 Hughes RAC (2005) Disease of the fifth cranial nerve. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, pp 1207–1218, Chapter 49 Kampa RJ, Prasthofer A, Lawrence-Watt DJ, Pattison RM (2007) The internervous safe zone for incision of the capsule of the hip. J Bone Joint Surg 89B:971–976 Kierner AC, Zelenka I, Burian M (2001) How do the cervical plexus and the spinal accessory nerve contribute to the innervation of the trapezius muscle? Arch Otolaryngol Head Neck Surg 127:1230–1232 Lambert SM (2005) Further opinion: radial nerve palsy associated with fractures of the shaft of the humerus. Doi: 10.1302/0301-620X. 87B12.17720 Langley JN (1903) The autonomic nervous system. Brain 26:1–26 Langley JN (1926) The autonomic nervous system. Part 1. Heffer, Cambridge Last RJ (1949) Innervation of the limbs. J Bone Joint Surg 31B:452–464 MacAvoy MC, Rust SS, Green DP (2006) Anatomy of the posterior antebrachial cutaneous nerve: practical information for the surgeon operating on the lateral aspect of the elbow. J Hand Surg 31A:908–911 MacNicol MF, Kelly M (2002) Identification of the saphenous nerve at arthroscopy. Arthroscopy 14:312–314 Mahakkanukrauh P, Somsarp V (2002) Dual innervation of the brachialis muscle. Clin Anat 15:206–209 Mok D, Nikolis A, Harris PG (2006) The cutaneous innervation of the dorsal hand: detailed anatomy with clinical implications. J Hand Surg 31A:563–574 Nathan PW, Smith MC (1987) The location of descending fibres to sympathetic pre-ganglionic vasomotor and sudomotor neurones in man. J Neurol Neurosurg Psychiatry 50:1257–1262 Pereira MT, Williams WW (1999) The spinal accessory nerve distal to the posterior triangle. J Hand Surg 24B(3):368–369 Schott GD (1994) Visceral afferents: their contribution to “sympathetic-dependent” pain. Brain 117:397–413 Solan MC, Lemon M, Bendall SP (2001) The surgical anatomy of the dorsomedial cutaneous nerve at the hallux. J Bone Joint Surg 83B:250–252 Srinivasan R, Rhodes J (1981) The median-ulnar anastomosis (Martin Gruber) in normal and congenitally abnormal fetuses. Arch Neurol 38:418–419 Standring SM, Bunting S (2000) The fibre content of the spinal accessory nerve. Personal Communication Suseki K, Takahishi Y, Takahishi K, Chiba T, Yamagata M, Moriya H (1998) Sensory nerve fibres from lumbar intervertebral discs pass through rami communicates. J Bone Joint Surg 80B:737–742 Tennent TD, Birch NC, Holmes MJ, Birch R, Goddard NJ (1998) Knee pain and the infrapatellar branch of the saphenous nerve. J R Soc Med 91:573–575 Tindall MA, Patel M, Frost A, Parkin I, Shetty A, Compson J (2006) The anatomy of the dorsal cutaneous branch of the ulnar nerve – a safe zone for positioning of the 6R portal in wrist arthroscopy. J Hand Surg 31B:203–205 Weddell G, Harpman JA, Lambley DG, Young L (1940) The innervation of the musculature of the tongue. J Anat 74:255–267 Wilbourn AJ (2005) Brachial plexus injuries. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, Chapter 55. pp 1339–1373
2
The Microscopic Structure of the Nervous System: Its Function
The neurone and neurotrophins; transitional zone and roots; the Schwann cell and myelin; the connective tissue sheaths; conduction; axonal transport; blood supply; nervi nervorum; changes with aging; motor path with central and peripheral connections; sympathetic outflow; sensory path with superficial and deep peripheral connections and with central connections; the role of visceral afferents; neurotransmitters. The essential component of the system is the nerve cell with its dendrites and its prolongation, the axon (Figs. 2.1–2.3). Young (1945) characterized the axon as “a very long cylinder of a semi fluid nature.” It is a column of neuronal cytoplasm, the axoplasm enclosed by a cell membrane, the axolemma. Thomas et al. (1993a) described the axoplasm as a “fluid cytosol in which are suspended formed elements.” The most conspicuous of the latter is the cytoskeleton consisting of neurotubules, neurofilaments and matrix. In addition, there are mitochondria, axoplasmic reticulum, lamellar and multivesicular bodies, and membranous cisterns, tubes and vesicles. It is the cytoskeleton that provides the apparatus for axoplasmic transport. Berthold et al. (2005) describe the axolemma as a three-layered unit membrane about 8 nm thick and consider that it: “conveys signals between the neurone and its Schwann cells that control the proliferative and myelin
producing functions of the Schwann cells and partly regulate axon size.” The glial cells of the peripheral nervous system are essential for the development, maturation, survival and regeneration of the neurone. The relationship between the axon and the Schwann cell is lifelong. The myelinating and non myelinating Schwann cells are the main peripheral glial cells. There are others, which include the satellite cells surrounding cell bodies in the dorsal root and autonomic ganglia, the glia of the enteric system; the teloglia (terminal Schwann cells) at the terminals of somatic motor axons and the glia associated with sensory terminals such as the Pacinian corpuscle. Mirsky and Jessen (2005) observe: “evidence to date suggests that the molecular and morphological differences between these various cells depend on the specific location and cellular environment in which they are found and that the glial cells of the PNS retain unusual plasticity throughout
Fig. 2.1 Cultured human dorsal root ganglion neurone immunostained for Gap 43 (growth associated protein) showing the cell body and neurites arising from the cell body, x40 (Courtesy of Dr. Uma Anand).
Fig. 2.2 Phase contrast image of a single DRG neurone showing the large rounded cell body, x63 (Courtesy of Dr. Uma Anand).
R. Birch, Surgical Disorders of the Peripheral Nerves, DOI: 10.1007/978-1-84882-108-8_2, © Springer-Verlag London Limited 2011
43
44
Surgical Disorders of the Peripheral Nerves
2.1 The Neurotrophins 2.1.1 Nerve Growth Factor
Fig. 2.3 Cultured human DRG neurones immunostained for Gap 43 (green) and for the vanilloid receptor TRPV1 (red), the nuclei of satellite cells are stained blue (DAPI). Bar = 25 mm (Courtesy of Dr. Uma Anand).
life.” King (Berthold et al. 2005) estimated that about 10% of nuclei within the endoneurium of a normal peripheral or spinal nerve root are fibroblasts, and that endogenous macrophages account for between 2% and 9%. Most of the remainder are Schwann cells. Whilst mast cells are also seen their function is not well understood. “Neuron theory” Cajal (1954) asserts that “the neuron, a nerve cell with its processes, is the structural unit of nervous tissue, and the neurons are the only elements in the nervous system which conduct nervous impulses” (Brodal (1981b)). There is no continuity between nerve cells: the termination of the axon on a cell is no more than a contact, a contact to which Sherrington (1897) gave the name, synapse. Brodal (1981b) goes on: “not only is the neuron a structural unit; it also, in most cases, behaves as a trophic unit”; but the neurone itself requires trophic support during maturation and for survival after injury to the axon (Fig. 2.4).
Fig. 2.4 NGF immuno reactivity in neurones in a human dorsal root ganglion which was avulsed from spinal cord six weeks earlier. Immuno reactivity, identified by green fluorescence staining, is localized to cells of the small type. The yellow intracellular granules are lipofucsin deposits which exhibit auto fluorescence. Indirect immunofluorescence method, x50 (Courtesy of Professor Praveen Anand).
Cajal (1928) proposed that there was no autogenous regeneration of the peripheral stump. His belief in the presence of some neurotrophic factor in the distal stump is expressed in the phrase “The penetration into the peripheral stump implies a neurotrophic action, or the exercise of electrical influence by the latter.” He named contact guidance “tactile adhesion.” Young (1942) seemed to conclude that “successful nervous regeneration must depend mainly on the chances provided for adequate numbers of outgrowing fibers to establish connections resembling their original ones.” A hint of belief in neurotrophins had, however, been given in his work with Holmes (Young and Holmes 1940). It was during the 1940s that Hamburger and his colleagues (Hamburger and Keefe 1944) showed that the removal of a limb bud from a chick embryo led to a reduction in the number and size of the neurones destined for that limb whereas addition of extra target tissue during embryonic development was followed by an increase in the number and the size of the relevant neurones. “These observations led directly to the discovery of nerve growth factor (NGF)” (Windebank and McDonald 2005). Levi-Montalcini et al. (1954) and Cohen et al. (1954), described an agent found in mouse sarcomata which markedly promoted growth in the sympathetic and posterior root ganglia of chick embryos. They found that treatment with snake venom enhanced the activities of this agent, and that snake venom itself contained a potent growth promoting agent. The results of partial purification and characterisation of this nerve growth factor (NGF) were presented and suggested that the active material was a protein or bound to a protein. Levi-Montalcini and Angeletti (1968) indicated the specific action of NGF on sensory and sympathetic nerve cells: “the control exerted by the NGF on sensory and sympathetic nerve cells stands out by virtue of the magnitude of its effects, its target specificity, and the plurality of its actions.” Windebank and McDonald (2005) defined growth factors as “soluble extracellular macromolecules that influence the proliferation, growth and differentiation of target cells by a cell surface receptor mediated mechanism.” Most neurotrophins are polypeptides which are produced in tissues such as skin or muscle from whence they are transported to the neuronal cell body by the fast centripetal component of axonal transport. Three major families of growth factors are recognized. 1. The classic neurotrophins include nerve growth factor (NGF), brain derived nerve factor (BDNF) and the neurotrophins 3–7 (NT3, 4, 5, 6, 7). NGF is produced by cells including keratinocytes, melanocytes, vascular and smooth muscle cells, testis and ovarian cells, and endocrine and exocrine tissue. NGF interacts with the high
The Microscopic Structure of the Nervous System: Its Function
affinity receptor p140 tyrosine receptor kinase (TrkA) which is expressed by sympathetic neurones and by small diameter neurones in the dorsal root ganglia. After nerve injury, cells in other tissues, including Schwann cells and fibroblasts, synthesize and release NGF. Mice experimentally engineered to be deficient in TrkA do not develop thermoceptive or nociceptive neurones. BDNF apparently supports the development of motor neurones and their survival after axonotomy. After nerve transection, BDNF messenger RNA increases in muscle and the messenger RNA of one of the receptors to this nerve growth factor, tyrosine receptor kinase B (TrkB) increases in motor neurones. Neurotrophin 3(NT3) is mainly expressed in muscle spindles, Merkel cells and the Golgi tendon organs. This neurotrophin specifically binds to tyrosine receptor kinase C (TrkC). Mice which have been genetically engineered to lose this receptor lack proprioceptive organs. 2. Other neurotrophins are synthesized by the glial cells. It is likely that these factors support the embryonic midbrain and motor neurones in the spinal cord. The glial derived nerve factor (GDNF) binds with its high affinity receptor, also the tyrosine kinase receptor c-Ret. The ciliary neurotrophic factor (CNTF) first binds to its receptor and also the leukemia inhibitory factor receptor beta (LIFR ß). CNTF supports neurones in the ciliary ganglion, dopaminergic neurones, retinal rods and sympathetic and motor neurones. 3. The third family include the insulin growth factor (Igf) which structurally resembles insulin and binds with the tyrosine kinase IGF-I receptor, which is itself homologous to the insulin receptor. This receptor is expressed throughout the nervous system. Nerve growth factors are synthesized by the target organs of the nerves and are conveyed centrally to the neuronal cell bodies. This transport ceases after axonotomy. The interruption of this flow contributes to cell death amongst central neurones, an effect which is more severe in the immature nervous system and after axonotomy close to the neuronal cell bodies (Figs. 2.5 and 2.6). Anand and his colleagues have extensively investigated neurotrophic factors and their receptors in the normal human nerve, and in nerves affected by diabetes, leprosy, and injury. These studies have been extended to neurones from human dorsal root ganglia, (Anand et al. 1996, Bar et al. 1998,Saldanha et al. 2000, Yiangou et al. 2000, Durrenberger et al. 2004, Chessell et al. 2005, Facer et al. 2007). More recently, Uma Anand and her colleagues have perfected methods in vitro for the study of living neurones from human dorsal root ganglia and this has permitted close study of the effects of neurotrophins and molecular mechanisms. (Anand et al. 2006, Sanchez et al. 2007, Anand et al. 2008a, Anand et al. 2008b, Anand et al. 2008c) Anand et al. (2006) summarize the effect of neurotrophin factors on the morphology and expression of some receptors in cultured human dorsal
45
Fig. 2.5 Brain derived neurotrophic factor (BDNF) immunoreactivity in medium sized neuronal cell bodies and associated axons in human dorsal root ganglion 6 weeks after avulsion of spinal nerves, x40 (Courtesy of Professor Praveen Anand).
Fig. 2.6 Glial derived nerve factor (GDNF) receptor Cret immunoreactivity in neuronal cell bodies of human dorsal root ganglion two weeks after avulsion of spinal roots, x40 (Courtesy of Professor Praveen Anand).
root ganglion sensory neurones: the factors NGF, NT3 and GDNF are produced by peripheral target tissues such as epidermal keratinocytes and affect the phenotype of cultured sensory neurones. In the mature nervous system neurotrophins switch from providing trophic support for neuronal survival to maintenance of a specific neuronal phenotype thereby facilitating modality specific sensory function. For example GDNF and NGF regulate the normal function of two distinct classes of nociceptors via their receptors Ret and TrkA respectively. The levels of these factors are altered by injury. Administration of NGF induces thermal and mechanical hyperalgesia, it is upregulated by inflammation and plays a key role in the pathophysiology of nociception (Figs. 2.7–2.10). Evidently, important questions regarding the role of neurotrophic factors both in regeneration of peripheral nerves and in causation of pain and cutaneous hyperaesthesia and
46
Fig. 2.7 Neurotrophin 3 (NT3) immunostaining in suprabasal epithelial cells of human glabrous arm skin, x40 (Courtesy of Professor Praveen Anand).
Surgical Disorders of the Peripheral Nerves
Fig. 2.10 Subepithelial fascicles of NGF positive axons in the tongue of patient with burning mouth syndrome (BMS), x40 (Courtesy of Professor Praveen Anand).
hyperalgesia are raised by the continuing work in this field. Although trials using factors for the treatment of amyotrophic lateral sclerosis and diabetic neuropathy failed, their therapeutic roles continue to be explored and developed.
2.2 The Peripheral Nerve Fibres
Fig. 2.8 Nerve growth factor (NGF) immunostaining of basal epithelial cells in human glabrous arm skin, x40 (Courtesy of Professor Praveen Anand).
Fig. 2.9 GDNF immunoreactivity in Schwann cells of healthy human sural nerve, x150 (Courtesy of Professor Praveen Anand).
In the central nervous system the neurones are supported in a network of oligodendrocyte and astrocyte processes, with very little extracellular space. The structure of peripheral nervous tissue is one of nerve fibres (axons – Schwann cell units) suspended in a collagen rich extra cellular space (Berthold et al. 1984, 1993). The transition from central to peripheral nervous structures takes place in the rootlets or less often in the roots of the spinal nerves. This is the transitional region or transitional zone (TZ). The development of the transitional region in dorsal rootlets was described by Carlstedt in 1981. The extension of CNS structure into the base of the rootlet is cone-shaped. Thus, “each transitional region can be subdivided into an axial CNS compartment and a surrounding PNS compartment” (Berthold et al. 1993) (Figs. 2.11 and 2.12). The myelin sheath distal to the transitional zone is formed by the transitional Schwann cell and that central to it by the transitional oligendendrocyte. The basal lamina of the axon remains continuous through the TZ. Some myelinated nerve fibres become nonmyelinated centrally. Fraher (2005) states that during development: “the CNS-PNS interface oscillates and continually changes its form and position as the two tissue classes establish their mutually exclusive territories.” The astrocyte barrier, which is at first flush with the glia limitans becomes pushed
The Microscopic Structure of the Nervous System: Its Function
peripherally by the central tissue process. Myelination is delayed in the transitional zone and that in the proximal rootlet segment lags behind the rest of the root. The ventral root
47
of the rat has a rich blood supply unlike the dorsal root; in the cat the blood vessels do not accompany the axons in the dorsal root, instead they deviate from the endoneurial space and join vessels on the surface of the cord. Fraher suggests that this arrangement causes the dorsal roots to be more susceptible to ischemia than ventral roots. His findings that the mechanical arrangements lead to rupture at the rootlet rather than in the transitional zone has been confirmed in the human by Schenker (Schenker and Birch 2001) who examined biopsies of the tips of avulsed roots in 12 patients by light and by transmission electron microscopy. Of the ten biopsies taken within 4 weeks of injury the level of rupture was central to the TZ in two of the roots and peripheral to it in the remaining eight (Figs. 2.13 and 2.14). Schenker also studied the corpora amylacea (CA), round homogenously staining
Fig. 2.11 Morphology of the normal human spinal cord. A transverse light microscopic section at C7 showing a ventral root. A single large root in transition is demonstrated with central islands of autolysed glia. Numerous corpora amylacea are seen on the central side of the transitional zone. Toluidine blue, x100 (Courtesy of Editor Journal of Bone and Joint Surgery [British]).
Fig. 2.13 Transverse section of the tip of an avulsed dorsal root at C6 4 days after injury showing a central avulsion. The tissue in the centre of the section is CNS tissue in which glial cells show post traumatic lytic changes. Corpora amylacea are indicated by arrows. Toluidine blue, x100 (Courtesy of Editor Journal of Bone and Joint Surgery [British]).
Fig. 2.12 Normal anatomy of the transitional zone. Schwann cell processes are seen above, astrocyte processes lie closely to an unmyelinated axon (arrows) which is surrounded by a basal lamina(arrow head) Electron microscopy by Michael Kayser, Institute of Orthopaedics (Courtesy Michael Schenker).
Fig. 2.14 Avulsed ventral root of C6 4 days after injury showing a peripheral intradural rupture. The nerve tissue at the site of the rupture showed no CNS features. The tip is covered by organized blood clot and erythrocytes which interweave with fibrin strands. The myelinated fibres show early signs of Wallerian degeneration. Toluidine blue, x200.
48
Surgical Disorders of the Peripheral Nerves
bodies 15–50 mm in diameter, which are found in the subpial zone of the spinal cord in post mortem material. The CA mark the watershed between the central and the peripheral nervous systems. The basal lamina surrounding the Schwann cell – axon unit extends into the spinal cord and cannot be used as a reliable indicator of peripheral nervous tissue. Most of the fibres of the ventral roots have their cells in the ventral horn of the grey matter. They can, perhaps, be regarded as outposts of the peripheral system in the central system. The fibres of the dorsal roots have their cells in the dorsal root ganglion: possibly, outposts of the central in the peripheral nervous system? These neurones are “unipolar in form with a single axon and no true dendrites” (Thomas et al. 1993b). Each axon bifurcates into peripherally and centrally directed axons after leaving the cell body at a variable distance from the cell. The centrally directed branch is of smaller caliber than the peripheral one. The central processes enter the spinal cord along the posterolateral sulcus. In the cord the fibres bifurcate into ascending and descending branches. The branches of the smaller fibres in the lateral part of the root reach the dorsal horn of the grey matter, where both soon terminate having traversed between three and five segments. The branches of the larger fibres in the medial part of the root, mostly myelinated, similarly bifurcate after entering the white matter just medial to the dorsal horn. Some ascending fibres reach as high as the gracile and cuneate nuclei in the caudal part of the medulla. Other fibres of this group have short ascending and descending branches which enter the grey matter of the dorsal horn to establish synapses with nerve cells in its different laminae (Fig. 2.15). Sherrington (1894) showed that ventral roots of cats and monkeys contained intact myelinated nerve fibres after
transection of both the ventral and dorsal roots and he concluded that some afferents reached the spinal cord the “wrong way” through the ventral roots. Unmyelinated afferent fibres certainly enter the spinal cord in the ventral roots (Coggeshall et al. 1974). They may be concerned with the transmission of painful impulses (Clifton et al. 1976) though White and Sweet (1969) failed to produce pain in man by stimulation of ventral roots. On the other hand, Brindley (1986) described three paraplegic patients who experienced pain whilst stimulating the ventral roots of S2 and S3 or of S3 and S4 to induce micturition by an implanted stimulator. Brindley says: “the Bell-Magendie law is certainly nearly true, but on published evidence it seems likely that it is not exactly true,” pointing out that pain could be evoked by antidromic impulses in efferent fibres. We were able to take advantage of the opportunity offered by intradural damage to nerves of the brachial plexus to demonstrate the presence of afferent fibres in the ventral roots of man. Stephen Gschmeissner examined by electron microscopy the ventral root of the eighth cervical nerve avulsed with its dorsal root and found with it at operation for exploration of the plexus in the posterior triangle of the neck. He found several surviving small myelinated and unmyelinated axons (Fig. 2.16). Clearly, the cell bodies of these axons must have been in the nerve or in the dorsal root ganglion; the fibres must have belonged to the afferent system. These findings were confirmed by Schenker (Schenker and Birch 2000) who examined biopsies from the tip of the avulsed rootlets in nine patients. The tips of the dorsal and of the ventral rootlets of 5 spinal nerves were examined within 8 days of injury, the other patients were operated at intervals ranging between 4 and 50 weeks from injury. Schenker found intact myelinated fibres in all ventral root specimens. The majority of those Direct ascending tract Lissauer’s tract I
Lateral corticospinal tract
X
II III IV V VI VII VIII
Fig. 2.15 The paths of the afferent fibres entering and efferent fibres leaving the spinal cord. Note (right) the laminae of the grey matter.
IX
Crossed ascending tract
The Microscopic Structure of the Nervous System: Its Function
49
Fig. 2.16 Afferent and efferent fibres in the ventral root. Large healthy myelinated axons in the ventral root of the 8th cervical nerve avulsed from the spinal cord 6 weeks previously. The myelinated efferent fibres have undergone Wallerian degeneration and there is much collagenisation (Electron microscopic study by Mr. Stephen Gshmeissner, x5000).
identified in the later biopsies were regenerating, perhaps signifying collateral sprouting from the intact cell bodies of the adjacent avulsed dorsal root ganglion. There were a small number of intact fibres in the early specimens and Schenker suggests that: “it is therefore likely that these few fibers of the ventral root represent afferent fibers in continuity with the cell body in the DRG.” Less than 5% of the central processes of all dorsal afferents survive the traction injury and it was assumed that “wrong way” ventral afferents do just the same so that the few surviving ventral root afferents that were observed may represent only a very small proportion of the population. Based on this assumption Schenker proposed that the proportion of afferent fibres in the ventral root of man is similar to that found by Loeb (1976) who, by microelectrode recordings in cats, calculated that 3.9% of all afferent fibres reached the cord through the ventral root (Fig. 2.17). In the peripheral nervous system the axons are closely associated with the Schwann cells (Schwann 1839). Sanders (1942) established the central role of Schwann cells in regeneration through grafts: “autografts remain alive and myelin fragmentation and Schwann multiplication go on very much as in a normal peripheral stump.” Sanders rejected methods of repair which do not enrich the environment with Schwann cells. Schwann cells arise from the neural crest, from the same cells that differentiate into peripheral neurones; they provide essential trophic support to the neurone during development and also during regeneration. Our present understanding of the part played by these cells in repair owes a
great deal to the work of Susan Hall of Guy’s Hospital, and now editor of Gray’s Anatomy. (Hall 1997, 1999, 2001, 2005, Li et al. 1998) (Fig. 2.18). Mirsky and Jessen (2005) described how two important growth factors, neuregulin 1 and endothelin regulate early Schwann cell development. The Schwann cell separated from its axon survives for rather longer in the mature than in the immature nervous system. The most important component of the basal lamina is laminin which interacts with receptors in the plasma membrane of the Schwann cell which include the integrins and alphadystroglycan. Laboratory mice which have been genetically engineered to produce defective laminin or a defective receptor for laminin, develop profound nerve pathology and muscle dystrophy (Uziyel et al. 2000). Schwann cells provide another important protein, the tumor suppressor protein, Merlin (schwannomin), which links membrane proteins to the actin cytoskeleton in epithelia and other cell types. Mutations in the gene controlling Merlin lead to an increased frequency of schwannomas. In leprosy, components of the cell wall of M.leprae interact with laminin 2, the major isoform of laminin, and this interferes with the normal interaction between laminin and alphadystroglycan leading to demyelination and to axonal degeneration. During myelination Schwann cells radically transform their phenotype in response to signals from the larger axons; as Mirsky and Jessen (2005) say: “this response represents one of the most striking examples of cell-cell interaction that is known.”
50
Surgical Disorders of the Peripheral Nerves
Fig. 2.19 A large myelinated nerve fibre within the posterior root of a 7th cervical nerve which had been avulsed from the spinal cord six weeks previously. The axoplasm contains neurofilaments and a few microtubules. The Schwann cell cytoplasm it is enveloped by a well defined basal lamina. There are processes from fibroblasts from within the endoneurium, x16,200. Fig. 2.17 Rupture of the ventral root of C5 peripheral to the transitional zone examined 5 months after avulsion. Two thinly myelinated axons are seen. x 17,000 (Electron microscopic study by Mr. Michael Kayser, Institute of Orthopaedics. By courtesy editor Journal of Anatomy).
Fig. 2.18 In vitro cultures of mouse DRG neurones and Schwann cells. The Schwann cells are immunostained for S100 (red) the neurones for nerve growth factor (green), x50 (Courtesy of Professor Susan Standring).
Scherer and Arroyo (2002) provide an extensive review of the molecular organisation of myelinated axons. The larger axons are enwrapped along their length by a continuous series of contiguous Schwann cells into which they are invaginated. The nodes of Ranvier represent the points of contiguity of adjacent Schwann cells (Fig. 2.19). The fibre is contained within a basal lamina. The basal lamina separates nerve fibres from the endoneurial space and it runs without interruption from the central nervous system to the termination of the axon. Thomas (1963) defined this structure in an electron microscope study. Schwann cells are surrounded by a basal lamina approximately 250 Å in thickness which is separated from the plasma membrane of the Schwann cell by a gap of 250 Å. The endoneurium is organized in two layers which surround the basal lamina. The inner layer is composed of collagen fibres of smaller diameter than those in the outer layer which run longitudinally, circularly and obliquely. This layer is inflected at the nodes with the basement membrane. The outer layer consists solely of longitudinal collagen fibres and it is not inflected at the node. Bunge et al. (1986) considered the basal lamina essential in the linkage
The Microscopic Structure of the Nervous System: Its Function
51
Fig. 2.20 Clusters of unmyelinated axons (long arrows), enveloped by Schwann cell cytoplasm. Short arrows indicate basal lamina, x26,220.
between the Schwann cell and the axon and with the extracellular matrix. The smaller fibres are contained in bundles by columns of Schwann cells. Eames and Gamble (1970) showed the ensheathing arrangement of successive Schwann cells which overlapped and interdigitated: the Schwann cells of unmyelinated axons in these nerves give off multiple cytoplasmic processes, which form close relationships with axons, with other processes, and with bundles of collagen. A ramifying “network” system of Schwann processes is thereby present throughout the endoneurium.
Eames and Gamble recognize areas of specialisation of the Schwann membranes which consisted of “a short length of increased plasma membrane thickness and electron density.” Later studies using scanning electron microscopy and freeze fracture replication have generally confirmed these earlier observations (Stolinski and Breathnach 1982). It is at times very difficult to distinguish between a nonmyelinated axon and a Schwann cell process in histological sections of regenerating peripheral nerves! (Fig. 2.20). The diameter of the axon is one important factor which determines whether the Schwann cells will lay down a myelin sheath around it. Webster (1993) proposes that the sheath is laid down in spiral layers by the Schwann cell, or part of its surface, moving around the axon (Webster 1993). The multilamellar sheath has a high lipid content and some protein components. Suter and Martini (2005) say that the major component of the protein components of myelin is Po myelin protein zero (MPZ), which accounts for 50–60% of all myelin protein. Peripheral myelin protein 22 (PMP 22) comprises from 2% to 5% of myelin proteins and mutations of the controlling gene lead to inherited myelin disorders. Suter and Martini comment that mutation of this gene was: “the first identified culprit gene for inherited neuropathies of Charcot-Marie Tooth (CMT) type.” Myelin basic protein
(MBP) accounts for 5–15% of myelinated proteins. The myelin associated glycoprotein (MAG), which forms no more than 0.1% of the myelin proteins, may play a pivotal role during myelination because of its early expression and because of its location to the axon-Schwann cell interface. Mice genetically engineered to be deficient in MAG showed extensive axonopathy and degeneration of myelin in motor fibres. The myelin sheath is traversed by cytoplasmic channels – the incisures of Schmidt-Lanterman (Hall and Williams 1970) (Fig. 2.21). The meeting points of consecutive Schwann cells are the nodes of Ranvier. These are short, about 1 micron in length and the axon here is constricted,
Fig. 2.21 A widened Schmidt-Lanterman incisure in a teased mouse sciatic nerve. Normal saline in vitro, x100 (Courtesy of Professor Susan Standring).
52
free of myelin but enveloped by projections of Schwann cytoplasm. The node is bordered by an adjacent paranode, which is dilated and which contains an increasing amount of mitochondrial rich Schwann cell cytoplasm outside a more or less crenated myelin sheath. Berthold et al. (2005) characterize the node thus: “these parts of the myelinated nerve fibers, the paranode-node-paranode (PMP) regions, constitute, structurally as well as functionally, the most spectacular parts of a myelinated nerve fiber.” The PMP regions are responsible for the generating and propagation of the action potential and they are the centers for activity in the early phases of Wallerian degeneration and collateral sprouting (Figs. 2.22–2.24). The axon-myelin sheath–Schwann cell complexes are arranged in bundles otherwise known as fascicles or funiculi. In so small a nerve as the fourth cranial there may be as many as 3,400 fibres. In the roots inside the spinal canal endoneurial collagen is scanty in contrast with the abundant content in the nerves outside the foramen (Gamble 1964; Eames and Gamble 1964) The surgeon who has had dealings with nerves inside and outside the spinal canal will appreciate the distinction: the spinal roots and rootlets are fine and fragile and very susceptible to trauma; the peripheral nerves are strong and have much greater resistance to handling. Outside the intervertebral foramina the three supporting structures, epi, peri, and endoneurium are clearly established. The epineurium, in effect the prolongation of the dural sleeve of the nerve roots, is composed of longitudinally directed collagen fibres, fibroblasts and fat cells. (Gamble and Eames 1964). The perineurium, which ensheaths the bundles, is composed of flattened cell processes alternating with layers of collagen. It provides a barrier to diffusion (Thomas 1963). The perineurium is strong; the intrafascicular pressure can be raised to more than 300 mmHg before it ruptures (Selander and Sjostrand 1978). The contents of the perineurium are under tension so that when it is cut they are extruded, rather like
Fig. 2.22 Node of Ranvier, mouse sciatic nerve in vivo, oblique incident illumination, x400(Courtesy of Professor Susan Standring).
Surgical Disorders of the Peripheral Nerves
toothpaste. This is most clearly seen on the day of injury in nerves which have been transected or ruptured and it is one of the indications that the level of section of the stump is adequate. The outflow rapidly diminishes over the course of several days. In the endoneurium, supporting the fibres
Fig. 2.23 Longitudinal section through a node of Ranvier, showing a remyelinated heminode (left) adjacent to a normal heminode (right). Compare the complexity of the paranodal fingers of the normal myelin sheath with the simple arrangement of the paranodal loops of the thinner, remyelinated sheath, x5,000 (Courtesy of Professor Susan Standring).
Fig. 2.24 Double immunostaining of sural nerve showing nodes of Ranvier (black) stained with antibodies to junction adhesion molecule (JAM – c) and axons (red) stained with antibodies to neurofilaments, x40 (Courtesy of Professor Praveen Anand).
The Microscopic Structure of the Nervous System: Its Function
53
themselves, there is a return to longitudinal direction of cells and fibres; there are abundant collagen fibrils (Thomas 1963; Gamble and Eames 1964). Stolinski (1995) describes the epineurium as an outer layer of alveolar connective tissue and a more compact inner layer containing collagen and elastic fibres which are arranged in a wavy pattern. Perineurial cells are also arranged in a wavy manner and Stolinski suggested that the spiral bands of Fonata represent the wavy or zig zag organisation of nerve fibres. These arrangements provide a degree of protection to the nerve against traction. The nerve can be stretched by as much as 20% before the wavy arrangement is converted into a linear array. Tillett and her colleagues (2004) offer the concept of a distinct core and sheath in the rat sciatic nerve, and proposed that the interactions between the core and the sheath involve physical connections rather than a viscous fluid interface. The anatomical features of this interface were characterized using transmission electron microscopy and it appeared that the sheath was derived from the epineurium and most of the perineurium, whilst the core consisted of the endoneurium and a small proportion of perineurium: the plane of cleavage involved the innermost perineurial cell layer. A normal peripheral nerve trunk exposed at operation is enveloped in a well defined translucent envelope. This is the external epineurium (Fig. 2.25). Normal nerve trunks are easily distinguished from other longitudinal structures by the appearance of white spiral bands on their surface, the spiral bands of Fontana (Clarke and Bearn 1972, Stolinski 1995). The individual bundles or fascicles are seen within. These are enclosed by the perineurium with some condensation of the innermost epineurium forming a white, opaque layer (Fig. 2.26). The tissues surrounding the bundle form the epineurium, rather loose in texture, and rich in blood vessels which pass longitudinally along the axis of the nerve. However, the observer will see adventitial material outside the epineurium which is more clearly defined in some nerves
than in others and in different locations within the limb. There are, for example, translucent connective tissue arcades accompanying the median nerve in the forearm where it passes between the superficial and the deep flexor muscles of the fingers. Such vessels provide an alternative collateral pathway to the part; they also supply the nerve trunk so permitting the use, for example, of the ulnar nerve as a free vascularized graft. This tissue plane not only conveys vessels to the nerve but it also permits gliding of that nerve across joints and against the adjacent tissues. Thomas (1963) has estimated that 45% of nuclei seen in transverse sections of nerves are those of fibroblasts. Sunderland (1968) mapped the arrangement of bundles along the course of nerve trunks, showing branching, fusion and changes in number. He further showed that the cross-sectional area of the nerve occupied by connective tissue was variable, ranging from 60% to 85% (Fig. 2.27). These findings, especially those concerning re-arrangement of bundles, have been used to cast doubt on the feasibility of achieving accurate co-aptation of the ends of divided nerves. However topographical organisation is one essential quality of the nervous system and this is shown by the considerable topographical segregation of neurones involved in the somatic afferent pathways in the dorsal root ganglia, dorsal horn of the spinal cord, the thalamus and the sensory cortex. Perhaps predictably injury anywhere in the nervous system provokes considerable reorganisation. Sunderland himself recognized that there was a degree of topographical segregation of nerve fibres according to function over considerable lengths of the median and ulnar nerves. Microneurographic studies (Torebjörk and Ochoa 1980) confirmed these findings. Specific organisation (aggregation) of sensory and motor fibres occurs in the median nerve in the arm, so it does in all peripheral nerves. This segregation permits transfers such as that of Oberlin et al. (1995), in which one bundle of the ulnar nerve
Fig. 2.25 The extrinsic epineurial vessels of the ulnar nerve, x40.
Fig. 2.26 The bundles and epineurial vessels of the ulnar nerve after displacing the adventitia, x40.
54
Surgical Disorders of the Peripheral Nerves
Fig. 2.27 Sural nerve undergoing Wallerian degeneration 4 days after a proximally sited crush lesion showing macrophages stained positively for ED1, x100 (Courtesy of Professor Susan Standring).
is anastomosed to the nerve to the biceps. Stimulation of individual bundles, whose number ranges from 6 to 12, permits separation of those passing to the flexor muscles of the wrist and fingers from those passing to the small muscles of the hand. The ability to “map” the stump of a divided nerve allied to the ease of matching individual bundles by their size and disposition is but one of the great advantages of urgent repair of nerves (Fig. 2.28).
2.2.1 Conduction The special property of the nerve fibre is that of conducting a signal in the form of a propagated action potential (Landon 1985). Rasminsky (1985) opens the subject thus: “Nerve fibres are specialized processes of nerve cells that have the unique property to propagate action potentials, the currency of information in the nervous system.” The action potential is a brief, self propagating reversal of membrane polarity and it depends on an initial influx of sodium ions which cause a reversal of polarity to about +40 mV followed by a rapid return towards the resting potential as potassium ions flow out. In the unmyelinated fibre, a wave of depolarisation spreads continuously along the axon, attenuated by the large capacitance of the axolemma, which limits the velocity of conduction to about 1m/s. Standring (2008) likens this to: a flame moving along a fuse. Just as each segment of the fuse is ignited by its upstream neighbor, each segment of axon membrane is driven to threshold by the depolarization of neighboring membrane. Sodium channels within the newly depolarized segment
open and positively charged sodium ions enter, driving the local potential inside the axon towards positive values. This inward current in turn depolarizes the neighboring, downstream, non depolarized membrane, and the cyclic propagation of the action potential is completed.
The action potential is evoked by a stimulus which exceeds threshold by the all or nothing law of Adrian (1928); the cell body, on the other hand, responds in a graduated manner to stimuli transmitted across synapses which either inhibit or facilitate by raising or lowering the threshold respectively. In the myelinated fibre the myelin sheath acts as a capacitor and limits radial resistance at the internode, so that most of the current flows axially along the fibre. It is, says Bostock (1993) “powered by inward ‘kicks’ of inward membrane current at the nodes of Ranvier.” This method of “saltatory” conduction was so named by Tasaki and Takeuchi (1941; 1942), and further confirmed by Huxley and Stämpfli (1949). The myelin sheath thus enables the fibre to conduct rapidly without the necessity for a very large increase in axonal diameter. The caliber of unmyelinated axons varies from 0.4 to 1.25 mm (Gasser 1955); that of myelinated fibres from 2 to 22 mm (Ranson 1915; Greenfield and Carmichael 1935; Sunderland et al. 1949). The largest, fastest conducting elements are the myelinated fibres of around 20 mm diameter concerned with somatic afferent and efferent activity; the smallest and slowest conducting are the fibres of around 1 mm diameter that subserve autonomic activity and delayed pain sensibility (Galbraith and Myers 1991). Conduction velocity ranges from about 0.7 m/s in small unmyelinated fibres to about 80 m/s in the largest
The Microscopic Structure of the Nervous System: Its Function
55
Fig. 2.28 Fascicular arrangement of nerve fibres and their supporting structures, the vascular systems of the peripheral nerve.
Epineurium Extrinsic vessel
Perineurium
Endoneurium Regional feeding vessel
Unmyelinated fibres
Node of Ranvier
Schwann cell nucleus
Myelin sheath
Axon
Myelinated fibre
56
myelinated fibres. Omer (1980) gives a range of 40–75 m/s in large myelinated fibres. The electrical changes associated with the wave of depolarisation can be measured through electrodes placed on the skin over the nerve, on the nerve, in the nerve or in individual fibres. These reactions form the basis for electrophysiological examination and for microneurography.
2.2.2 The Basis of the Action Potential: Ion Channels Hodgkin and Huxley (1939) made the first intracellular measurements of the resting potential across the cell membrane in the unmyelinated giant axon of the squid. In 1952 Hodgkin and Huxley described the cycle of depolarisation and repolarisation which underlies the high speed transmission of nerve action potentials and showed that the reversal of polarity was brought about by the influx and efflux of sodium and potassium ions across the axon membrane through individual parallel pathways which are controlled by independent gating particles or charges. Conduction of an action potential was blocked by pressure or by cold (Hodgkin 1937a, 1937b) Chiu (2005) defines the ion pathways, now known as channels: “voltage gated ion channels are like membrane lodged proteins that mediate rapid ion flux (106 ions/s) across cell membranes.” Chiu describes some of the methods that have been developed to define these entities. The patch clamp technique permitted study of the electrical events associated with the opening and the closing of a single ion channel. Later came the cloning of ion channels which permitted the recognition of 50 potassium channel genes and 10 sodium channel genes (by 2005). Later still the pore structure of potassium channels was studied by x-ray crystallography at a resolution of between 2.4 and 3.2 Å. Scholz, Reid, Vogel and Bostock (1993) were amongst the first to study, by patch clamping, sodium and potassium channels in the human nerve. The voltage-gated sodium ion channels are uniformly distributed along the membrane of nonmyelinated axons but they are densely concentrated at the nodes of Ranvier in the myelinated nerve axons. The potassium channels, on the other hand, are concentrated in the membrane at the juxta paranode. Ion channel function is energy dependent, it is ATP driven and this function is curtailed or altogether blocked by anoxia. Distortion of the myelin sheath adjacent to the node of Ranvier may unmask the potassium ion channels to such an extent that prolonged conduction block ensues. Evidently, demyelination is bound to lead to decrease of conduction velocity (McDonald 1963; McDonald and
Surgical Disorders of the Peripheral Nerves
Sears 1970) and eventually to conduction block. These facts will not escape the attention of the clinician faced with an ischaemic limb or with a deepening nerve lesion caused by expanding haematoma or entrapment. Once again we are indebted to Praveen Anand and his colleagues who have provided us with illustrations of sodium and potassium channels in the normal and in the injured nerve. Their extensive studies on the behavior of the ion channels in the injured nerve and in painful lesions of nerves are described in Chaps. 3 and 12. One of their observations reveals an important difference between the immature and the mature peripheral nerve in the human. Two voltage-gated sodium channels, Nav 1.8 and Nav 1.9 play a key role in neuropathic pain. The sodium channel Nav 1.9 could not be demonstrated in the nerves of infants (Yiangou et al. 2000) (Figs. 2.29 and 2.30).
Fig. 2.29 Sodium channel staining normal sural nerve, x40 (Courtesy of Professor Praveen Anand).
The Microscopic Structure of the Nervous System: Its Function
57
Fig. 2.30 Nerve fibres within the mucosa and submucous plexus in human rectum stained with antibody to Protein Gene Product 9.5 (PGP9.5), a pan neuronal marker, x10 (Courtesy of Professor Praveen Anand).
2.2.3 Axonal Transport The axon functions as part of the neurone as a whole in transporting materials to and from the cell body. Lasek (1982) goes further than this: neurones exhibit a remarkable form of locomotion when they extend axons over great distances without moving the cell body. This capacity of neurones to extend axons independently of the movement of the perikaryon is one of the distinctive properties of the neuronal linkage because it distinguishes neurones from other migratory cell types.
Lasek (1982) further proposes that the unusual feature of neurones – their ability to translocate the axonal skeleton independently of the perikaryon – is accomplished by the continuous addition of cytoskeletal proteins at the proximal end of the cytoskeleton in the perikaryon. So, the neurone is able to extend its process “without towing the cell body along.” Ochs and Brimijoin (1993) define axonal transport as: “a system of intra cellular motility enabling nerve cells to deliver essential proteins and membrane components to the periphery, and to receive from these chemical signals and materials for disposal.” Two forms of transport, fast and slow, are recognized (Brimijoin 2005). The former may be orthograde (centrifugal) or retrograde (centripetal). The fast retrograde (centripetal) component conveys material to the cell body in microvesicles at the rate of 150– 300 mm/day. The fast orthograde (centrifugal) component
transports proteins, peptides, and neurotransmitters from the cell body at a rate of 200–400 mm/day. All systems are ATP dependant, and the microtubules are critical for fast axonal transport. Landon (1985) puts it this way: fast transport “is concerned with the orthograde transport of particular constituents of the axoplasm and materials such as some transmitter synthesizing enzymes, glycoproteins and membrane components, and the retrograde transport of membranous prelysosomal structures, and extra cellular materials such as nerve growth factor ingested at the axon terminal.” The process is sensitive to temperature; it is sensitive to deprivation of oxygen. Slow transport is uni-directional, orthograde (centrifugal). Rates of transport are from 1 to 4 mm daily; it is concerned with the transport of the neurotubule-neurofilament network of the cytoskeleton. Brimijoin recognizes two distant components: 1. Slow component A (SCa averaging about 1 mm/day) 2. Slow component B (SCb averaging 2–10 mm/day) The rate of transport of the SCa component is of course about the same as the rate of peripheral regeneration after axonotomy. The significance of axonal transport in disorders of peripheral nerves is plain: interference with the centrifugal process is likely to lead to defect or cessation of conduction; interference with the centripetal process will ultimately lead to degeneration of the nerve cell.
58
2.2.4 The Blood Supply of Nerves Nerves have a very good blood supply: they need it. The surgeon notes the segmental blood supply from without and the axial vessels within the nerve. There are indeed (Lundborg 1979,1988) intrinsic epineurial perineurial and endoneurial plexuses, and extrinsic regional vessels in the “paraneurium.” These form “separate but extensively interconnected microvascular systems” providing a wide margin of safety. As McManis and colleagues (1993) remark: “these anastomotic vessels confer a resistance to ischemia in peripheral nerves so that nerve suffers functional or structural changes only when there is widespread vascular or microvascular damage.” Dyck et al. (2005) show that arterioles and small arteries are detectable only in the epi- and perineurium, and that these vessels may range in diameter from 50 to 400 mm. The arteries penetrating the perineurium are small, only rarely do they exceed 15 mm in diameter and small arteries are rarely detectable within the endoneurium. Lymphatics have not been detected within the perineurium. The endothelial junctions of the endoneurial vessels are tight, unlike those in the epineurium (Low 1976). Some nerves are better supplied than others. Kadiyala and his colleagues (2005) showed that the common peroneal nerve at the knee has far fewer extrinsic vessels than the tibial nerve at that level. Seddon (1975) commented on the extent to which it was permissible to mobilize the nerve in order to facilitate suture. His rather sanguine views of the effect on the blood supply are not confirmed by the later injection studies of Bell and Weddell (1984); indeed, later clinical experience has shown that it is preferable to bridge a gap by interposition than by mobilisation.
Surgical Disorders of the Peripheral Nerves
Weerasuriya (2005) proposed the concept of a “bloodnerve interface” instead of the “blood-nerve barrier.” Weerasuriya comments: “the structure of the perineurium combines features of the dura mater in terms of mechanical strength and the arachnoid with respect to impermiance.” Endoneurial fluid is in contact with the cerebro-spinal fluid through the subarachnoid angle. Flow is proximal to distal, the hydrostatic pressure of cerebro-spinal fluid is 10 mmHg, it is 3.5 mmHg in the dorsal root ganglion and it drops to between 1 and 2 mmHg within the peripheral nerves. Endoneurial hydrostatic pressure (EHP) rises in the aged nerve and it is possible that external pressure increases EHP by obstructing flow. This may be one factor underlying the effect of entrapment or compression. Hill and Hall (1999) suggest that the aggregation of Renaut bodies at sites of potential entrapment represents a response to local injury to the endoneurial capillaries. The blood supply to the roots of the spinal nerves is much less robust (Figs. 2.31a–c and 2.32). Two distinguished Cambridge anatomists, Woollam and Millen (1958), considered that: “man has just as much nervous system as he can supply with oxygen and no more.” The most important spinal vessel, the anterior spinal artery, was studied in the fetus and in the guinea pig and rat. Relatively few radicular arteries survived into adult life, the average number of those so doing was eight. Two of these seemed to be particularly important: a cervical vessel, arising from the vertebral artery and entering into and sustaining the anterior spinal artery at C6, C7 or C8, and the artery of Adamkiewicz 1881a,b, in the upper lumbar region. Figure 1 in Woollam and Millen’s paper shows, in a 24 week human fetus, that the radicular vessel adjacent to
Fig. 2.31 The dorsal vessels to the spinal cord and cauda equina: (a) upper thoracic segment (b) thoraco lumbar junction, (c) cauda equina.
The Microscopic Structure of the Nervous System: Its Function
a
59
b
Fig. 2.32 Resin cast of the blood vessels of the neck, seen from the front (a) and from the side (b). The arteries are shown red, the veins blue showing the segmental arrangement of vessels. The anterior spinal artery is shown (arrow) with an important segmental feeder vessel.
the seventh cervical nerve was the major supplier for the anterior spinal artery (Fig. 2.33). Dommisse (1974, 1975) confirmed that the number of radicular arteries (which he termed the medullary feeders) reinforcing the anterior longitudinal arterial channel was eight and that those reinforcing the dorsal arterial columns were 17. Only 8% of those passing to the cervical spinal cord arose from the vertebral artery. The pattern was variable: “but the principle of a rich supply for the cervical and lumbar enlargements was confirmed” (Dommisse 1975). In the 6 drawings from cadaver dissections a total of 12 vessels are described, 4 of these at the level C6, C7 and 4 more at C4. The anterior spinal artery is the most important of the longitudinal channels. Its central branches, which are end arteries, supply about two-thirds of the cross-sectional area of the cord. The rest of the dorsal grey and the white columns are supplied by branches arising from the dorsal arterial system. Occlusion of the anterior spinal artery leads to the catastrophe of infarction of the anterior cord, the anterior spinal cord syndrome. The significance of damage to major feeder arteries is emphasized by the work of Svensson (2005) who addressed the risk of severe cord lesion after aortic surgery. Svensson achieved a rate of paralysis of 3.8% on patients most at risk, those with complex thoracoabdominal aneurysms, by a number of measures which included: “sequential segmental repair with repositioning and moving the clamp upon the grafts sequentially downwards
and reattaching intercostal and lumbar arteries.” Disruption of the radicular arteries entering the spinal canal with the spinal nerves which form the brachial plexus probably underlies the partial Brown Séquard syndromes which are seen in cases of avulsion. Occlusion of flow through those vessels by tamponade explains the catastrophe of spinal cord infarction or even death complicating spinal nerve or interscalene block.
2.2.5 The Nervi Nervorum Curiously but perhaps predictably, nerves have their own nerve supply in the shape of the nervi nervorum, derived from their own fibres. There are free endings in the epi-, peri- and endoneurium, and some encapsulated endings of Pacinian type in the endoneurium. (Hromada 1963). These are probably one factor underlying the exquisite sensitivity of nerves trapped in fibrosis after operations for “entrapment” neuropathy.
2.2.6 Changes in Nerves with Ageing Corbin and Gardner (1937) examined the dorsal and ventral roots of the eighth and ninth thoracic spinal nerves of 34
60
Surgical Disorders of the Peripheral Nerves Posterior spinal aa.
Posterior segmental medullary a. Spinal branch Anterior segmental medullary a.
Posterior segmental medullary a. Spinal branch Anterior segmental medullary a.
Anterior spinal a.
Fig. 2.33 The segmental medullary (radicular) arteries and the anterior spinal artery in the lower cervical cord (after Woollam and Millen 1958).
cadavers whose ages ranged from 1 day to 89 years. The highest number of myelinated fibres, found in persons in their second and third decades, was 187% larger than that found in the nerves of a 1 day old infant. They found that after the third decade there was a gradual loss of myelinated fibres, up to 32% for the person of 89 years. Cottrell (1940) examined the median, femoral, sciatic and common peroneal nerves of 30 persons coming to necropsy at ages ranging from 3 h to 81 years. She found changes in the vessels and the connective tissue, with “alteration and loss of the nerve fibers.” Ochoa and Mair (1969) examined portions of the sural nerves of seven volunteers aged 5–59, and were the first to show that active destruction of unmyelinated fibres started early in life, whereas loss of myelinated fibres was found only in the 59-year-old. Tohgi and colleagues (1977) examined the sural nerves of 79 persons coming to necropsy after “acute death,” whose ages ranged from 1 week to 88 years. The average density of small myelinated fibres decreased rapidly from the age of 1 week (26,300/mm2) to the second decade (9,560/mm2), and continued to decrease gradually with age, to reach at the eighth decade an average of 74% of that at the second decade. There seems to be an error in the printing of the figure for the eighth decade. Large myelinated fibres appeared first in an infant aged 3 months. Their density increased rapidly, to reach at 3 years the level found in the young adult. The average of the
third decade was 6,480/mm2; at the ninth decade it was 3,480/ mm2. In this extensive study there is mention of the ageing changes in the vasa nervorum and the perineurium. Jacobs and Love (1985) made a qualitative and quantitative study of 27 sural nerves obtained within 24 h of death from human subjects without history of disease or of ingestion of drugs known to affect peripheral nerve. Densities of myelinated and unmyelinated fibres decreased from birth to the end of the eighth decade, because of increasing size and separation of fibres during the first decade, and an increase in endoneurial collagen in the older persons. Also, the slope of the internodal length-fibre diameter increased progressively during the first decade, but then remained virtually constant until the age of 60. Then, degeneration, regeneration, demyelination and remyelination caused increasing variation in internodal length. Thomas et al. (1993b), summarising previous findings, refer to “mild peripheral neuropathy” of older humans. Norris and colleagues (1953) examined age changes in the maximum conduction velocity of motor fibres of the ulnar nerves of 175 ambulatory male patients, employees and staff members of the Baltimore City Hospitals. There was steady reduction in conduction velocity from the third to the ninth decade. They canvassed possible causes for this regression. Taylor (1984) studied the effects of age on conduction and amplitude in motor and sensory fibres of adult nerves. He indicated that tables of normal data of the rise and
The Microscopic Structure of the Nervous System: Its Function
fall of conduction and amplitude could be constructed for use in clinical investigation. Kimura (1993) stated that: “nerve conduction velocities are roughly half the adult value in fullterm infants, but increase rapidly, reflecting the process of myelination, to the adult value at age 3–5 years.” He noted the slow decline of conduction velocity after the fourth decade, and drew attention to the contemporaneous increase of the latencies of the F wave and somatosensory evoked potentials. Cowen et al. (2005) describe the greater vulnerability with increasing age of large, long, myelinated nerve fibres, and of larger sensory neurones. There also appears to be selective vulnerability of some autonomic neurones, notably within the enteric system. Evidence is provided to support the proposal that “interactions between sensory neurones and their receptor targets are crucial initiators of an age related nervous deterioration.” These changes must evidently concern clinicians treating the very young and the rather old: in the former, they may have a bearing on diagnosis; in the latter, they may be relevant to the susceptibility of a nerve or nerves to damage by pressure or traction.
61 V
Re
V
C Re Ru
2.3 The Somatic Motor System The motor pathway begins in the neurones in the pre-central gyrus of the cerebral cortex. Their axons pass by the internal capsule to the mid-brain and to the pyramids of the medulla. From each side most fibres cross the mid line at the decussation of the pyramids to descend in the lateral part of the white matter of the cord as the lateral corticospinal tract. At various segmental levels impulses from this tract activate, through internuncial neurones, the motor cells in the anterior part of the grey matter (Fig. 2.34). “Extrapyramidal” tracts from the red nucleus, the vestibular nuclei and the reticular formation also influence the activity of the ventral horn neurones. Brodal (1981c) prefers to discard the designation “extrapyramidal,” on the grounds that “those cortical regions which give rise to fibers in the pyramidal tract also give rise to fibers to a number of nuclei which project further caudally….” The cell bodies of the motor neurones are in Lamina IX (Rexed 1954) of the ventral horn of the grey matter. There are large (alpha) and small (gamma) cells (Fig. 2.35). They are acted on by primary sensory fibres and by fibres descending from the cortex and from nuclei in the brain stem. The axons from the large cells are destined for the extrafusal fibres. By correlating the distribution of paralysis with the sites of loss of cells in the ventral horn, Sharrard (1955) was able to show how the cells were grouped in the grey matter. Broadly, the medial group supply the muscles of the trunk and neck; the lateral group supply the muscles of the limbs.
Fig. 2.34 The major descending tracts in the spinal cord and their overlapping zones of termination in the grey matter. C corticospinal, V vestibulospinal, Re reticulospinal, Ru rubrospinal.
Thus, cells of the latter group are present chiefly in the cervical and lumbar enlargements, whereas those of the medial group are found throughout the length of the cord. Nathan and Smith (1958) studied the spinal cord in patients who had earlier undergone cordotomy, and defined the descending
62
Fig. 2.35 Motor neurone cell bodies in the ventral horn of human spinal cord stained with antibodies to TRPV 4, a novel ion channel, x40 (Courtesy of Professor Praveen Anand).
pathways for the control of the bladder: “in man, the majority of descending fibers concerned with micturition lie in the lateral column, on an equatorial plane passing through the central canal. This location remains the same in the cervical, thoracic, lumbar and sacral segments.” The distribution of nerves within the muscles of the upper limb has been described by Lim and his colleagues (2004). The staining technique of Sihler was used which renders muscle translucent and stains the myelin in the nerves a dark blue. The intramuscular distribution of the nerves was then mapped by photographing the superficial and deep surfaces of the muscle using a back light technique. In flat, triangular or trapezoid muscles (class 1) the main nerve runs perpendicular to the muscle fibres giving off side branches that run parallel with them. The spindle shaped or fusiform muscles (class 2) were subdivided into unipennate or bipennate muscles. In the bipennate muscles the aponeurosis of the tendon splits the muscle into two compartments and in these the primary nerve divided into two secondary branches passing each side of the tendon. In muscles with more than one head of origin (class 3) the pattern of innervation is more complex. The pattern of innervation is determined by three factors: “the shape of the muscle, its position and orientation in relation to the passage of the nerve, and the muscle tendon morphology.” These findings support the idea of transfer of part of a muscle and they emphasize the requirement for the repair of intramuscular nerves in lacerated muscles. Contact with, and transmission to muscle is effected through the motor end-plates (see Fig. 2.42). There are two components of each end plate: neural and muscular. They are separated by a cleft of about 30 nm. The muscular soleplate contains a number of muscle cell nuclei; it is not itself
Surgical Disorders of the Peripheral Nerves
contractile. There are two types of neural endings: the en plaque terminal on extrafusal (alpha nerve fibre) muscle fibre, and the plate endings on intrafusal (gamma nerve fibre) muscle fibres. Transmission at en plaque endings initiates action potentials which are rapidly conducted to all parts of the muscle fibres, whereas transmission at plate endings of “trail” and “en grappe” types excites the fibres at several points. Acetylcholine released at the nerve ending interacts with receptors to produce depolarisation of the muscle membrane and trigger the action potential in the muscle. The ventral roots from the first thoracic to the second lumbar segments of the spinal cord contain also the efferent pre-ganglionic fibres of the sympathetic nervous system: those of the second to fourth sacral nerves contain the efferent pelvic parasympathetic outflow.
2.4 The Somatic Sensory System The afferent pathways of the peripheral nervous system considerably exceed the efferent pathways in numbers and in complexity of organisation. By no means all lead to conscious sensation. Amongst the somatic afferents the Golgi organs and the muscle spindles are examples; the whole array of the visceral afferents is one more.
2.4.1 Cutaneous Sensibility The long debate about the mechanism of cutaneous sensibility begun by Blix (1884) and Goldscheider (1884) and continued by von Frey (1894, 1896), Head et al. (1905), Head and Sherrin (1905), Head (1920),Adrian and Zotterman (1926), Adrian (1928), Zotterman (1939), Weddell (1941), Sinclair (1955) and many others is now drawing to its close: as Iggo (1985) remarks “the long-standing argument of ‘specificity’ versus ‘pattern’ in the operation of cutaneous receptors has been settled in favor of the ‘specificity’ hypothesis.” Iggo also states that: “it is now clear that no cutaneous receptors have an absolute specificity; they have a high degree of selective sensitivity, that is, a much reduced threshold to one form of stimulation” (Fig. 2.36). Adrian pioneered methods in Cambridge in the 1920s (Adrian and Zotterman 1926) which enabled electrophysiological studies of single afferent units. This led to work with microelectrodes and with intraneural microstimulation done by, among others, Hallin and Torebjörk (1973), Torebjörk and Ochoa (1980), Vallbo and Hulliger (1981) Hagbarth et al. (1993), and Torebjörk Schmelz (2005). However, as Wall
Fig. 2.36 Cutaneous sensory receptors: glabrous skin (left) and hairy skin (right). Aα Aδ C
‘G’ unit ‘D’ unit Cutaneous (C) unit
Aδ/C
Aα
Field receptor
Thermoceptors
Aα
Lanceolate
Aδ/C
Aα
Aα
Aα
Aα
Paciniform
Type II (Ruffini)
Type I (Merkel)
Vibrissa
Nociceptors
Rapidly adapting mechanoreceptors
Slowly adapting mechanoreceptors
Hairless skin
Aα
Aδ/C
Aα
Meissner
Thermoceptors
Aα
Pacinian
Aδ/C
Aα
Type II (Ruffini)
Nociceptors
Aα
Type of ending: Type I (Merkel)
Type of fibre:
Rapidly adapting mechanoreceptors
Slowly adapting mechanoreceptors
The Microscopic Structure of the Nervous System: Its Function 63
Hairy skin Guard hair Down hair
Touch dome
64
(1961) pointed out: “specific modality and patterning theories are supplementary. The recognition of receptor specialisation for transduction of particular kinds and ranges of cutaneous stimulation does not preclude acceptance of the concept that information is coded in a pattern of impulses.” Mountcastle (1980) assigns to the specialized receptors a role as transducers responsible for amplification. He points out that in areas of skin subjected to sensory testing and later marked and excised, histological examination has shown only free nerve endings. No specialized mechanosensor transducers transmitting through unmyelinated fibres have been identified yet; the same is not so for nociceptors and most thermoreceptors, where the transient receptor potential (TRP) channels have been identified as transducing specific ranges of temperature. All these receptors seem to be represented in fine branching unmyelinated nerve endings in the cell layers of the epidermis (Fig. 2.37). The basis of stereognosis is a combination of stimuli from skin, tendons, muscles and joints relaying sensory information centrally where comparison is made from memories of movement. The role of movement is vital: a blindfolded person cannot identify the nature of a material if it is simply placed on the finger. Identification is aided if the material is drawn across the finger tip. Recognition is, however, immediate if the subject is allowed to create temporal and spatial patterns by feeling the texture between the moving finger and thumb (Melzack and Wall 1962). The fibres of afferent neurones are classed by their conduction velocity. Afferent fibres from the skin are divided into A-ab, A-d and C; muscle afferents are classed I, II, III and IV (Light and Perl 1993, Lawson 2005). There is some correlation between fibre diameter and the characteristics of the soma within the dorsal root ganglion. These are classed as large light (neurofilament rich) and small dark (neurofilament poor) neurones.
Surgical Disorders of the Peripheral Nerves
Fig. 2.38 Small diameter nociceptor cell bodies from a human dorsal root ganglion 2 weeks after avulsion stained with antibodies for prostaglandin receptor sub type EP1, x40 (Courtesy of Professor Praveen Anand).
Fig. 2.39 Small diameter nociceptor cell bodies and axons in the human dorsal root ganglion immunoreactive for the heat and capcaicin receptor TRPV1 2 weeks after avulsion injury, x20 (Courtesy of Professor Praveen Anand).
The neurones of C-fibres are small; those with Ad fibres are small to medium size; those with A-ab fibres are medium to large (Lawson 2005) (Figs. 2.38 and 2.39).
2.4.2 The Skin The introduction of immunohistochemical staining of nerve antigens has provided new insights into innervation of the skin (Kennedy et al. 2005). Kennedy says: Fig. 2.37 PGP9.5 immunoreactive somatic nerves in skin of patient with small fibre neuropathy, x 40 (Courtesy of Professor Praveen Anand).
bundles of nerves enter the skin deep in the dermis and course towards the skin surface, giving off axons to innervate the associated end organs. Unmyelinated nerve fibers comprise the vast
The Microscopic Structure of the Nervous System: Its Function
65
morphologically unidentified receptors responding to very slow displacement of hair or skin are innervated by small slowly conducting C fibres. The principal mechano-sensor in hairy skin is the hair follicle receptor; in hairless (glabrous) skin the two principal types are the Meissner corpuscle, rapidly adapting, and the Merkel receptor, slowly adapting. Beneath the skin the rapidly adapting organ is the Pacinian corpuscle; the slowly reacting organ is the Ruffini’s corpuscle which is found in deep dermal layers and is characterized by large diffuse receptive fields. The Ruffini corpuscles provide information about finger position by responding to stretching of the skin (Lundborg 2004b). Stark et al. (1998) studied the distribution of Pacinian corpuscles in the hand: the total number identified was around 300 and most were located in the pulp skin of the fingers. Fig. 2.40 PGP9.5 immunoreactive autonomic nerve fibres surrounding sweat glands in the skin of a patient with small fibre neuropathy, x40 (Courtesy of Professor Praveen Anand).
majority of cutaneous innervation to the above dermal structures. The few myelinated nerve fibers terminate at hair follicles, Meissner corpuscles and Merkel complexes. The vertically orientated nerve bundles form a horizontal sub epidermal neural plexus. Epidermal nerve fibers branch from this plexus and, while penetrating the dermal-epidermal basement membrane to enter the epidermis, they lose their Schwann cells ensheathment and collagen collar.
The sweat glands are carpeted by a dense pattern of autonomic nerves (Fig. 2.40). Fine unmyelinated nerve endings form a network covering larger arteries in the deep dermis. The density of innervation of the epidermis is greatest in the proximal segment of the limb. There is little change between the 20th and 80th year.
2.4.3 Cutaneous Sensory Receptors
2.4.3.2 Thermoreceptors So long ago as 1884 Blix postulated that there were two types of thermoreceptor in the human skin: one for cooling and one for warming. Cooling receptors are served by unmyelinated and fine myelinated fibres, usually serving receptor fields about 100 mm in diameter (Light and Perl 1993). They are very sensitive to decrease in skin temperature from the normal or “neutral” level of 30–35°C. Fowler and colleagues (1988) indicate a conduction velocity of up to 2.1 m/s. Davis and Pope (2002) found that the sensation of cooling is replaced by an ache below 17.5°C and by pain below 14°C. Warming receptors, less common than cooling receptors, have receptive fields of less than a millimetre in diameter. Warm sense is a function of unmyelinated fibres within the epidermis; as we have seen the TRP channels transduce specific ranges of temperature. Temperatures above 43°C induce firing in C-fibre polynociceptors. Temperatures above 53°C evoke responses in fast conducting myelinated mechano-heat fibres (Light and Perl 1993, Lawson 2005).
Three types of cutaneous receptor are defined: low threshold mechano-sensors; thermoreceptors; nociceptors. 2.4.3.3 Nociceptors 2.4.3.1 Low Threshold Mechanosensors The distinction is made between slowly adapting receptors responding to sustained displacement such as continuous pressure; rapidly adapting receptors responding to the beginning or withdrawal of a stimulus or by a moving stimulus, and receptors responding to brief mechanical disturbances such as vibration and tapping. The first group includes the Merkel cells; the second includes the Meissner corpuscles and the third, the Pacinian corpuscles. Most are innervated by large and medium sized fibres conducting at rates of from 20 to 90 m/s. A few
The term is applied to primary afferent units which “uniquely signal stimuli intense enough to threaten physical damage to the tissue” (Light and Perl 1993). Some respond to intense mechanical stimuli; some to strong thermal stimulation, and some are polymodal. Impulses travel in myelinated fibres in the Ad to Aab ranges and in unmyelinated C fibres. Nociceptor fibres are widely distributed in the skin, muscle, joints, the epineurium of trunk nerves and the wall of blood vessels as an extensive plexus of free nerve endings. These pass to fine myelinated and non myelinated fibres and also to the largest (Aab) fibres (Light and Perl 1993, Lawson 2005). Ad nociceptors are high threshold mechano-sensors.
66
Some respond to damaging heat. They conduct impulses from receptive fields of about 5 mm2 at about 20 ms. Many of the C-fibres are polymodal, responding to a range of noxious stimuli including histamine and other chemicals, heat, cutting and crushing. C-fibres are responsible for the triple response of Lewis and they are the basis of the axon reflex. They are less than 2 mm in diameter and conduct at between 0.5 and 2 m/s from fields which range from 1 to 10 mm2. Microneurography has clarified the physiological characteristics of the nociceptors in humans. Sharp, well localized pain follows stimulation of Ad afferents. Stimulation of single C-afferents induces dull, burning, poor localized and delayed pain (Torebjörk and Ochoa 1980). Head (with Rivers and Sherrin in 1905) observed two types of pain after the division and suture of his own superficial radial and lateral cutaneous nerves of the forearm. There was epicritic pain which was sharp and localized and protopathic pain which was dull, burning, delayed and unpleasant. It is tempting to relate these two pain types to the now proven characteristics of the two main groups of nociceptors. The reader can experience the two modalities of pain by stimulating the skin on the front of the wrist with a sharp pin. First, and almost immediately, a sharp, well localized pain is experienced. A little later the delayed response–slightly unpleasant, a little longer lasting and a little diffuse – is felt. The A-ß and A-d nociceptors have punctate superficial receptive fields and respond to noxious mechanical or noxious mechanical and thermal stimuli (Mechano-heat units) (Lawson 2005).
2.4.4 Deep Sensibility Sensation is conveyed from muscles, ligaments and tendons from specialized receptors and from free nerve endings in those structures. The receptor organs are: in muscle, muscle spindles and free nerve endings; in tendons, the Golgi organs, and in capsules and ligaments various endings, some similar to Ruffini endings, Pacinian corpuscles and Golgi organs. There are also plexuses of unmyelinated fibres (Fig. 2.41). Joints are innervated by a network of rapidly conducting myelinated fibres some of which are associated with encapsulated mechanosensors and by high threshold, slowly conducting fibres many of which are perhaps nociceptors (Takebayashi et al. 1997, Petrie et al. 1997, Hogervorst and Brand 1998, Chen et al. 2000). Takebayashi et al. (2006) recognized sympathetic afferents innervating the lumbar intervertebral discs. Tomita et al. (2007) investigated the distribution of nerve endings in the human dorsal radio-carpal ligament by fluorescence immunohistochemistry, with confocal laser microscopy and by Kontron image analysis “to rebuild” the endings, so providing data about morphological characteristics as well as incidence, density and distribution.
Surgical Disorders of the Peripheral Nerves
2.4.5 The Muscle Spindles It is over a 100 years since Sherrington (1894) demonstrated by ventral root section that “in a muscular nerve-trunk from one-third to one-half of the myelinated fibers are from cells of the spinal root-ganglion.” The size of these fibres was from 1.5 to 20 mm; they were not the largest fibres in the nerve; the largest came from the ventral root. On the other hand, the largest of these fibers were larger than any fibres in the cutaneous nerves. “It was shown that about two-thirds of all the afferent fibers measure above 7 mm. Of these I imagine that considerably more than one-half may be apportioned to the muscle spindles, the majority of the rest belonging to Golgi’s tendon organs.” Sherrington found too that “the smallest myelinated fibers in the muscular nerve are for the most part, perhaps entirely, root ganglion fibers.” Sherrington further showed that there were “recurrent” (afferent) fibres in the ventral roots. He identified the special end-organs of the afferent fibres in the muscle-spindles (muskel-spindel) of Kühne (1864). Ruffini (1897), in his observations on sensory nerve endings in muscles, defined the “sensorial” end organs of muscle as (1) the muscle spindles; (2) the tendon organs (Golgi organs) and (3) the Pacinian corpuscles. He concluded: “in my opinion it is upon these three levels of sense organ that physiology must turn its attention if it will resolve the problem of the muscular sense.” Batten (1897) further examined the muscle spindle and its behavior in various pathological conditions including injury to the brachial plexus. Horsley (1897) noted their “preservation in conditions of extreme muscular atrophy.” Since Sherrington’s time, a very great deal of work has been done on the structure and function of the muscle spindles (Banks et al. 1982; Boyd 1962, 1966, Boyd and Smith 1984; Goodwin and colleagues 1972; Cooper and Daniel 1963; Matthews 1981). Their behavior after nerve injury and on their regeneration is described in Chaps. 3 and 4. Boyd and Smith (1984) state that “The whole human body must contain about 20,000 spindles, each of them a marvel of micro-engineering.” Most spindles lie deep in muscle near branches of its nerve or blood vessels. Each contains small muscle fibres (intrafusal fibres) within a cellular and connective tissue capsule. Banks (2005) says: “there are two principle kinds of encapsulated sense organs in skeletal muscle.… the tendon organ, which senses the force of contraction and the muscle spindle which responds to muscle length.” Both have a copious innervation by large afferent fibres and the muscle spindle and the Golgi tendon organ account for nearly all group 1 afferent axons from muscles. There are three types of specialized intrafusal muscle fibres: the nuclear bag fibres (bag 1, bag 2), with a central accumulation of nuclei, and nuclear chain fibres, smaller and with a single row of nuclei. The spindles vary in length from a few millimeters to a centimeter. Each spindle receives up to 25 terminal branches
The Microscopic Structure of the Nervous System: Its Function
67
Ib afferent
Golgi tendon organ
Primary (annulospiral) Ia afferent
Secondary (flower spray) II afferent
Aα efferent Muscle sole plate
Aγ efferent Motor end-plate with synaptic vesicles Muscle spindle
Fig. 2.41 The afferent and efferent innervation of skeletal muscle.
Motor end-plate
68
of motor and sensory axons, together with autonomic innervation. The motor axons are the b and g axons of the ventral root; the sensory axons are myelinated fibres of groups I and II. The combination of motor and sensory innervation is reflected in the complexity of the function of the spindles. The conception of a servo action is evidently too simple; rather, it is evident that the spindles play several roles in the “feedback” mechanism for regulating muscle contraction and for appreciation of body position. Stacey (1969) reckoned that in a “motor” nerve of the cat the distribution of fibres was one-third myelinated motor, one-third myelinated sensory and one-third unmyelinated sensory. Banks (2005) studied the nerve to the soleus muscle of the cat. The nerve contains 180 myelinated sensory and 270 myelinated motor fibres. Most of the myelinated afferents arose from 56 spindles and 45 tendon organs. There were 115 fusimotor gamma efferent fibres, which means that the 25,000 extrafusal skeletal muscle fibres are innervated by only one-third of the total of myelinated nerve fibres. The human longissimus capitis is the most densely spindled muscle and the density of muscle spindles is 25 times more in the lumbrical muscles than in the gastrocnemius (Cooper and Daniel 1963). The other afferent endings in skeletal muscle are free endings innervated by unmyelinated and small myelinated fibres. Iggo (1961) found that these responded to sustained pressure but not to stretch or contraction. They responded to hypertonic saline, but that stimulus was sufficient to excite the spindles too. Other studies of these organs and their afferent fibres were made by Mense and Schmidt (1974) and by Mense (1977) who studied the response to chemical noxious stimulation by single fibre recording. Pomeranz and colleagues (1968) traced fine myelinated afferents from viscera, muscle and skin to Lamina V of the dorsal horn of the spinal cord.
2.4.6 The Golgi Tendon Organs Scott (2005) characterizes the second of the two encapsulated mechano-sensors in muscle. The Golgi organ senses length of muscle and impulses project from it to the cerebellum and cortex. Stimulation of the fibres from the Golgi apparatus in hand muscles causes cortical potentials and “illusions of muscle stretch.” The Golgi organ is about 0.1 mm in diameter and between 0.2 and 1.5 mm in length. It contains collagen strands which continue into muscle fibres at one end and into the tendon at the other. There are between 3 and 50 of these organelles in each muscle, and Scott says that the ratio of Golgi organs to spindles is less than 0.3. The myelinated afferent fibre is a little smaller than the largest afferent from the muscle spindle, and in the cat it conducts at the rate of 60–110 m/s. The terminals interweave amongst the collagen strands as sprays or spirals. The capsule contains capsular cells which
Surgical Disorders of the Peripheral Nerves
are continuous with the Schwann cells. The receptor is slowly adapting; it responds to the whole range of muscle contraction and the firing rate is proportionate to active tension. In humans the fibres are silent at physiological rest and there are progressive steps in the firing rate with increasing steps in muscle contraction. Recovery is virtually complete after a crush lesion inducing axonotmesis, it is very much worse after repair of divided nerves. The poor recovery of the two main encapsulated mechano-sensors in muscle after repair of divided nerves may account for the common complaints of weakness, lack of stamina, and poor coordination and also for the failure of musculotendinous transfer using reinnervated muscles. As MacQuillan (2006) says: “the normal function of muscles is dependent on their sensory apparatus.” Up to now, most clinical work on sensation and on recovery of sensibility after nerve injury and repair has been directed to cutaneous sensibility. Yet function such as stereognosis and proprioception must depend principally on signals from endings in muscle, tendons and ligaments. It is perhaps inadequately appreciated that there may be good recovery of sensory function of the hand with very imperfect cutaneous reinnervation, and that pain is just as likely to follow damage to a “purely motor” nerve as it is to follow damage to a “mixed” or “sensory” nerve. There is in fact no such entity as a “purely motor” nerve, except perhaps for the hypoglossal or facial. The signals from the muscles supplied by those nerves probably proceed by other cranial nerves: the lingual in the case of the hypoglossal nerve and the trigeminal and the auricular branch of the vagus in the case of the facial nerve. There are indeed a few peripheral nerves without a cutaneous sensory component: the spinal accessory; the phrenic; the anterior and posterior interosseous; the deep branch of the ulnar; the suprascapular. The content of afferent fibres in all such nerves is about 30%. Laviano (1992/93) in his thesis on “transfer of the spinal accessory nerve for the suprascapular in avulsions of the brachial plexus” shows EM photographs of portions of the suprascapular nerve taken at operation as long as 94 days after injury (Fig. 2.42). There are numerous medium-sized and small myelinated fibres in good condition: evidently, the distal processes of dorsal root ganglia. Our work along the same lines also suggests a proportion of myelinated afferent fibres for the “motor” nerve of around 30%. Laviano’s work confirms the earlier supposition (Bonney 1959) that the dorsal root ganglion cells survive for a long time the interruption of their central processes. As to the last: in a case of extensive intradural damage to the plexus we examined a specimen of suprascapular nerve taken at the time of “neurotisation” well after the period required for degeneration, and found that it contained not less than 30% of myelinated fibres. It is perhaps best to drop the terms “purely motor” and “purely sensory,” and even drop the term “mixed” applied to nerves with both motor and cutaneous sensory components. The terms
The Microscopic Structure of the Nervous System: Its Function
69
2.4.7 Central Connections
Fig. 2.42 Intact (afferent) myelinated and unmyelinated fibres in the suprascapular nerve after avulsion lesion of the brachial plexus. Specimen taken 6 weeks after injury, when efferent fibres had degenerated, x6,600(Electron microscopic study by Mr Stephen Gshmeissner).
“nerves with motor and cutaneous sensory components” and “nerves without somatic motor components” are, unfortunately, cumbersome, but they do say what they mean.
The great array of sensory receptors in the skin and deep tissues sends back to the centre the signals of the stimuli received. Most afferent fibres, with their cell bodies in the dorsal root ganglia, enter the cord by the dorsal roots. Others, with cell bodies in the dorsal root ganglia or actually in the ventral roots, enter the cord by the latter (Fig. 2.43). The first analysis of incoming signals takes place in the spinal cord and medulla where all fibres terminate. Most of the large myelinated fibres ascending in the dorsal columns terminate in the gracile and cuneate nuclei in the medulla. Some smaller fibres of the dorsal columns terminate and relay in the cord: these are the propriospinal fibres. Although the classical view of the function of the dorsal column has been challenged (Wall 1970) it is broadly true that, as Brodal (1981d) states, they “mediate sensory signals necessary for rather complex discrimination tasks” (Nathan et al. 1986). Other afferent fibres terminate and relay in the grey matter of the dorsal horn. Each lamina (Rexed 1952, 1954; Price and Meyer 1974) of the grey matter receives afferents of specific functional modalities; each has a particular neuronal structure. Small myelinated nociceptor and thermoreceptor fibres terminate in lamina I; C fibres, nocithermo- and mechanosensors, in lamina II (substantia gelatinosa). Larger mechano-sensor fibres terminate in laminae III and IV. These relay to cells whose axons either ascend in the dorsal columns or reach the dorsal column nuclei by the dorsolateral fasciculus. Some fibres pass through the dorsal horn to relay with the large cells in the motor apparatus in lamina IX. Some unmyelinated and small myelinated fibres enter the dorsolateral fasciculus (Lissauer’s tract) just lateral to the tip of the dorsal horn to join fibres from cells in the substantia gelatinosa. Direct ascending tract
Lissauer’s tract Lateral corticospinal tract
III IV V VI
II
Dorsal root ganglion
I
X VII VIII
Fig. 2.43 The laminae of the grey matter with direct ascending, crossed ascending and internuncial tracts.
Crossed ascending tract
IX
70
Surgical Disorders of the Peripheral Nerves
Fig. 2.44 Ascending tracts in the cord, brain stem and cortex. Note the relay of the directly ascending tracts in the gracile and cuneate nuclei in the medulla (after Brodal 1981a).
Leg
Trunk
Gyrus post. centr. Arm Caps. int. Face
Pes. pedunculi Nucl. ruber Subst. nigra Nucl. sens. princ. n. v.
Nucl. vent. post. lat. thalami (VPL) Mesencephalon
Pons Lemn.med.
Nucl. fun. gracil. Nucl. fun. cun. N. tract. spin. n. v. Tr. sp. cereb. dors. Tr. sp. cereb. ventr.
Med. obl. Tr. spinothal.
Med. spin.
Some fibres cross the midline to terminate in laminae I and V of the contralateral dorsal horn. There is a complex network of interconnecting fibres in the dorsal horn and in the substantia gelatinosa in particular. Sensory input is first analyzed and modified here. Secondary neurones in the dorsal horn give rise to fibres which ascend or descend for a few segments in the cord. They give rise principally to the fibres that, crossing the midline, ascend in the long tracts in the anterolateral segment of the spinal cord (Fig. 2.44). The transmission of impulses from the nuclei of the dorsal column is influenced by fibres descending from the sensory motor cortex (Gordon and Jukes 1964). This influence is predominantly inhibitory. The ascending fibres of those nuclei form the medial lemniscus of the brainstem, crossing the midline in the medulla to end in the thalamus. The final resolution of sensory impulses takes place in the somatosensory areas of the cerebral cortex (Mountcastle 1957). Not
even in this last analysis are afferent functions separated completely from motor function: stimulation of any of these areas produces motor effect (Woolsey 1964).
2.4.8 Visceral Afferents Schott’s (1994) principal proposition was that “sympathetically determined” pain was so determined because the afferent pathways were in the autonomic nerves. He noted the finding by Varro and colleagues (1988) of calcitonin generelated peptide (CGRP) in visceral afferents, admitting that selective histochemical markers for specifically pain-serving afferents were not, at that time, available. In developing his thesis Schott reverted to the conception of a system of visceral afferents which comprised not only fibres from organs
The Microscopic Structure of the Nervous System: Its Function
Fig. 2.45 Afferent fibres in the enteric division of the autonomic system. Nerve fibres within the mucosa and submucous plexus of human rectum stained with antibodies to sensory sodium channel (Nav1.7), x10 (Courtesy of Professor Praveen Anand).
generally classed as “viscera” but also afferents from blood vessels (Fig. 2.45) The evidence for the presence of a system for conveying from all viscera sensations both perceived and unperceived rests largely on indirect observations: the truly autonomic functioning of viscera: the production of pain by mechanical stimulation of peripheral arteries and veins; the operation of “referred pain” mechanisms; pain after operations on the sympathetic chain; the lack of “visceral sensibility” when the function of visceral afferents is impaired by age. There is evidence from the work of Sugiura and colleagues (1989) that visceral afferents terminate in laminae IV, V and X of the dorsal horn as well as in laminae I and II. They were also traced up and down in the cord and crossing the midline. In their final “conclusion,” Sugiura and colleagues (1989) proposed that their morphological observations suggested that “the somato-visceral convergence could occur in the superficial dorsal horn of the spinal cord,” and that “the scattered and extensive distribution of the terminal fields of single visceral C-afferent fibres may be one basis for the poor localisation of visceral sensation.”
2.5 Cortical Maps The fact that individual movements were controlled by specific areas of the cerebral cortex was recognized in the latter part of the nineteenth century and a “body map” of the sensory cortex was developed by Penfield and Boldrey (1937). The extensive observations made by these early workers do not support the notion of a rigid motor and sensory homunculus. Graham Brown and Sherrington (1912) and Sherrington and Leyton (1917) recognized the functional instability of
71
cortical motor points. A motor response might be elicited by stimulation of the post central cortex after preliminary stimulation of the precentral gyrus, a process that they termed facilitation. Repeated stimulation of the face area of the motor cortex in chimpanzees was followed by swift expansion of the region to embrace the territory originally representing the hand. Repeated stimulation sometimes induced a change from flexion to extension, a phenomena that they termed “reversal.” A third phenomenon, of “deviation of response” was recognized: “a cortical point can also influence the motor response of another whose response is neither diametrically opposed to nor identical with or very closely similar to its own” (Graham Brown and Sherrington 1912). Penfield and Boldrey (1937) mapped the areas of motor function by direct stimulation of the cerebral cortex during operations in 163 patients which were conducted under local anesthetic for the purpose of removing tumor or epileptogenic foci. Localisation was sharply defined for finger movements but very much less so for movements of the tongue and the jaws. Stimulation of the post central gyrus usually evoked a sensory response but this was by no means sharply defined. Even in the best defined map of finger sensation over one-sixth of the 158 points of stimulation actually lay in the pre central gyrus (Fig. 13 in the original paper). On 11 occasions stimulation of the cortex evoked a sensation of pain and in 13 more a sensation of cold. Patrick Wall (Wall 1977) provided clear evidence of plasticity within the adult somatosensory system by recording from single neurones and dorsal column nuclei and finding striking changes in the size of the receptive fields soon after partial deafferentation. Wall suggested that these changes might be caused by the unmasking of synapses normally ineffective or silent. The concepts of brain plasticity are reviewed by Lundborg (2004a) in his excellent monograph . Lundborg says that: brain plasticity implies the capacity of cortical synapses to change their function as circumstances require. In a short term perspective, they may rapidly alter their function, as a result of unmasking or potentiation of already existing synapses. In a more extended perspective, the synapses may increase or decrease in actual numbers and new dendrites may be formed.
(Lundborg 2004c). He continues: “cells that fire together, wire together i.e. neurones become involved in accomplishing the same function and learn to work together efficiently. This phenomenon is named Hebbian learning. Conversely, cells that fire apart, wire apart – i.e. neurones that are not involved in accomplishing the same function learn to ignore each other.” (Hebb 1947) We touch upon other aspects of brain “plasticity” in later chapters but should not forget the salutary observations of Ramachandran and Hirstein (1998): “it is an embarrassing fact that despite five decades of single unit physiology in animals, studied in excruciating detail, we still have no clear idea of how the brain works or why cortical maps exist.”
72
2.6 Synaptic Activity The transmission of impulses at synapses is chemical, by the release of neurotransmitters causing a change in the permeability and hence the electrical polarisation of the post-synaptic membrane. Such changes may be excitatory or inhibitory; they are usually short-lived, because of early inactivation of the neurotransmitter. This is not of course the whole process: the effect of some neurotransmitters may be more prolonged or even permanent. In addition, some substances released at synapses may simply modify the response of the post-synaptic membrane to neurotransmitters. The general term “neuromediators” has been applied to substances released at synaptic endings; “neurotransmission” implies a direct effect on post-synaptic membrane; “neuromodulation” implies alteration of its response to a neuromediator ( Wigley et al. 2008). The best-known mediator and the one that has longest been known, is of course acetylcholine, synthesized by motor neurones and released at the motor terminals in skeletal muscle and at the synapses in sympathetic and parasympathetic ganglia. The other well-known mediators belong to the monoamine group: they are noradrenaline, adrenaline, dopamine, serotonin and histamine. Noradrenaline is the chief transmitter at the endings of sympathetic ganglionic neurones. Adrenaline too is present in peripheral neural pathways. Nitric oxide (NO) mediates smooth muscle relaxation at autonomic synapses. The other monoamines are chiefly present in the central nervous system. Gamma amino butyric acid (GABA) is a major inhibitory transmitter which is released at the terminal of such local circuit neurone systems as the inhibitory Renshaw loop. Glycine is another example of an inhibitory transmitter which is particularly prominent in the lower brain stem and spinal cord. Glutamate and aspartate are widely distributed excitatory transmitters. The range of neuropeptide modulators is very wide, including those associated with the function of the hypothalamus and hypophysis, corticotrophin, beta-endorphin, the enkephalins, calcitonin-related gene peptide and nerve growth factors. In the field of peripheral nerves, the last is of great and growing importance; the beta-endorphins and enkephalins are important in the consideration of the mechanism and treatment of pain.
References Adamkiewicz AA (1881a) Die Blutgefässe des Menschlichen Ruckemarkes; 1) Die gefässe der Ruckenarksubstanz Situngsb.d.k Acad.d. Wissensch. Math Naturw C1 84:469–502 Adamkiewicz AA (1881b) Die Blutgefässe des Menschlichen Ruckemarkes; 2) Die gefässe der Ruchenmarksoberfläche Situngsb.d.k Acad.d. Wissensch. Math Naturw C1 85:101–130 Adrian ED (1928) The action of the sense organs. In: The basis of sensation. Hafner Publishing Co Ltd, London
Surgical Disorders of the Peripheral Nerves Adrian ED, Zotterman Y (1926) The impulses produced by sensory nerve endings. Part 2. The responses of a single end organ. J Physiol (Lond) 61:151–171 Anand P, Terenghi G, Warner G, Kopelman P, Sinicropi DV, WilliamsChestnut RE (1996) The role of nerve growth factor in human diabetic neuropathy. Nat Med 2:703–707 Anand U, Otto WR, Casula MA, Day NC, Davis JB, Bountra C, Birch R, Anand P (2006) The effect of neurotrophic factors in morphology TRPVI expression and capsaicin responses of cultured human DRG sensory neurons. Neurosci Lett 399:51–56 Anand U, Otto WR, Facer P, Zebda N, Selmer I, Gunthorpe MJ, Chessell IP, Sinisi M, Birch R, Anand P (2008a) TRPA1 receptor localisation in the human peripheral nervous system and functional studies in cultured human and rat sensory neurons. Neurosci Lett 438:221–227 Anand U, Otto WR, Bountra C, Chessell IP, Sinisi M, Birch R, Anand P (2008b) Cytosine arabinoside affects the heat and capsaicin receptor TRPV1 localisation and sensitivity in human sensory neurons. J Neurooncol 89:1–7 Anand U, Otto WR, Sanchez-Herrera D, Facer P, Yiangou Y, Korchev Y, Birch R, Benham C, Bountra C, Chessell IP, Anand P (2008c) Canabinoid receptor CB2 localisation and agonist-mediated inhibition of capcaisin responses in human sensory neurons. Pain 138: 667–680 Banks RW (2005) The muscle spindle. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, Chapter 6, pp 131–150 Banks RW, Barker D, Stacey MJ (1982) The sensory innervation of cat muscle spindles. Philos Trans R Soc Lond B299:329–364 Bar KJ, Saldanha GJ, Kennedy AJ, Facer P, Birch R, Carlstedt T, Anand P (1998) GDNF and its receptor component Ret in injured human nerves and dorsal root ganglia. NeuroReport 9(1):43–47 Batten FB (1897) The muscle-spindle under pathological conditions. Brain 20:138–179 Bell MA, Weddell AGM (1984) A descriptive study of the blood vessels of the sciatic nerve in the rat, man and other animals. Brain 107:871–898 Berthold CH, Carlstedt T, Corneliuson O (1984) Anatomy of the nerve root at the central-peripheral transitional region. In: Dyck PJ, Thomas PK, Lambert EH, Bunge R (eds) Peripheral neuropathy, 2nd edn. Saunders, Philadelphia, pp 156–170 Berthold CH, Carlstedt T, Corneliuson O (1993) Anatomy of the mature transitional zone. In: Dyck PJ, Thomas PK, Griffin JW, Low PA, Poduslo JF (eds) Peripheral neuropathy, 3rd edn. WB Saunders, Philadelphia, pp 75–80 Berthold CH, King RHM, Rydmark M (2005) The myelinated nerve fibres. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, Chapter 3, pp 42–62 Blix M (1884) Experimentelle Beitrage zur Lösung der Frage über die specifische Energie der Hautnerven. Z Biol 20:141–156 Bonney G (1959) Prognosis in traction lesions of the brachial plexus. J Bone Joint Surg 41B:4–35 Bostock H (1993) Impulse propagation in experimental neuropathy. In: Dyck PJ, Thomas PK, Griffin JW, Low PA, Poduslo JF (eds) Peripheral neuropathy, 3rd edn. WB Saunders, Philadelphia, pp 109–120 Boyd IA (1962) The structure and function of the nuclear bag muscle fibre system and the nuclear chain muscle fibre system in mammalian muscle spindle. Philos Trans R Soc Lond B 245:81–136 Boyd IA (1966) The behaviour of isolated mammalian muscle spindles with intact innervation. J Physiol (Lond) 186:109P–110P Boyd IA, Smith RS (1984) The muscle spindle. In: Dyck PJ, Thomas PK, Lambert EH, Burge R (eds) Peripheral neuropathy, 2nd edn. WB Saunders, Philadelphia, pp 171–202 Brimijoin S (2005) Axonal transport: properties, mechanisms and role in nerve disease. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, Chapter 18, pp 387–409
The Microscopic Structure of the Nervous System: Its Function Brindley GS (1986) The action of parasympathetic and sympathetic nerves in human micturition, erection and seminal emission and their restoration in paraplegic patients by implanted electrical stimulators. The Ferrier lecture. Proc R Soc Lond B 235:111–120 Brodal A (1981a) Neurological anatomy. Oxford University Press, New York/Oxford Brodal A (1981b) Neuron theory. Ibid. Oxford University Press, New York/Oxford, p 9 Brodal A (1981c) Extrapyramidal tracts. Ibid. Oxford University Press, New York/Oxford, pp 194–211 Brodal A (1981d) The fibers in the dorsal funiculi and the medial lemniscus. Ibid. Oxford University Press, New York/Oxford, pp 74–84 Bunge RD, Bunge MB, Eldridge CF (1986) Linkage between axonal ensheathment and basal lamina production by Schwann cells. Annu Rev Neurosci 9:305–328 Cajal RS (1928) Neurotrophic action of the distal stump. In: Degeneration and regeneration in the nervous system. Oxford University Press, London/New York, p 196 (May RM (trans ed)) Cajal RS (1954) Neuron theory or reticular theory? Purkiss WU, Fox CA (trans). Consejo Superior de Investigaciones Cientificas Instituto “Ramon y Cajal”, Madrid Carlstedt T (1981) An electron microscopical study of the developing transitional region in feline S1 dorsal rootlets. J NeurolSci 50:357–372 Chen YG, McClinton MA, Dasilva MF, Shaw Wilgis EF (2000) Innervation of the metacarpophalangeal and interphalangeal joints: a microanatomic and histological study of the nerve endings. J Hand Surg 25A:128–133 Chessell IP, Hatcher JP, Bountra C, Michel AD, Hughes JP, Green P, Egerton J, Murfin M, Richardson J, Peck WL, Grahames CB, Casula MA, Yiangou Y, Birch R, Anand P, Buell GN (2005) Disruption of the P2X7 purinoceptor gene abolished chronic inflammatory and neuropathic pain. Pain 114:386–396 Chiu SY (2005) Channel function in mammalian axons and support cells. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, Chapter 4, pp 95–112 Clarke E, Bearn JS (1972) The spiral nerve bands of Fontana. Brain 95:1–20 Clifton GL, Coggeshall RE, Vance WH, Willis WD (1976) Receptive fields of unmyelinated ventral root afferent fibres in the cat. J Physiol (Lond) 256:573–600 Coggeshall RB, Coulter JD, Willis WD (1974) Unmyelinated axons in the ventral roots of the cat lumbosacral enlargement. J Comp Anat 153:39–58 Cohen S, Levi-Montalcini R, Hamburger V (1954) A nerve growth stimulating factor isolated from sarcomas 37 and 180. Proc Nat Acad Sci USA 40:1014–1018 Cooper S, Daniel PM (1963) Muscle spindles in man: their morphology in the lumbricals and the deep muscles of the neck. Brain 86: 563–588 Corbin KB, Gardner ED (1937) Decrease in number of myelinated nerve fibres in human spinal roots with age. Anat Rec 68:63–74 Cottrell L (1940) Histologic variations with age in apparently normal nerve trunks. Arch Neurol Psychiatry (Chicago) 43:1138–1150 Cowen T, Ulfhake B, King RHM (2005) Aging in the peripheral nervous system. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, Chapter 22, pp 483–507 Davis KD, Pope GE (2002) Noxious cold evokes multiple sensations with distinct time cause. Pain 98:179–185 Dommisse GF (1974) The blood supply of the spinal cord. A critical vascular zone in spinal surgery. J Bone Joint Surg 56B:225–235 Dommisse GF (1975) The arteries and veins of the human spinal cord from birth, 1st edn. Churchill Livingstone, Edinburgh Durrenberger PF, Facer P, Gray RA, Chessell IP, Naylor A, Bountra C, Banati RB, Birch R, Anand P (2004) Cyclooxygenase – 2 (COX-2) in injured human nerve and a rat model of nerve injury. J Peripher Nerv Syst 9:15–24
73 Dyck PJ, Dyck PJB, Engelstad J (2005) Pathological alterations of nerves. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, Chapter 32, pp 733–829 Eames R, Gamble G (1970) Schwann cell relationships in normal human cutaneous nerves. J Anat 106:417–435 Facer P, Casula MA, Smith GD, Benham CD, Chessell IP, Bountra C, Sinisi M, Birch R, Anand P (2007) Differential expression of the capsaicin receptor TRPVI and elated novel receptors TRPV3, TRPV4 and TRPM8 in normal human tissues and changes in traumatic and diabetic neuropathy. BMC Neurol 7(1):11 Fowler CJ, Sitzoglou K, Ali Z, Halonen P (1988) The conduction velocities of peripheral nerve fibres conveying sensations of warming and cooling. J Neurol Neurosurg Psychiatry 51:1164–1170 Fraher JP (2005) The CNS-PNS transitional zone. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, Chapter 3, pp 67–78 Galbraith JA, Myers RR (1991) Impulse conduction. In: Gelberman RH (ed) Operative nerve repair and reconstruction. JB Lippincott (2 volumes), Philadelphia vol 1, Chapter 2, pp 19–45 Gamble HJ (1964) Comparative electron-microscopic observations on the connective tissue of a peripheral nerve and a spinal nerve root in the rat. J Anat 98:17–24 Gamble HJ, Eames RA (1964) An electron microscope study of the connective tissues of human peripheral nerve. J Anat 98:655–662 Gasser HF (1955) Properties of dorsal root unmedullated fibres on the two sides of the ganglion. J Gen Physiol 38:709–728 Goldscheider A (1884) Die spezifische Energie der Gefühlsnerven der Haut. Monatshefte für praktische Dermatologie 3:283–300 Goodwin GM, McCloskey DI, Matthews PBC (1972) The contribution of muscle afferents to kinaesthesia shown by vibration induced illusions of movement and by the effect of paralysing joint afferents. Brain 95:705–748 Gordon G, Jukes MJM (1964) Descending influences on the exteroceptive organisations of the cat’s gracile nucleus. J Physiol (Lond) 173: 291–319 Graham BT, Sherrington CS (1912) On the instability of a cortical point. Proc R Soc 85B:250–277 Greenfield JG, Carmichael EA (1935) The peripheral nerves in cases of subacute combined degeneration of the cord. Brain 58:483–491 Hagbarth K-E, Torebjörk HE, Wallin BG (1993) Microelectrode exploration of human peripheral nerves. In: Dyck PJ, Thomas PK, Griffin JW, Low PA, Poduslo JF (eds) Peripheral neuropathy, 3rd edn. WB Saunders, Philadelphia, pp 658–671 Hall SM (1997) Axonal regeneration through acellular muscle grafts. J Anat 190:57–71 Hall SM (1999) The biology of chronically denervated Schwann cells. Ann NY Acad Sci 883:215–233 Hall SM (2001) Nerve repair: a neurobiologists view. J Hand Surg 26B:129–136 Hall SM (2005) The response to injury in the peripheral nervous system. J Bone Joint Surg 87B:1309–1319 Hall SM, Williams PL (1970) Studies in the “incisures” of Schmidt and Lanterman. J Cell Sci 6:767–791 Hallin RG, Torebjörk HE (1973) Electrically induced A and C fibre responses in intact human skin nerves. Exp Brain Res 16: 309–320 Hamburger V, Keefe EL (1944) The effects of peripheral factors on the proliferation and differentiation in the spinal cord of chick embryos. J Exp Zool 96:223 Head H (1920) Studies in neurology, vol 2. Oxford University Press, London Head H, Sherrin J (1905) The consequences of injury to the peripheral nerves in man. Brain 28:116–338 Head H, Rivers WHR, Sherrin J (1905) The afferent nervous system from a new aspect. Brain 28:99–116 Hebb DO (1947) The effects of early experience on problem solving at maturity. Am Physiol 2:737–745
74 Hill S, Hall S (1999) Microscopic anatomy of the posterior interosseous and median nerves at sites of potential entrapment in the forearm. J Hand Surg 24B:170–176 Hodgkin AL (1937a) Evidence of electrical transmission in nerve. Part II. J Physiol (Lond) 90:211–232 Hodgkin AL (1937b) Evidence of electrical transmission in nerve. Part I. J Physiol (Lond) 90:183–210 Hodgkin AL, Huxley AF (1939) Action potential recorded from inside a nerve fibre. Nature 144:710–711 Hodgkin AL, Huxley AF (1952) A quantitative description of membrane current and its application to conduction and excitation in nerve. J Physiol 117:500–544 Hogervorst T, Brand RA (1998) Mechanoreceptors in joint function. Current concept reviews. J Bone Joint Surg 80A:1365–1378 Horsley V (1897) Short note on sense organs in muscle and on the preservation of muscle spindles in conditions of extreme muscular atrophy. Brain 20:375–376 Hromada J (1963) On the nerve supply of the connective tissue of some peripheral nervous system components. Acta Anat 55:343–351 Huxley AF, Stämpfli R (1949) Evidence for saltatory conduction in peripheral myelinated fibres. J Physiol (Lond) 108:315–339 Iggo A (1961) Non-myelinated fibres from mammalian skeletal muscle. J Physiol (Lond) 155:52P–53P Iggo A (1985) Cutaneous sensation. In: Swash M, Kennard C (eds) The scientific basis of clinical neurology. Churchill Livingstone, Edinburgh, pp 153–162 Jacobs JM, Love S (1985) Qualitative and quantitative morphology of human sural nerve at different ages. Brain 108:897–924 Kadiyala RK, Ramirez A, Taylor AE, Saltzman CL, Cassell MD (2005) The blood supply of the common peroneal nerve in the popliteal fossa. J Bone Joint Surg 87B:337–342 Kennedy WR, Wendel-Schafer-Crabb G, Polydefkis M, McArthur J (2005) Pathology and quantitation of cutaneous innervation. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, Chapter 34, pp 869–895 Kimura J (1993) The effects of age. In: Dyck PJ, Thomas PK, Griffin JW, Low PA, Poduslo JF (eds) Peripheral neuropathy, 3rd edn. WB Saunders, Philadelphia, p 605 Kühne W (1864) Über die Endigung der nerven in den Nervenhügeln der Muskeln. Archiv fur pathologische anatomie and Physiologie und für klinische Medicin (Virchow) 30:187–220 Landon DN (1985) Structure and function of nerve fibres. In: Swash M, Kennard C (eds) Scientific basis of neurology. Churchill Livingstone, Edinburg, London pp 375–389 Lasek RJ (1982) Translocation of the neuronal cytoskeleton and axonal locomotion. Philos Trans R Soc Lon B 299:313–327 Laviano M (1992/93) Il nervo accessorio spinale pro sovrascapolare nelle avulsioni del plesso brachiale. Tesi de Specializzazone, Universita delgli studi de Roma “La Sapienza”, Facolta di Medicina e Chirurgia Lawson SN (2005) The peripheral sensory nervous system: dorsal root ganglion neurones. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, Chapter 8, pp 163–202 Levi-Montalcini R, Angeletti PU (1968) Nerve growth factor. Physiol Rev 48:534–569 Levi-Montalcini R, Meyer H, Hamburger V (1954) In vitro experiments on the effects of mouse sarcomas 180 and 37 on the spinal and sympathetic ganglia of the chick embryo. Cancer Res 14:49–57 Li H, Wigley C, Hall SM (1998) Chronically denervated rat Schwann cells respond to GGF in vitro. Glia 24:290–303 Light AR, Perl ER (1993) Peripheral sensory systems. In: Dyck PJ, Thomas PK, Griffin JW, Low PA, Poduslo JF (eds) Peripheral neuropathy, 3rd edn. WB Saunders, Philadelphia, pp 149–165 Lim AYT, Pereira BP, Kumar VP, de Coninck C, Taki C, Baudet J, Merle M (2004) Intramuscular innervation of upper limb skeletal muscles. Muscle Nerve 29:523–530
Surgical Disorders of the Peripheral Nerves Loeb GE (1976) Ventral root projections of myelinated dorsal root ganglion cells in the cat. Brain Res 106:159–165 Low FN (1976) The perineurium and connective tissue of peripheral nerve. In: Landon DN (ed) The peripheral nerve. Chapman & Hall, London, Chapter 3 Lundborg G (1979) The intrinsic vascularization of human peripheral nerves: structural and functional effects. J Hand Surg 4:34–41 Lundborg G (1988) Vascular systems. In: Lundborg G Nerve injury and repair. Churchill Livingstone, Edinburgh, pp 32–63 Lundborg G (2004a) Nerve injury and repair: regeneration, reconstruction and cortical remodelling, 2nd edn. Elsevier, Churchill Livingstone Lundborg G (2004b) Sensation and sensorimotor integration in hand function. Ibid 2nd edn. Elsevier, Churchill Livingstone, Chapter 8, pp 198–210 Lundborg G (2004c) Brain plasticity and cortical remodelling. Ibid 2nd edn. Elsevier, Churchill Livingstone, Chapter 9, pp 211–223 MacQuillan A (2006) The variance of nerve axon to muscle fibre ratio and its effect on outcome in functional muscle transfer. Thesis, MD University of London Matthews PBC (1981) Evolving views of the internal operation of and functional role of the muscle spindle. J Physiol (Lond) 320:1–30 McDonald WI (1963) The effects of experimental demyelination on conduction in peripheral nerves: a histological and electrophysiological study. Brain 86:501–524 McDonald WI, Sears TA (1970) The effects of experimental demyelination on conduction in the central nervous system. Brain 93:583–598 McManis PG, Low PA, Lagerlund TD (1993) Microenvironment of nerve: blood flow and ischaemia. In: Dyck PJ, Thomas PK, Griffin JW, Low PA, Poduslo JF (eds) Peripheral neuropathy, 3rd edn. WB Saunders, Philadelphia, pp 433–440 McManis PG, Low PA, Lagerlund TD (2005) Nerve blood flow and microenvironment. In: Dyke PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, Chapter 30, p 671 Melzack R, Wall PD (1962) On the nature of cutaneous sensory mechanisms. Brain 85:331–356 Mense S (1977) Nervous outflow from skeletal muscle following chemical noxious stimulation. J Physiol (Lond) 267:75–88 Mense S, Schmidt RF (1974) Activation of group IV afferent units from muscle by algesic agents. Brain Res 72:305–310 Mirsky R, Jessen KR (2005) Molecular signalling in Schwann cell development. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, Chapter 16, pp 341–376 Mountcastle VB (1957) Modality and topographic qualities of single neurones of cat’s somatic sensory cortex. J Neurophysiol 20:408–434 Mountcastle VB (1980) In: Mountcastle VB (ed) Medical physiology, 14th edn. CV Mosby, St. Louis, pp 61–63 Nathan PW, Smith MC (1958) The centrifugal pathway for micturition within the spinal cord. J Neurol Neurosurg Psychiatry 21:177–189 Nathan PW, Smith MS, Cook AW (1986) Sensory effects in man of lesions of the posterior columns and some other afferent pathways. Brain 109:1003–1041 Norris AH, Shock NW, Wagman IH (1953) Age changes in the maximum conduction velocity of motor fibres of human ulnar nerves. J Appl Physiol 5:589–593 Oberlin C, Beal D, Bhatia A, Dauge MC (1995) Nerf cubital (ulnaire), neurotisation du muscle biceps. In: Alnot J-Y, Narakas A.(eds) Les Paralysies du plexus brachial, 2nd edn. Monographie de la Société Française de Chirurgie de la Main, pp 46–49 Ochoa J, Mair WGP (1969) The normal sural nerve in man. II. Changes in the axons and Schwann cells due to ageing. Acta Neuropathol (Berlin) 13:217–239 Ochs S, Brimijoin WS (1993) Axonal transport. In: Dyck PJ, Thomas PK, Griffin JW, Low PA, Poduslo JF (eds) Peripheral neuropathy, 3rd edn. WB Saunders, Philadelphia, pp 331–360
The Microscopic Structure of the Nervous System: Its Function Omer G (1980) Range of conduction velocities. In: Omer GE, Spinner M (eds) Management of peripheral nerve problem. WB Saunders, Philadelphia, pp 6–7 Penfield W, Boldrey E (1937) Somatic motor and sensory representation in the cerebral cortex of man as studied by electrical stimulation. Brain 60:389–443 Petrie S, Collins J, Solomonow M, Wink C, Chuinard R (1997) Mechano receptors in the palmar wrist ligaments. J Bone Joint Surg 79B: 494–496 Pomeranz B, Wall PD, Weber WV (1968) Cord cells responding to fine myelinated afferents from viscera, muscle and skin. J Physiol (Lond) 199:511–532 Price DD, Meyer DJ (1974) Physiological laminar organisation of the dorsal horn of M. Mulatta. Brain Res 79:321–325 Ramachandran VS, Hirstein W (1998) The perception of phantom limbs. The DO Hebb lecture. Brain 9(121):1603–1630 Ranson SW (1915) Unmyelinated nerve fibres as conductors of protopathic sensation. Brain 38:381–389 Rasminsky M (1985) Conduction in normal and pathological nerve fibres. In: Swash M, Kennard C (eds) Scientific basis of clinical neurology, vol 28. Churchill Livingstone, Edinburgh, pp 390–399 Rexed B (1952) The cytoarchitectonic organisation of the spinal cord in the cat. J Comp Neurol 96:415–495 Rexed B (1954) A cytoarchitectonic organisation of the spinal cord in the cat. J Comp Neurol 100:297–379 Ruffini A (1897) Observations on sensory nerve-endings in voluntary muscle. Brain 20:367–374 Saldanha G, Bar K, Yiangou Y, Anand P, Birch R, Carlstedt T, Burrin JM (2000) Marked increase of interleukin – 6 in injured human nerves and dorsal root ganglia. J Neurol Neurosurg Psychiatry 69:693–694 Sanchez D, Anand V, Gorelik J, Benham CD, Bountra C, Lab M, Klenerman D, Birch R, Anand P, Korchev Y (2007) Localized and non-contact mechanical stimulation of dorsal root ganglion sensory neurons using scanning ion conductance microscopy. J Neurosci Methods 159:26–34 Sanders FK (1942) The repair of large gaps in the peripheral nerves. Brain 65:281–337 Schenker M, Birch R (2000) Intact myelinated fibres in biopsies of ventral spinal roots after preganglionic traction injury to the brachial plexus. A proof that Sherrington’s “wrong way afferents” exist in man? J Anat 197:383–391 Schenker M, Birch R (2001) Diagnosis of level of intradural ruptures of the rootlets in traction lesions of the brachial plexus. J Bone Joint Surg 83B:916–920 Scherer SS, Arroyo EJ (2002) Recent progress on the molecular organisation of myelinated axons. 3 rd Bunge Memorial lecture. J Peripher Nerv Syst 7:1–12 Scholz A, Reid G, Vogel W, Bostock H (1993) Ion channels in human axons. J Neurophysiol 70:1274–1279 Schott GD (1994) Visceral afferents: their contribution to “sympatheticdependent” pain. Brain 117:397–413 Schwann T (1839) Mikroskopiche Untersuchungen über die Übereinstimmung in der Strukter und dem Wachstum der Tiere und Pflanzen. Sander, Berlin Scott JJA (2005) The Golgi tendon organ. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, Chapter 7, pp 151–161 Seddon HJ (1975) Surgical disorders of peripheral nerves. Churchill Livingstone, 2nd edn. Edinburgh, London/New York, pp 267–270 Selander D, Sjostrand J (1978) Longitudinal spread of intraneurally injected local anaesthetics. An experimental study of the initial distribution following intraneural injections. Acta Anaesthesiol Scand 22:622–634 Sharrard WJW (1955) The distribution of the permanent paralysis in the lower limb in poliomyelitis. J Bone Joint Surg 37B:540–558
75 Sherrington CS (1894) On the anatomical constitution of nerves of skeletal muscles: with remarks on recurrent fibres in the ventral spinal nerve-root. J Physiol 17:211–258 Sherrington CS (1897) The synapse in: the central nervous system. In: Foster M, Sherrington CS A textbook of physiology, vol 3, 7th edn. Macmillan, London, p 929 Sherrington CS, Leyton ASF (1917) Observations on the excitable cortex of the chimpanzee, orang-utan and gorilla. Quart J Exp Physiol 11:135–222 Sinclair DC (1955) Cutaneous sensation and the doctrine of specific energy. Brain 78:584–614 Stacey MJ (1969) Free nerve endings in the skeletal muscle of the cat. J Anat 105:231–254 Standring S (2008) Conduction of nervous impulses. In: Standring S (ed in chief) Gray’s anatomy, 40th edn. Churchill Living stone, Elsevier, pp 63–64 Stark B, Carlstedt T, Hallin RG (1998) Distribution of human pacinian corpuscles in the hand. J Hand Surg 25B:370–372 Stolinski C (1995) Structure and composition of the outer connective tissue sheaths of peripheral nerve. J Anat 186:123–130 Stolinski C, Breathnach AS (1982) Freeze-fracture replication of mammalian peripheral nerve – a review. J Neurol Sci 57:1–28 Sugiura Y, Terui N, Hosoya Y (1989) Difference in distribution of central terminals between visceral and somatic unmyelinated (C) primary afferent fibres. J Neurophysiol 62:834–840 Sunderland S (1968) Intraneural topography. In: Sunderland S Nerve and nerve injuries. E & S Livingstone Edinburgh, London. … Median nerve pp 758–769; ulnar nerve pp 816–825; radial nerve pp 905–914; sciatic nerve pp 1029–1046 Sunderland S, Lavarack JO, Ray LJ (1949) The caliber of nerve fibers in human cutaneous nerves. J Comp Neurol 91:87–101 Suter U, Martini R (2005) Myelination. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, Chapter19, pp 411–431 Svensson LG (2005) Paralysis after aortic surgery: in search of lost cord function. The James IV lecture. Surgeon 3:396–405 Takebayashi T, Yamashita T, Minakiy, Ushii S (1997) Mechano sensitive afferent units in the lateral ligament of the ankle. J Bone Joint Surg 79B:490–493 Takebayashi T, Cavanagh JM, Kallukuri S, Chen C, Yamashita T (2006) Sympathetic afferents from lumbar intervertebral discs. J Bone Joint Surg 88B:554–557 Tasaki I, Takeuchi T (1941) Der am Ranvierschen Knoten entstehende Aktionström und seine bedeutung für die Erregungsleitung. Pflügers Archiv 244:696–711 Tasaki I, Takeuchi T (1942) Weiters Studien über den Aktionsstrom der markhältigen Nervenfäser und über die elektrosaltatorische. Uberträgung des Nervenimpulses. Pflügers Archiv 245:764–782 Taylor PK (1984) Non-linear effects of age on nerve conduction in adults. J Neurol Sci 66:223–234 Thomas PK (1963) The connective tissue of peripheral nerve; an electron microscope study. J Anat 97:35–42 Thomas PK, Berthold C-H, Ochoa J (1993a) Microscopic anatomy of the peripheral nervous system. In: Dyck PJ, Thomas PK, Griffin JW, Low PA, Poduslo JF (eds) Peripheral neuropathy, 3rd edn. WB Saunders, Philadelphia, Chapter 3, pp 28–92 Thomas PK, Scaravilli J, Belai A (1993b) Pathological alterations in cell bodies of peripheral neurons in neuropathy. In: Dyck PJ, Thomas PK, Griffin JW, Low PA, Poduslo JF (eds) Peripheral neuropathy, 3rd edn. WB Saunders, Philadelphia, Chapter 29, pp 476–513 Tillett RL, Afoke A, Hall SM, Brown RA, Phillips JB (2004) Investigating mechanical behaviour at a core-sheath interface in peripheral nerve. J Peripher Nerve Syst 9:255–262 Tohgi H, Tsukagoshi H, Toyokura Y (1977) Quantitative changes with age in normal sural nerves. Acta Neuropathol (Berlin) 38: 213–220
76 Tomita K, Berger EJ, Berger RA, Kraisarin J, An KN (2007) Distribution of nerve endings in the human dorsal radio carpal ligament. J Hand Surg 32A:466–473 Torebjörk HE, Ochoa JL (1980) Specific sensations evoked by activity in single identified sensory units in man. Acta Physiol Scand 110:445–447 Torebjörk E, Schmelz M (2005) Single-unit recordings of afferent human peripheral nerves by microneurography. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. WB Saunders, London, Chapter 38, pp 1003–1014 Uziyel Y, Hall S, Cohen J (2000) Influence of laminin-2 on Schwann cell-axon interactions. Glia 32:109–121 Vallbo AB, Hulliger M (1981) Independence of skeletomotor and fusimotor activity. Brain Res 223:176–180 Varro A, Green T, Holmes S, Dockray GJ (1988) Calcitonin generelated peptide in visceral afferent nerve fibres: quantification by radio immunoassay and determination of axonal transport rated. Neuroscience 26:927–932 Von Frey M (1894) Beitrage zur Physiologie der Schmerzsinns. Berichte der Königliche Sächsiche Gesellschaft der Wissenschaften 46:185–196 Von Frey M (1896) Untersuchungen uber die Sinnesfunktionen der menschlichen Haut. 1. 1. Druckempfindung und Schmertz. Berichte der Köningliche Sächsiche Gesellschaft der Wissenschaften 48:175–264 Wall PD (1961) Two transmission systems for skin sensation. In: Roseblith WA (ed) Sensory communication. MIT press, Cambridge, pp 475–496 Wall PD (1970) The sensory and motor role of impulses travelling in the dorsal columns towards the cerebral cortex. Brain 93:505–524 Wall PD (1977) The presence of ineffective synapses and the circumstances which unmask them. Philos Trans R Soc Lond B278: 361–372 Webster HdeF (1993) Development of peripheral nerve fibers. In: Dyck PJ, Thomas PK, Griffin JW, Low PA, Poduslo JF (eds)
Surgical Disorders of the Peripheral Nerves Peripheral neuropathy, 3rd edn. WB Saunders, Philadelphia, pp 243–266 Weddell G (1941) The pattern of cutaneous innervation in relation to cutaneous sensibility. J Anat 75:346–366 Weerasuriya A (2005) Blood-nerve interface and endoneurial homeostasis. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, Chapter 29, pp 651–665 White JC, Sweet WH (1969) Pain and the neurosurgeon: a forty year experience. Charles C Thomas, Springfield, pp 895–896 Wigley C, Felts P, Standring S (2008) Mechanism of synaptic activity. In: Standring S (ed in Chief) Grays anatomy, 40th edn. Churchill Livingstone, Elsevier, pp 46–48 Windebank AJ, McDonald ES (2005) Neurotrophic factors in the peripheral nervous system. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, Chapter 17, pp 377–386 Woollam DHM, Millen JW (1958) Discussion on vascular disease of the spinal cord. Proc R Soc Med 51:540–543 Woolsey CN (1964) Cortical localization as defined by evoked potential and electrical stimulation studies. In: Scaltenbrand G, Woolsey CN (eds) Cerebral localization and organisation. University of Wisconsin Press, Madison, pp 17–32 Yiangou Y, Birch R, Sangeswaram L, Eglen R, Anand P (2000) SNS/PN3 and SNS2/NaN sodium channel like immunoreactivity in human adult and neonate injuries of sensory nerves. FEBS Lett 467:249–252 Young JZ (1942) Functional repair of nervous tissue. Physiol Rev 22:318–374 Young JZ (1945) The history of the shape of a nerve fibre. In: Le Gros Clark WE, Medawar PB (eds) Essays on growth and form. Oxford University Press, Oxford, p 41 Young JZ, Holmes W (1940) Nerve regeneration. Lancet 2:128–130 Zotterman Y (1939) Touch, pain and tickling: an electrophysiological investigation on cutaneous sensory nerves. J Physiol 95: 1–28
3
Reactions to Injury
Reactions to injury: the reaction of the nerve cell and myelin/Schwann cell complex; differential reaction of fibres of different sizes; two main types of nerve injury, degenerative and non-degenerative; Wallerian degeneration; the special case of the brachial plexus; reactions to various physical agents; the effects of denervation. Nerves can be damaged in a number of ways: (1) ischaemia; (2) physical agents such as traction or stretching which may be sudden, intermittent or prolonged, pressure, distortion, cold, heat, severance, electric shock, injection of noxious substances, ionising radiation; (3) infection and inflammatory processes; (4) ingestion of drugs and metals; (5) infiltration by or pressure from tumours; (6) the effects of systemic disease. The damage to the nerve may be “closed,” or “open” through a wound of the skin. Damage may be acute or chronic; single, repeated or continuing. The lesion may affect the whole nerve or only part of it. The depth of affection may vary from fibre to fibre or from one part of the nerve to another. The nerve affected may be entirely healthy or may be the subject of a neuropathy from hereditary or systemic causes or from a more proximal affection. Nerve injury may be associated with damage to one or more important structures: artery, vein, viscus, bone, muscle or ligament. It is helpful to the clinician to bear in mind two facts: (1) a nerve which has been transected or ruptured cannot recover until it is repaired; (2) a lesion of a peripheral nerve which remains in continuity but which continues to be subjected to the cause of that lesion will deepen until the cause is removed (Fig. 3.1a, b). The speed of that deepening is related to the cause. A nerve crushed by a plate or by an encircling suture may recover if that cause is removed within a minute or two. It is unlikely to recover if it is not relieved for more than 2 or 3 h. A nerve subjected to continuing traction and ischaemia from an expanding haematoma will recover if the situation is corrected within 6–8 h. After that recovery will probably be imperfect and may not occur at all. The nerve subjected to compression and ischaemia within a swollen ischaemic limb will almost certainly recover if the cause is corrected within 4 h; the chances of full spontaneous recovery diminish with the passage of every hour after that time. It may be years before the situation becomes irretrievable for a nerve subjected to radiotherapy or exposed to continuing traction from a malunited fracture. The cardinal symptom of the persistence of a noxious agent is pain (Fig. 3.2).
a
b
Fig. 3.1 Deepening of lesion. (a) Median nerve extricated from supracondylar fracture in a 9 year old girl at 3 days from injury. There was complete recovery. (b) Median nerve extricated from supracondylar fracture in a 13 year old girl 8 weeks after injury. There was no recovery.
In the mildest form of damage – that produced by transient ischaemia – there is transient failure of conduction affecting principally the large myelinated fibres. Lewis et al. (1931) described this form of centripetal paralysis produced by the application round the arm of a cuff inflated to suprasystolic pressure. First, there is loss of superficial sensibility. This is succeeded by a gradual loss of motor power. The first
R. Birch, Surgical Disorders of the Peripheral Nerves, DOI: 10.1007/978-1-84882-108-8_3, © Springer-Verlag London Limited 2011
77
78
Fig. 3.2 Conduction block. The radial nerve shown tented over the tip of the screw 10 days after operation for fracture. There was a painful deep radial palsy. Conduction in the distal segment was preserved but there was no conduction across the lesion. Conduction in the nerve to brachioradialis (left hand sling) was preserved. Stimulation of the radial nerve above the level of lesion did not evoke response through the nerve to extensor carpi radialis longus (right hand sling). The screw was shortened. Recovery was complete by 24 h.
pain response is lost soon after superficial sensibility fails, but the delayed pain response can still be elicited after 40 min of ischaemia. Pilomotor and vasomotor functions are scarcely affected. This differential paralysis reflects, of course, the differing reactions of the fibres of different sizes. Large myelinated fibres are first affected; C fibres and autonomic fibres escape. Recovery of all modalities occurs within a few minutes of release of the cuff. By a simple but ingenious method Lewis and his colleagues demonstrated that the lesion was caused by ischaemia of the segments of the nerves underlying the cuff. MacKinnon and Dellon (1988) rightly suggest that all those intending to take up peripheral nerve work should try this experiment on themselves. The experience gives a very clear indication of what is meant by the depth of a nerve injury. That knowledge is certain to be important in clinical practice. Further, the experience of the unpleasant quality of the residual delayed pain sensation gives a good insight into the feelings of patients affected by dysaesthesia. It is interesting to note Merrington and Nathan’s (1949) observation that the paraesthsiae occurring during recovery arise not from the periphery but from the nerve trunks recovering from ischaemia. The effect of temporary ischaemia is particularly well illustrated by the behaviour of the main nerves during emergency repair of a main artery. Conduction persists for about an hour whilst the artery is being controlled and repaired. Then, conduction diminishes before altogether ceasing. Conduction returns within a few minutes after removal of clamps and restoration of flow. The conduction block caused by the penetrating missile wounds of war was recognised by Mitchell et al. 1864: “this condition of local shock is very curious. A man is shot in the thigh, the ball passes near the sciatic nerve, and instantly the
Surgical Disorders of the Peripheral Nerves
limb is paralysed; within a few minutes or at the close of a day or a week the volitional control in part returns, but finally they may be left with some single group of muscles permanently paralysed.” Seddon (1943) observed the characteristic features: paralysis exceeds loss of sensation; the nerves responsible for proprioception are more deeply affected than those conveying light touch sensation; vasomotor and sudomotor function is least affected. It is likely that this lesion is provoked by a momentary displacement, or stretching, of the nerve trunks. This explanation cannot account for the conduction block of blast injuries in which the patient is exposed, at close range, to the shock wave of an explosion without any wound, or fracture and with no signs of significant injury to the soft tissues. Case Report: A rocket propelled grenade landed 5 ft from an infantryman who was thrown through the air by the blast, landing heavily on his left side. He immediately noted loss of sensation of the whole of the left forequarter including the skin of the neck and of the chest wall. There was complete paralysis of the whole of the left upper limb. There was no local injury to the posterior triangle, the shoulder or to the cervical spine. Sensation started to recover after 3 days, the muscles started to work after 5 days. Recovery was nearly complete by about 3 weeks although subsequent QST revealed elevated warm thresholds. The mechanism underlying this increasingly common form of conduction block is now under investigation. It is possible that the shock wave causes distortion of the paranodal myelin and the membrane of the axon at the node of Ranvier. As Lewis et al. (1931) noted, the differential response of nerve fibres is reversed in the conduction block produced by infiltration of local anaesthetic agent. Here, the first sign of developing paralysis is the drying and warming of the extremity. It is probable that here conduction block is caused by interference with axonal conduction: so, the small unmyelinated fibres are the most susceptible and the first to succumb. By contrast compression of nerves by haematoma or aneurysm produces a characteristic pattern: autonomic paralysis is early and deep; loss of power extends over hours or days; deep position sense and limited joint position sense persist. With severe and prolonged pressure there is local demyelination and more prolonged conduction block (Fowler et al. 1972; Gilliatt 1981) (Fig. 3.3). The pressures used by
Proximal
Cuff
Distal
Axon
Fig. 3.3 Effect of pressure on nerve: squeezing of myelin with invagination at the node of Ranvier. Narrowing of axon with extrusion of its contents (After Ochoa J et al.1972).
Reactions to Injury
Gilliatt could not reasonably be used in experiments on human subjects. The myelin is squeezed proximally and distally from underneath the tourniquet so as to invaginate into the proximal and distal sheaths of the nodes of Ranvier. The structural affects of focal compression were analysed by Dyck et al. (2005a) in a series of experiments on the peroneal nerve of the rat. The nerve was subjected to a compression force of 300 mmHg for periods ranging from 2 min to 2 h, after which the nerve was rapidly stabilised by perfusion fixation. There was a sequence of events deep to the cuff: (1) the endoneurial fluid was squeezed out so that the nerve fibres and cells became more closely apposed; (2) fluid was squeezed from out of the axon leading initially to compaction of the formed elements and later to their being expressed from out of the axon at the margins of the zone of compression, and (3) the internodes were lengthened with shearing between the lamellae of the myelin sheath. Additional changes occurred at the edges of the cuff, at the intersection between the compressed and the non compressed nerve fibres. Here the area of the endoneurium increased, the axons became distended, especially at the nodes of Ranvier because of the increase of axonal fluid and loops of myelin which had been sheared off and came to overlap the nodes. The structural alterations revealed by this method cannot be caused by ischaemia, for these take several hours to develop and are, in appearance, quite different. A conduction block which may last for weeks or months results. If, in the clinical situation, the cause of the local demyelination persists in the form of (say) a bony projection causing pressure and distortion or of external pressure by haematoma, the block persists. This was so in the three cases reported by Birch and St Clair Strange (1990): in these, removal of the external pressure was rapidly followed by recovery in lesions which had persisted for up to 3 years. The bestowal of the name axonamonosis on this type of lesion by Birch and St Clair Strange, abetted by Bonney, was sharply criticised by Gilliatt in a communication which he was kind enough to keep private. In the experimental situation conduction recovers and motor and sensory functions are restored, though not necessarily in an orderly manner. In this type of paralysis, the autonomic fibres are usually spared, though not necessarily so. We have seen lesions of the median, the tibial and the common peroneal nerves produced by prolonged pressure in which there was prolonged vasomotor and sudomotor paralysis. All these forms of paralysis are examples of neurapraxia (Seddon 1943); all are examples of the non-degenerative lesion. Seddon gave the credit for the naming of his three types of nerve lesion to Sir Henry Cohen, later Lord Cohen of Birkenhead. “Neurapraxia” is perhaps the most admirable but potentially the most misleading of the three: derived from neuron (neuron – nerve or tendon) and apraxia (apraxia – non function), it signifies loss of nerve function – no more and no less. It will not do as a diagnosis unless the clinician using it is aware that it never signifies a degenerative lesion.
79
The use of the negating a (alpha) is or should be familiar to clinicians, since it occurs in terms in common use such as aphasia, aphonia, astigmatism, anorexia, amyotrophy and so on. It is all the sadder that declining standards of literacy have combined with rising looseness of thought to render the use of the term neuropraxia a commonplace even in formerly reputable Journals and text books. Punishment should be reserved for clinicians using this meaningless neologism; lawyers using it should lose their fees; classicists using it should suffer an appropriate loss. The situation is wholly different when the lesion is serious enough to cause interruption of the axon. Then follow the changes of a degenerative lesion, caused by the damage to the neurone itself. The changes were first described by Waller (1850) and have since that time been known as those of Wallerian degeneration. We know now that they affect not only the axon but also the cell body; not only the neurone but also its Schwann cell ensheathment and its myelin sheath. There are changes too in the endoneurial cells and, over longer periods of time, in the motor and sensory end-organs. Distal to the site of injury the axon degenerates; there is a granular disintegration of the cytoskeleton and axoplasm, which are converted over succeeding days into amorphous debris (Figs. 3.4a–c and 3.5). Although Adrian (1916, 1917) and Pollock et al. (1946) thought that peripheral neural conduction might survive for up to a month after nerve section in man, Landau (1953) found that the motor response after section persisted for a much shorter period. The interval between injury and the last observation of a neuromuscular response ranged from 66 to 121 h. Gilliatt and Taylor (1959) investigated motor response after section of the facial nerve in man, and found that visible twitch in response to stimulation disappeared within 3–4 days; though an electrical response persisted for a further 48–72 h. Our own observations on motor conductivity after pre-ganglionic injury to the brachial plexus suggest that the motor response ceases about 3 days after injury. In one case stimulation of the ventral roots of spinal nerves which had been avulsed from the spinal cord evoked a motor response 132 h after the injury. The detection of persisting conduction in the distal segment of a nerve which has been ruptured or detached from the spinal cord excludes a more distal lesion of the nerve and excludes ischaemia of the nerve in cases where the axial artery has been damaged. Conduction may cease first in the most proximal part of the fibres. Gilliatt and Hjorth (1972) studied the matter in baboons. The motor response to stimulation disappeared after 4–5 days, but ascending nerve potentials could be recorded for a further 2 or 3 days. Evidently, failure of transmission at the neuromuscular junction precedes failure of conduction along the degenerating axon. Landau (1953) made the perceptive comment that “the distinction between complete Wallerian degeneration and less severe injury can be made on the basis of the disappearance of excitability in the peripheral nerve segment at this
80
a
Surgical Disorders of the Peripheral Nerves
b
c
Fig. 3.4 Changes in the distal stump of ulnar nerve transected 3 weeks previously. The gap between the proximal and distal stumps was 5 cm. (Electronmicroscopic studies prepared by Mr Stephen Gshmeissner.) (a) Disintegration of the axon and the myelin sheath. Axoplasm and
neurofilaments are seen in the lower fibre ×2,210. (b) Another part of the same specimen. Myelin debris within a macrophage (asterisk), probable Schwann cell processes (arrows) ×5,525. (c) Another part of the same specimen. Many Schwann cells, some with active nuclei ×5,525.
time,” that is, the time of disappearance of neuromuscular function. Many misdiagnoses of “neurapraxia” could have been and could be avoided if this simple procedure were more widely used. The essential difference between neurapraxia and the different types of degenerative lesion is the persistence of conduction in the distal segment of the nerve in the former and
its disappearance in the latter. Conduction disappears within 1 h of an ischaemic lesion but persists for hours or even days after transection or rupture. The early disappearance of conduction is, of course, the hall mark of impending or actual “critical” ischaemia. It is unwise to make a diagnosis of neurapraxia in the presence of persisting pain for that pain signifies that the
Reactions to Injury
81
Node of Ranvier
Fig. 3.5 Wallerian degeneration in the distal stump of fifth cervical nerve ruptured 3 weeks previously. The axon is collapsing and it is surrounded by a macrophage although the myelin sheath appears intact x 1,100.
noxious agent is continuing to act. A diagnosis of neurapraxia should not be made in the presence of a strong Tinel sign for this indicates that axons have been ruptured. A diagnosis of neurapraxia should never be made in limb rendered pulseless by injury: conduction block may prove to be but the first step towards something much worse.
3.1 Axonotmesis – Neurotmesis So long as the lesion is not severe enough to interrupt the continuity of the Schwann cell basal laminae from proximal to distal segment the original pathways for re-growth of axons remain (Young 1949; Causey and Palmer 1952; Thomas 1964; Haftek and Thomas 1967) (Fig. 3.6). As Gutmann and Sanders (1943) noted in their experimental work, “only after crushing (as opposed to section) was the nerve fully reconstituted.” Seddon (1943) described the appearance of axonotmesis in rabbits following firm squeezing of a nerve with fine smooth bladed forceps: “the immediate lesion is of striking appearance: the central and peripheral parts are united only by a fine ribbon of translucent connective tissue. The main substance of the nerve is broken and separated by an appreciable interval. Within a few minutes it flows together again, and the fine connecting ribbon is filled out so that the gap is no longer visible.” Seddon described similar experiments, which were performed with JZ Young, in patients in whom part of a limb required amputation: “on several occasions patients have allowed us to crush or divide a nerve, in the part to be sacrificed, at an appropriate interval before the final operation.” The general pattern of the nerve was retained: “the axons were completely interrupted and the conjugating tissue
Basal lamina
Schwann cell nucleus
Myelin sheath
Axon
Fig. 3.6 Axonotmesis (centre) and neurotmesis (bottom) at the moment of injury. Note the preservation of the Schwann cell basal lamina in axonotmesis.
appeared to consist of collapsed endoneurial tubes. In all but one case the lesion had been allowed to proceed to some measure of regeneration; there was a noticeable absence of the axonal branching and crisscrossing that are always found after suture. As Fig. 3.4b shows, the fibres have grown without hindrance into their old paths, exactly as described by Stroebe, Cajal and Langley.” Seddon and Young recognised that the integrity of the Schwann tubes provided important support for regeneration: “if, as we believe, the process of regeneration is a protoplasmic outflow from the central stump (Holmes and Young 1942) progress will be faster when the flow is confined to one main channel and not dissipated in many separate protoplasmic streams.” (Seddon 1975). It is the difference between preservation and destruction of continuity that underlies the division of degenerative lesions between those with the potential for spontaneous recovery and those that will not recover unless action is taken. Seddon (1943) named the first axonotmesis, from azwnw (axono – shaft, or axon) and TmhsiV (tmesis, cutting). That term is still valuable, indicating as it does the vital element of the degenerative lesion: the interruption of the axon and the consequent damage to the neurone. Seddon (1943) named lesions in which interruption of the axon was associated with interruption of the basal lamina, neurotmesis. He, of course derived the term from neuron (neurone, nerve or tendon) and TmhsiV (tmesis, cutting). It rightly indicates interruption of continuity of all elements of the nerve. As our colleague Gereis (2005) points out it is never easy to distinguish between paralysis (par lnsiV) an unrecoverable nerve lesion, and a palsy or paresis (paaresiV), a lesion which may recover spontaneously.
82
Surgical Disorders of the Peripheral Nerves
Table 3.1 Classification of focal mechanical nerve injury. I. Focal conduction block Transient Ischaemic Other More persistent Demyelinating Axonal constriction II. Axonal degeneration With preservation of basal laminal sheaths of nerve fibers With partial section of nerve With complete transection of nerve From Thomas and Holdorff (1993). With permission.
Sunderland (1951) introduced a rather more elaborate system of classifying injury. Five degrees of severity were named, ranging from simple conduction block to loss of continuity. The views of this pioneer in the field command respect; it may well be, that some clinicians find Sunderland’s classification an improvement on Seddon’s and of more practical use than the earlier method. We, on the contrary, have tended towards a further simplification: to classification as “degenerative” and “non-degenerative.” This, we think, is the manner in which clinicians should regard nerve injuries: the first question to be asked is “is this lesion degenerative or non degenerative?” Thomas and Holdorff (1993) elaborate this conception of “degenerative” and “non-degenerative” lesions in a manner helpful to clinicians and experimentalists. We reproduce by permission their table 52–1, in which the subdivisions of “focal conduction block” and “axonal degeneration” are set out Table 3.1. We think that the question mark placed against “axonal constriction” is removed by the case shown to one of us by Campbell Semple (personal communication) (Fig. 3.7). In that case of hour-glass constriction of a main nerve trunk external neurolysis and decompression was followed by early and complete recovery, there must have been axonal constriction. Nagano et al. (1996) and Nagano (2003) studied “spontaneous” paralysis of the anterior interosseous nerve. They found during interfascicular neurolysis of the median nerve that there was “hourglass” constriction of the fascicles belonging to the anterior interosseous nerve. There was no external compression. Recovery after interfascicular neurolysis alone was “generally good.” Akira Nagano has summarised his experience in a personal communication (Nagano 2008). He has identified “hourglass” like fascicular constriction involving the median nerve near the elbow in 24 patients. Five of these presented with symptoms and signs suggestive of a diagnosis of neuralgic amyotrophy. Oberlin et al. (2006) describe a similar lesion involving the circumflex nerve. The possibility that this represents a twisting or torsion of the bundles is raised. Dyck et al. (2005a) show that
Fig. 3.7 Hour glass “constriction” of the lateral root of the median nerve, the result of traction injury. There was full spontaneous recovery (Mr Campbell Semple’s case).
the axon itself may be damaged or even transected by infolded loops of the myelin sheath thrust in by external pressure or other processes. Is it possible that this occurs in cases of “hour glass” constriction? In clinical practice, most injuries of nerves short of transection inflict damage of all three grades of severity. The lesion is pure only in cases of transient ischemia or complete transection. Seddon based his classification on a study of 650 nerve lesions. The injury was in continuity in 537 of these amongst which only 117 cases of “pure” lesions could be indentified; the rest were mixed. This is the usual pattern for the sciatic nerve stretched but not ruptured by dislocation of the femoral head and for the cords of the brachial plexus stretched but not ruptured by anterior dislocation of the head of the humerus. The decision about exploration in cases such as these is never easy but the difficulties may be eased if the clinician bears in mind the possibility that removal of the cause of the lesion may prevent deepening or deterioration and that it may convert a situation inimical to spontaneous recovery to one that is more favourable. The persistence of pain is an important indication for exploration of the nerve.
3.2 The Cell Body and Proximal Stump The central and the peripheral effects of Wallerian degeneration are profound and they are ultimately irreversible. The cell body is separated from the supply of neurotrophins and it may be drained or exhausted by the process of regeneration. The sooner the cell body is reconnected with the periphery the better. Proximal to the lesion changes occur in the axon, the myelin sheath, and in the nerve cells. Within a few days there is a reduction in the calibre of the proximal axon; there may be atrophy of the whole axonal “shaft.” Gutmann and Sanders (1943) thought that this might be caused by the
Reactions to Injury
outflow of axoplasm during regeneration. Aitken and Thomas (1962) showed that the reduction in the diameter of the fibre was still evident 300 days after suture of the nerve. Cragg and Thomas (1961) found that nerve conduction velocity in the proximal segment fell to between 60% and 70% of normal by 200 days and it did not improve in nerves which had not been repaired. Hoffer et al. (1979) recorded nerve conduction potentials in the roots of spinal nerves in cats whose leg nerves had been cut and prevented from regrowing. A progressive decline in nerve conduction velocity from the dorsal root fibres was demonstrated; the decline for the motor fibres stabilised. A clear correlation between the reduction in the calibre in the larger myelinated nerve fibres and a drop in the peak to peak amplitude of the nerve compound action current was demonstrated by Walbeehm et al. (2003). In the cell body itself there may be chromatolysis, a process characterised by Groves and Scaravilli (2005) as one associated with a regenerative and not a degenerative response to an insult. Curiously enough, transection of the central branches going to the central nervous system does not produce such clear cut changes in the cell bodies in the dorsal root ganglia. Lawson (2005) describes changes in the behaviour and in the expression of ion channels in the dorsal root ganglion (DRG) cell bodies which can be detected within a few minutes of an injury to the nerve. Anand and his colleagues (Rabert et al. 2004) studied the expression of genes in the neurones within adult human dorsal root ganglia in cases of avulsion lesions of the brachial plexus, taken at intervals ranging from 1 to 100 days after injury. There was a clear alteration in the expression of 91 genes particularly for those known to be involved in neurotransmission, trophism, cytokine function, signal transduction, myelination, transcription regulation and apoptosis. In other studies of the behaviour of human DRG neurones Boettger et al. (2002) measured the changes in potassium channel behaviour and the influence of neurotrophic factors, Bar et al. (1998) recorded significantly higher levels of GDNF and Saldhana et al. (2000) found a remarkable increase in interleukin-6 (IL-6), which is a member of the neuropoietic cytokine family (Fig. 3.8). The chromatolysis following axonotomy1 may continue to actual dissolution of the cell body: “the nucleus becomes unidentifiable, all basophilia has disappeared and what remains is a seemingly empty sac containing the condensed remnants of neuronal DNA, a so-called ghost cell” (Groves and Scaravilli 2005). On the other hand rhizotomy scarcely affects the neurone, it “does not produce any observable morphological change in the perikarya of the affected neurones.” Chromatolysis does not occur in newborn mammals, instead the affected neurones become smaller, more intensely stained and many die. Groves and Scaravilli go on to define apoptosis
1
he common usage is of course axotomy, but we believe that axonotomy T is the more correct term
83
Fig. 3.8 TRPV1 (the heat and capsaicin receptor) in a human dorsal root ganglion 6 weeks after avulsion showing immunostaining in small diameter neuronal cell bodies and axons ×40 (Courtesy of Professor Praveen Anand).
as a controlled and deliberate form of neurone destruction that is largely a secondary effect of an insult. The loss of cells is more severe in more proximal axonotomy; axonotomy in the neonate produces a more rapid and much greater incidence of sensory and motor neurone death than in the adult and it is particularly severe after avulsion of the ventral root: “ventral root avulsion being the most severe and acute form of injury, because of the significant trophic support being supplied by surrounding glia and neurones in the adult spinal cord which is disrupted in avulsion” (Fig. 3.9a, b). The loss of sensory neurones was studied in the rat by Ygge (1989a,1989b) and by Schmalbruch (1987). Romanes (1946) observed the changes in the anterior horn cells of mice after permanent axonotomy by amputation. Dyck et al. (1984) were able to study the effect of permanent axonotomy on the spinal cords of two patients who had undergone respectively, partial and total amputation of a lower limb. They found that “loss of target tissue by axotomy leads to atrophy and then loss of motor neurones.” Suzuki et al. (1993) examined the cervical cord and the roots and ganglia of the spinal nerves in a patient who died 38 years after amputation of the upper limb. They demonstrated a loss of neurones in the DRG, in the anterior horn, and a diminution of the large myelinated nerve fibres in the ventral and dorsal roots. Perhaps the electrophysiological findings in a cohort of old polio survivors described by Sorenson et al. (2006) represent an attrition of surviving neurones in the anterior horn. Carlstedt (2007) says that: “motor neurones are rapidly killed by ventral root avulsions. Disconnection from the periphery means that interrupted supply of neurotrophic factors, together with vascular trauma leading to cytotoxicity drastically reduces the number of motor neurones up to about 90% of the normal population.” Carlstedt reckons that about one half of all motor neurones in the affected spinal cord segment have disappeared by 2 weeks after avulsion of the ventral root and he suggests that “a swift intervention to re
84
Surgical Disorders of the Peripheral Nerves
a
b
Fig. 3.9 Delay: the proximal stump of C5 ruptured at the same level, 1 cm distal to the foramen, but exposed at operation at different intervals. (a) Six weeks after injury. Clusters of myelinated and unmyelinated nerve fibres forming minifascicles. Bar = 20 mm, (b) Eight months
after injury. Some unmyelinated axons lie within debris laden Schwann cell cytoplasm. There is extensive fibrosis. Fibroblast processes surround the Schwann cell- axon units ×3,000.
establish contact between the injured nerve cells and the periphery with its supply of neurotrophic substances would counteract nerve cell loss in these injuries.” The same phenomenon, of central cell death, follows more peripheral axonotomy although not to the same degree. One example of the depletion of neurotrophins is provided by Anand et al. (1997) who showed, in human nerves, depletion of NGF proximally and rapid depletion of CNTF both proximally and distally after section of a nerve. As long ago as 1913 Cajal (1928) noted
intradural damage to the roots of the brachial plexus. At delayed operation (hemi laminectomy) the ipsilateral cord is seen to be atrophic, doubtless from the damage to the motor and sensory pathways. Further, magnetic resonance imaging plainly shows atrophy of the cord after birth injury of the brachial plexus. It is a matter of interest that, despite the degeneration of the proximal process of the cells of the dorsal root ganglion, those cells appear to retain for a long time their ability to sustain their distal processes, apparently without serious functional impairment (Bonney and Gilliatt 1958) (Fig. 3.10).
as shown by the older observations of Dickinson (1868) and by the more modern ones of Gadden, Ford and Monakow (1882), there are cases where the neuron whose axon is mutilated or even functionally isolated shows morbid changes, and may even die.
In the words of McComas et al. (1978) “the periphery is essential for the functional integrity of motor and sensory nerve axons.” Further evidence about the changes in proximal neural structures is available from studies of the spinal cord after
3.2.1 Wounds of the Perineurium Wounding of the perineurium causes demyelination and Wallerian degeneration (Thomas and Jones 1967). Spencer et al. (1975) studied the process of demyelination and
Reactions to Injury
85
3.2.2 Contralateral Effects
Fig. 3.10 Right side of the spinal cord at the level C5 to C8 seen at hemilaminectomy 6 months after intradural lesion of the plexus. Note the shrinking of the middle part of the cord at the level of the emergence of the seventh cervical root and the mouths of the pseudo meningocoele.
For some years it has been our impression that the regenerative capacity of the central nervous system is diminished by the most severe injuries to the brachial plexus and we attributed this to an affect extending beyond the injured segments of the spinal cord. A dramatic example of the involvement of regions beyond the area of injury is provided by Suzuki et al. (1993) who, in addition to the ipsilateral effects already described, also demonstrated atrophy of the contralateral anterior horn, where there was a loss of the medium size cells and a reduction in the numbers of the medium and small myelinated nerve fibres in the contralateral ventral roots. This phenomenon has been closely examined by Oaklander and Brown (2004) who used the pan neuronal marker protein (PGP9.5) to measure the density of innervation in the skin of the paws of the rat after transecting one tibial nerve. There was almost complete loss of innervation within the plantar skin in the injured limb. However, a persisting loss of innervation, in excess of 50%, was noted in the skin of the contralateral hind paw. The significance of these findings will not be lost on the clinicians treating patients with nerve injuries.
3.3 The Distal Stump
Fig. 3.11 Injection injury. The tip of the needle lacerated the lateral part of median nerve at the elbow and a neuroma occupied about four of the bundles. Small fascicles of thinly myelinated axons and Schwann cells occupy the proximal stump of one bundle ×2,000.
subsequent remyelination induced by making a window in the perineurium and Thomas and Bhagat (1978) scrutinised the effects of performing a window in the perineurium and extracting its contents at that level. The perineurial cells separated from one another and also from their basal lamina and came to resemble fibroblasts. Regrowth occurred by bundles of axons and Schwann cells which became surrounded by fibroblasts which later developed into perineurial cells leading to the formation of many small fascicles, otherwise known as mini fascicles. This process may occur when the perineurium is lacerated by the tip of a needle and it leads to the formation of a small neuroma within the nerve trunk. The initiation of degeneration is inevitable after lacerating the perineurium which is, of course, deliberately incised during “end to side” repair or during transfer of one bundle from one undamaged nerve to the distal stump of another (Fig. 3.11).
Waller (1850) described the microscopical changes in the hypoglossal and glossopharyngeal nerves of the frog following division. He noted no change for the first 4 days but then progressive varicosity and irregularity of the myelin sheath. About the twentieth day the medullary particles are completely reduced to a granular state…. where we found a presence of the nervous element indicated by numerous black granules, generally arranged in a row like the beads of a necklace. In their arrangement it is easy to detect the wavy direction characteristic of the nerve.
Hall (2005) conceives Wallerian degeneration as an active process in which the environment of a normal nerve so inimical to regeneration of axons is transformed into one which is actively receptive to that regeneration, at least for a limited period. The earliest changes affect the cytoskeleton: “the earliest evidence of fibre alteration is the dissolution and clumping of the neurofilaments and microtubules of axons” (Dyck et al. 2005a). Schwann cells in the distal nerves, both in myelinated and unmyelinated fibres, begin a process of proliferation and the fact that the Schwann cells of undamaged unmyelinated fibres seem to share the proliferation of Schwann cells of damaged myelinated fibres suggest that diffusible, trophic factors are the stimulants for this proliferation(Archer and Griffin 1993, Griffin and Hoffman 1993). Within 48 h of injury, denervated
86
myelinating Schwann cells down regulate expression of those genes encoding myelin associated proteins and other proteins which are important for maintaining the organisation of nodes, paranodes, and incisures (Hall 2005). The denervated Schwann cell columns lie within the original basal lamina forming Schwann tubes, formerly known as bands of Büngner. There is proliferation too of endoneurial fibroblasts in the distal nerve. The last important feature is the increase of macrophages; some derived from “resident” cells, others “recruited from the circulation” (Stoll et al. 1989, Griffin and Hoffman 1993). Omura et al. (2005a) estimated that the increase in macrophages reached a peak of 150-fold by day 14 after breaking down of the blood brain barrier. Macrophages clear the debris of myelin and axoplasm during which process a Schwann cell mitogen is liberated from the debris. Significantly, they remove proteins such as the myelin associated glycoprotein (MAG) which normally inhibit axonal growth (Hall 2005). Omura et al. (2005b) measured the changes in expression of BDNF, NT3 and NT4 in muscle and nerve after different types of injuries to the peripheral nerves. The significant changes occurred only after transection of the nerve. By 7 days there was a dramatic increase in the expression of messenger RNA for BDNF in muscle and the distal nerve. The messenger RNA for NT-4 decreased. Durrenberger et al. (2004, 2006) measured increase in the expression of chemical mediators of inflammation by macrophages and glial cells after injury to human nerves. Thomas (1964) examined by electronmicroscopy the distal stumps of nerves transected 7 days earlier and found that cords of Schwann cells, often several cells thick and surrounded by a common basal lamina, extended from the distal stump into the outgrowth by about 2 mm. The endoneurial tubes shrink, more collagen is deposited within the endoneurium and there is progressive fibrosis within the distal stump (Holmes and Young 1942). As time passes the number of Schwann cells in the distal stump diminishes and they become less receptive to regenerating axons because of the decrease in the expression of receptors which are normally important in Schwann cell-axon signalling (Li et al. 1997, 1998, Hall 1999, 2005). Calder and his colleagues (Terenghi et al. 1998) examined the distal stumps from ten patients in whom the delay before repair ranged from 8 to 53 months: “denervated S-100 positive Schwann cells remained within the distal stumps for at least 53 months, organised into typical bands of Büngner, i.e. delicate processes of densely staining Schwann cell cytoplasm enclosed by a basal lamina.” Axons were demonstrated in all specimens; some of these were myelinated (Fig. 3.12). The effects of delay on the behaviour of the Schwann cell is summarised by Niels et al. (2007) who write: “the reduced capacity to regenerate seems to be
Surgical Disorders of the Peripheral Nerves
Fig. 3.12 Distal stump of median nerve in a 25 year old man, transected 6 months previously by bullet from military rifle. Numerous pale processes of Schwann cell cytoplasm and occasional axonal sprouts. Extensive endoneurial collagenisation ×3,245 (Electronmicroscopic studies by Mr Michael Kayser).
linked to progressive inability of chronically denervated Schwann cells to respond to axonally-derived signals, a response that has been shown to decrease significantly after only two months of denervation” (Li et al. 1998). It is now established that delay before repair is harmful not only because of the effects upon the target organs, not only because of the effects upon the neurones themselves but also because of the falling away of the capacity of the Schwann cells, the essential supporting cells of peripheral nerves, to support regeneration. These facts will not escape the attention of clinicians (Fig. 3.13a–c). The special case of the brachial plexus – A wholly different system of thought has to be employed in lesions of the brachial plexus, or for that matter of the lumbo-sacral plexus, in which there is intradural damage to the roots (Fig. 3.14). In spite of the damage to the proximal branches, the axons, whose cells of origin are outside the cord in the posterior root ganglion, remain healthy for a long time when they are avulsed from the cord or ruptured intradurally. Such axons include all those in the dorsal root; also, of course, many
Reactions to Injury
a
87
b
c
Fig. 3.13 The results of rupture complicated by sepsis. Electron microsopic study of serial (2 cm interval) biopsies of proximal stump of sciatic nerve of 34 year old man examined 8 months after rupture
c omplicated by fracture and subsequent sepsis. (a) 2 cm from the tip of the proximal stump ×2,340. (b) 4 cm from the tip of the proximal stump ×3,000. (c) 6 cm from the tip of the proximal stump ×1,500.
“recurrent” fibres in the anterior root whose cells of origin are in the dorsal root ganglion. These axons, their Schwann cells and myelin sheaths remain intact and functional, detached not only from central connection but also from Seddon’s, Sunderland’s, Thomas and Holdorff’s and even our systems of classification (Figs. 3.15–3.17). This is in fact, so far as afferent neurones are concerned, a lesion of the central nervous system. Somatic efferent fibres undergo degeneration, being separated from their cells; post ganglionic autonomic efferent fibres also degenerate, because of damage to their grey rami communicantes. We did think of applying the term neuradetosis (from neuron: neurone; nerve and a etoV : adetos; unbound, loose, free) to the lesion of the afferent fibres in avulsion of the plexus. However, we suffered so severe a mauling from the late Roger Gilliatt in connection with our previous attempt at naming that we have decided not to press the point.
3.4 Types of Lesion Produced by Different Physical Agents 3.4.1 Acute Ischaemia The first effect of ischaemia upon peripheral nerves is the loss of conduction caused by anoxic block of fast axoplasmic transport systems and the paralysis of ion channel function. We have seen the effect of transient ischaemia from a short period of pressure from a suprasystolic cuff and this can be observed during exposure of limb nerves with an inflated cuff in position. For about the first 20 min stimulation of the nerve evokes a brisk muscular response by transmission through the neuromuscular junction. This response diminishes and disappears after about 30 min. Conduction within the nerve itself can still be detected for up to another 20 min.
88 Fig. 3.14 The nerve and roots detached from the spinal cord. Note the intact dorsal root ganglion cell, with healthy axons in the detached parts of the roots, the degeneration of the efferent fibre in the ventral root and of the central projections of the afferent fibres.
Surgical Disorders of the Peripheral Nerves Direct ascending tracts
Dorsal root ganglion
Dorsal root ganglion
Cells and sensory nerves
Degenerate efferent Ventral horn cell and axon
Fig. 3.15 Dorsal root ganglion 6 months after avulsion from the spinal cord. The two neuronal cell bodies appear healthy and there are numerous myelinated fibres. Solochrome cyanin ×960.
On the other hand direct stimulation of the muscle provokes a twitch which can be elicited for up to several hours. Indeed, it is the loss of that direct response which signifies impending death of the muscle, and with it, the death of the limb. Whilst it is hard to separate the effects of acute ischaemia from those of other physical agents there are examples of the effects of ischaemia alone upon conducting tissue. Harriman (1977) examined the lower limb soon after amputation in a case where femoral embolectomy successfully restored flow but failed to relieve severe pain. There were areas of muscle infarction but these were confined to the thigh, the muscles below the knee were normal in colour. The nerves in the leg were infarcted. The stump of the sciatic nerve appeared normal but as the nerves passed distally they became soft and
Afferent fibre in ventral root
grey and sections showed swelling of the myelin sheaths and axons with only a scanty cellular reaction in the epineurium. Case report: In October 1993 a man then aged 73 underwent an injection of steroid into the right forearm. There was instant severe pain and loss of function in the distribution of the right ulnar nerve. Persistence of pain and alteration of sensibility combined with failure to detect sensory nerve action potentials led to exploration (August 1995). There were slight localised changes in the flexor carpi ulnaris muscle. The ulnar nerve appeared healthy, but the ulnar artery was tortuous, calcified and distally occluded. Evidently, steroid had been injected into the ulnar artery which at this point was the principal nutrient of the ulnar nerve. An ischaemic nerve lesion had followed, chiefly affecting the larger myelin fibres with very little and much circumscribed affection of muscle. The necrotising angiopathy which complicates disorders such as rheumatoid arthritis and polyarteritis nodosa illustrate the effects of occlusion of smaller vessels in the epineurium. Dyck et al. (2005b) show that many of the epineurial arterioles become occluded and this leads to: “quite dramatic multifocal fibre degeneration, fibre loss and regeneration, extending to involve the perineurium.” Sander (1999) described the case of a diver returning from a depth of 195 ft who developed a lesion of the deep peroneal nerve affecting the sensory fibres and suggested that there had been an infarction of part of the nerve by gas bubbles within the vasa nervorum. The cases of cranial nerve palsies reviewed by Greer (1997) were thought to be caused by compression of the facial and the trigeminal nerves by gas bubbles within the osseous canals.
Reactions to Injury
89
Fig. 3.16 Rupture and avulsion of the spinal nerves forming the brachial plexus: response to intradermal injection of histamine. (left) There is a flare, mediated by the axon reflex in the territory of the avulsed first thoracic nerve. (right) There is a wheal but no flare in the dermatome of C7 which was ruptured.
Fig. 3.17 Wallerian degeneration in the ventral root of the eighth cervical nerve 6 weeks after avulsion from the spinal cord. A degenerate efferent myelinated fibre (right) compared with a non degenerated myelinated afferent fibre (left) ×11,115 (Electron microscopic studies by Mr Stephen Gshmeissner).
3.4.2 Ischaemia from Tamponade The most dramatic examples of this are provided by the catastrophic cases of infarction of the spinal cord and the roots of the spinal nerves by the injection of local anaesthetic and other agents into the invertebral foramina. Flow through radicular vessels passing into the spinal canal to the anterior
spinal artery is occluded. One such case (Brouwers et al. 2001) was caused by a diagnostic block of the right sixth cervical nerve in a 58 year old man. Under radiological control a 22 gauge needle was positioned in the posterior caudal corner of the foramen of C6 so that the tip of the needle lay well within the foramen. No cerebrospinal fluid was aspirated and radio contrast medium showed spreading alongside the nerve root. A mixture of 0.5 ml of bupivicaine and 0.5 ml triamcinolone was injected around the nerve root over a 1 min period. At about 1 min after this the patient suddenly developed flaccid paralysis and severe breathing difficulties. It soon became clear that there was a complete lesion of the cord from C3. An MR scan at 6 h showed increased signal intensity from C2 to T1 and a further MR scan at 24 h confirmed infarction of the spinal cord. There was, later, some recovery in the function of dorsal column but little else. The patient later died. In an informed and extensive comment on this paper Nash (2001) describes other cases of severe cord ischaemia following radiofrequency lesioning of dorsal root ganglia and root sleeve injection and he emphasised the importance of the blood vessels passing with the spinal nerves through the intervertebral foramina. “The intervertebral foraminae are of critical value in the blood supply of the cord, as the feeder vessels pass through the foraminae alongside the nerve root.” Similar cases of infarction mainly of the anterior part of the cord but without a fatal outcome, are described by Houten and Errico (2002) and Somayaji et al. (2005). These cases are thoroughly presented and it seems more likely that ischaemia was caused by tamponade of the radicular vessels rather than by direct damage. We believe that a
90
similar mechanism underlies the cases of anterior cord infarction following interscalene block in which the infusion of relatively large volumes of fluid deep to the unyielding prevertebral fascia disturbs flow within the radicular vessels accompanying the spinal nerves. The permanent defects in eight patients are set out in Table 3.2. We are aware of five more cases. Case Report: A 48 year old woman underwent arthroscopic decompression for her painful left shoulder. An interscalene block was performed after induction of anaesthetic using 20 ml of 0.5% bupivicaine under stimulator control. She developed hypotension and was slow to breathe spontaneously. On awakening she had numbness and weakness in all four limbs. There was flaccid paralysis in the left upper limb and of the C5/6 and C7 muscles in the right upper limb. There were severe defects in light touch and temperature sense in the left forequarter with lesser abnormalities in the right hand. Joint position and vibration sense were maintained. There was vasomotor and sudomotor paralysis in the right hand. There was no Bernard Horner syndrome. There was some affection of both lower limbs with weakness of the muscles about the hip and defects in light touch and temperature sense in the right lower limb. Her reflexes were brisk but there was no clonus. A dissociated sensory loss was evident on the right side of the trunk, the level for light touch was T4, that for pinprick sense was T6. An MR scan done on the day after injury was reported as normal but when this was repeated at 5 days bilateral linear patchy areas of high signal were seen, extending from C3 to T5. Somatosensory evoked potentials (SSEPs) at 11 days were normal; sensory action potentials were maintained in all four limbs and motor conduction in both lower limbs was preserved. Motor conduction was absent in the median and ulnar nerves in both upper limbs. By 6 months the lower limbs had recovered. There was complete paralysis of C8 and T1 muscles on the right and of C7,C8 and T1 muscles on the left. By now she experienced severe burning pain in the shoulder girdles and chest. Quantitative sensory testing revealed normal vibration sense in all four limbs, a marked elevation of thermal thresholds especially so in the left forequarter and reduced sweating in both hands. It seems that motor neurones in the anterior horns of the segment C7, C8 and T1 were infarcted, that there was some involvement of the sympathetic outflow to both of the upper limbs and that there was a primary spinothalamic syndrome. She remains in severe pain. Case Report: a previously fit 4 year old boy fell 3 or 4 ft from the steps of a slide sustaining a fracture through the extreme lateral end of the right clavicle. There was no loss of consciousness. The clavicular fragments were markedly displaced with dropping of the forequarter, but there was no evidence of neural affection. Six days after injury, operative reduction and internal fixation were done. The distal clavicular fragments had pierced the trapezius and platysma; after
Surgical Disorders of the Peripheral Nerves
reduction the fragments were united by two sutures. A right interscalene injection of 10 ml of 0.5% bupivicaine was done. It was clear early that there was a lesion of the cord and brain stem with palsies of right cranial nerves III, IV, V, VI, VII, VIII, IX, X, XI, XII, right phrenic palsy and right sided Claude Bernard Horner syndrome. There was flaccid paralysis in the right upper limb but sweating was present in the hand. Light touch sensation was present. Magnetic resonance imaging 2 days after operation showed swelling of the right side of the brain stem and cervical spinal cord. Vertebral angiography showed no abnormality. Neurophysiological investigations were done at 16 days after the operation and showed sensory conduction in the nerves of the upper limbs whilst motor conduction was absent in the right median and ulnar nerves. Somatosensory evoked potentials were recorded at Erb’s point and also from electrodes placed over the lower cervical vertebrae by stimulating the median and ulnar nerves but the higher cervical potentials and all cortical responses were absent, suggesting a lesion at the root entry level of the cervical cord (Dr. Steven Jones, Queen Square). By 5 months the cranial nerves and the phrenic nerve had recovered but the Bernard Horner sign persisted. There was complete paralysis of the muscles innervated by the right eighth cervical and first thoracic nerves. Quantitative sensory testing 1 year after the event revealed preservation of dorsal column function and of light touch and cotton wool sensation in both upper limbs but the thresholds for cooling and warming were very elevated in the right upper limb and also in the contralateral T3 and T4 dermatomes. Sweating in both palms was normal. It seems likely, in view of the absence of abnormal findings on angiography, that the responsible lesion here was a true spasm of the right vertebral artery, causing ipsilateral lesion of the cervical cord and brain stem and that there was also occlusion of flow through the radicular artery accompanying C8. The state of this boy’s hands, 7 years after the event, is illustrated in Fig. 3.18. Evidently, interscalene block is not lightly to be undertaken: the price of avoidance of general anaesthesia or relief of pain after operation may be too high. The increasing use of ultrasonography is likely to reduce the risk to the vertebral artery but it is unlikely to remove altogether the risk of tamponade of critical radicular vessels. The blocks were given with the patient awake in two of our cases.
3.4.3 Ischaemia and Acute Compression Within Neurovascular Fascial Compartments This is caused by bleeding or infusion of fluid into a fascial compartment which encloses the nerve and axial vessels but not muscle. Nerves especially at risk include the femoral
F
F
F
M
F
55
46
46
60
4a
48
R
L
R
R
L
R
L
No
No
Yes
Yes
No
No
Yes
a
The bulbar palsy had recovered by 5 months in case 4.
F
F
60
C7,8, T1
C7,8, T1
C7,8,T1
C5,6,7,8,T1
C7,8
C5,6,7,8,T1
C5,6,7,8,T1
C4,5,6
C8,T1
T1
–
–
–
C6,7
C5
No
Yes
No
No
No
Yes
No
No
Bilateral
No
Bilateral
No
No
Bilateral
No
No
Yes
M
52
R
Upper limb sympathetic
Table 3.2 Permanent defects in eight patients after interscalene block, followed for at least 3 years. Age Sex Side Phrenic palsy Lower motor neurone Cervical sympathetic Ipsilateral Contralateral
C5,6,7,8,T1
C5,6,7,8,T1
C5,6,7,8,T1
–
–
C6,7,8,T1
C7,8,T1
C5,6,7
C7,8,T1
T3,T4
Upper limb C7,8,T1 Lower limb
–
–
Upper and lower limbs
Lower limb
Spinothalamic tract Ipsilateral Contralateral
PNI3:VAS8 Shoulders and neck
No
PNI3: VAS 8 Neck. Shoulders and right hand
PNI 3: VAS 8 neck and shoulder
No
PNI 3: VAS 8 neck and shoulders
Resolved by day 7
Neck and shoulders PNI 3; VAS 8
Pain
Reactions to Injury 91
92
Surgical Disorders of the Peripheral Nerves
Fig. 3.18 The hands of an 11 year old boy 7 years after anterior cord infarction caused by interscalene block.
nerve in the groin, the ulnar nerve in the forearm and the tibial nerve in the leg (Fig. 3.19). The syndrome is a common complication of skeletal injury, of nerve blocks and of vessel puncture. The anatomical arrangements putting certain nerves at risk are outlined in Chap. 1 and examples are described in Chap. 8. The medial brachial fascial compartment syndrome (Tsao and Wilbourn 2003) is an important example, for as Wilbourn (2005) suggests it is responsible for the majority of infraclavicular plexopathies following axillary regional block and also for many of the neurovascular injuries which result from closed or penetrating injuries in this region. The progression of the lesion is characteristic: there is, almost always, pain accompanied by dysaesthesiae; loss of sensation soon follows and then, over the next 2–3 h, paralysis ensues. Wilbourne’s comment bears repeating: “distal pulses are normal as they are with most compartment syndromes because the elevated pressure, although sufficient to collapse the vasae nervori, is far below mean arterial pressure. Ultrasound, MR and CT may reveal the vascular lesion, but, considering the very brief time available for surgical decompression before irreversible nerve damage occurs, obtaining these is rarely justified.” The 16 patients described by Stenning et al. (2005) probably fall into the syndrome described by Wilbourn. There was, in all of these cases, an injury to the axillary artery or one of its offsets caused by dislocation of the shoulder or fracture of the proximal humerus. The diagnosis of continuing bleeding into the axillary sheath was made by the delayed onset of nerve palsy or the deepening of the lesion whilst under observation. There were 87 nerve palsies. A favourable outcome was seen in all cases where urgent repair of the artery and decompression of the axillary sheath was performed (see Table 8.5). The
Fig. 3.19 Recovering femoral palsy from haematoma in the femoral triangle in a 69 year old farmer who was taking warfarin after aortic valve replacement. Two months previously he had injured his thigh whilst vaulting a gate. He experienced severe pain for 24 h but this recovered spontaneously. The area of sensory loss is outlined and the site of the Tinel sign marked. Recovery was good but not complete by 6 months.
outcome for function within the hand in another patient in whom the diagnosis was delayed by 8 weeks is shown in Fig. 3.20.
3.4.4 Ischaemia by Acute Compression from Swollen Muscle The effect upon the nerves is at least as rapid as it is in cases of compression within a neurovascular sheath. The response to correction within 3 hours is almost always gratifying and the consequences of delay before that correction are particularly severe. The vascular arrangements of nerves are such that injuries of main arteries are more likely to produce infarction of muscle than necrosis of nerve trunks. Even when ischaemic degeneration of myelinated fibres was produced by multiple arterial ligation (Hess et al. 1979) Schwann cells and their lamina tubes and fibroblasts survived and provided conditions for recovery. The evidence is distorted by
Reactions to Injury
93
Fig. 3.21 Volkmann’s ischaemic contracture. Below: the ulnar nerve exposed during flexor muscle slide 8 weeks after supracondylar fracture. The epineurial vessels and also the ulnar recurrent collateral vessels are occluded and the nerve is compressed by the swollen infarcted muscle. Above: the appearance of the hand 14 years later.
Fig. 3.20 Fracture of left proximal humerus was complicated by expanding haematoma in a 63 year old man. Two attempts to occlude the torn posterior circumflex artery by interventional radiology failed. He was seen at 8 weeks by which time he was in right heart failure, in great pain and he had a complete infraclavicular plexopathy on the left side. Six litres of altered blood were removed from the axilla. His pain was relieved. Recovery was particularly poor in the radial and median nerves. (a) MR angiogram before operation. (b) The left hand 4 years after operation.
the circumstance that ischaemia is rarely complete. Even nerves which have been seriously ischaemic for 36 h have been seen to recover adequate function. In one case of ischaemia after supracondylar fracture of the humerus complicated by thrombosis of the brachial artery, the median nerve was seen at operation to lack all vascular pedicles from elbow to wrist. It lay in the middle of the completely infarcted flexor muscles of the forearm. Three years later there was recovery of sweating and of impaired sensation in its area of distribution. The effect of increasing pressure within the osseo-fascial compartment upon the vessels running with the nerves at the elbow is illustrated in Fig. 3.21. These vessels provide the main pathway for collateral circulation at the elbow after cessation of flow through the axial artery; the consequences
of neglect are especially severe. We shall return to this serious, and as it seems, increasingly common problem in Chap. 8. It is somewhat reassuring that adequate decompression, even when delayed for some months, usually leads to relief of pain and considerable improvement in sensation. Leonard (1967) described five such cases and he observed that: “ the vasa nervorum were seen to fill upon release of the tourniquet” (Fig. 3.22). Case Report: A 33 year old man suffered injury to the popliteal artery in the course of arthroscopic meniscectomy. We saw him 5 months later when he complained of severe pain. There was a vicious equino varus deformity of the ankle and foot. Below knee amputation had been recommended. The deep flexor compartment was exposed and wide step elongation of the heel cord and the flexor tendons brought the foot into a plantigrade position. The tibial nerve and posterior tibial artery were strangled within a thickened sheath. Both were narrowed to less than one half their normal diameter, the artery was not pulsatile and was there was no flow through the epineurial vessels. They were widely decompressed. Pulsatile flow returned at the end of the procedure. The patient reported relief of his pain and a sensation of warmth in the sole of his foot on the first post operative day. Over the next
94
Surgical Disorders of the Peripheral Nerves
Fig. 3.22 The popliteal artery was lacerated during arthroscopic reconstruction of the posterior cruciate ligament. There was 24 h delay before the vessel was repaired. This was followed by rhabdomyolysis and acute renal failure. Most of the muscle in the leg was excised. Nine months later step elongation of the flexor tendons improved the posture of the foot and the decompression of the tibial nerve was followed by considerable recovery of sensation and some recovery into the small muscles of the foot.
few days he regained cutaneous sensibility in the skin of the sole of the foot. Recovery into the small muscles of the foot was apparent by 6 months and some recovery of vasomotor and sudomotor function in the plantar skin was observed 2 months later. Some nerve fibres were affected by conduction block; the greater number sustained a degenerative lesion. The strangulation of nerves by scar tissue is a common occurrence in war wounds, especially when damaged or repaired nerves are covered by split skin graft. “Nothing less than a full thickness flap or tune-pedicle graft will suffice since it is important that the nerve graft should lie, a far as possible, in healthy, well vascularised tissue” (Seddon 1954). Case Report: A 38 year old soldier sustained severe soft tissue damage to the right posterior triangle from bomb fragments. The wound was closed with split skin graft. He experienced significant pain and much loss of function from a deep palsy of the spinal accessory nerve. Operation was performed 4 years later in the hope of easing his pain. The scar was excised then the cervical plexus and the spinal accessory nerves were exposed. The spinal accessory nerve appeared as a narrowed cord from the posterior margin of the sternocleidomastoid to three fingers breadth above the clavicle. The epineurial circulation was almost nonexistent. Stimulation of the nerve above, when it was first exposed, evoked no distal response. After the nerve had been liberated from scar stimulation evoked an increasingly powerful muscular response in the fibres of the upper trapezius. A free full thickness skin flap was used to cover the defect (Roderick Dunn, Salisbury).
On the day after operation the patient demonstrated powerful activity in his right trapezius (Fig. 3.23). This case is by no means exceptional. It seems there must have been prolonged conduction block involving a significant number of the efferent fibres in the spinal accessory nerve. This type of lesion is common in nerves strangled in scar. Phang et al. (2009) described the case of a 26 year old woman who developed pain in her left hip. This was attributed to previously undiagnosed bilateral hip dysplasia. Pelvic osteotomy of both hips was done. The pain in the left hip became much worse and both hips were resurfaced 6 years later. The pain in her left hip deepened so that she could walk only with crutches. When she was reviewed 10 years after the onset of her symptoms it was clear that there was a focal lesion of the left femoral nerve. There was a strong and painful Tinel sign over the nerve at the groin crease, motor and sensory conduction was impaired and electromyography confirmed a mild degree of degenerative lesion of the fibres to the quadriceps muscle. The femoral nerve was exposed (Marco Sinisi RNOH). It was tethered and sharply compressed by scar tissue over a 4 cm segment. The nerve was liberated. Her pain was improved, there was, later, improvement in the range of movement at the hip and improvement to the power of extension of the knee. By 1 year she was able to walk freely without any aid. Hindsight is always easy but it does seem that the lesion of the nerve occurred during the operation of pelvic osteotomy and her description of the new symptoms which occurred after that operation indicated neuropathic pain.
Reactions to Injury
95
Fig. 3.23 The spinal accessory nerve after excision of scar during operation 4 years after missile injury. The transverse cervical nerve (1) and the proximal accessory (2) and distal accessory (3) are exposed. A full thickness skin flap was inserted. Recovery of function was evident on the day after operation (Courtesy of Mr Roderick Dunn, Salisbury).
Fig. 3.25 A 20 year old man developed staphylococcal septicaemia after an operation for pilonidal sinus. Rhabdomyolysis led to multiple organ failure. Early fasciotomy and wide excision of the muscles of the anterior compartment of both legs was performed. There was recovery of sensation and sympathetic function in the foot: the small muscles also recovered.
Fig. 3.24 An 18 year old man with Marfan’s syndrome suffered severe and extensive rhabdomyolysis with organ failure after an operation for aortic valve replacement. Urgent and extensive fasciotomies were performed. There was considerable recovery in the peripheral nerves of both lower limbs so that he was able to walk independently by 3 years after operation.
“Critical illness neuropathy” may develop in patients who develop multi organ failure or sepsis. (Donaghy 2009a). It is possible that the intense compression of nerve trunks within oedematous and swollen limbs contributes to this disorder (Figs. 3.24 and 3.25).
3.4.5 Ischaemia Caused by Traction Lundborg and Rydevik (1973) showed that 8% elongation of a segment of a nerve could cause impairment to vascular flow and that an elongation of 10–15% could arrest all blood
flow. Relaxation within 30 min would, in most cases, lead to restoration of flow and conductivity. Nerves which are stretched over an expanding haematoma sustain the most severe lesions, especially so when the false aneurysm is pulsatile. This type of injury is associated with causalgia. Recovery for nerves embedded within the wall of the sac is generally poor (Stewart and Birch 2001). Gardiner et al. (2006) reported a fatal outcome in a 71 year old man with a 3 month history of “sciatica.” The pain increased in severity and it was attended by rapidly deteriorating function in the lumbar plexus. The underlying cause was an aneurysm of the common iliac artery (Fig. 3.26). Ischaemia is the common factor in many of the lesions inflicted upon nerves in continuity. The situation can usually be retrieved by urgent and accurate action. The recovery of some function through nerves strangled by fibrosis for months or even years, provides an indication of their resilience: on the other hand relief of the cause within 3 or 4 h offers the only real prospect for complete recovery. The
96
Fig. 3.26 A 68 year old woman developed an intensely painful and rapidly deepening lesion of the sciatic nerve on the evening of operation for total hip arthroplasty. The nerve was exposed 3 months later and it was found stretched over an organised haematoma. Her pain improved but the nerve did not recover.
different responses evoked by acute compression of a nerve trunk within a fascial sheath versus the insidious effect of a haematoma is illustrated by the following example. Case report: A healthy 28 year old woman suffered a fracture of her left tibia and fibula. This was treated by open reduction, internal fixation and bone graft from the ipsilateral iliac crest 7 days later. A femoral nerve block was given before induction of general anaesthesia. In spite of her complaints of intense searing, shooting pain radiating down the front of her thigh and leg the anaesthetist persisted with the infusion of local anaesthetic into the groin crease. Upon awakening the patient continued to experience intense pain which required high doses of morphine during the next 3 days. She became aware that she could not feel the sole of her foot and could not move her ankle. Her haemoglobin was measured at less than 8 g per dL on the third post operative day but the significance of this was not grasped. By 7 months it was appreciated that she had developed not only a significant femoral neuropathy but also a profound sacral plexopathy. The femoral neuropathy remained very painful and she continued to experience dynamic mechanical allodynia to stimulation of the skin of the front of the thigh. She was unable to walk because of pain, weakness and loss of position sense throughout the left lower limb. NPI were performed 9 months after her injury. Sensory and motor conduction studies were normal. Electromyo graphy revealed a few small polyphasic motor units in the right vastus medialis, otherwise the motor units were normal. QST found that joint position sense was very poor throughout the left lower limb, indeed it could not be detected for the hip. On the other hand vibration thresholds at the left great toe were within normal limits. The thresholds to monofilament stimulation and warm sensation were
Surgical Disorders of the Peripheral Nerves
increased throughout the left lower limb and there was a rather patchy loss of pinprick sense. The pain was exacerbated by extension at the hip. The femoral nerve was explored 11 months after her injury. The fascia surrounding the nerve was greatly thickened. The nerve was narrowed, and inflamed with a much diminished epineurial circulation. There was no sign of injury to any of the individual bundles within the trunk. The nerve was decompressed over some 10 cm. A tissue catheter was placed for the infusion of local anaesthetic about the proximal part of the femoral nerve for 48 h. There was considerable improvement in her pain and in her ability to walk. By 9 months after operation the muscles about both hips and knees were graded 5 by the MRC system but she still had extremely poor joint position sense throughout the left lower limb. It seems likely that the injury to the femoral nerve was provoked by injection of bupivicaine into the epineurium but not, in all likelihood, into the perineurium and that the nerve became strangled within its fascial sheath. The chief effect of this injury was pain, a pain so severe that the more insidious and painless lesion of the sacral plexus caused by continuing bleeding from the donor site at the ipsilateral iliac crest escaped attention. The effect of the haematoma was particularly severe for the largest myelinated afferent fibres: some never recovered. The myelinated efferent fibres and the smaller myelinated afferent fibres recovered over the course of 3 months: in these the lesion was one of conduction block. The experience of severe pain at the beginning of the femoral nerve block should have led to instant cessation of the injection.
3.4.6 Reperfusion Injury Bywaters et al. (1941) described the fatal outcome after freeing people whose limbs had been crushed for some hours by the debris of bombed houses. The renal failure was caused by the release of myoglobin from ischaemic muscle. The effects upon the nerves of reperfusion are of lesser consequence than the systemic effects but as McManis et al. (2005)point out reperfusion aggravates the ischaemic injury to the nerve fibre. Nukada et al. (1997) described the stages of reperfusion injury of nerves. The blood nerve barrier (BNB) breaks down at about 3 h; then follows oedema, demyelination and axonal degeneration. There is an infiltration of polymorphonuclear leucocytes which peaks at about 24 h. This is followed by an increase in macrophages which invade the Schwann cell and the myelin sheaths. McManis et al. (2005) consider the function of the BNB as a physiological index of oxidative injury, which causes microvascular permeability and an environment in which there is generation of oxygen free radicals.
Reactions to Injury
3.4.7 Chronic Ischaemia Eames and Lange (1967) studied clinically and by light and electron microscopy of biopsy specimens, the effect of chronic ischaemia on the sural nerves of 30 patients with arteriosclerotic obliterative disease and two with Buerger’s disease. There was a good correlation between the severity of the arterial disease and the incidence of neuropathy. The incidence of neuropathy was high: 87.5%. There was found extensive segmental demyelination and remyelination in addition to Wallerian degeneration and regeneration. The unmyelinated fibres were least affected. In all cases – not only in those with Buerger’s disease – there were marked occlusive changes in the small arterial vessels of the epineurium. The relation of these changes to “ischaemic pain” and “trophic” changes in the skin was canvassed.
3.4.8 Crush Crush injuries stand with inadvertent transection, traction or neglected ischaemia as one of the four main causes of iatrogenous lesions. Although a crush injury might be considered as an extreme form of compression the clinical evidence suggests that it is rather more than that. Unless the cause is relieved swiftly the lesion of the nerve rapidly becomes a neurotmesis. The most extreme examples of this of course is when a nerve is encircled by a suture. Much depends on the material used and the tightness of the strangulation. We have seen nerves divided by stainless steel wire used for the fixation of displaced fragments of bone. Whilst recovery might be anticipated after removal of a braided suture, within an hour or two, nerves do not recover after this time (Fig. 3.27).
Fig. 3.27 A 48 year old woman experienced severe pain and an incomplete lesion of the sciatic nerve after operation of total hip arthroplasty. The nerve was exposed 8 weeks later. It had been transfixed by a braided suture. There was some moderation in her pain but little recovery for the nerve.
97
Not only do catgut sutures crush, they also provoke a brisk inflammatory response. Not only does an encircling suture strangle a nerve it also tethers it so that there is the added element of stretch. Severe pain is usual and it is often related to posture. Attempted stretching of the limb against the tethered nerve provokes intense pain (Fig. 3.28). We have seen more than 40 patients in whom main nerves had become inadvertently strangled by a suture passed around, or through, the nerve. When the suture was removed within hours of the operation there was instant relief of pain and a high level of recovery. Relief of pain was usual after removal of the suture up to about 2 weeks. In these cases recovery was always incomplete and when the ligature was removed later there was very little useful recovery. Hindsight suggests that that the better course might have been resection and suture of the relatively short damaged segment. Nerves entrapped within fractures or joints certainly pass through a period of ischaemic conduction block which may last for as long as 2 or 3 days. After that there is demyelination but recovery may be anticipated if the nerve is set free within 7–10 days. The situation is made very much worse if a compression plate or tension band is applied to the fracture without extricating the nerve. The resected material in such cases shows transection of normal structures with an interposed zone of dense fibrosis. The effect of compression between the bone and the plate is illustrated by a case treated by an alert surgeon who, drawing to the close of a difficult operation of internal fixation of a fractured shaft of humerus, realised that the hitherto protected radial nerve had inadvertently slipped between the plate and the bone. The nerve was
Fig. 3.28 A 38 year old woman experienced intense pain and a partial lesion of the ulnar nerve after repair of the capsule of the joint using an arthroscope. Her pain was worsened by attempted extension at the elbow. The nerve was exposed 4 days later, it had been caught by a clip. The clip was removed and her pain was relieved. There was considerable recovery in the ulnar nerve but she did not recover vasomotor and sudomotor control and there was lasting weakness of the small muscles.
98
Fig. 3.29 The appearance of a radial nerve after extrication from beneath a compression plate 48 h after first operation. There was relief of pain but only incomplete recovery so that later flexor to extensor transfer was necessary (Courtesy of Marco Sinisi, RNOH).
at once extricated. The duration of compression was, at the most, 5 min. The nerve was re-explored at that surgeon’s insistence some 6 weeks later even though an advancing Tinel sign offered the prospect of spontaneous recovery. The nerve had reconstituted, the epineurium was thickened but the epineurial vessels were patent and the bundles within had not been severed. Recovery proceeded uneventfully as an axonotmesis. Other examples of a good outcome after urgent correction of the situation are outlined in Chap. 8. Nerves extricated from within a fracture 2 or 3 days of the injury usually recover. There is certainly some narrowing of the nerve trunk in these cases but the perineurium is not breached. All may not be well within the perineurium: the lesion is mixed with elements of ischaemic and demyelinating conduction block interspersed with varying degrees of axonal degeneration (Fig. 3.29).
3.4.9 Compression – Acute Thomas and Holdorff (1993) consider crush and compression separately. It is likely that in the latter type of injury the lesion is localised – a mixture of demyelination and axonal interruption (Richardson and Thomas 1979). Certainly, with more severe pressure there is axonal interruption but that interruption is within the Schwann cell basal lamina tubes. Thomas and Holdorff show a figure which well illustrates this point. Dyck, Dyck and Engelstad (2005a) reserve the term compression for a “monophasic application of force exerted to the nerve from the outside so that it is compressed against a rigid underlying structure. Acute compression may also occur repeatedly.” They rightly point out that “entrapment” should be reserved for a nerve passing through an opening that is too small for it. This effectively tethers the
Surgical Disorders of the Peripheral Nerves
nerve so that it becomes subject to stretch. Platt (1928) described precisely this lesion in relation to the radial nerve stretched over an ununited fracture of the humeral shaft. Closed compression lesion, in which an external force is applied to the limb of a conscious patient usually leads to a conduction block. Such is the event in radial palsy caused by a badly applied plaster of Paris splint or the older style of crutch favoured by the cook in Treasure Island. Case report: A fit 23 year old woman fell deeply asleep lying on her left side for about 2 h and awoke with a complete left sided radial palsy. She had no pain, there was no Tinel sign and there was some preservation of cutaneous sensibility within the distribution of the nerve. She was fitted with a dynamic extension splint and at 6 weeks the first evidence of recovery into the extensor muscles of the wrist was apparent. Neurophysiological investigations were performed 9 weeks later which revealed normal conduction and a normal recruitment to a full pattern of motor units of normal appearance. Her recovery was complete by 12 weeks from the incident. The lesion was one of pure conduction block. More sustained compression leads to deeper lesions (Tait and Danton 1991). Case Report: A slim healthy 22 year old woman sustained a mid shaft right femoral fracture through an area of fibrous dysplasia. The operation of internal fixation was difficult and lasted for 7 h during which time the contralateral left leg was kept in a flexed abducted position to permit the use of an image intensifier. A muscle relaxant (Rocuronium) was used. She awoke with a complete left sciatic palsy. There was no bruising in the left thigh or the buttock. NPI at 2 months after operation showed a complete degenerative lesion of the nerves and muscles of the leg and foot with extensive denervation in the hamstring muscles. She was examined again 8 months after the incident, when she reported increasingly severe pain but some recovery of feeling into her foot. Recovery of power of the knee flexor muscles was measured at MRC Grade 4 and there was perceptible activity in the flexor muscles of the heel. There was complete vasomotor and sudomotor paralysis in the foot. There were strong Tinel signs for both divisions of the sciatic nerve in the leg indicating a rate of regeneration of a little more than 2 mm per day. By 14 months all muscle groups in the leg and foot were recovering and she could localise light touch to the skin of the foot. There was still sympathetic paralysis in the sole of the foot and she still had pain. NPI were repeated. No sensory conduction could be demonstrated in the nerves of the leg nor was motor conduction demonstrable for the tibial nerve. Motor conduction in the common peroneal nerve was reduced and slowed. Electromyography of the leg muscles revealed persisting denervation with reinnervation by collateral sprouting with many wide polyphasic units and irregular recruitment. By 18 months she showed further recovery and some improvement of her pain. Quantitative sensory testing (Praveen Anand Imperial College) showed elevated thermal
Reactions to Injury
thresholds in the plantar skin but the threshold to cooling fell within normal limits. The threshold to monofilament sensation was elevated and pinprick was felt as an unpleasantly sharp sensation from the mid calf down. Sweating in the left sole was reduced to about one half of that on the right. The threshold to vibration was markedly elevated. It seems that the largest and the smallest fibres suffered most. Such lesions can prove very serious in the growing child because of the disturbance of growth. Case Report: A 12 year old boy fell from a swing and experienced much pain. There was deep bruising in the left buttock and he presented with a high sciatic palsy which involved the superior gluteal nerve. The common peroneal nerve was more severely affected, NPI at 6 weeks revealing loss of conduction and denervation of all of the muscles innervated by that nerve. Sensory conduction was diminished in the tibial division and there were signs of partial denervation of all tibial muscles. There was considerable muscular recovery by 1 year. Sensory conduction in the tibial nerve had recovered. There was no sensory conduction in the common peroneal nerve and motor conduction was slowed and reduced in amplitude to just under 10% of the uninjured side. At 2 years the nerve had recovered but the left lower limb was shorter by 1.5 cm and the heel was fixed in equino varus deformity. The lesion in the more severe compression injuries is usually much more than a conduction block. The tempo of recovery varies from one population of nerve fibres to another
Fig. 3.30 A 50 year old man who required crutches developed “carpal tunnel syndrome” in his right hand. Two operations were performed with no lasting relief of his symptoms. His symptoms were improved after modification of his crutches and by the use of a silicone pad.
99
and it is not unusual to see poor recovery for somatic unmyelinated fibres. As a rule, the sympathetic efferent fibres suffer least. Some nerves never recover.
3.4.10 Chronic Nerve Compression Entrapment syndromes provide most of the examples of chronic nerve compression, though the latter is one of the causes of neuropathy after ionising radiation and in other conditions in which a nerve is deformed over a tumour or bony abnormality (Fig. 3.30). Clearly, there are a number of mechanisms. Episodic ischaemia accounts for the nocturnal paraesthesiae in “carpal tunnel syndrome,” and possibly continued ischaemia accounts for the continuous symptoms (Hongell and Mattson 1971). Evidence about the size of fibres affected comes from Thomas and Fullerton (1963), and of segmental demyelination from Neary and Eames (1975). The systematic studies of MacKinnon and her colleagues (MacKinnon et al. 1985; Dellon and MacKinnon 1988; MacKinnon and Dellon 1986) showed that the earliest changes occurred in the small vessels of the endo- and perineurium. The presence of numerous Renaut corpuscles (Renaut 1881; Jefferson et al. 1981) was also observed, consistent with the observation made by Hill and Hall (1999) that these aggregate at sites of entrapment, and represent a response to local injury to the endoneurial
100
Surgical Disorders of the Peripheral Nerves
Fig. 3.31 Intraneural ganglion of the common peroneal nerve. Top left: the patient presented with a painful foot drop of 9 months duration. The area of sensory loss is shown. Top right: magnetic resonance scan
shows the extent of the lesion and the communication with the proximal tibio-fibular joint. Below left: the ganglion is shown at operation. Below right: the resected specimen includes the articular branch.
capillaries. Later, there is perineurial and epineurial fibrosis, followed by loss of fibres and thinning of myelin sheaths. At the end of the process there is damage to unmyelinated fibres. Compression may be intermittent as it was in the case described by Mannan et al. (2008). A 25 year old woman experienced intense but intermittent sciatica which was closely related to the menstrual cycle. She endured this for many months and became dependent upon crutches and even a wheel chair. Her pain was relieved by removal of a focus of endometrial cysts which had infiltrated the sciatic nerve in the upper thigh. Bendszus et al.(2003) described two cases of severe but intermittent sciatica, provoked by sitting or lying and eased by standing or walking, which were caused by varicose veins enveloping the sciatic nerve. Pain was relieved by excision of the varicosities. The perineurium and the intraneurial ganglion. This rather mysterious disorder is often a source of needless maiming. Spinner et al. (2003), Spinner et al. 2006) have shown that the pedicle of the ganglion is, in fact, the articular branch of the nerve and that this is a pathway for synovial
fluid passing from an adjacent joint into the trunk of the nerve. It seems that the pathway is epineurial, and the perineurium acts as a barrier between the endoneurium and the increasing volume of synovial fluid. The perineurium, with its contents, is subjected to prolonged and varying compression. This may explain the high success following successfully conducted operations which involve not only decompression of the nerve but also excision of the articular branch through which synovial fluid tracks. The early recovery after months or even years of pain and weakness is remarkable indicating that the lesion of the nerve fibres is predominantly one of conduction block (Fig. 3.31).
3.4.11 Traction or Stretch Injury Peripheral nerves outside the spinal canal have considerable tensile strength, but their function is damaged by an elongation of 12% or more, the extent of damage varying with the
Reactions to Injury
suddenness and the length of time during which that elongation is maintained. Lundborg and Rydevik (1973) showed that venous flow was blocked when a nerve was stretched by 8% of its resting length, and that stretching by 15% produced ischaemia. At first, elongation is permitted by the elongation of the epineurium and the straightening of the irregular course of the fibres within the fascicles (Haftek 1970; Clarke and Bearn 1972). The latter drew attention to the significance of the “spiral bands of Fontana,” confirming his conclusion that the banding appearance of the peripheral nerve is due to the wave-like alignment of its individual nerve fibres. As the fibres are straightened the banding disappears. It is difficult to separate the effects of stretch upon the conducting tissue from those imposed upon the vessels. Ochs et al. (2000) studied the effects of stretch upon isolated segments of nerve placed within an oxygen chamber. A very light stretch straightened out the zig zag disposition of the nerve fibres. The spiral bands of Fontana were erased when the nerve was elongated to about 15%. This effect was achieved with an applied tension of less than 0.1 g. An applied tension of 2 g or more induced beading of the nerve fibres which was maximal at 4–5 g of tension. The change in the nerve was rapid and it was also rapidly reversible. The compound nerve action potential was actually augmented in the earliest stages before it fell away. Haftek (1970) added the observation that “before rupture of the perineurium the damage to the nerve fibres is either neurapraxia or axonotmesis, because the endoneurial sheaths and Schwann fibres remain intact.” Next, the calibre of the fibres is diminished, the endoneurial space is diminished and myelin is disrupted (Chalk and Dyck 1993). Then rupture starts: first the epineurium and lastly the nerve fibres. The longitudinal extent of damage is considerable. The process of damage by a bullet passing through soft tissue near a nerve is comparable. Such lesions are usually conduction blocks. Seddon (1975) illustrates one such case, where the epineurium was split by the passage of a bullet but the bundles were intact. It is usual, in closed traction lesion, to see that the epineurium has ruptured but the perineurium within remains intact, albeit stretched. These injuries are usually complicated by bleeding into the epineurium, extending over many centimetres. However we have encountered cases where the perineurium was ruptured even though the epineurium remained intact in nerves sharply angulated over a fragment of bone. It is much easier to recognise this pattern when the nerve is explored within 24 h of the injury (Figs. 3.32a–c). In the extreme traction injury the nerve is ruptured or avulsed from muscle. The wide recoil of the stumps be reduced only by urgent operation. We have seen nerves destroyed over a length of 15 cm by the action of a drill during operation. The manipulation of any joint which has been fixed for some time in a position of deformity always carries the risk of damage to nerves and vessels passing across it and accustomed to the position of deformity. In one case of attempted correction of long standing flexion deformity of both knees
101
in an adult with cerebral palsy loss of pulses at the ankles was recognised by the staff of the recovery ward. No action was taken and above knee amputation of one lower limb proved necessary. Aspden and Porter (1994) reported a case in which the sciatic nerve was damaged by straightening knees that had long been flexed because of spastic diplegia. Birch et al. (1991) described three cases in which the brachial plexus was damaged during manipulation of the shoulder for “frozen shoulder.” In one of these cases, manipulation had caused anterior dislocation. The plexus was explored in this case; there was no interruption of continuity, and recovery followed. Case report: A fit man, in age 64 years, developed severe osteoarthritis in his right knee complicated by severe fixed flexion deformity. Capsulotomy was done to regain extension as the first step in total arthroplasty of the joint. The surgeon avoided a suprasystolic cuff because of anticipated difficulties. A complete lesion of the tibial and common peroneal nerves was recognised on the evening of operation. Neurophysiological investigations at 22 months revealed no sensory or motor conduction. Considerable reinnervation of the tibial muscles was demonstrated and the distal muscles of the anterior compartment were also recovering (EHL, EDC) However, tibialis anterior was fibrosed, the muscle was silent and the concentric needle met with the characteristic, gritty resistance. In addition to the traction injury of the nerves it is likely that flow through the anterior tibial artery was interrupted. The patient had been discharged from the first hospital in spite of severe disability and continuing pain. Case report: A 13 year old girl with severe cerebral palsy was treated by anterior transfer of the hamstring muscles of the left knee with the object of correcting flexion deformity. On the day of operation she developed intense pain in the leg and foot. This did not respond to opiates and neither the child nor her mother were able to sleep for 8 weeks. The child lost a good deal of weight. We saw her at 8 weeks when it was clear that the nerve lesion, although deep, was not complete. The child had causalgia. The extent of discoloured skin matched the areas of intense mechanical allodynia. At operation both tibial and common peroneal nerves were found stretched and compressed, indeed strangled, by fascia and by scar in the popliteal fossa. The common peroneal nerve was reduced to about one half of normal diameter and the epineurial vessels were obliterated. The tibial nerve was inflamed and embedded in vascular adhesions. An external neurolysis was done, and a tissue catheter placed to permit infusion of local anaesthetic for 48 h after operation. The pain from the tibial nerve was improved, the pain from the common peroneal nerve persisted. Both nerves recovered. By about 9 months the child was able to tolerate shoes and weight bearing. Her mother described how vasomotor and sudomotor disturbance in the foot persisted even after pain had improved. In this case stretching of the nerves damaged the epineurial vessels, the myelin sheath and the axons, and this was compounded by compression from
102
a
Surgical Disorders of the Peripheral Nerves
b
c
Fig. 3.32 Traction injury. (a) The circumflex nerve exposed 4 days after anterior dislocation of shoulder. The bundles had ruptured and retracted within the intact epineurium. (b) Traction injury of the common peroneal nerve from varus injury at the knee. There was extensive recovery over the course of 9 months (axonotmesis). (c) A more violent
traction injury of the right upper limb. The median nerve was elongated by more than 100%. Over the course of the next 3 years there was recovery of cutaneous sensation in the hand, and some recovery of the flexor muscles of the forearm. The sympathetic fibres never recovered.
bleeding and by the persistence of inflammation induced by haematoma. A wide range of drugs had been used in an attempt to control the pain without considering the possibility of a persisting focal noxious agent at work. There may be a case, not only for monitoring of nerve conduction, but also the flow through adjacent main arteries before, during and after operations for the correction of severe flexion deformity at the knee and other joints as recommended by Martin et al. (2006). Somewhat similar circumstances obtain in limb lengthening operations, even though it is usual to effect a change slowly, over a period of weeks. It has been argued that, if during the process, signs or symptoms of actual or impending neural damage appear, the process can be stopped to permit recovery. Galardi et al. (1990) are less reassuring: they studied events in the limbs of five patients whose tibiae were lengthened at the rate of 1 mm per day over a period of 53–107 days. The ages of the patients ranged from 3 to 24 years. The conduction velocity of nerves and the electrical activity in muscles during lengthening were examined
and evidence of damage to myelin sheaths and axons was found in all cases. These findings evidently raise the question of whether nerve function should always be monitored during correction of major deformities of long standing. This concept was used by Nogueira et al. (2003) who used a pressure sense monitoring device during 814 limb lengthening procedures. Seventy-six (9.3%) nerve lesions occurred. Nerves were most at risk in double level lengthening of the tibia and in skeletal dysplasia. Most of the affected nerves were decompressed; 74 of the 76 recovered. Clear conclusions are drawn from this admirable work: the rate of lengthening should be slow; affected nerves should be decompressed as soon as possible; monitoring of nerve function by a pressure sense device is more sensitive than clinical examination and the largest myelinated fibres are the most vulnerable. Paley (1990) provides sound advice about prevention. Analysis of the deformity provides information about the potential distortion of neurovascular bundles; the drill hole for the frame screws should be made opposite to the bundle and neuromuscular blocking agents should not be used.
Reactions to Injury
Severe pain indicates damage to a nerve or vessel or both. If a Tinel sign is evoked by tapping the transfixion wire then it has certainly passed through, or close to, a trunk nerve. Polo et al. (1997) lengthened the lower limbs in 14 patients with achondroplasia and found that the common peroneal nerve was most vulnerable and that the lesion was detectible in the first days of elongation. The prognosis for nerves injured during correction of deformity or limb elongation is poor. Case report: A 12 year old child with Stüve-Weidemann syndrome had osteotomy of the femur for flexion deformity of the lower limb. The operation was done under epidural anaesthetic and a suprasystolic cuff was placed about the upper part of the thigh. There was a complete, but painless, sciatic palsy. NPI confirmed deep axonopathy without any evidence of generalised neuropathy. The tibial and common peroneal nerves were exposed 5 months later. They were unblemished. The amplitude of CNAP’s across the presumed level of lesion was diminished. At 36 months, there was some recovery for the tibial nerve but very little for the common peroneal nerve. Case report: A 35 year old woman with severe deformity at the knee and in the leg after operations for Blount’s disease, was treated by two level osteotomy of the tibia and fibula stabilised by an Ilizarov frame. This was followed by a deep, but painless, palsy of the tibial and common peroneal nerves. The nerves were exposed at 14 days to exclude the possibility of haematoma. They were unblemished. There was little recovery for either nerve at 18 months.
3.4.12 Thermal Injury The effects of cold have been studied extensively (Franz and Iggo 1968; Lundberg 1948; Douglas and Malcolm 1955; Dodt 1953; Denny Brown et al. 1945). Bickford (1939)), cooling the ulnar nerve, found a sequential loss of the different modalities of sensation: first, appreciation of cold, and last, sensibility to pain, touch and warmth. As Thomas and Holdorff (1993) remark “observations on non freezing cold injury of nerve have largely stemmed from the exigencies of war”.
In the First World War “trench foot” was a common cause of disablement; in the Second World War “immersion foot” was a more common occurrence (Ungley et al. 1945). Donaghy (2009b) describes the freezing injury of frost bite, in which there is tissue necrosis often with a clear demarcation between living and dead tissue. Wallerian degeneration is an early feature of these injuries. One of us experienced this injury to a relatively minor degree and noted the loss of sensation in the affected toes which persisted for more than 10 years, a
103
sensory loss which increased and became painful when the foot was exposed to mild cooling. The terrible pain which often complicates more severe frost bite is described by Maurice Herzog from his experience with the French expedition to Annapurna. The second type of cold injury recognised by Donaghy follows prolonged immersion in cold water or prolonged exposure to cold around freezing point. Irwin et al. (1997) had the opportunity of investigating the neural damage in a case of “trench foot” or “non freezing cold injury.” A 40 year old Asian man, homeless in the UK through the operation of free market principles, was subjected to exposure lasting 19 days, during 8 of which the ambient temperature was below 0°c. He lost all function and sensibility in his feet, and, ultimately, several toes. Sensory nerve function was carefully measured and recorded; at the time of removal of dead tissue, 4 weeks after admission, a biopsy of viable tissue was taken from the plantar surface of the right great toe. There was evidence of damage to both myelinated and unmyelinated nerve fibres, possibly, the authors’ speculate, “in a cycle of ischaemia and reperfusion.” In civil practice at the present time, it is rather the effects of heat that concern the clinician, principally because of damage to the sciatic nerve by the heat of polymerising cement during arthroplasty of the hip. The effects of heat seem to have received little attention. Nerves can be destroyed by extremes of heat, or by diathermy during operation. Hoogeveen et al. (1991) investigated the effect of heating a 5 mm segment of the sciatic nerve of the rat for 30 min at 45°C. They found that this produced swelling of the endothelium of the neural blood vessels and Wallerian degeneration of all the nerve fibres. There was of course total loss of motor function. The process in the nerves was, however, reversible: regeneration occurred over 4–5 weeks. The vascular changes were not apparently reversible. Xu and Pollock (1994) examined physiologically and morphologically the effect of heat ranging from 47°C to 58°C on rat sciatic nerve. Unmyelinated fibres showed a greater direct vulnerability to hyperthermia, first manifest as a reversible conduction block and at higher temperatures by immediate and selective axonal degeneration. Lower grade thermal injury caused a delayed selective loss of myelinated fibres. Xu and Pollock remark in relation to the latter effect: “Evidence from this study suggests that this is secondary to a heat-induced angiopathy, immediately and diffusely manifest in the vasa nervorum and giving rise to a progressive and ultimately severe reduction in nerve blood flow.” The relative sparing of unmyelinated fibres was attributed to their greater resistance to ischaemia. It is necessary in considering risk to the sciatic nerve in operation for hip replacement, to recall that the temperature of polymerising cement rises to 95°C about 15 min after mixing, and remains above 70°C for another 12 min. Birch et al. (1992) had the opportunity of examining a length of sciatic nerve damaged by the heat of polymerising cement. They later took the
104
a
Surgical Disorders of the Peripheral Nerves
a
b b
Fig. 3.33 A case of damage to the sciatic nerve by cement during operation for arthroplasty of the hip. (a) The sciatic nerve seen at operation a year after initial operation. The extruded cement can be seen in relation to the nerve. (b) The severely damaged common peroneal component was resected and grafted. The proximal stump appeared to be relatively healthy at 1 cm proximal to the cement. Toluidine blue ×100.
opportunity of studying the effect of the heat of polymerising cement on the median nerve of an arm recently amputated because of a complete pre-ganglionic injury of the brachial plexus. The remarkable feature was the localised nature of the lesion: although at the site of burning there was destruction of axoplasm and disruption of myelin, a normal pattern of myelinated and unmyelinated fibres was found 10 mm from the margin of the cement (Figs. 3.33a, b and 3.34 a, b). Wilkinson and Clarke (1992) reported two cases of burns to the brachial plexus in patients who fell asleep with their arms draped over the back of heated towel rails. Later decompression was performed in one case, revealing extensive fibrosis of the nerves and adjacent tissues. Recovery was poor in both these cases. Tubiana (1988) made rather sombre observations on the evolution of the treatment of these injuries during a long and illustrious career, noting little significant improvement. Salisbury and Dingeldein (1988) and Salzberg and Salisbury (1991) provide extensive reviews emphasising the necessity for early incision of the encircling eschar (Fig. 3.35). A clear description of the technique, with
Fig. 3.34 Thermal damage to the median nerve after exposure to the heat of setting cement, showing the limited longitudinal extent of the neural lesion. (Electronmicroscopic studies by Mr Stephen Gschmeissner.) (a) Virtual destruction of axoplasm and cellular elements at the site of the lesion ×3,600. (b) Healthy axons and collagen 10 mm from site of injury ×3,000.
emphasis on avoiding damage to underlying nerves, is given. Chalain and Clarke(2000) review the field of thermal injuries, their pathophysiology and the treatment of the consequences in children. Bonney (2002) describes the tragic consequences of the flash burns incurred during the Battle of Jutland which affected the face and the hands and were caused by the momentary flame of high explosives.
3.4.13 Electric Shock Hobby and Laing (1988) recognised four groups of electrical injury in 169 cases from a total of 3,300 patients: true electrical injury, from current passing from the conductor through
Reactions to Injury
105
Fig. 3.35 Urgent and extensive division of the skin and fascia in a full thickness burn from petrol fire (Courtesy of Major K Brown RAMC).
the skin to the tissues; “arc” burn, where a current passes external to the body to the ground; secondary flame burn, from ignited clothing; and direct burn, from a hot electrical element. Reasonable spontaneous recovery occurred in those cases where the blood supply to the adjacent tissues was not destroyed. Di Vincenti et al. (1969) reported 69 cases of electrical burns, mostly from high tension injuries. The median and ulnar nerves were the most frequently damaged and the lesions were often severe and irreparable. Because nerves are good conductors the neurovascular bundles may be destroyed. Clifton et al. (2000) advise immediate repair of trunk nerves charred in electrical burns using nerve grafts which are covered by free full thickness skin flaps or by a vascularised nerve graft, such as the lateral cutaneous nerve of forearm within a free lateral arm flap. When electrical contact is brief there may be little burning but the passage of electrical current causes break down of the cell membranes. Clifton and colleagues describe this process of electroporation which is caused: “by polar water molecules being driven into molecular scale defects in the lipid bilayer component of the cell membrane, causing the defects to enlarge, with an increase in permeability, and the membrane to rupture.” Muscle and nerve cells are particularly susceptible to this injury which is responsible for some of the immediate clinical signs such as muscle spasms and episthotonus. The case described by Thaventhiran et al. (2001) provides one example of this phenomenon. The patient was injured by contact with a 20,000 voltage AC supply, the current entering at the right arm and exiting through the left elbow. He was completely paralysed below the neck but over
the subsequent 5 years there was extensive recovery although he remained doubly incontinent: “the remarkable degree of peripheral nerve regeneration and recovery seen in this patient suggests that the axons were selectively injured, leaving the surrounding tissue including the Schwann cells intact to enable subsequent regeneration.” The recovery of the very severe and generalised sensory motor polyneuropathy unmasked the extent of the myelopathy. There are some similarities between this case that of a 34 year old man who was struck by lightning. He suffered immediate loss of consciousness and cardiac arrest but was successfully revived. He was wearing a heavy gold chain at the time and a deep entry burn was seen on the right side of his neck with an exit burn on the left side. There were second degree flash burns to the chest, groin, upper and lower legs and there were superficial burns over the whole of his torso which formed a feathery or fern leaf pattern. He had a C5 tetraplegia with both urinary and faecal incontinence. His rehabilitation was directed by Dr Lal Landham (Rochester) and we were given the opportunity to examine him on several occasions. By 2 years after injury he had regained considerable recovery of power in all four limbs so that he could walk and he had regained urinary and faecal control. Quantitative sensory testing showed normal sensory thresholds for all modalities in all four limbs except for an elevated vibration threshold in the right lower limb. Neurophysiological investigations confirmed normal sensory and motor conduction but persisting denervation of the C5 and C6 myotomes. He had sustained a severe, diffuse sensory motor neuropathy. The permanent defect lay with the anterior horn cells at C5
106
Surgical Disorders of the Peripheral Nerves
findings did nothing to explain the transient interruption of nerve function after percussive injury described by DennyBrown and Brenner (1944).
3.4.15 Injection Injury
Fig. 3.36 Function in the upper limbs 2 years after lightning strike. Note the atrophy of the muscles of the shoulder and in the arm.
and C6 and it seems likely that the site of the entry and exit burns in the neck is relevant. Perhaps these neurones were destroyed by electroporation? (Fig. 3.36). Quantitative sensory testing may reveal abnormalities which cannot be detected by classical neurophysiological investigations. A 40 year old man was electrocuted (AC current), by a faulty fuse in his house. The current passed through his right ear down to his right hand. He experienced very considerable pain and was unable to use his right upper limb. NPI were performed, first, 1 month after the injury and then again after 6 months which showed no disturbance either of sensory or motor conduction. Electromyography showed a pattern consistent with conduction block. However, QST at 6 months revealed diminished pinprick sensation in the right forearm and hand but also in the right trunk. Thermal thresholds were elevated in the right thenar eminence, with paradoxical burning on cooling by 7.2°C. There was also a suggestion of elevated thresholds in his right T2 region. Vibration thresholds were elevated at the right thumb and middle finger. In the same area there was elevation of monofilament threshold. Cotton wool sensation was reduced in the same distribution as pinprick sense. His sweating was preserved.
3.4.14 Percussion Injury Richardson and Thomas (1979) studied the effects of percussion injury. The clinical application in the peripheral nerve field is perhaps slight, but the authors indicate that their results could give some indication of events in injury of the spinal cord, in spite of the difference in structure between central and peripheral nervous tissue. The lesion appears principally to be a localised axonal interruption with later degeneration. Richardson and Thomas noted that their
The perineurium may be lacerated by the point of a needle an event which causes severe pain in the conscious patient. If that complaint of pain is ignored and through an injection into the nerve, the consequences may be severe. The damaging substances commonly injected into nerves are: steroid preparations; anaesthetic agents for intravenous use such as thiopentone; non steroidal anti-inflammatory drugs; anxiolytic agents such as diazepam; antibiotics; and local anaesthetics. The nerves most commonly affected are the brachial plexus in the neck and axilla, the radial nerve in the arm, the median at the elbow and the sciatic in the buttock. Usually, the occurrence of severe local and radiated pain makes it plain that the drug has been injected into the nerve. The onset of later and sensory paralysis may, however, be delayed for some hours. Although Hudson (1984) suggested that the ill effects arise from intraneural injection rather than from injecting near the nerve and later diffusion, it is hard to resist the speculation that the latter must sometimes be the responsible mechanism, a suggestion which is supported by Kline and Hudson (1995). The response of the nerve to the needle and to the substances injected is illustrated by some cases that we have treated. 1. A 40 year old man of slender physique came to operation which required exposure of the common peroneal nerve at the knee. After induction of general anaesthesia the line of incision was infiltrated with 0.25% bupivicaine. This provoked a twitch in the extensor muscles of the ankle indicating that the needle had been passed too deeply. The nerve was then exposed and the epineurium was greatly distended by the injected fluid. The bundles within were intact but the epineurial circulation had disappeared over a segment of some 4 cm. The epineurium was incised to decompress the nerve; the operation was then completed. On awakening he had a foot drop with sensory loss but no pain. By 6 h, sensation had recovered and there was the first evidence of recovery into the dorsiflexor muscles. Recovery was complete by 36 h. 2. A 45 year old woman with rheumatoid arthritis, for which she was taking prednisolone, experienced sudden severe shooting pain into her hand during venepuncture at the elbow. Pain persisted and sensation remained abnormal in the index and middle fingers whereas muscle power and sympathetic function remained intact. At 24 months the amplitude and velocity of sensory conduction from the affected digits was reduced to less than one half of
Reactions to Injury
normal. The median nerve was explored at the level of lesion which was marked by a strong Tinel sign. The epineurium was thickened and adherent to adjacent structures; epineurotomy revealed a neuroma of two bundles. After external neurolysis local anaesthetic was infused about the nerve by catheter for 24 h. Her pain was greatly improved. 3. The sciatic nerve of a 60 year old woman was blocked by 30 ml of 0.5% bupivicaine during operation for total knee replacement. Numbness and weakness were apparent within a few minutes but pain did not develop until 7 days had passed, increasing in intensity for the next 14 days. This was expressed as burning pins and needles in the sole of the foot and as shooting or electrical pain to the dorsum of the foot. The common peroneal nerve recovered by 6 weeks but the lesion of the tibial nerve remained dense. NPI at 14 weeks confirmed loss of conduction for the tibial nerve but recovery of motor conduction in the common peroneal nerve. Normal motor units were demonstrated in gastrocnemius but there were fibrillations in the abductor hallucis. Pain resolved spontaneously by 16 weeks, recovery was extensive by 36 weeks and by this time conduction was now detectable in the sural nerve, but the tibial nerve action potential was still absent and sympathetic paralysis persisted. It seems that recovery was better for the larger myelinated fibres than it was for the finely myelinated and non myelinated fibres. 4. A 6 year old boy was given an injection of antibiotics into the buttock and he experienced instantaneous and severe pain. When he was seen 2 years later there was a severe equinovarus deformity with shortening of the femur, the tibia and the foot on the ipsilateral side. Neurophysiological investigations at that time revealed an extensive degenerative lesion. There was no motor or sensory conduction in the common peroneal nerve, sensory conduction was preserved in the sural nerve. Denervation of the muscles was widespread, but rather deeper for the common peroneal nerve than it was for the tibial divisions. The posture of the foot was improved by an extensive operation which included elongation of the tendo Achillis, slide of the plantar muscles, shortening of the lateral column of the foot, calcaneal osteotomy and anterior transfer of tibialis posterior (Mr Fergal Monsell, RNOH). Experience with injection injuries in human beings does not always match with that of injection into the nerves of rats, just as the behaviour of the former is not regularly matched by that of the latter. Full recovery is by no means invariable; noxious injection is often followed by epineurial fibrosis and sometimes by dense intraneural scarring. Persistence of pain following an injection of steroid for carpal tunnel syndrome led to exploration of the median nerve in four patients at intervals ranging from 2 weeks to 8 weeks after injection. The nerves were found inflamed and swollen over a length of some 3 cm and they were surrounded by filmy adhesions, but
107
there were no signs of penetration of the trunk by the needle. Injection of steroids into the buttock of a 64 year old man with psoriasis and scleroderma was followed, at about 1 hour, by complete common peroneal palsy and a painful partial lesion of the tibial nerve. The pain resolved spontaneously at about 3 months. Neurophysiological investigations at 1 year showed extensive demyelination with some axonopathy. Sensory conduction was absent: motor conduction was reduced in amplitude and in velocity. Polyphasic potentials were widespread in the muscles sampled but there were no fibrillations. It is possible that diffusion of the agent provoked vasculitis within and about the nerve in this case. The prognosis is not at all good. Pandian et al. (2006) followed 65 lesions of sciatic nerve and radial nerves caused by intramuscular injections of various drugs. Axonopathy was confirmed in all cases, and reinnervation was demonstrated in only one third. Pain was usual. The consequences for the growing limb were particularly severe.
3.4.16 Vibration Injury Extensive studies from the Department of Hand Surgery in Malmo have improved understanding of this difficult and controversial field (Stromberg 1997). The regular use of hand held vibrating tools may lead to a complex of symptoms, the hand–arm vibration syndrome. Stromberg defines three groups of symptoms: sensory-neural; vasospastic, and a combination of both. Cold intolerance presents as a significant symptom in one half of the patients with sensoryneural symptoms. There was impairment of nerve conduction, vibro tactile sense, and temperature sense in all patients, changes which were more strongly expressed in the median nerve. Biopsies of the dorsal interosseous nerve revealed demyelination, endoneurial and perineurial fibrosis, and loss of axons.
3.5 The Perineurium and Neoplasm or Infiltration Tumours arising from cells within the endoneurium are contained within the perineurium which acts as a compartment. The tumour may be a benign schwannoma but malignant peripheral nerve sheath tumours whose cell of origin is the Schwann cell, are contained within the perineurium, at least initially, and spread within it. We have seen intraneural spread of malignant tumours extending up to 18 cm from the main tumour (Fig. 3.37). The brachial and lumbo sacral plexus are common sites for malignant invasion by direct spread or by metastases through the lymphatics. However
108
Surgical Disorders of the Peripheral Nerves
3.6 Radiation and Peripheral Nerves
Fig. 3.37 Malignant peripheral nerve sheath tumour in the medial cord of the brachial plexus in a patient with NF1. A biopsy of C8, 15 cm proximal to the main tumour mass, shows one bundle replaced by tumour and some degenerative changes in adjacent bundles. Haemotoxylin and eosin ×160 (Courtesy of Dr Jean Pringle).
there are no lymphatics within the endoneurium and the two cases of lumbosacral radiculoplexopathy in prostate cancer, described by Ladha et al. (2006a) at the Mayo Clinic, represent, in all likelihood, direct extension of the tumour within nerves from the prostate. Both patients had experienced worsening pain with a deepening sensory motor loss for a number of years. In one case this led to an increasing defect of bladder and bowel control. Biopsy of the affected nerves showed tumour cells organised into a glandular appearance, surrounded by S-100 positive cells which were contained within the perineurium. These workers comment that magnetic resonance imaging is extremely helpful in localising the disease and defining its extent and they observe that: “the clinician should always consider the entity of neoplastic radiculoplexopathy in the differential diagnosis of progressive, painful, lower extremity weakness as treatments such as radiotherapy may be palliative and spare further neurological decline.” Ladha et al. (2006b) described two cases of isolated amyloidosis presenting as lumbosacral radiculoplexopathy. The first patient experienced a painless progressing sensory and motor loss. There was no systemic disease. Biopsy revealed diffuse infiltration of the endoneurium and perineurium. The MR scan in this case showed enlargement of the sciatic nerve with a denervation signal in the gluteal muscles. In the second case the lesion was more severe, with incontinence of bladder and bowel. This patient had successfully been treated for a lymphoma some years previously and there were no signs of active disease when he presented to the Mayo Clinic. Biopsy of the cauda equina was performed which revealed replacement by amyloid, and his neurological state was stabilised by excision of an “amyloidoma” from L3. The neurological effects of generalised amyloidosis are protean, the subject is reviewed by Kyle et al. (2005).
The statement by Janzen and Warren (1942) that “nerve tissue is extremely resistant to radiation” is, alas, true only so far as the function of conducting an impulse is concerned, and in so far as doses up to 10,000R (100 Gy) “do not cause overt damage to this normally static cell population” Vujaskovic (1997) found that peripheral nerves were damaged by exposure exceeding 20 Gy. The more deeply seated larger fibres were worst affected: the first changes occurred within the axon, where there was increased density of microtubules and neurofilaments. The lesion is perhaps best considered as (1) a lesion from external compression exerted by fibrosis of soft tissue and (2) an intrinsic lesion of the nerve. The latter affects the axon, the Schwann cells and the myelin sheath; it is associated with vasculitis, which leads eventually to fibrosis (Spiess 1972) (Fig. 3.38a, b). The affection may extend to the main vessel (Fig. 3.39). It seems likely that the dose of radiation tolerated by neural tissue depends broadly on the total dose and the period of time over which it is given, but evidently there are individual variations and there may indeed be individual susceptibilities.
3.7 Envenomation In the singular case reported by Laing and Harrison (1991) a young man’s right ulnar nerve was affected by the venom expressed from the tentacle of a box jellyfish (Chironex fleckeri) and presumably conveyed transcutaneously. The nerve was explored and found to be oedematous, but the period elapsing between injury and exploration is not related. There was spontaneous recovery. The case evokes echoes of that of the Lion’s Mane (1909), though because of the absence of Dr Watson from the latter we are condemned to remain in the dark about the presence of neuropathy in the surviving victim of Cyanea capillata. The fascinating field of poisons and evanomation is reviewed by Donaghy (2009a). The active agents cause death by interfering with ion channel function, by blocking the neuromuscular junction and by blocking synaptic transmission within the central nervous system. Cicutoxin, the agent in water hemlock, blocks ion channels and also receptors to gamma amino butyric acid (GABA) in the brain and the symptoms bear little relation to those attributed to Socrates in the Phaedo. The Puffer fish contains tetradotoxin which blocks sodium channels, a phenomenon which has enabled the analysis of these channels. The Krait injects bungarotoxins which block the post synaptic acetyl choline receptors at the neuromuscular junction; whereas the venom injected by the Mamba and the Taipan block pre synaptic nerve endings.
Reactions to Injury
109
a
Fig. 3.39 Radiation induced thrombosis of the third part of subclavian artery. A 38 year old woman developed severe pain, swelling and paralysis 6 weeks after completing a course of radiotherapy for breast cancer. She lost all sensation in the tips of her fingers. She could recognise light touch but could not localise it in the palm of the hand. Sensation was normal proximal to the wrist crease. There was no recovery of skeletal muscle or of smooth muscle function in her forearm and hand.
b
Fig. 3.38 Radiation neuropathy. Electron microphotographs of specimens from biopsy of the lateral trunk of the brachial plexus taken two and a half years after radiotherapy for cancer of the breast. (a) Extensive collagenisation, loss of axons and myelin ×4,125. (b) Extensive demyelination and Schwann cell loss ×4,125 (Electron micrographs prepared by Mr Michael Kayser).
3.8 The Peripheral Effects of Denervation These effects follow degenerative lesions in which recovery fails or is long delayed. If regeneration of axons into the distal stump fails, changes, which over time become
irreversible, develop in the target organs and in the proximal neurone. Motor end-plates atrophy and disappear (Bowden and Gutmann 1944); the denervated muscles undergo fibrosis. Bowden and Gutmann (1944) recorded that after 3 months denervation it became increasingly difficult to identify the end plates. From 3 years on they found fragmentation of structures, and reckoned that this represented irreversible change. Luco and Eyzaguirre (1955) found that the nearer the lesion was to the point of denervation the more rapidly did the changes in the end plates appear. Cutaneous sensory end-organs atrophy more slowly. Dellon (1981) found that they degenerate and eventually atrophy. The reaction of muscle spindles to denervation has been studied extensively. Sherrington (1894) found the spindles “very obvious amid atrophied extrafusal muscle fibres five months after denervation.” Batten (1897) found in a case of complete paralysis after a lesion of the brachial plexus that atrophy of spindles occurred but took place much later than was the case in ordinary muscle fibres. Tower (1932) found that the nuclear bag fibres showed only a slight reduction in diameter after denervation, whereas the chain fibres atrophied almost as rapidly as did the extrafusal fibres. Tower noted in particular the thickening of the capsule. De Reuck (1974) reported rather similar findings. Swash and Fox (1974) noted capsular thickening and atrophy of intrafusal fibres after sensorimotor denervation and Myles and Glasby (1992) conclusively showed that atrophy of muscle spindles followed uncorrected denervation. Loss of nervous control of the small peripheral blood vessels leads to defects of nutrition and atrophy in the subcutaneous tissues. This is compounded by unnoted injury, which can lead to the loss of part or the whole of the limb or even death as is seen in congenital insensitivity to pain, in leprosy
110
Surgical Disorders of the Peripheral Nerves
or in septicaemia from an unnoted sore in neglected cases of sciatic palsy. The effects of denervation are particularly severe in the growing child especially so where there is persisting imbalance of muscle forces acting across a joint (Fig. 3.40). It is our impression that the shortening of the limb is more severe after complete lesions of main trunk nerves such as the sciatic than it is in cases of poliomyelitis.
3.9 Changes at the Higher Levels: The Phantom Limb
Fig. 3.40 Atrophy of the left foot in an 11 year old boy 4 years after transection of the tibial and common peroneal nerves at the knee. The repair of the common peroneal nerve was successful, that of the tibial nerve failed.
The changes in the functional organisation of the somatosensory cortex are rapid and extensive. Structural or, at least, synaptic reorganisation at a subcortical level was probably shown by Banati et al. (2001) who demonstrated changes in the activity of glial cells in the thalamus in patients with intractable pain after injury to the brachial plexus or peripheral nerves (Fig. 3.41). Ramachandran and Hirstein (1998) described their findings in a patient (DS) who had his arm amputated 1 year after an avulsion lesion of the brachial plexus. Stimulating the skin of the face evoked a sensation which was referred to the amputated digits: “when the cotton swab was moved continuously from the angle of the mandible to the symphysis menti, the referred sensation felt as if ‘it was moving from the ball of the thumb to the tip in an arc like motion’. This observation was replicated several times.” Ramnachandran and Hirstein suggest that the sensory input from the face is being received in two different cortical areas: the original face area and the area that previously only received information from the arm. Kew et al. (1997) studied
Fig. 3.41 Positron emission tomographic scan showing an in vitro marker of increased glial cell activity indicating synaptic reorganisation in the thalamus in a patient with persisting pain 4 years after brachial plexus lesion (Courtesy of Professor Praveen Anand).
Reactions to Injury
two patients with avulsion lesions of the brachial plexus using positron emission tomographic (PET) scanning. Both of these patients had referred sensations on the anterior and posterior chest wall where stimulation evoked a response in the amputated hand. PET scans showed a precise correlation between the finger where the referred sensation was felt and the disposition of the cortical maps. The phenomena of the “phantom” limb provides opportunities for analysis of the changes in the somatosensory cortex after injury. Mitchell (1872) wrote: “nearly every man who loses a limb carries about with him a constant or inconstant phantom of the missing member, a sensory ghost of that much of himself and sometimes a most inconvenient presence, faintly felt at times, but ready to be called up to his perception by a blow, a touch or a change of wind.” To Lord Nelson the severe phantom limb pain which he experienced in his right arm, lost during the attack at Santa Cruz provided a “direct proof of the existence of the soul” (Riddoch 1941). One extraordinary example of the persistence of a painful phantom is provided by a patient treated by our colleague, Christopher Wynn Parry, in the Rehabilitation Unit of the Royal National Orthopaedic Hospital. A 33 year old woman presented with persistent and severe phantom pain following amputation of her right arm in a road traffic accident at the age of 11 years. She described her clenched fist tightly holding a florin, for she had been going shopping at the time of her accident. There was a dramatic response to transcutaneous nerve stimulation and she described that, as the pain diminished, she felt the fingers of her clenched fist open and the coin dropped from her hand. The transcutaneous nerve stimulator was used on several occasions, ultimately with lasting relief. On the last occasion she described how she “saw” the coin disappearing down a drain. There are a number of possible explanations for this result; we might add that Christopher Wynn Parry often succeeded in relieving pain in patients otherwise abandoned.
References Adrian ED (1916) The electrical reactions of muscles before and after injury. Brain 39:1–35 Adrian ED (1917) Physiological basis of electrical tests in peripheral nerve injury. Arch Radiol Electrother 21:379–393 Aitken JT, Thomas Pk (1962) Retrograde changes in fibre size following nerve section. J Anat 96:121–136 Anand P, Birch R, Terenghi G, Wellmer A, Cedarbaum J, Lindsay RM, Williams-Chestnut RE, Sinicropi DV (1997) Endogenous nerve growth factor and ciliary neurotrophic factor levels in human peripheral nerve injury. NeuroReport 8:1935–1938 Archer DR, Griffin JW (1993) “Bystander” proliferation of cells in the sciatic nerve is induced by lumbar ventral radiculotomy.Cited by Griffin JW, Hoffman PN In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy. 4th edn. WB Saunders, Philadelphia, p 365 Aspden RM, Porter RW (1994) Nerve traction during correction of knee deformity. J Bone Joint Surg 76B:471–473
111 Banati RB, Cagnin A, Brooks DJ, Gunn RN, Myers R, Jones T, Birch R, Anand P (2001) Long term transynaptic glial responses in the human thalamus after peripheral nerve injury. NeuroReport 12:3439–3442 Bar KJ, Saldanha GJ, Kennedy AJ, Facer P, Birch R, Carlstedt T, Anand P (1998) GDNF and its receptor component Ret in injured human nerves and dorsal root ganglia. NeuroReport 9(1):43–47 Batten FB (1897) The muscle-spindle under pathological conditions. Brain 20:138–179 Bendszus M, Rieckmann P, Perez J, Koltzenburg M, Reiners K, Solymosi L (2003) Painful vascular compression syndrome of the sciatic nerve caused by gluteal varicosities. Neurology 61:985–987 Bickford RG (1939) The fibre dissociation produced by cooling human nerves. Clin Sci 4:159–164 Birch R, St Clair Strange FG (1990) A new type of peripheral nerve lesion. J Bone Joint Surg 72B:312–313 Birch R, Jessop J, Scott G (1991) Brachial plexus palsy after manipulation of the shoulder. J Bone Joint Surg 73:172 Birch R, Wilkinson MCP, Vijayan KP, Gschmeissner S (1992) Cement burn of the sciatic nerve. J Bone Joint Surg 74B:731–733 Boettger MK, Till S, Chen MX, Anand U, Otto WR, Plumpton C, Trevise DJ, Tate SN, Bountra C, Coward K, Birch R, Anand P (2002) 2002 calcium-activated potassium channel SK1 – and IK1-like immunoreactivity in injured human sensory neurons and its regulation of neurotrophic factors. Brain 125:252–263 Bonney G (2002) The battle of Jutland. Sutton, Phoenix Mill, Stroud. Glos. GL5 2BU. ISBN ISBN O-7509-2926-X Bonney G, Gilliatt RW (1958) Sensory nerve conduction after traction lesion of the brachial plexus. Proc RSoc Med 51:365–367 Bowden REM, Gutmann E (1944) Denervation and re-innervation of human voluntary muscle. Brain 67:273–313 Brouwers PJAM, Kottiuk EJBL, Simon MAM, Prevo RL (2001) A cervical anterior spinal artery syndrome after diagnostic blockade of the right C6 nerve root. Pain 91:397–399 Bywaters EGL, Ge D, Remington C, Smiles J (1941) Myo haemoglobin in the urine of air-raid casualties with crushing injury. J Biochem 35:1164–1168 Cajal RY (1928) Degeneration and regeneration in the nervous system. Oxford University Press, London/New York, Translated and Edited by May RM Carlstedt T (2007) Central nerve plexus injury. Imperial College Press, London pp 83–100. Causey G, Palmer F (1952) Early changes in degenerative mammalian nerves. Proc R Soc Med London B 139:597–609 Chalk CH, Dyck PJ (1993) Ischemic neuropathy. In: Dyck PJ, Thomas PK, Griffin JW, Low PA, Podusio JF (eds) Peripheral neuropathy, 3rd edn. WB Saunders, Philadelphia, pp 980–989 Clarke E, Bearn JS (1972) The spiral nerve bands of Fontana. Brain 95:1–20 Clifton JF, Hall FJ, Guerroro JS (2000) Electrical burn injuries. In: Gupta A, Kay SPJ, Scheker LR (eds) The growing hand, 1st edn. Mosby ISBN OT 234 21331, pp 693–700 Cragg B, Thomas PK (1961) Changes in conduction velocity and fibre size proximal to peroneal nerve lesions. J Physiol 157:315–327 Chalain T de Clarke HM (2000) Thermal and chemical injuries. In: Gupta A, Kay SPJ, Scheker LR (eds) The growing hand, 1st edn. Mosby, ISBN OT 234 21331, pp 665–692 De Reuck J (1974) The pathology of human muscle spindles. Acta Neuropathol 30:43–50 Dellon AL (1981) Results of nerve repair in the hand. In: Dellon AL (ed) Evaluation of sensibility and re-education of sensation in the hand. Williams and Wilkins, Baltimore, pp 193–202 Dellon AL, MacKinnon SE (1988) Human ulnar neuropathy at the elbow: clinical, electrical and morphological correlations. J Reconstr Microsurg 4:179–184
112 Denny Brown D, Adams RD, Brenner G, Doherty MM (1945) The pathology of injury to nerve induced by cold. J Neuropathol Exp Neurol 4:305–323 Denny-Brown D, Brenner C (1944) The effect of percussion on nerve. J Neurol Neurosurg Psychiatry 7:76–95 Di Vincenti FC, Moncrieff JA, Pruitt BA (1969) Electrical injuries: a review of 65 cases. J Trauma 9:497–507 Dodt E (1953) Differential sensitivity of mammalian nerve fibres. Acta Physiol Scand 29:91–108 Donaghy M (2009a) Polyneuropathy. In: Donaghy M (ed) Brain’s diseases of the nervous system, 12th edn, Oxford University Press, Oxford Chapter 21, pp 541–602 Donaghy M (2009b) Toxic and environmental disorders. In: Donaghy M (ed) Brain’s diseases of the nervous system, 12th edn, Oxford University Press, Chapter 5, pp 141–164 Douglas WW, Malcolm JL (1955) The effect of localized cooling on conduction in cat nerves. J Physiol 130:63–71 Durrenberger PF, Facer P, Gray RA, Chessell IP, Naylor A, Bountra C, Banati RB, Birch R, Anand P (2004) Cyclooxygenase – 2 (COX-2) in injured human nerve and a rat model of nerve injury. J Peripher Nerv Syst 9:15–24 Durrenberger PF, Facer P, Casula MA, Yiangou MY, Gray RA, Chessell IP, Day NC, Collins SD, Birnham S, Wilson AW, Elliot D, Birch R, Anand P (2006) Prostanoid receptor EP1 and COX-Z in injured human nerves and a rat model of nerve injury: a time course study. BMC Neurol 6:1 Dyck PJ, Nukada H, Lais CA, Karnes J (1984) Permanent axotomy: a model of chronic neuronal degeneration produced by axonal atrophy, myelin remodelling and regeneration. In: Dyck PJ, Thomas PK, Lambert EH, Bunge R (eds) Peripheral neuropathy, 2nd edn. WB Saunders, Philadelphia, pp 660–690 Dyck PJ, Dyck J, Engelstad PJB (2005a) Pathological alterations of nerves. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, pp 733–829, Chapter 32 Dyck PJB, Engelstad J, Dyck PJ (2005b) Microvasculitis. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, pp 2405–2414, Chapter 106 Eames R, Lange LS (1967) Clinical and pathological study of ischaemic neuropathy. J Neurol Neurosurg Psychiatry 30:215–236 Fowler TJ, Danta G, Gilliatt RW (1972) Recovery of nerve conduction after a pneumatic tourniquet: observations in the hind limb of a baboon. J Neurol Neurosurg Psychiatry 35:638–647 Franz DN, Iggo A (1968) Conduction failure in myelinated and nonmyelinated axons at low temperatures. J Physiol 199:319–345 Galardi G, Comi G, Lozza L, Marchettini P, Novarina M, Facchini R, Paronzini A (1990) Peripheral nerve damage during limb lengthening. J Bone Joint Surg 72B:121–124 Gardiner MD, Mangwani J, Williams WW (2006) Aneurysm of the common iliac artery presenting as a lumbosacral plexopathy. J Bone Joint Surg 88B:1524–1526 Gereis V (2005) Paresis and paralysis, pers com Gilliatt RW (1981) Physical injury to peripheral nerves. Mayo Clin Proc 56:361–370 Gilliatt RW, Hjorth RJ (1972) Nerve conduction during Wallerian degeneration in the baboon. J Neurol Neurosurg Psychiatry 35:335–341 Gilliatt RW, Taylor JC (1959) Electrical changes following section of the facial nerve. Proc Royal Soc Med 52:1080–1083 Greer HD (1997) Neurological consequences of diving. In: Bove AA, Davis JC (eds) Diving medicine. WB Saunders, Philadelphia, pp 258–269 Griffin JW, Hoffman PN (1993) Degeneration and regeneration in the peripheral nervous system. In: Dyck PJ, Thomas PK, Lambert EH, Bunge R (eds) Peripheral neuropathy, 3rd edn. WB Saunders Company, Philadelphia, pp 361–376 Groves MJ, Scaravilli F (2005) Pathology of peripheral neurone cell bodies. In: Dyke PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, pp 683–732, Chapter 31
Surgical Disorders of the Peripheral Nerves Gutmann E, Sanders FK (1943) Recovery of fibre numbers and diameters in the regeneration of peripheral nerves. J Physiol 101:489–518 Haftek J (1970) Stretch injury of peripheral nerve. J Bone Joint Surg 52B:354–365 Haftek J, Thomas PK (1967) An electron-microscopic study of crush injuries to peripheral nerves. J Anat 102:154–155 Hall S (1999) The biology of chronically denervated Schwann cells. Ann NY Acad Sci 883:215–233 Hall S (2005) The response to injury in the peripheral nervous system. J Bone Joint Surg 87B:1309–1319 Harriman DGF (1977) Ischaemia of peripheral nerve and muscle. J Clin Pathol 30(supplII):94–99 Hess K, Eames RA, Darveniza P, Gilliatt RW (1979) Acute ischaemic neuropathy in the rabbit. J Neurolog Sci 44:19–43 Hill S, Hall S (1999) Microscopic anatomy of the posterior interosseous and median nerves at sites of potential entrapment in the forearm J Hand Surg 24B: 170–176 Hobby JAE, Laing JE (1988) Electrical injuries of the upper limb. In: Tubiana R (ed) The hand, vol III. WB Saunders Company, Philadelphia, pp 779–787 Hoffer JA, Stein RB, Gordon T (1979) Differential atrophy of sensory and motor fibres following section of cat peripheral nerves. Brain Res 178:347–361 Holmes W, Young JZ (1942) Nerve regeneration after immediate and delayed suture. J Anat 77:63–96 Hongell A, Mattson HS (1971) Neurographic studies before, after and during operation for median nerve compression in the carpal tunnel. Scand J Plast Reconstr Surg 5:103–109 Hoogeveen JF, Wondergem J, Troost D, van der Kracht AHW, Haveman O (1991) Blood vessel damage followed by Wallerian degeneration in the rat sciatic nerve after local hyperthermia. Neuropathol Appl Neurobiol 17:250–251 Houten JK, Errico TJ (2002) Paraplegia after lumbosacral nerve root block: report of three cases. Spine 2:70–75 Hudson AR (1984) Injection injuries of nerves. In: Dyck PJ, Thomas PK, Lambert EH, Burge R (eds) Peripheral neuropathy, 2nd edn. WB Saunders, Philadelphia. pp 429–431 Irwin MS, Gilbert SEA, Terenghi G, Smith RW, Green CJ (1997) Cold intolerance following peripheral nerve injury. Brit J Hand Surg 22:308–316 Janzen AH, Warren S (1942) Effect of roentgen rays on peripheral nerves of rats. Radiology 38:333–337 Jefferson D, Neary D, Eames RA (1981) Renaut body distribution at sites of human peripheral nerve entrapment. J Neurolog Sci 49:19–29 Kew JJ, Halligan PW, Marshall JC, Passingham RE, Rothwell JC, Ridding MC et al (1997) Abnormal access of axial vibrotactile input to deafferented somatosensory cortex in human upper limb amputees. J Neurophysiol 77:2753–2764 Kline DG, Hudson AR (1995) Nerve injuries,WB Saunders, Philadelphia Injection injury p 46–50 Kyle RA, Kelly JJ, Dyck PJ (2005) Amyloidosis and neuropathy. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. WB Saunders, Philadelphia. Chapter 108, pp 2427–2452 Ladha SS, Spinner RJ, Suarez GA, Amrami KK, Dyck PJB (2006a) Neoplastic lumbosacral radiculoplexopathy in prostate cancer by direct perineurial spread: An unusual entity. Muscle Nerve 34: 659–665 Ladha SS, Dyck PJB, Spinner RJ, Perez DG, Zeldenrust SR, Amrami KK, Solomon A, Klein CJ (2006b) Isolated amyloidosis presenting with lumbosacral radiculopathy: description of two cases and pathogenic review. J Peripher Nerv Syst 11:346–352 Laing JHE, Harrison DH (1991) Envenomation by the box-jellyfish - an unusual cause of ulnar nerve palsy. J Royal Soc Med 64:115–116 Landau WM (1953) The duration of neuromuscular function after nerve section. Neurosurgery 10:64–68 Lawson SN (2005) The peripheral sensory nervous system: dorsal root ganglion neurones. In: Dyck PJ, Thomas PK (eds) Peripheral
Reactions to Injury neuropathy (in two volumes), 4th edn. Elsevier Saunders, Philadelphia, pp 163–202, Chapter 8 Leonard MH (1967) Immediate improvement of sensation on relief of extraneural compression. J Bone Joint Surg 51A:1282–1284 Lewis T, Pickering GW, Rothschild P (1931) Centripetal paralysis arising out of arrested bloodflow to the limb. Heart 16:1–32 Li H, Terenghi G, Hall SM (1997) Effects of delayed reinnervation on the expression of C-er6B receptors by chronically denervated rat Scwann cells in vitro. Glia 20:333–347 Li H, Wigley C, Hall SM (1998) Chronically denervated rat Schwann cells respond to GGF in vitro. Glia 21:1–14 Luco JV, Eyzaguirre C (1955) Fibrillation and hypersensitivity to ach in denervated muscle: effects of length on degenerated nerve fibers. J Neurophysiol 18:65–73 Lundberg A (1948) Potassium and the differential thermosensitivity of membrane potential, spike and negative after potential in mammalian A and C fibres. Acta Physiol Scand Suppl 50(15):1–67 Lundborg G, Rydevik B (1973) Effects of stretching the tibial nerve of the rabbit: preliminary study of the intraneural circulation of the barrier function of the perineurium. J Bone Joint Surg 55B:390–401 MacKinnon SE, Dellon Al (1986) Experimental study of chronic nerve compression. Hand Clin 2:639–650 MacKinnon SE, Dellon AL (1988) Ischaemia of nerve: loss of vibration sensibility. In: Surgery of the peripheral nerve. Thieme Medical, New York, Georg Thieme Verlag Stuttgart New York p 57 MacKinnon SE, Dellon AL, Hudson AR, Hunter DA (1985) A primate model for chronic nerve compression. J Reconstr Surg Microsurg 1:185–195 Mannan K, Altaf F, Maniar S, Tirabosco R, Sinisi M, Carlstedt T (2008) Cyclical sciatica. J Bone Joint Surg 90B:98–101 Martin JN, Vialle R, Denormande P, Sorriaux G, Gad H, Harding I, Dizien O, Judet T (2006) Treatment of knee flexion in contracture due to central nervous system disorders in adults. J Bone Joint Surg 88A:840–845 McComas AJ, Sica REP, Banarjee S (1978) Long term effects of partial limb amputation in man. J Neurol Neurosurg Psychiatry 41: 425–432 McManis PG, Low PA, Lagerlund TD (2005) Nerve blood flow and microenvironment. In: Dyke PJ, Thomas PK (eds) Peripheral neuropathy. Elsevier Saunders, Philadelphia, pp 671–680, Chapter 30 Merrington WR, Nathan PW (1949) A study of post-ischaemic paraesthesiae. J Neurol Neurosurg Psychiatry 12:1–18 Mitchell SW (1872) Injuries of nerves and their consequences. JB Lippincott, Philadelphia Mitchell SW, Morehouse GR, Keen WW (1864) Gunshot wounds and other injuries to nerves. JB Lippincott, Philadelphia Myles M, Glasby MA (1992) The fate of muscle spindles after various methods of nerve repair in the rat. Neuroorthopaedics 13:15–23 Nagano A (2003) Spontaneous anterior interosseous nerve palsy review article. J Bone Joint Surg Br 85B:313–318 Nagano A (2008) “Hour glass” like fascicular constriction. Pers com Nagano A, Shibata K, Tokimura H, Yamamoto S, Tajiri Y (1996) Spontaneous anterior interosseous nerve palsy with hourglass-like fascicular constriction within the main trunk of the median nerve. Hand Surg 21A:266–270 Nash TP (2001) Comment on: “A cervical anterior spinal artery syndrome after diagnostic blockade of the right C6-nerve root” Browers PJAM et al. Pain 91:397–399 Neary D, Eames RA (1975) The pathology of ulnar nerve compression in man. Neuropathol Appl Neurobiol 1:69–88 Niels OB, Thomsen, Lars Dahlin B (2007) Injury to the radial nerve caused by fracture of the humeral shaft: timing and neurobiological aspects related to treatment and diagnosis. Scand J Plast Reconstr Surg Hand Surg 41:153–157 Nogueira MP, Paley D, Nave A, Herbert A, Nolente C, Herzenburg JE (2003) Nerve lesions associated with limb lengthening. J Bone Joint Surg 85A:1502–1510
113 Nukada H, Anderson GM, McMorran PD (1997) Reperfusion nerve injury: pathology due to reflow and prolonged ischaemia. J Peripher Nerv Syst 2:60–69 Oaklander AL, Brown JM (2004) Unilateral nerve injury produces bilateral loss of distal innervation. Ann Neurol 55:639–644 Oberlin C, Shafi M, Diverres JP, Silberman O, Adle H, Belkheyar Z (2006) Hourglass like constriction of the axillary nerve. J Hand Surg 31A:1100–1104 Ochoa J, Fowler TJ, Gilliatt RW (1972) Anatomical Changes in peripheral nerves compressed by pneumatic tourniquet. J Anat 113:433–455 Ochs S, Pourmand R, Si K, Friedman RN (2000) Stretch of mammalian nerve in vitro: effect on compound action potentials. J Peripher Nerv Syst 5:227–235 Omura T, Omura K, Sano M, Sawada T, Hasegawa T, Nagano A (2005a) Spatio temporal quantification of recruit and resident macrophages after crush nerve injury utilizing immuno-histo chemistry. Brain Res 1057:29–36 Omura T, Sano M, Omura K, Hasegawa T, Doi M, Sawada T, Nagano A (2005b) Different expressions of BDNF, NT3 and NT4 in muscle and nerve after various types of peripheral nerve injuries. J Peripher Nerv Syst 10:293–300 Paley D (1990) Problems, obstacles and complications of limb lengthening by the Ilizarov technique. Clin Orth Rel Res 250:81–104 Pandian JD, Bose S, Daniel V, Singh Y, Abraham AP (2006) Nerve injuries following intramuscular injections: a clinical and neurophysiological study from North West India. J Periph Nerve Syst 11:165–171 Phang IS,Biant LC, Jones TS (2009) Neurostenalgia of the femoral nerve A treatable cause of intractable hip pain in a young adult. J Arthroplasty. PMID: 19577882 Platt H (1928) On the peripheral nerve complications of certain fractures. J Bone Joint Surg 10:403–414 Pollock LJ, Golseth JG, Arieff AJ (1946) The response of muscle to electrical stimuli during degeneration, denervation and regeneration. Military Neurophysiol 25:236–257 Polo A, Aldegheri R, Zambito A (1997) Lower limb lengthening in short stature. J Bone Joint Surg 79B:1014–1018 Rabert D, Xiao Y, Yiangou Y, Kreder D, Sangameswaran L, Segal MR, Hunt CA, Birch R, Anand P (2004) Plasticity of gene expression in injured human dorsal root ganglia revealed by gene chip oligonucleotide microarrays. J Clin Neuroscience 11:289–299 Ramachandran VS, Hirstein W (1998) The perception of phantom limbs. The DO Hebb lecture. Brain 9(121):1603–1630 Renaut J (1881) Systéme hyalin de souténement des centres nerveux et de quelques organes de sens. Archives de Physiologie Normale et Pathologique 8:845–860 Richardson PM, Thomas PK (1979) Percussive injury to peripheral nerve in rats. J Neurosurg 51:178–187 Riddoch G (1941) Phantom limbs and body shape. Brain 64:197–222 Romanes GJ (1946) Motor localization and the effects of nerve injury on the ventral horn cells of the spinal cord. J Anat 80:117–131 Saldhana G, Bar K, Yiangou Y, Anand P, Birch R, CarlstedT T, Burrin JM (2000) Marked increase of interleukin – 6 in injured human nerves and dorsal root ganglia. J Neurol Neurosurg, Psychiatr 69:693–694 Salisbury RE, Dingeldein GP (1988) The burnt hand and upper extremity. In: Green DP (ed) Operative hand surgery, vol II, 2nd edn. Churchill Livingstone, New York, pp 2135–2164 Salzberg CA, Salisbury RE (1991) Thermal injury of peripheral nerve. In: Gelberman RH (ed) Operative nerve repair and reconstruction. JB Lippincott, Philadelphia, pp 671–677 Sander HW (1999) Mononeuropathy of the medial branch of the deep peroneal nerve in a scuba diver. J Periph Nerv Syst 4:134–137 Schmalbruch H (1987) Loss of sensory neurones after sciatic nerve section in the rat. Anat Rec 219:323–329 Seddon HJ (1943) Three types of nerve injury. Brain 66:237–288 Seddon HJ (1954) Nerve grafting. In: Peripheral nerve injuries by the Nerve Injury Committee of the Medical Research Council MRC Special Report Series 282. London HMSO, pp 402–403
114 Seddon HJ (1975) Surgical disorders of peripheral nerves, 2nd edn. Churchill Livingstone, Edinburgh, pp 67–88 Sherrington CS (1894) On the anatomical constitution of nerves of skeletal muscles: with remarks on recurrent fibres in the ventral spinal nerve-root. J Physiol 17:211–258 Somayaji HS, Saifuddin A, Casey ATH, Briggs TWR (2005) Spinal cord infarction following therapeutic computed tomography-guided left L2 nerve root injection. Spine 30:E106–E108 Sorenson J, Daube JR, Windebank AJ (2006) Electrophysiological findings in a cohort of old polio survivors. J Peripher Nerv Syst 11:241–246 Spencer PS, Weinberg HJ, Raine CS, Prineas JW (1975) The perineurial window – a new model of focal demyelination and remyelination. Brain Res 96:323–329 Spiess H (1972) Schädigungen am peripheren Nervensystem durch iioniserende Strahlen. In: Schriftenreihe neurologie, Band 10. Springer Verlag, New York Spinner RJ, Atkinson JLD, Teil RL, Birch R et al (2003) Peroneal intraneural ganglia: the importance of the articular branch. A unifying theory. J Neurosurg 99:330–343 Spinner RJ, Amrami KK, Kliot M, Johnston SP, Casanas J (2006) Suprascapular intraneural ganglia and glenohumeral joint attachments. J Neurosurg 104:551–557 Stenning M, Drew S, Birch R (2005) Low energy arterial injury at the shoulder with progressive or delayed nerve palsy. J Bone Joint Surg 87B:1102–1106 Stewart M, Birch R (2001) Penetrating missile injuries. J Bone Joint Surg 83B:517–524 Stoll G, Griffin JW, Li CY, Trapp BD (1989) Wallerian degeneration in the peripheral nervous system: participation of both Schwann cells and macrophages in myelin degeneration. J Neurocytol 18:671–683 Stromberg T (1997) Vibration induced neuropathy in the hand. Doctoral thesis. Wallin and Dalholm Boktryckeri AB ISBN. 91-628-2539-9 Sunderland S (1951) A classification of peripheral nerve injuries producing loss of function. Brain 74:491–516 Suzuki H, Oyanagi K, Takahishi H, Kono H, Yokoyama M, Ikjta F (1993) A quantitative pathological investigation of the cervical cord, roots and ganglia after long term amputation of the unilateral upper arm. Acta Neuropathol (Berl) 85:666–673 Swash M, Fox P (1974) The pathology of the human muscle spindle: effect of denervation. J Neurol Sci 22:1–24 Tait GR, Danton M (1991) Contralateral sciatic nerve palsy following femoral nailing. J Bone Joint Surg 73B:689–690 Terenghi G, Calder JS, Birch R, Hall SM (1998) A morphological study of Schwann cells and axonal regeneration in chronically transected human peripheral nerve. J Hand Surg 23B:583–587 Thaventhiran J, O’leary MJ, Coakley JH, Rose M, Nagendran K, Greenwood R (2001) Pathogenesis and recovery of tetraplegia after electrical injury. J Neurol Neurosurg Psychiatry 71:535–537
Surgical Disorders of the Peripheral Nerves Thomas PK (1964) Changes in the endoneurial sheaths of peripheral myelinated nerve fibres during Wallerian degeneration. J Anat 98:175–182 Thomas PK, Bhagat S (1978) The effect of extraction of the interfascicular contents of peripheral nerve trunks in perineurial structure. Acta Neuropathol (Berl) 43:135–141 Thomas PK, Fullerton PM (1963) Nerve fibre size in the carpal tunnel syndrome. J Neurol Neurosurg Psychiatry 26:520–527 Thomas PK, Holdorff B (1993) Neuropathy due to physical agents. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 3rd edn. WB Saunders, Philadelphia, pp 990–1014 Thomas PK, Jones DG (1967) The cellular response to nerve injury. 2. Regeneration of the perineurium after nerve section. J Anat 101:45–55 Tower SS (1932) Atrophy and degeneration in the muscle spindles. Brain 55:77–90 Tsao BE, Wilbourn AJ (2003) The medial brachial fascial compartment syndrome following axillary arteriography. Neurology 61:1037 Tubiana R (1988) Burns. In: Tubiana R (ed) The hand, 3rd edn. WB Saunders, Philadelphia Ungley CC, Channell GB, Richards RL (1945) The immersion foot syndrome. Br J Surg 33:17–31 Vujaskovic Z (1997) Structural and physiological properties of peripheral nerves after intra-operative irradiation. J Peripher Nerv Syst 2:343–349 Walbeehm ET, von Heel EBM, Kuypers PDL, Terenghi G, Hovius SER (2003) Nerve compound action current (NCAC) measurements and morphometric analysis in the proximal segment after nerve transection and repair in a rabbit model. J Peripher Nerv Syst 8:108–115 Waller A (1850) Experiments on the section of the glossopharyngeal and hypoglossal nerves of the frog, and observations of the alterations produced thereby in the structure of their primitive fibres. Philos Trans R Soc Lond 140:423–429 Wilbourn AJ (2005) Brachial plexus injuries. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, pp 1339–1373, Chapter 55 Wilkinson M, Clarke J (1992) Thermal Injuries to the brachial plexus. Injury 23:342–343 Xu D, Pollock M (1994) Experimental nerve thermal injury. Brain 117:375–384 Ygge J (1989a) Neuronal loss in lumbar dorsal root ganglia after proximal compared to distal sciatic nerve resection: a quantitative study in the rat. Brain Res 47B:193–195 Ygge J (1989b) Central projections of the rat radial nerve investigated with transganglionic degeneration and transganglionic transport of horse radish peroxidase. J Com Neurol 279:199–211 Young JZ (1949) Factors influencing the regeneration of nerves. Adv Surg 1:165–220
4
Regeneration and Recovery
Regeneration: the cellular proliferation; importance of integrity of the basal lamina; neurotropins and neurotrophins; regeneration after intradural injury; regeneration of end organs; predictors of recovery. But the journey of the axon tip to an end organ is only the most dramatic of the phases in the process of regeneration, and its arrival is alone no guarantee of the return of useful function. J.Z. Young 1942
Much of what is now known about the regeneration of nerves is based on the work of Ramon Y. Cajal (1928) who developed concepts about the regrowth of divided axons, the formation of growth cones, the re-ordering of growing axons into new fascicles (Cajal 1928a, Cajal 1928b), and the guidance of new axons to their target tissues. He postulated the existence of “allowing or attracting substances” (Cajal 1928c) and gave credit to Forsmann (1900) for introducing the term neurotropism. Cajal described neurotropism as an effect which was initially one of general attraction emanating from the distal stump which later became more individual and specific as the new axon approached its end organ and in particular, the muscle spindle or the Golgi tendon organ. He also outlined the likelihood of a neurotrophic influence provided by the Schwann cells: “the nutritive and tutorial functions of the cells of Schwann had already been recognized by the embryologists, many of whom believed that the success of the thickening of young fibres and the development of the medullary sheath were the principal functions of these cells. We may especially cite Graham Kerr (1904) who studied the behaviour of these cells in the normal development of nerves, and Mott and Halliburton (1901), who attributed to the bands of Büngner of the peripheral stump of cut nerves the function of feeding and protecting recently arrived sprouts. There is only a step between the recognition of these cells as a nutritive placenta and the admission in them of secretions capable of stimulating and orientating the sprouts that are wandering in the scar.”(Cajal 1928c) These concepts have, in effect, been confirmed by later work. Early in the Second World War a Nerve Injuries Committee was again appointed by the Medical Research Council. George Riddoch was appointed chairman and Hugh Cairns, then Professor of Surgery in Oxford, was placed in charge of injuries to the nervous system. Five hospitals were designated for the treatment of nerve injuries. A team was set up in the Zoology Department of Oxford University which included,
amongst others, Young, Medawar, Holmes, Sanders, Gutmann, Guttmann and Ruth Bowden. Young (1993) summarized some of this work. Nerve grafting in all its forms was studied. The first evidence of axonal transport was revealed. The rate of nerve regeneration, the effects of disuse, the regeneration of proprioceptors, the maturation of regenerated fibres and the significance of retrograde influence were amongst the fields of enquiry. Collectively this work marked a most important stage in improving understanding of regeneration and of applying those lessons to clinical practice: here is an example of what has become known, in fashionable circles, as “translational research.” The decades since the appearance of the MRC report in 1954 have seen extensive advances in understanding cellular and molecular mechanisms. These include the role of neurotrophic factors; the dramatic and urgent changes in the expression of genes which transform, at least in the early days, an environment from one which inhibits new axonal growth to one which protects the cell body and facilitates new growth. The functions of the Schwann cell and its basal lamina, and the mechanisms of guidance, selection, and discarding of new axons have been illustrated. Much of this new work is reviewed in the comprehensive and excellent works of Hall (2005a, b), and of Lundborg (2003, 2004). For an authoritative description of the cellular and molecular events which underlie the process of Wallerian degeneration and regeneration the reader is referred to Dyck and Thomas (2005). It is important to remember the differences between the laboratory investigation, in which a controlled, precise and limited lesion is inflicted upon a nerve and the situation faced by the clinician presented with a patient with a massive wound involving the soft tissues, the skeleton, the vessels and sometimes by other injuries which threaten life and limb. The demonstration of regenerating axons across a lesion of a nerve inflicted in the laboratory does not necessarily translate
R. Birch, Surgical Disorders of the Peripheral Nerves, DOI: 10.1007/978-1-84882-108-8_4, © Springer-Verlag London Limited 2011
115
116
to the recovery of function in the human. The phenomena of pain and recovery of sensation are rather poorly revealed by experiments upon small mammals or for that matter by barely justifiable experimentation upon primates. “The proper study of man is man.” We have been singularly fortunate over the years in the work of colleagues who meticulously examined tissues removed from injured nerves in more than 1,000 patients. At St Mary’s Hospital all biopsies were examined by Dr Arthur Boylston, later Professor Boylston at Leeds and at the Royal National Orthopaedic Hospital material was examined by Dr Jean Pringle. Dr Yusuf Ali, of the Institute of Orthopedics actively encouraged study of material by transmission electronmicroscopy in investigations performed by Mr Stephen Gshmeissner and by Mr Michael Keyser. The work of the joint diagnostic clinic in the Department of Neurophysiology at Queen Square is related in Chaps. 5 and 6. From 1990 Professor Praveen Anand and his colleagues at the Royal London Hospital and then later at Imperial College investigated, by novel techniques, changes in expression of nerve growth factors and of ion channels, correlating these with the measurement of small and large fibre function using methods which are described in Chap. 5. There is no doubt that the fundamental cellular processes which underlie regeneration are similar in the laboratory, in the injuries of civilian practice and in the wounds of war, but they are modified by the violence of the injury, by the effects of injury on associated tissues and in particular by ischemia and by delay before repair. We offer some general conclusions which are based on our clinical and laboratory investigations. 1. The extent of longitudinal damage of the nerve is greater in a rupture caused by traction that it is in “tidy” transections caused by knife or glass and this effect rapidly worsens with increasing delay before repair. If healing has been complicated by sepsis then it is often necessary in a late repair to resect 3 cm or more from both proximal and distal stumps of the divided nerve before reaching a recognizable architecture and even then microscopic examination reveals considerable abnormality. The concept of retrograde degeneration of the axon extending to the first internode applies only to the most benign lesions, that of crushing the nerve between the tips of a jeweller’s forceps. 2. As time goes by the cellular response in both stumps changes from one friendly to regeneration to one less favorable. Dense collagenisation and a profusion of fibroblasts are characteristics of the distal stump in late cases. 3. The normal architecture of the nerve is most closely restored to normal in a well executed primary suture (Birch 1989). 4. It seems that regeneration through a graft falls away along its length and not solely at the suture lines. 5. Delay before repair leads to increasing fibrosis and to shrinking of the distal segment so that it becomes impossible to ensure an accurate topographical match.
Surgical Disorders of the Peripheral Nerves
4.1 The Response of the Nerve and Axon to Transection The cellular response. The cellular process of regeneration involves both proximal and distal stumps and any gap between them. Wallerian degeneration is an essential preliminary to regeneration in peripheral nerves for it is a process which transforms the environment of the peripheral nerve from one which is actively hostile to the sprouting and growth of axons into one which actively supports that process. The disappearance of the distal axon is more than a withering away, rather it is an active self destructive process, “triggered by events which occur before any morphological or electrophysiological changes can be detected” (Hall 2005a). These very early events, which lead to cytoskeletal disintegration, were likened to the lighting of a fuse by Tsao et al. (1999). Within a few hours the transected axon seals off. The proximal end is transformed into a growth cone which has been defined by Lundborg (2004a) as: “a swelling at the tip of the regenerating axon and possesses multiple needlelike extensions, filopodia and broader sheet like extensions, lamellopodia.” The filpodia are rich in actin, and they may extend or retract within a matter of minutes. The axon forms new branches or sprouts: collateral sprouts arise from nodes of Ranvier at levels at which the axons are still intact; terminal sprouts arise from the tip of the surviving axon. Seddon (1975a) described this process: “the individual outgrowing streams of axoplasm – even when confined, as they may sometimes be, within one Schwann tube – have been referred to as fibers or even axons. They are, in fact, only parts of the axon or nerve fibres. The Nile is still the Nile even when the body of water, diversified by rocks or islands, breaks up into many streams.” The axon response is followed, within a few days, by a dramatic increase in the numbers of supporting cells. The influx of haematogenous macrophages into the endoneurium is accompanied by intense mitotic activity within the Schwann cells and the fibroblasts so that the number of nuclei increases by as much as six times or more (Griffin and Hoffman 1993, Dyck et al. 2005) (Figs. 4.1–4.4). The beautiful work of Mirsky and Jessen (2005) describes the development of Schwann cells from the neural crest to Schwann cell precursors, then to immature Schwann cells and finally to myelinating and nonmyelinating cells. This work provides important insights into the behaviour of the Schwann cell after nerve transection. A remarkable feature of the Schwann cell is the rapid reversibility of the last stage of differentiation following separation of the Schwann cells from the axon. This “can be achieved either by injuring nerves in vivo or by dissociating cells from adult nerves and placing them in culture without neurons. Both in vivo and in vitro, the process entails the developmental regression and dedifferentiation of individual Schwann cells and myelin
Regeneration and Recovery
Fig. 4.1 Wallerian degeneration in the ulnar nerve at 3 weeks after section. (Electronmicroscopic study by Stephen Gshmeissner) Distal stump. Macrophage (top left) clears myelin debris. A degenerating large myelinated axon (bottom left). Schwann cells x 2880.
Fig. 4.2 Distal stump. Degenerating large myelinated axons with proliferating Schwann cells x 11340.
breakdown. The characteristic molecular markers and structural features of myelinating and nonmyelinating cells are lost and the cells reenter the cell cycle and regress to a phenotype similar to that of immature Schwann cells. Such denervated Schwann cells redifferentiate readily on reassociation with axons during nerve regeneration” (Mirsky and Jessen 2005). This redifferentiation is associated with the renewed expression of phenotypes which are specific to myelinated and non myelinated axons and also to motor and sensory axons (Hall 2009, Höke et al. 2006). The proliferation, indeed the survival, of Schwann cells is modified by the severity of
117
Fig. 4.3 Proximal stump. Retrograde degeneration of one large myelinated axon, proliferating Schwann cells and sprouting axons, forming a regenerating unit x 6960.
Fig. 4.4 Wallerian degeneration in the ulnar nerve at 3 weeks after section (Electronmicroscopic study by Stephen Gshmeissner). Distal stump. Schwann cell proliferation x 6840.
the injury to the distal trunk. Examination of biopsies taken during late repairs, especially when the injury was complicated by arterial injury or by sepsis, often show remarkably few Schwann cells. Instead it seems that the cellular population is largely composed of fibroblasts, surrounded by much collagen. Myelin fragments are detectable in such cases many months after the injury (Figs. 4.5 and 4.6). Transection of neonatal sciatic nerves in the rat results in the death of teloglia at the neuromuscular junction (Tra chtenburg and Thompson 1996). Mirsky and Jessen (2005) point out that there is no comparable Schwann cell death
118
Fig. 4.5 The proximal stump of the median nerve 3 months after rupture of the nerve and of the axillary artery, which was not repaired 8 mm from the tip of the neuroma. Disintegration of axon and of myelin with very scanty evidence of regeneration activity. x 4880.
following nerve transection after the first post natal week. In adult animals Schwann cells survive in the distal stumps for several months although their numbers slowly decrease and they respond less well to extrinsic signals: “although most Schwann cells within the main nerve trunks survive, there is also increased Schwann cell death within the nerve, indicating a partial dependence on axonal signals. Since the cells can be rescued by application of exogenous neuregulin-1, it is likely that the death is caused, at least in part, by loss of contact with axon associated neuregulin as axons degenerate after transection.” These findings provide further evidence about the harmfulness of delay before repair of divided nerves an also offer a possible explanation for the rather poor recovery of muscle function in the neonate (Anand and Birch 2002). The axon. The axon sprouts form clusters which are surrounded by the cytoplasm of one Schwann cell and its basal lamina: “the regenerating units” of Morris et al. (1972a, b). Dyck et al. (2005) say: “sometimes in abortive regeneration many, (as many as 25–20 or more) neurites will be found in a cluster - regenerative nerve fibre clusters.” The sprouts within one regenerating unit represent the regenerative effort of one
Surgical Disorders of the Peripheral Nerves
Fig. 4.6 The proximal stump of the median nerve 3 months after rupture of the nerve and of the axillary artery, which was not repaired 2 cm proximal to the rupture. One surviving myelinated axon embedded within dense collagen. x 3400.
neurone and its axon (Figs. 4.7–4.9). The proximal segments of regenerating axons become ensheathed by Schwann cells. Enlargement of the diameter of the axon is one important signal for myelination. The regenerating sprouts of unmyelinated fibres remain so. The ensheathment of the growing axon is brought about by complex interactions involving the laminin component of the basal lamina and receptors such as the integrins in the plasma membrane of the Schwann cell and also between the adjacent plasma membranes of the axon and the Schwann cell (Mirsky and Jessen 2005). Reorganisation of regenerating nerve fibres by compartmentation. Cajal (1928b) described the process called “compartmentation” by Morris et al. (1972c) in which the proximal stump becomes divided into “mini fascicles,” replacing the original large bundles (Cabaud et al. 1982). Griffin and Hoffman (1993) described the process by which endoneurial fibroblasts proliferate, migrate, and form a partial basal lamina: “adopting the appearance of perineurial cells; in extreme cases they divide into “mini fascicles”.” We have seen this pattern in many grafts in small mammals and in failed grafts coming to revision in the human. Even after an ideal
Regeneration and Recovery
Fig. 4.7 Regenerating units. Proximal stump of ulnar nerve at 3 weeks after section. (Electronmicroscopic study by Stephen Gshmeissner) Numerous axon sprouts, Schwann cell processes, commencing remyelination with some retrograde degeneration. x 2784.
119
Fig. 4.9 Regenerating units. Proximal stump of ulnar nerve at 3 weeks after section. (Electronmicroscopic study by Stephen Gshmeissner) Axonal sprouts adjacent to a Schwann cell nucleus x 3190.
Fig. 4.8 Regenerating units. Proximal stump of ulnar nerve at 3 weeks after section. (Electronmicroscopic study by Stephen Gshmeissner) Axonal sprouts and Schwann cells x 17980.
laboratory repair, immediate suture after clean transection, regenerating nerve fibres burst out from the perineurium at the suture line and form mini fascicles in the epineurium. Perhaps the process of compartmentation into mini fascicles should be seen as an aberrant or incomplete form of regeneration? (Figs. 4.10–4.13). Guidance and selection. In a carefully performed crush by jeweller’s forceps only the axon is interrupted and growth cones elongate along former Schwann tubes, “the cluster of axonal sprouts Schwann cells and macrophages enclosed in a Schwann cell basal lamina tube are the counterpart of what light microscopists call bands of Büngner” (Dyck et al. 2005). Unfortunately this most favorable situation is rarely
Fig. 4.10 Compartmentation or mini fascicle formation in the proximal stump of the common peroneal nerve ruptured 9 weeks previously in an 8 year old boy. Myelinated and unmyelinated axons and their ensheathing Schwann cells are loosely associated with perineurial like fibroblasts. x 3400.
120
Surgical Disorders of the Peripheral Nerves
Fig. 4.11 Failure of regeneration through a graft inserted 3 years p reviously to bridge a gap in the median nerve. The proximal suture line. Many small bundles (minifascicles) of regenerating myelinated fibres. Scale bar: 100 mm.
Fig. 4.12 The middle of the graft. Increasing fibrosis and loss of neural elements. Scale bar: 100 mm.
encountered in clinical practice. Other mechanisms become important in the control of the exuberant but chaotic regeneration which occurs after nerve repair. Witte and Bradke (2005) describe the guidance of the growth cone: “guidance cues act either by direct contact (e.g., between the axonal growth cone and a repelling environment) or as diffusible factors that are secreted by the target regions. Thus they act as short range or long range cues.” If the “growth cone” encounters non neural tissue, axonal elongation still occurs and may even be followed by myelination. Witzel and Brushart (2003) studied the behaviour of axons traversing a gap. Some projected directly into distal Schwann tubes even though they meandered about in the gap and continued to produce collateral sprouts. Others “arborized,” sampling a number of distal tubes, whilst still others ended in the gap whilst sending out minute sprouts with multiple growth cones. Lundborg (1991) helpfully defines the terms neurotrophic and neurotropic. “A neurotrophic factor (nemron (neurone)
Fig. 4.13 Failure of regeneration through a graft inserted 3 years previously to bridge a gap in the median nerve. Distal segment of graft. Poor regeneration. Scale bar: 100 mm.
and trofh (trophe): nutrition) is a substance influencing survival and growth of nerve cells. A neurotropic factor (nemron (neurone) and tropoV (tropos): way, conduct) is a substance exerting an attraction, at a distance, on growing axons.” He later refined this view (Lundborg 2004a) because neurotrophic factors such as nerve growth factor (NGF) also exert a tropic influence. Trophic and tropic influences are preferable terms. Experiments were undertaken in which axons growing from the proximal stump were “offered” a choice between nerve, tendon and granulation tissue as distal targets. Nerves almost exclusively grew towards nerves (Lundborg et al. 1986, MacKinnon et al. 1986a). An even further degree of specificity may lead motor axons to motor endings and sensory axons to sensory endings (Brushart 1988). Brushart and Seiler (1987) studied regeneration in the rat femoral nerve using retrograde labeling techniques and demonstrated that motor axons entering sensory pathways were “pruned” whilst those entering motor pathways were
Regeneration and Recovery
maintained. Redundant axon sprouts atrophy (Bennett et al. 1986, MacKinnon et al. 1986b, Brushart 1993). There are limits to neurotropism: in experimental repairs where the stumps were deliberately malaligned axonal dispersion was determined by fascicular size, not by fascicular identity (Brushart et al. 1995). Hall (2005a) puts it thus: “there is no doubt that, providing the gap length does not exceed 1 cm, adult rat axons and their associated cells behave as though they are responding to “tropic” signals emanating from a distal stump. The extent to which specificity exists at the fascicular or nerve trunk level has been difficult to establish, perhaps because it is only unmasked in particular experimental protocols, and cannot be demonstrated when axons are challenged to enter grafts rather than intact distal stumps.” Clinicians must strive for accurate topographical alignment during repair whether by suture or by graft, and must accept the limitations of neurotropism exerted by a cutaneous nerve graft. Maturation. Initially, there are more axonal sprouts than there were axons in the intact nerve, but only those that establish connections with end organs survive (Sanders and Young 1946, Aitken et al. 1947). This can cause difficulty in estimating recovery by the numbers of myelinated nerve fibres (MNF) crossing a lesion. In a series of experiments, the sciatic nerve of the rat was transected and repaired by immediate (primary) or delayed suture at 2 weeks, or by immediate grafting or delayed grafting. Predegenerate grafts were also tested. (Birch 1989). The nerves were examined 1 year later. Even at this interval the numbers of MNF at the suture line and in the proximal segment of grafts were increased by about one third more than the number in the proximal trunk or in control nerves. The numbers of MNF in the distal trunk were reduced by about 25% in the primary sutures, by 33% in the secondary sutures and primary grafts and by 40% in the delayed and predegenerate grafts. These differences corresponded to such functional measures of outcome as power and weight of the reinnervated muscles. As connections are established and the regenerating nerve matures, the original profusion of Schwann cells gives way to the more orderly arrangement characteristic of the healthy nerve. Sanders and Young (1946) and Vizoso and Young (1948) studied internodal length and fibre diameter in the developing and regenerating nerves of rabbits. In the adult the former relationship between fibre length and internodal length did not reappear: internodal lengths were always shorter in the regenerated than in the healthy nerve. Griffin and Hoffman (1993) suggest that this is because longitudinal growth in regeneration occurs at the growing tip, whereas the internodal length increases by true longitudinal elongation during growth after birth. The production of nerve growth factors in the distal nerve by endoneurial fibroblasts, macrophages, Schwann cells and by muscle and skin provide an environment friendly to regenerating axons. Indeed, Griffin and Hoffman (1993) remark
121
“given the environment of the distal stump, it is perhaps surprising that functional restoration is so seldom the consequence of nerve injury in humans.” The perspective of most clinicians is perhaps rather different: they think that the results they obtain are not all that bad considering the natural difficulty encountered by regenerating nerve fibres in reaching their targets. When there is a lesion “in continuity” or when healthy nerve end has been opposed to healthy nerve end, the elongating axon enters the environment most favorable to it: the denervated distal stump. Early and accurate suture of a divided nerve is followed by reconstitution of the trunk which is closest to normal. On two occasions the suture lines of the median and ulnar nerves 12 months after reattachment of the amputated hand were inspected during operations of tenolysis. The perineurium surrounding the bundles had reformed, there was only slight thickening of the epineurium and there was no visible neuroma. One indication about the accuracy of repairs completed within 3 or 4 h of the injury is the temporary restoration of conduction. On several occasions we have evoked the appropriate muscular response by stimulating the nerve trunk proximal to the repair. After even the most careful primary suture of a cleanly divided nerve it seems that there is: (1) a diminution in the number of nerve fibres which have made successful reconnection with their target tissues; (2) reduction in the calibre of those nerve fibres; (3) shortening of the internode and (4) slowing of conduction. These facts should not permit any sense of therapeutic nihilism: the best chance for the restoration of useful function rests on following the fundamental principles in the treatment of wounds by prevention of sepsis, restoration of flow through an injured axial artery, restoration of perfusion of the tissues, stabilisation of any fracture and providing full thickness skin cover over the repair. The nerve itself should be repaired as soon as reasonably possible. Whilst there may be a limit to technique much more needs to be done to improve the organisation of the handling of severe nerve injuries. The neuroma. Neuroma is a common example of the fate of the regenerating axons which are unable to form any connections with original targets. Griffin and Hoffman (1993) describe the behaviour of the growth cones when they enter “trackless regions of connective tissue or other foreign tissue even in this setting, the impulse to growth and extend continues; it may result in a meandering elongation followed by at least partial axonal maturation and myelination, culminating a neuroma near the site of the injury, or fronds of regenerating fibres extending over connective tissue planes.” A neuroma is an expression of the vitality of the neurones producing new axons. After amputations of the lower limb it is common to find neuromas of the main nerves 3 cm or more in length, which contain abundant myelinated nerve fibres rather chaotically organised into mini fascicles. Barnes et al. (1945) revealed the capacity for robust regeneration many months
122
after injury. An ulnar nerve was exposed more than 600 days after it had been divided. It could not be repaired. A piece of the distal stump of the nerve was removed and sutured to the proximal stump after the neuroma had been resected. It was a predegenerate graft. This segment was later removed and examined. The graft was reinnervated throughout but there was no myelination of the fibres. Barnes and his colleagues thought that this lack of myelination was due to endoneurial collagenisation, with consequent narrowing of the Schwann tubes, a suggestion made earlier by Holmes and Young (1942). Another reason is provided by Li et al. (1998) who showed that Schwann cells separated from axons for a long time lose their capacity to support regeneration and remyelination. The only way to avoid inflicting neuromas upon patients during the course of treatment is to avoid cutting nerves. Sometimes neuromas are responsible for severe and debilitating pain particularly so when the nerve is only partially injured. It seems strange that wounding a terminal branch of a cutaneous nerve so often leads to severe pain whilst the planned and deliberate section of that nerve at its origin very rarely does so. (Fig. 4.14) It is usual to find numerous axons, some of them myelinated, in the tissues bridging nerve stumps in the human: so it was in the cases studied by Terenghi et al. (1998). These findings illustrate the difference between regeneration and recovery. The extent of reinnervation of the skin by collateral sprouting after resection of a cutaneous nerve was examined by Healy et al. (1996). Twenty patients were studied in whom upper limb nerves had been used as grafts. The area of donor nerve sensory defect shrank by rather small amounts: 6 mm for light touch, 7 mm for sharp pain, and 11.5 mm for heat pain. There was no change in the area of loss for warming and cooling sense. Healy and his colleagues
Fig. 4.14 NGF immunoreactive fibres in a human painful neuroma, x 40 (Courtesy of Professor Praveen Anand).
Surgical Disorders of the Peripheral Nerves
found that sweating did recover, and caution against using this as an indication of nerve recovery (Figs. 4.15–4.17). The rate of regeneration. The rate of peripheral outgrowth of regenerating fibres is reckoned at about 1 mm a day in the human adult: that is about the rate of slow axoplasmic transport. The rate is substantially faster in children, as indeed it is in young experimental animals. It is almost certainly faster after primary than after secondary repair (Birch and Raji 1991) and is faster proximally than distally. We suggest a rate of 2 mm a day after suture of clean wounds of nerves at clavicular level and in the proximal part of the lower limb. Indeed, we have seen a rate of 3 mm a day after suture, or even after graft of ruptures performed within 24 h of injury, a rate similar to that recorded by Holmes and Young (1942) in their experiments on rabbit nerves. Why not? Such wounds are nearer to the nerve cell than are more distal injuries. Case report. A 24 year old soldier sustained a missile injury to the upper thigh which caused an extensive comminuted fracture of the sub trochanteric region of the femur. The fracture was stabilised and the sciatic defect, which measured more than 5 cm, was grafted 5 days after injury by Wing Commander Ian Sargent and Mr Garth Titley, Birmingham. By 180 days there were strong Tinel signs for both divisions
Fig. 4.15 Useless regeneration. Failure of recovery in a median nerve sutured 3 years previously. Proximal to the suture line. Clusters of myelinated axons extensive endoneurial collagenisation x 4300.
Regeneration and Recovery
Fig. 4.16 Useless regeneration. Failure of recovery in a median nerve sutured 3 years previously distal to the suture line. There was no recovery of function in spite of the presence of regenerating well myelinated axons x 4300.
of the sciatic nerve in the upper third of the leg at a distance of 44 cm from the proximal suture line. Flexion of the heel was powerful and there was early recovery into tibialis anterior suggesting that motor axons had traversed at least 35 cm (2 mm a day) of the length of the limb. The patient had no pain. This example of strong regeneration after urgent repair is by no means unusual, indeed we consider it the rule.
4.2 The Repair of Large Gaps Direct apposition may be impossible because of the fixed retraction of stumps in delayed cases or because so much of the nerve has been destroyed by the injury. Methods of bridging this gap have been extensively studied for more than 100 years. They include: 1. Autogenous grafts of cutaneous nerves 2. Autogenous grafts of main nerves. These may be vascularised as a pedicle or as a free graft 3. Homografts or allografts 4. Non neural material
123
Fig. 4.17 Useless regeneration. Failure of recovery in a median nerve sutured 3 years previously. Biopsy of the skin of the index finger. Schwann cell processes and some unmyelinated axons x 18000.
Sanders (1942) recognized the essential role of Schwann cells when he classed operations for the repair of large gaps in the peripheral nerves into two groups: those which provide orientated live Schwann cell columns, down which large numbers of fibres can grow and become mature,........ and those which provide some form of artificial scaffolding down which....... the new fibres and Schwann cells can grow in a regular manner.
The Schwann cells promote, sustain and guide regenerating axons. Regeneration in the peripheral nervous system is possible only because of the interaction between the Schwann cell and the axon (Hall 2005a,Hall 2005b). Some of these methods are now considered. The technical aspects are set out in Chap. 7 and results in Chaps. 8–10. Regeneration through grafts. Tinel (1917) wrote: “when the distance between the segments of the nerve trunk is too great to permit direct suture the only legitimate operation is nerve grafting as recommended by J and A Déjerine and Mouzon.” He further added “nothing but nerve tissue can serve as a conductor for regenerating axis cylinders” and roundly castigated “mischievous” operations such as lateral implantation, transplantation of a motor nerve into a sensory one and isolation of nerve by foreign body. Only the
124
interposition of a “muscular- or better still, a fatty-layer” between a nerve and callus or bony projection is permitted. It seems difficult to add to this. Sanders (1942) thought that the earliest adequately studied case of repair by fresh autograft was that performed by Dean in 1896 and reported by Sherren (1906), and he commended the series of Foerster (1929), Bunnell and Boyes (1939) and Joyce (1919) for their thoroughness. Sanders (1942) confirmed and extended observations made by workers in the first decades of the 20th century in concluding that: “autografts remain alive and myelin fragmentation and Schwann multiplication go on in them much as in a normal peripheral stump.” Human cutaneous nerve grafts could be fully reinnervated, presenting an appearance barely distinguishable from normal (Seddon et al. 1942, Gutmann and Sanders 1942, Seddon 1954. The ulnar nerve could be used with success but nerves of larger caliber were much less successful; a grafted segment of the common peroneal nerve simply necrosed (Seddon 1947). These observations stimulated the development of the vascularised ulnar nerve as a pedicle graft (Strange 1947) and a similar method for the common peroneal nerve MacCarty (1951). Millesi (1973) provides further evidence about the ill effects of tension and also about the necessity of accurate topographical disposition of the grafts. It is often said that grafting is the “gold standard” in the repair of nerves with defects but adherence to the gold standard led to severe financial difficulties for more than one nation and a considerable set back in Churchill’s career whilst he was Chancellor of the Exchequer in the 1920s. We have seen a high rate of failure amongst the more than 4,000 grafts of trunk nerves carried out since 1965. We are not alone in recording a success rate for repair of the more severe injuries to nerves which is little better than that set out in the Medical Research Council Special Report Series No. 282 (1954). The chief advantage of grafting is that it overcomes tension, which, as Highet and Sanders (1943) demonstrated, leads to a slow traction injury and ischaemia of the nerve after removing protecting splints. Seddon (1975b) commented that: “suture under tension… is not practiced by any knowledgeable surgeon today” and he referred to the normal elasticity of nerve which is not a negligible factor. Wide retraction of the stumps in closed traction lesion is usual and it is a simple matter to draw the nerves back to their normal position during operation performed within the first 1 or 2 days. It is always easy to be wise after the event but it now seems that suture would have been the better way forward in some of our cases particularly for nerves injured at the elbow, the knee and the wrist. In an earlier study of repair of the median and ulnar nerves at the wrist and forearm (Birch and Raji 1991) there was no detectable difference in the results between the 25 sutures and the 35 grafts. The gap after resection of the neuroma was measured with the elbow extended and the wrist at 20° dorsi flexion. The average gap was 2.5 cm in the sutures and 5.5 cm in the grafts. Trumble (1991) provides sound advice about reasonable methods of diminishing
Surgical Disorders of the Peripheral Nerves
the gap between nerve stumps by careful mobilisation and by positioning of adjacent joints and he emphasizes that the gap is always longer in delayed repairs. Wherever possible the stumps of transected or ruptured nerves should be approximated at the first operation in cases where circumstances prevent primary repair. The shortage of available donor graft means that the sciatic nerve should be sutured wherever possible, even though the patient must endure a hip spica with gradual release for some weeks. Clinicians must never forget that nerve grafts obtain their blood supply from the bed in which they lie. The admonitions of Tinel and Delorme were restated by Seddon (1947): “since the wounds of nerves that call for repair by grafting are usually extensive the need for the replacement of skin scars by healthy tissue arises with corresponding frequency. Nothing less than a full thickness flap or tube pedicle graft will suffice since it is important the graft should lie, so far as is possible, in healthy well vascularised tissue.” Leaving a nerve graft within scarred muscle or underneath split skin graft just will not do. The limitations of conventional grafting. 1. There is only a limited amount of cutaneous nerve available. In an adult with a complete lesion of the brachial plexus it is possible to collect about 180 cm of nerve by using both sural nerves with the cutaneous nerves of sensation from the injured limb. 2. The added defect imposed upon the patient may be too severe. The lateral cutaneous nerve of forearm and the superficial radial nerve provide significant innervation to the skin of the thumb, the thenar eminence and the palm of the hand. We use them only when the parent nerve is irreparably damaged. The supraclavicular nerves provide important sensation from the skin above the clavicle, over the shoulder and the upper part of the chest. The sural nerve innervates the heel and it should not be used for repair of low lesions of the tibial nerve. Pain is a common complication after deliberate wounding of the terminal branches of the nerves of cutaneous sensation and it is advisable always to section the donor nerve proximally, deep to the deep fascia. We agree with Seddon (1975c) in favoring the medial cutaneous nerve of forearm over other cutaneous nerves. 3. The architecture of a cutaneous nerve bears little resemblance to that of a main nerve trunk. The fifth cervical nerve contains between four and eight bundles. The largest of these requires one strand of cutaneous nerve which may contain between 8 and 20 such bundles. Dyck et al. (2005) examined the cross sectional area occupied by myelinated nerve fibres (MNF) in the roots of the spinal nerves, in the spinal nerves and more peripheral nerves. Myelinated fibres accounted for between 46% and 70% of the cross sectional areas of the ventral root of L5; the area fell to between 23.8% and 34.5% in the proximal sural nerve. The area occupied by MNF in the dorsal roots of
Regeneration and Recovery
L5 lay between 35.8% and 50.1%. In equally painstaking studies Dyck et al. (2005) showed that the median diameter of MNF in the ventral root of L5 was 12 mm. It was about 5 mm in the tibial nerve and just under 4 mm in the sural nerve. The median diameters in the fibres of the dorsal root resembled those found in the peripheral nerves. These findings suggest that there are serious defects in cutaneous nerve grafts particularly when they are used for repair of the ventral roots of the brachial plexus or the most proximal parts of the spinal nerves because of the disproportion in the volumes occupied by conducting tissue and the calibre of nerve fibres. Seddon (1975c) had this to say about the possible effects of fibre size: “it might therefore be thought that motor recovery through the cutaneous nerve graft would be appreciably inferior to the sensory recovery. The evidence presented by our series of cases shows that this possible objection has little substance; and Simpson and Young (1945) have shown that large fibres in the proximal stump have no difficulty in inflating rather smaller distal Schwann tubes, and so attaining a diameter sufficient for their effective function.” However Young (1942) recognized the limitations of experimentation on small mammals and in particular, the robust regeneration characteristic of the rat and the rabbit. The problem resurfaces when regenerating axons having traversed the graft reach the distal trunk of the nerve for here, particularly in late or neglected cases, the Schwann tubes are embedded in dense collagen. 4. The recognition that Schwann cells may be specific either to motor or to sensory axons casts a shadow over the use of cutaneous nerves for the repair of main nerves (Hall 2009, Höke et al. 2006). Nichols et al. (2004) and Lago et al. (2007) showed that the regeneration of motor axons is better promoted by a graft of a “motor” nerve whilst the regeneration of sensory axons is better through a graft from a cutaneous nerve. Whilst there is no such thing as a purely motor or sensory nerve within the peripheral nervous system these findings may account for the relative success of vascularized main nerve grafts. We reproduce a section through such a graft showing numerous myelinated and unmyelinated fibres but, perhaps disappointingly, we show sections through a failed vascularised sural nerve graft in which the vascular anastomosis remained patent and yet regeneration failed (Figs. 4.18–4.22). 5. The nerve pedicle operations remain extremely useful in the most severe cases especially so in otherwise irreparable injury of the sciatic nerve. It is strange that these innovative techniques seem to have been forgotten by so many clinicians. 6. The most elaborate method of nerve repair cannot overcome the baleful effect of delay. Allografts. The advantages of the autogenous cable graft are now so well known that we tend to forget that earlier last
125
Fig. 4.18 Regeneration through a vascularised ulnar nerve graft (fifth cervical nerve to lateral cord). Although the biceps muscle recovered well, the appearance of the hand led the patient to demand amputation. Note the numerous myelinated and unmyelinated fibres without excess of collagenisation x 2142.
century surgeons obtained what we would now consider remarkably good results with grafts which would now not be used. Thus, Mayo-Robson (1917) reported a good result after repair of a lesion of the median nerve with an allograft of median nerve, and another good result after repair of another lesion of the median nerve with a length of the spinal cord of a rabbit. Barton (2005) has provided us with details of the use of calf’s nerve for repair of a war wound to the right (dominant) median nerve, in 1917. The operation was done at Baschurch Military Hospital (in Shropshire). The patient lived to the age of 90 years and recovery was good enough for him to act as organist to various churches and chapels. The method was thoroughly and extensively reviewed by Sanders (1942), and by Sunderland (1978). Seddon (1975d) pointed out that the immune reaction may be avoided either by reducing the antigenicity of the graft or by suppression of the response in the host: “it must be remembered that repair of a nerve is not a life saving operation.” A number of methods were used including treating the graft with alcohol or with other agents, freeze drying and irradiation. The results were generally very poor but there was the occasional success. The introduction of newer methods of immuno suppression
126
Fig. 4.19 Failure of regeneration through a vascularised sural nerve graft used to repair a long defect in the median nerve. At re-exploration two years after operation the vascular anastomosis was found to be patent. (Electronmicroscopic study by Michael Kayser) Just distal to the proximal suture line. Mini fascicles (arrows) of myelinated and unmyelinated nerve fibres x 2108.
allowed MacKinnon (1996) to give a qualified recommendation: “In the carefully selected patient with an otherwise irreparable nerve injury, consideration for nerve transplantation should be given.” Hand transplantation provides one dramatic example of allograft. Schuind et al. (2006) described the first Belgian patient at 37 months after operation, showing very impressive recovery for motor, sensory and sympathetic functions. Larsen et al. (2007) sound a warning note in reporting Epstein Barr virus infection complicating transplantation of a nerve allograft from a living donor.
4.2.1 Other, Non Neural, Material for Grafts: Entubation Surgeons have for many years sought a source of material for graft to supplement or to replace the rather meagre stock provided by dispensable cutaneous nerves.
Surgical Disorders of the Peripheral Nerves
Fig. 4.20 Just distal to the proximal suture line. A mini fascicle (small arrows) of myelinated and unmyelinated nerve fibres lie just outside the perineurium of one of the vascularised strands (large arrows) x 6270.
The freeze thawed muscle graft (FTMG) introduced by Glasby et al. 1986 is a method which has been used extensively in laboratory and clinical studies. Hall (1997) showed that axonal regrowth through short acellular muscle grafts requires a sustained supply of Schwann cells, perineurial cells, and fibroblasts. Schwann cells may migrate from the stumps of the nerve on either side of the acellular graft or which may be “seeded” in that graft (Calder and Green 1996). Thomas et al. (1994) defined a particular role for FTMG in the repair of damaged cutaneous nerves, in the treatment of the painful states arising from such damage. Although in such cases an adequate supply of nerve is available for grafting it seems perverse to inflict a lesion on a cutaneous nerve in a patient suffering from the results of damage to another. A paper of importance appeared from Pereira et al. (2008) who treated the leprotic hand and foot by replacing the damaged segment of the median or tibial nerve with FTMG. Most patients recovered protective sensation; their ulcers healed. Doubtless, a number of amputations were prevented. One
Regeneration and Recovery
Fig. 4.21 The middle part of the graft. Endoneurial fibrosis; myelinated and unmyelinated axons; a suggestion of new perineurium formed from fibroblasts (arrows). x 1770.
intriguing finding was the improvement in the condition of the skin in one third of the contralateral feet. Entubation. In 1997 Lundborg and colleagues (Lundborg et al. 1997) published a significant paper describing a prospective randomized trial comparing silicone entubution and suture of median and ulnar nerves in the forearms of 18 patients. Recovery was studied most thoroughly and no significant difference in outcome between the two groups was noticed. They wrote “in one case, the tube was removed at 11 weeks. Upon opening the tube, it was found that the former empty space was now occupied by newly formed nerve tissue in direct continuity with the proximal and distal ends..... it was not possible to define the exact level of the previous nerve injury.” The proposed advantages following placing of nerve stumps within a silicone tube, so providing a “chamber” separate from the surrounding tissues, include: the local accumulation of neurotrophic factors; the longitudinal orientation of fibrin matrix within the tube; and the possibility for the regenerating axons to be better guided into distal Schwann tubes across the gap. The tube must not be too tight. Silicone entubation may cause constriction and fibrosis of a sutured nerve (Birch 1979).
127
Fig. 4.22 Failure of regeneration through a vascularised sural nerve graft used to repair a long defect in the median nerve. At re-exploration 2 years after operation the vascular anastomosis was found to be patent (Electronmicroscopic study by Michael Kayser). The distal portion of the graft. This shows one of the few surviving clusters of neural elements. Several myelinated and unmyelinated axons lie within the perineurium x 4408.
With the exception of some cases using the freeze-thawed muscle graft, it seems that the upper limit of a defect reparable by these methods is 3 cm. Repair of the long gap by means other than autogenous graft remains elusive. De Ruiter et al. (2009) conclude that there is, as yet, little evidence demonstrating the superiority of empty, hollow biodegradable nerve tubes over suture or autografting and that: “the repair of all sorts of nerve lesions may lead to unnecessary failures and, again, a discontinuation of interest in the concept of the nerve tube. The extensions of the applications, especially in a repair of larger mixed or motor nerves, should be carefully evaluated.” Hall (2005b) defines the ideal conduit which should have: “walls that are biologically bio compatible and resorbable, sufficiently robust to resist collapse, and yet not so rigid that they compress the structures which they surround, and a lumen whose contents…. facilitate orientated and sustained axonal elongation. It goes without saying that it must also be easy to handle and to suture. It also goes without saying that
128
such a device does not currently exist.” Some show promise: one such is the bioresorbable glass fibre technique investigated by Bunting et al. (2005) in nerves of the rat. Axonal regeneration was indistinguishable from that seen in autografts. Hall’s more recent essay (2009) reviews the use of such natural polymers as collagen and various polysaccharides and a wide range of synthetic polymers. One essential requirement of the biodegradable conduit is that the degradation products must not be cytotoxic. Manipulating the behaviour of the growth cone will determine the path of the regrowing axon “since wherever the growth cone goes, the rest of the attached axon must surely follow.” Hall (2009) goes on: “given that axons and Schwann cells comigrate, it is perhaps surprising that little or no attention has thus far been paid to manipulating Schwann cell outgrowth from the nerve stumps.” The work of Caddick et al. (2006) opens the possibility of encouraging the differentiation of mesenchymal stem cells along a Schwann cell lineage. Cells were removed from the marrow of long bones of rates and cultured. Those cells cultured with glial growth factor (GGF) increasingly expressed such Schwann cell markers as S-100, p75, and glial fibrillary acidic protein (GFAP). The cells were then co cultured with neurones from the dorsal root ganglion and those which expressed Schwann cell markers enhanced neurite outgrowth from the sensory neurones to the same level as that produced by Schwann cells. This work represents a potentially important advance because the culture of autologous Schwann cells in the adult human is such a lengthy process. The development of micro-machined sieve electrodes may prove to be extremely important. These are made from silicone (Wallman et al. 2001) or polyimide (Zhao et al. 1997, Navarro et al. 1998) and are inserted into conduits to create a neural interface which can be used to control a limb prosthesis. Is it possible that these developments might enable the reconnection of stumps of the spinal nerves within the spinal canal, a concept put forward by Bonney (1977).
4.3 Nerve Transfer Nerve transfer, also known as neurotisation, or nerve crossing, involves the passing of nerve fibres from a healthy nerve to the distal stump of an injured nerve and this principle can be applied in a number of ways. 1. End to side transfer by suture of the distal stump of the injured nerve onto or within the epineurium of the healthy donor. 2. Transection of a healthy donor nerve and transfer onto the distal stump of the injured nerve. 3. Transfer of one or more healthy bundles within an uninjured donor to the recipient. This important technique was studied and used by Oberlin et al. (1994) and it rests on
Surgical Disorders of the Peripheral Nerves
the functional segregation and topographical organisation of nerve fibres within the donor nerve trunk so that it is possible to take one bundle from the ulnar nerve to reinnervate the nerve to biceps without inflicting any significant loss of function within the hand. The method has been extended to the transfer of intact bundles from C7 or C8 to the suprascapular nerve or to the avulsed ventral root of an adjacent spinal nerve. 4. Transfer of the proximal stump of a divided nerve onto the distal stump of another divided nerve. 5. Direct muscular neurotisation. Sometimes a nerve is avulsed from the muscle. The musculocutaneous and the circumflex nerves are most commonly affected. Sorbie and Porter (1969) took the nerve to the flexor carpi ulnaris muscle in dogs and implanted it into flexor carpi radialis. Muscle volume was restored and the twitch strength of the reinnervated muscle reached 50% of normal by 40 weeks. Brunelli and Monini (1985) developed a concept of implantation of a working nerve directly into the paralyzed muscle, calling this method “direct muscular neurotization.” Narakas (1988) described considerable experience with the method. The extent of afferent reinnervation of the muscle offers a fruitful field of study. End-to-side repair. Insertion of the stumps of damaged nerves into adjacent undamaged nerves is amongst the earliest of methods tested for bridging large defects. Sanders (1942) pointed out that the operation of “double lateral implantation” could be performed in one of two ways. The prepared stumps might be implanted into longitudinal slits between the bundles of the host nerve; or, some bundles within the host nerve could be cut at two levels and the prepared stumps sutured to these. In the first operation, axons from the implanted central stump might grow down through the epineurium: “an environment which contains no Schwann cells.” Kilvington (1907) investigated the second method, where bundles of the undamaged nerve were sectioned to act as a bridging graft and showed that there was histological evidence of reinnervation of the distal part of the damaged nerve. Ballance and Ballance (1903) used the technique in repair of the facial nerve. Kettle (2003) provides an extensive review and a careful analysis of the method using nerves in the sheep as a model. She demonstrated that end-to-side nerve repair “did support nerve regeneration” but her findings “do not support the view that this technique is associated with a functional outcome.” Carlstedt (2007a) sets out important objections to the technique. The internal environment of an uninjured nerve is not conducive to new axonal growth. It is only after injury to the donor nerve, by opening the perineurium that changes are induced in the Schwann cell phenotype more favorable to axonal growth. Regeneration is now end to end. One carefully studied clinical example comes from Irwin et al. (2006) whose patient sustained severe injuries to both legs. A below knee amputation proved necessary on the
Regeneration and Recovery
right side but the undamaged plantar skin of the right foot was used to successfully resurface the left foot in which the plantar skin had been avulsed. The right tibial nerve was transferred to the intact left tibial nerve by an end-to-side method without incision of the epineurium. Quantitative sensory testing (QST) at 9 months revealed that the patient could localize pressure but was not able to feel cotton wool or pinprick over the sole. She could perceive a graded monofilament of 7.5 g on the sole (normal, less than 0.08 g). Vibration sense was present but the threshold was elevated. She could not detect warm or cool sensations but sweating was close to normal. By 16 months there was clear improvement in the thresholds to vibration and light touch. Vibration thresholds now lay within normal limits and the monofilament threshold had improved to 1.66 g. She could not recognize cotton wool and a pin prick was felt as a touch. There was no change in the thermal threshold. She recovered some large sensory fibre function (vibration sense and touch) and autonomic function; there was no evidence of recovery of small sensory fibre function. Nerve transfer. When, as happens with serious injuries of the brachial plexus, nerves are so badly damaged that no repair is possible and no recovery can be expected, the question of “neurotizing” or reinnervating the distal stump with the proximal stump of an intact nerve or nerves is raised. Warren Low, working at St Mary’s Hospital, Paddington with Wilfred Harris, was probably the first to suggest this expedient and to put the suggestion into practice (Harris and Low 1903). The validity of the method was, however, open to question. The matter was raised again by Tuttle (1913), who reported on the use of the spinal accessory nerve as a donor. Transfer of the hypoglossal nerve to the facial nerve became a well recognized remedy for facial paralysis in the early years of the last century (Ballance and Ballance 1903) before it was superceded by grafting (Ballance and Duel 1932, Bunnell and Boyes 1939, Collier 1940). Probably the first recorded success from nerve transfers in the brachial plexus was that of Seddon (1975c), who with Yeoman transferred intercostal nerves into the musculocutaneous nerve. Seddon acknowledged an earlier contribution: “in a case of traumatic tetraplegia extending up to the fifth cervical segment Benassy and Robart (1966) helped the patient very considerably by severing an intact musculocutaneous nerve and uniting it to the neighbouring totally paralysed median.” Tsuyama and Hara (1973) adopted intercostal nerve transfer for patients with complete avulsion injury in 1965 and Nagano et al. (1989) described the results in 179 patients More than 80% of their patients regained functional flexion of the elbow. Kotani et al. (1972) and Allieu (1989) described the use of the distal part of the spinal accessory nerve in re-innervation of the upper limb preserving rami to the upper fibres of trapezius. A great deal of detailed and important work towards the study of potential nerves
129
available for transfer is set out in the volumes published by the Groupe d’Étude de la Main which were brought together by Alnot and Narakas (1989, 1996). Nerve transfer has developed as an essential part of the treatment of severe intradural lesions of the brachial plexus, and until the place of intradural re-attachment or repair is established, it is likely to remain so. It is also valuable in some peripheral lesions. Gore (1978) restored powerful wrist extension by transferring a nerve to the medial head of triceps to the radial nerve, 15 months after the injury. Gousheh (2002, personal communication) transfers the nerve to the lateral head of gastrocnemius to the deep division of the common peroneal nerve in severe lesions. The superficial division, which is usually more amenable to repair, can be grafted at the same time. Carlstedt (2007b) recommends the technique “in cases where intra spinal repair of a lumbosacral plexus lesion is impossible, eg. when there is no proximal, ventral root stump to use for reconstruction, it is possible to perform nerve transfers.” One example is described in the important paper from Lang et al. (2004). The patient, a 5 year old girl, presented with bilateral, asymmetric injuries to the lumbo-sacral plexus. Paralysis of the abductors of the hip on one side was treated by placing a graft between the superior gluteal nerve and two bundles within the femoral nerve. Paralysis of the extensor muscles of the knee on the other side was treated by transfer of intercostal nerves 10 and 11 to the femoral nerve. 18 months later the child was able to walk independently. Brunelli and Brunelli (1999) enabled a paraplegic patient to walk by transfer of the ulnar nerve to the superior gluteal and femoral nerves. The deficit in the hand was remedied by musculo-tendinous transfer. Contra-lateral transfer. A more daring idea comes from Gu and his colleagues in Shanghai. The contra-lateral seventh cervical nerve is used to re-innervate the upper limb after a complete lesion of the brachial plexus. A vascularized ulnar nerve is used as an interposed graft. The first operations were performed in 1986 and Gu and his colleagues described promising results in ten patients with adequate follow up in 1992 (Gu et al. 1992). The method is supported by a considerable amount of experimental work (Chen and Gu 1994). One of us was able to examine some of the patients operated by Professor Gu and his colleagues in Shanghai, and some details of one of these follows. Case report. The patient, a male aged 27, sustained a complete preganglionic lesion of the right brachial plexus in a motor cycle accident on the 18th June 1993. At 3 months accessory to suprascapular and phrenic to muculocutaneous transfers were performed. At four and a half months intercostal nerves were transferred to the thoracodorsal nerve and the contralateral C7 was sutured to the ulnar nerve from the damaged limb. Eight months later the distal stump of this vascularised graft was sutured to the median nerve. The patient was examined 18 months after re-innervation of the
130
median nerve. Power in the long digital and wrist flexors was MRC grade 4. There was some activity in abductor pollicis brevis and the patient was able accurately to localize touch to the thumb, the index and middle fingers. This patient had regained true hand function which could have come only from contralateral C7. Those of us examining him were unable to find any significant deficit of power or sensation in the donor limb. Closely scrutinized series have been reported by Waikakul et al. (1999) and Songcharoen et al. (2001). Our own experience of the transfer of contra-lateral spinal nerves is confined to two cases of bilateral and complete lesions of the brachial plexus. All available post ganglionic ruptures were used to reinnervate the less severely damaged upper limb. (Chap. 9). Limitations of nerve transfer: 1. There are far fewer MNF in the donor nerves than in the main trunks. The number of MNF in the spinal accessory nerve at the base of the posterior triangle is about 1,500; that number in C5, usually the smallest of the spinal nerves forming the brachial plexus, is at least 25,000. 2. The deficit imposed on the patient must not be too severe. No nerve of vital function should be used for the sake of regaining a non vital function. Phrenic nerve palsy at birth is life threatening. Some adults experience serious ventilatory impairment following injury to the phrenic nerve complicating lesions of the brachial plexus. Blaauw et al. (2006) condemn transfer of the hypoglossal nerve because of the high morbidity and disturbance of speech. We have seen one adult patient where both hypoglossal nerves were used in whom there was very serious disturbance of speech and swallowing for as long as 6 months after operation.
4.3.1 Recovery of Cutaneous Sensation after Nerve Transfer The infant. The recovery of sensation within the hand in children after repair of the brachial plexus in infancy is remarkably good. Anand and Birch (2002) studied 24 patients with severe birth lesions of the brachial plexus, in 20 of whom the plexus had been repaired. The recovery of sensation was far better than that of skeletal muscle and sympathetic function. Sensation was normal in all dermatomes for at least one modality in 16 examined hands and there was accurate localisation in the dermatomes of avulsed spinal nerves which had been reinnervated by intercostal nerves transferred from remote spinal segments. These findings are described in more detail in Chap. 10: there too are described the results of sensory recovery after nerve transfer in children and young adolescents. The reason for this difference has been sought in the better adaptation of the child’s central receptor mechanisms. The poor recovery of muscle function in these
Surgical Disorders of the Peripheral Nerves
children may reflect the greater vulnerability of the immature nervous system to avulsion or proximal axonotomy. The adult. Özkan et al. (2001) treated 20 cases of irreparable lesions of ulnar or median nerves by transfer of the digital nerve from the inner to the outer side of the appropriate digit. Extensive sensory retraining was used in all patients and accurate localisation was restored in 18. Battiston and Blanzetta(1999) reported useful recovery into six of seven cases of high ulnar lesion by transfer of the distal part of the anterior interosseous nerve and the palmar cutaneous branch of the median nerve to the ulnar nerve at the wrist. We have transferred the dorsal branch of the ulnar nerve or the medial cutaneous nerve of forearm to the median nerve with modest success. Whilst transfer of the intercostal nerves to the lateral or medial cords often improves the trophic state of the skin of the hand in adults with severe injuries to the brachial plexus it is usual to find that stimulation of the reinnervated skin is referred to the chest wall. Appropriate referral and accurate localisation has been seen in only nine cases and these were patients with substantial muscle recovery through the transfer. Htut et al. (2006) studied 76 patients with severe injuries to the brachial plexus repaired by conventional graft, nerve transfer or reimplantation. Stimuli were appreciated in the reinnervated skin of the hand in only a few patients. More than one half of patients experienced “wrong way” sensations such as pins and needles in the affected arm or hand while shaving, touching or tapping the ipsilateral cheek or lip. In some patients these sensations were evoked by warming or cooling or by transcutaneous nerve stimulation. Two patients experienced sensations in the affected arm or hand when the ipsilateral leg was stimulated. These referred sensations could be unpleasant, with a painful or gripping quality and in one patient coughing produced a squeezing or gripping feeling at the wrist. This “wrong way” sensation generally occurred within 2 months of the operation, some months before the development of “right way” sensation which followed regeneration from the donor nerve into its new territory. In the “right way” sensation stimulation of the hand evoked feelings in the chest wall, the territory of the donor nerve.
4.3.2 Recovery of the Deep Afferent Pathways after Nerve Transfer The capacity of the child’s central nervous system to adapt to a new peripheral situation is illustrated by the case of a 13 year old boy who suffered avulsion of C5, C6 and C7 in a road traffic accident (Fig. 4.23). It seems likely that there was some reinnervation of the deep afferent pathway from the muscle spindles and the tendon sensory organs so that an appropriate pattern of inhibition and facilitation was restored within the spinal cord.
Regeneration and Recovery
Fig. 4.23 A 13 year old boy. Right sided lesion. Avulsion of C5, C6 and C7.Repair at 2 months: accessory to suprascapular transfer, and one bundle of the ulnar nerve to the nerve to biceps and the nerve to the medial head of triceps; the medial cutaneous nerve of the forearm was transferred to the lateral root of the median nerve. Results at 18 months: full range of lateral rotation; abduction to 60°; power of elbow flexion MRC grade 4; and power of elbow extension, MRC Grade 3+. There was a full range of active flexion and extension without obvious cocontraction.
There is considerable adaptation of the central receptor and effector mechanisms in adults after nerve transfer. The association between activity in biceps with clenching of the fist after ulnar to biceps transfer is useful in the early stages of rehabilitation. Independent flexion of the elbow without associated activity in the flexor muscles of the forearm is usual by 24 months after operation. Kanamura et al. (1993) studied 14 patients after intercostal to musculocutaneous nerve transfer. Percussing the biceps tendon induced sensory evoked potentials in four cases and appropriate reflex activity was demonstrated by electromyography. These workers concluded that: “muscle afferent fibres in the intercostal nerves reinnervated mechano-sensors in the biceps, and these evoked both cortical sensory activity and muscle activity which represents the classical stretch reflex” Sai et al. (1996) confirmed these findings in a study of 15 patients. Recovery of the elbow flexion reflex was confirmed by electromyography in nine cases and in five “tapping at various frequencies induced gradual augmentation of the integrated electromyogram of reinnervated biceps,” implying reinnervation of the muscle spindles by 1a and 2 fibres. Mano et al. (1995) demonstrated remapping within the cortical motor cortex by magnetic stimulation in 44 patients. At intervals ranging from 4 to 33 months after the nerve transfer a change from the intercostal area to that of the arm was found, and this was associated with a clear separation between ventilatory effort and that of elbow flexion. Further observations about the changes within the central nervous system following intercostal nerve transfer were made by Malessy et al. (1998) and by Malessy et al. (2003) who used functional magnetic resonance imaging (FMRI) and cortical magnetic stimulation.
131
Voluntary elbow flexion became independent of breathing. In those patients who regained a powerful biceps the distribution of cortical activity did not materially differ between the injured and the uninjured arms. Regeneration of end-organs. There is, evidently, a limit to the period during which an end-organ can be denervated and yet recover after reinnervation. All end-organs have been studied: the earliest were perhaps the motor end-plates. Bowden and Gutmann (1944) did an extensive study of the response of the motor end-plates to denervation, finding that after 3 months it became increasingly difficult to identify the organs. Denervation for up to 3 years produced no degeneration nor disruption of the muscle fibres themselves. However, “shrinkage and increase of connective tissue may be too advanced to allow recovery after reinnervation.” After 3 years, disruptive changes occurred in the muscle fibres. Gutmann and Young (1944) investigated the reinnervation of muscle in rabbits and found that when muscles were kept denervated for increasing periods the proportion of old end-plates which became reinnervated was progressively reduced: “Most of the nerve fibres escape and run along between the muscle fibres, ultimately making contact with the sarcoplasm and forming new plates.” They had the opportunity of examining the extensor carpi radialis muscle of a patient whose radial nerve was repaired 5 years earlier after it had been interrupted. There was no recovery of motor power. Examination of a piece of this muscle showed that “although abundant nerve fibres were present they had failed to make connexions with the muchatrophied muscle fibres.” The regeneration of muscle spindles has been extensively studied by, amongst others, Barker et al. (1986), Barker et al. (1990), and Banks (2005). After crush injury (axonotmesis) regeneration and recovery of function is good although there is a reduction in the number of primary endings in the terminal bands coiled around nuclear bag fibres. This is related to the duration of denervation. Regeneration after repair of the nerve is much more variable. A muscle spindle may become reinnervated by afferents normally destined for the tendon organ although Barker et al. (1990) indicated that delaying repair for up to 8 weeks “did not give rise to any significant detrimental effect on such reinnervation.” The behaviour of the Golgi tendon organ has been investigated by Scott (1995, 2005). It appears that the tendon organ is relatively resistant to denervation atrophy: “after crush injury there is usually successful restoration of the 1b afferent ending, although it is reduced in extent” (Scott 2005). The pattern of recovery is worse following nerve repair. Many tendon organs remain uninnervated and the regenerated endings are frequently abnormal in appearance. Scott identified three deleterious factors: (1) the effects of denervation on the afferent axon; (2) the consequences of reinnervation by inappropriate axons, and (3) the effects caused by the reorganisation of the motor units after repair causing
132
s ignificant alterations in the mechanical input to an individual tendon organ. Changes in the number and distribution of muscle afferent neurons in the dorsal root ganglion of the rat following transection and repair of the sciatic nerve were investigated by Myles et al. (1992). The cell bodies were labeled by injecting horse radish peroxidase into the extensor digitorum longus 300 days after the first operation. The number of labeled cells was diminished, the cell size was reduced and the somatotopic organisation of neurones within the dorsal root ganglion was altered. Michael Glasby and his colleagues investigated recovery in the deep afferent pathway in the new born and 1 year old sheep (Fullerton et al. 2001). In the first experiment the sixth cervical nerve was ruptured and immediately grafted and in the second experiment the ventral root of C6 was avulsed from the spinal cord and immediately repaired by autologous coaxial freeze thawed skeletal muscle graft. The nerves were examined 1 year later, by electrophysiological and morphometric methods. Regeneration of the largest, fastest conducting nerve fibres was defective in both the new born and in the 1 year old sheep. The maximum conduction velocity never reached normal levels and the diameter of the fibres, the axons and the thickness of the myelin sheath were all reduced. These workers found: “a selective failure of regeneration of the largest diameter fibres..... it seems more likely that the failure of recovery of fine movements is due to the fact that the proprioceptive pathway involving 1a, 1b and group 2 fibres on the afferent side and the process of a and g coactivation on the efferent side, is lost.” The changes are not confined to the end-organs of the motor apparatus: the cutaneous sensory receptors similarly undergo a slow degenerative change after denervation. After 3 years they may in fact disappear. Reinnervation tends to reverse these changes, though the longer the period of denervation has been, the less complete will be the regeneration. It is not too surprising that correlation of histological evidence of regeneration of cutaneous nerves with clinical evidence of recovery is often imperfect. Thus, Jabaley et al. (1976) attempted to correlate sensory function with evidence of cutaneous reinnervation after section and repair of the median nerve. They found that there was no such correlation. Indeed, in one case, after division and repair of the median and ulnar nerves, two point discrimination was seriously defective in the thumb; yet, blindfold identification of objects was regularly correct and the pickup test of Moberg (1958) was performed in 1950s. Histological examination of the skin of the index and little fingers showed only an occasional nerve fibre and a few Meissner corpuscles. In contrast, a woman whose median nerve had been repaired by fascicular suture regained plentiful cutaneous innervation, but was unable to pick up objects because she could not feel them. On the other hand, Dellon and Munger (1983), using the partially denervated finger tips of three patients to correlate observations of sensibility with the presence of reinnervated
Surgical Disorders of the Peripheral Nerves
sensory corpuscles as seen by light and electron microscopy, found that in all cases the reinnervated receptors identified were appropriate to provide the neurophysiological basis for the observed results of quality of sensibility.
4.3.3 Complications of Regeneration Recovery of function after repair of nerves is often marred by cocontraction and pain, reflecting imperfect regeneration of both the afferent and the efferent pathways. Cocontraction. The normal movement of joints is brought about by the smoothly coordinated and controlled activity in muscles and precise and delicate regulation by inhibition and facilitation of the motor neurones. Beevor (1904) showed that the sensory motor cortex controls movements rather than individual muscles. The conversion of an antagonist to an agonist is the basis of musculotendinous transfer and it is common to see patients actively extending the knee or the ankle and toes as soon as the post operative splint is removed after hamstring to quadriceps transfer or anterior transfer of tibialis posterior. This control is damaged in lesions in continuity, especially of the brachial plexus; the joints of the shoulder girdle are especially vulnerable. It is possible that some of this is caused by the reinnervation of muscle units in antagonists by the same parent motor neuron but this explanation cannot account for the initial association between ventilatory and elbow flexion effort after intercostal transfer nor for the later separation of these functions. Muscles reinnervated by nerve repair or through a serious lesion in continuity usually fail to convert after muscle transfer irrespective of their power. Perhaps the defective reinnervation of the deep afferent pathways from the muscle spindles and the tendon receptors blinds muscles, which are, after all, sensory as well as effector organs. Cocontraction is common after reimplantation of the avulsed ventral root. Hallin et al. (1999) showed, in experiments on primates, that the motor neurones reinnervating the biceps were both abundant and scattered about in the anterior horn but that: “in double labeling experiments, simultaneous links to both antagonistic and agonistic muscles from the same neurone could not be demonstrated.” Carlstedt (2007c) thinks that: “the structural basis for deficient performance such as synkinesis or cocontraction was not due to the same neurones innervating antagonistic muscles, but rather was the effect of an unspecific reinnervation by inappropriate neurones under inappropriate supraspinal control.” Another possible mechanism is described by Camp and Allen (2008) who reported cocontraction between the trapezius and the swallowing muscles in a 27 year old woman who had sustained a basal skull fracture (Fig. 4.24). There was a severe but incomplete lesion of the spinal accessory nerve at or close to the jugular foramen and involuntary activity in
Regeneration and Recovery
133
Fig. 4.25 A 25 year old man. Penetrating missile wound to the shoulder, the exit wound close to the upper medial border of the scapula damaging the spinal accessory nerve at the level of the spine of the scapula. There was cocontraction between trapezius, levator scapulae and the rhomboid muscles.
Fig. 4.24 Cocontraction between platysma, trapezius, and other neck muscles (Courtesy of Camp and Allen).
sternocleidomastoid and trapezius with swallowing. There was abnormal synchronous activity between fibres within the spinal accessory the glossopharyngeal and vagus nerves. Camp and Allen suggest that there was aberrant reinnervation by motor fibres from the cranial root of the accessory nerve, or by motor fibres from the vagus, or glosso pharyngeal nerves. Other patients have been seen in whom a partial lesion of the peripheral nerve incited abnormal activity in motor fibres within adjacent peripheral nerves (Fig. 4.25). It is possible that disturbance at a spinal or supraspinal level underlies the abnormal muscle patterning seen in some cases of recurrent or habitual dislocation of the glenohumeral joint. Regeneration and Pain. Pain is, of course, a common feature of nerve injury particularly when the agent remains active but nerve regeneration itself may be associated with pain, particularly in adults. Anand (pers comm, 2009) has pointed out that there is a lifelong need for regeneration in the autonomic nerves to the smooth muscle of the skin and viscera and also in the somatic nerves of the skin. In traumatic
neuropathy the event is usually an isolated episode involving one nerve. The neuropathy induced by chemotherapy is cyclical, so that there are repeated sequences of axonopathy followed by regeneration. In diabetic neuropathy pain is usually associated with regeneration and in this disease the process is a lifelong event (Figs. 4.26a–d and 4.27a–d). Many patients experience flitting, deep, cramping pain in muscle, months after repair of the brachial plexus or high lesions of the sciatic nerve and shortly before obvious recovery of muscle function. We consider this a reassuring symptom and often suggest to patients that this is perhaps the only “good” form of neuropathic pain. The mechanism is obscure but the close relation to return of muscle activity suggests some reintegration of the deep afferent pathway. Anand and his colleagues have carried out extensive investigations in human nerve tissue and these provide some understanding about the mechanisms underlying pain associated with regeneration. In one study neurones from the dorsal root ganglion, which had been avulsed from the spinal cord, were cultured with a variety of neurotrophins including nerve growth factor (NGF), glial derived nerve factor (GDNF) and neurotrophin 3 (NT3) (Anand et al. 2006). The neurotrophins induced an increase in cell size and an increase in the proportion of cells exhibiting the receptor TRPV1, characteristic of nociceptor fibres. TRPV1 is the receptor for capsaicin, a potent nociceptor, and is expressed in DRG neurones, in painful neuromas and in the nerves of hypersensitive skin. The increase in this receptor was most marked in the larger cell bodies. Neurotrophin production is increased after nerve injury, indeed it is essential for the survival of the cell body and the regenerating axon. This finding provides one possible explanation for the mechanical sensitivity, at times so severe as to merit the term mechanical allodynia, which so often disturbs sensory recovery (Fig. 4.28). Durrenberger et al. (2004) studied the production of prostanoids, which are potent chemical mediators of inflammation
134
Surgical Disorders of the Peripheral Nerves
Fig. 4.26 A 26 year old woman developed back pain radiating to the left foot and an ankle a sprain whilst training for a run. Over a few weeks she developed allodynia to touch, changes in colour and sweating and an abnormal foot posture. Clinical examination and quantitative sensory testing revealed a mechanical allodynia, elevated cold threshold and reduced heat pain threshold over the affected limb. Below knee amputation was performed. Examples of various nerve markers in the skin of the
calf close to the ankle. (a) Preserved sub-epithelial nerve fibres stained with the marker PGP9.5 (arrowed) x 40. (b) Unusually dense PGP9.5 fibres around blood vessels in the skin (arrowed) x 40. (c) Increased intra-epithelial transient receptor potential TRPV1 (heat receptor) fibres (arrowed) x 40. (d) A few sub-epithelial TRPM8 (cool receptor) fibres, (arrowed) x 40 (Courtesy of Professor Praveen Anand).
associated with nerve injury, in the human and the rat. Prostanoids sensitise nerve terminals. The enzymes responsible for prostanoid production are the cyclooxygenases (Cox-1 and Cox-2). The immunoreactivity for Cox 2 increased in cells resembling macrophages especially in the nerve proximal to the injury. The number of cells labeled for the enzyme reached a peak at between 4 or 6 weeks after injury. Anand et al. (2008) studied cannabinoid receptors in human nerves and neurones cultured from avulsed dorsal root ganglia. The cannabinoid receptor 2 (CB2) inhibits nociception and reduces pain, possibly by inhibiting inflammatory cell action or by inducing the release of the endogenous opiate b endorphine and it does this without inducing effects on the central nervous system which are associated with activation of a cannabinoid
receptor 1 (CB1). It was shown that agonists to CB2 block the action of capsaicin by impeding the activation of inward cation currents and preventing the increase in intracytoplasmic Ca++. Anand and her colleagues concluded that: “CB2 receptor agonists functionally inhibited nociceptive signaling in human primary sensory neurones.” Studies of the sodium channels Nav1.9 and Nav1.8 by Yiangou et al. (2000) provides some understanding of the phenomenon of the “irritable” healing nerve and of the painful quality sometimes seen with Tinel’s sign. The channel Nav1.8 accumulated at the site of the injury in the proximal part of the nerve. One form of the channel Nav1.9 was not expressed at all in the infant nerves. Whilst much more needs to be done to explain the mechanisms underlying neuropathic pain and also the pain
Regeneration and Recovery
135
Fig. 4.27 Further examples of skin staining with various nerve markers in the patient illustrated in Fig. 4.26. (a) Sub-epithelial TRPM8 (cool and menthol receptor) fibres (arrowed) x 40. (b) PGP9.5 (structural marker) staining of injured nerve showing ovoid typical of Wallerian
degeneration. (c) Increased intra-epithelial (arrowed) and sub-epithelial TRPV1 (heat and capcaisin receptor) fibres x 40. (d) Basal keratinocytes (arrow head) and sub-epithelial NGF immunoreactive fibres (arrowed) x 40 (Courtesy of Professor Praveen Anand).
of regeneration these investigations of the cellular and molecular level in human tissue undoubtedly opens important lines of enquiry.
central nervous system. Although as long ago as 1907 Kilvington demonstrated the ability of the axons of the anterior horn cells to regenerate, and Tower (1943) showed the same, the matter was not systematically pursued. The latter indeed stated that the object of her work was “to prove unfounded the prevailing assumption that ventral nerve roots cannot regenerate if avulsed from the spinal cord.” She did just that in experiments on cats. Freeman (1952) used plasma clot suture for intradural transplantation of anterior spinal nerve roots in immature male guinea baboons. He reported functional restoration of a spinal reflex pattern of activity, and drew the conclusion that regeneration had occurred. Ochs and Barnes (1969) showed regeneration of ventral root fibres into dorsal roots in the cat, as shown by axoplasmic flow of radio-labeled leucine. Sanjuanbenito et al. (1976) demonstrated axonal regeneration in the ventral roots of the cat.
4.4 Regeneration after Intradural Injury The intradural, preganglionic lesions of the brachial plexus and to a lesser extent the similar lesions of the lumbosacral plexus have to be discussed separately in the matter of regeneration. It has for some time been clear that repair of such lesions offered the best chance of restoration of function and avoidance of pain, but the formidable technical difficulties of this have for years deterred most aspirants. There was, in addition, the discouraging thought that even if repair could be achieved, success required regeneration in the
136
Surgical Disorders of the Peripheral Nerves
Fig. 4.28 Small to medium sized neuronal cell bodies and nerve fibres from a human dorsal root ganglion 3 weeks after avulsion immunostained with antibody to TRPV1 (heat and capsaicin receptor, black). The cell bodies of large diameter neurones are weakly stained. x 40 (Courtesy of Professor Praveen Anand).
The demonstration of Nathaniel and Nathaniel (1973) that in adult rats the dorsal root fibres could regenerate into the spinal cord after crushing seemed to indicate that the proximal limb of the axon of the sensory cell could make its way into the substantia gelatinosa and posterior funiculi. Bonney and Jamieson (1979) achieved reattachment of dorsal roots in the case of a man with a severe traction lesion of the brachial plexus. That was followed by recovery of proximal muscle function. Then Jamieson and Eames (1980) studied in adult dogs the regeneration of motor and sensory axons after avulsion and reimplantation. They found that, whereas there was significant regeneration of motor axons across the zone of repair, there was no regeneration of sensory axons. It appeared that whereas the motor axons behaved as part of the peripheral nervous system, the central process of the axon of the sensory cell behaved as part of the central system. Regeneration through the ventral root. Carlstedt and his colleagues have taken the matter far forward in extensive experimental studies which confirmed the reinnervation of skeletal muscle after ventral root implantation into the spinal cord. It was not even necessary to re-implant the roots along the anterolateral sulcus; sub-pial implantation near the sulcus was all that was necessary. This far reaching work is outlined by Carlstedt (1991, 2007d). Carlstedt et al. (1995) reported the results of repair in man, of one root by direct implantation and of another by an intervening graft. There was secure evidence of functional reinnervation (Fig. 4.29). Regeneration through the dorsal root. The difference in behaviour of the nerve fibres within the dorsal and the ventral roots has been extensively studied by Carlstedt and his colleagues. Avulsion of the dorsal root causes proliferation of the asytrocytes which form a glial scar: “the astrocyte
Fig. 4.29 A horse radish peroxidase (HRP) filled motor neurone regenerated into an avulsed and implanted ventral root. (Above) Scale bar 20 mm. (Below). Scale bar 200 mm (Courtesy of Thomas Carlstedt).
barrier, developed as a response to nerve fibre degeneration, is most effective in preventing regeneration” (Carlstedt 1997, Carlstedt 2007d). However regeneration is possible in the immature animal before the establishment of the astrocyte population (Carlstedt 1988). It became clear that regeneration is enhanced if the transitional zone is bypassed by implanting the dorsal root directly into the spinal cord. This is followed by a substantial outgrowth of axons of the dorsal horn neurones to the injured dorsal root. Outgrowth from the secondary sensory neurones in the dorsal horn is further enhanced by excision of the dorsal root ganglion (Carlstedt 1997). Other methods have been used to enhance dorsal root fibre regeneration including glial cell transplants (Kliot et al. 1990), inflicting a preliminary or conditioning lesioning of the nerve before the implantation (Chong et al. 1999), and by transplanting olfactory ensheathing cells (Ramon-Cueto and Nieto-Sampedro 1994). Carlstedt and his colleagues applied olfactory ensheathing cells to reattached dorsal roots: the interposition enabled a “bridging channel” between the Schwann cells and the atrocytes, that is, between the peripheral and the central nervous systems (Li et al. 2004) (Fig. 4.30).
Regeneration and Recovery
137
avulsion of the ventral root. Nor can it, any longer, be a matter of doubt that this regeneration is capable of restoring useful function within the upper limb.
4.5 Recovery of Function after Nerve Repair a rule that a biologist, at any rate, finds useful is to suspect both those who lay down that results must be quantitative, and those who deny that they can be, but particularly to suspect the latter. (Young 1993)
Fig. 4.30 Regeneration through a reimplanted dorsal root. Olfactory ensheathing cells (green) in the proximal part of the dorsal root. Red stained regenerating nerve fibres passing between the root and the spinal cord. Original magnification x 180. Courtesy of Thomas Carlstedt.
Avulsion of the roots has a devastating effect on cells within the dorsal and ventral horns. It seems that the motor neurones can be rescued by reconnection with the periphery. Bergerot et al. (2004) combined reimplantation with intrathecal delivery of glial derived neurotrophic factor (GDNF), and the intraperitoneal injection of a neuroprotective agent, riluzole. The population of the anterior horn neurones was maintained at 80% of the contra lateral side. Hart et al. (2004) showed that dorsal horn neurones were protected by repair of the nerve within 24 h combined with the administration of acetyl-l-carbitine (ALCAR) or N-acetylcytseine (NAC). “As work in this field expands, it will be necessary to examine the effect of a solely motor innervation of muscles on their function. Doubtless, the search for sensory reinnervation will have to continue” (Birch and Bonney 1998a). Since that time we have confirmed the existence of myelinated afferent fibres in the ventral root of the human; their role awaits study. It is difficult to explain the level of function that has been seen in some successful cases after reimplantation without some elements of recovery of the muscle afferent pathway. In one case, which was demonstrated at the 7th International Brachial Plexus Meeting in London in 2008 the restoration of the biceps tendon reflex was demonstrated in a man who had undergone reimplantation of the ventral root of C6 six years previously. There can be no doubt that the motor neurones in the anterior horn are capable of regenerating new axons through the scar caused by
In clinical practice, most injuries of nerves short of transection inflict damage of all three grades of severity. The quality of reinnervation and hence of recovery depends first on the extent of damage causing degeneration and in particular loss of continuity of the basal lamina. A nerve fibre recovers completely after conduction block (neurapraxia), so long as the causative factor ceases to operate. Since this recovery does not have to be determined by axonal regrowth, it occurs over minutes, hours, days, weeks or – rarely– months. A nerve fibre will regenerate to its correct target after axonotomy so long as the basal lamina is preserved and the causative factor is removed. But recovery will take as long as it takes for the axon to re-innervate its target. Thus, recovery is the rule after “axonotmesis.” A nerve fibre will only regenerate to its correct target after axonotomy with interruption of the basal lamina if it is correctly directed to its target field or is drawn there by the operation of growth factors and other guiding mechanisms. Thus, in the case of individual nerves when the promptitude and quality of primary treatment is equal, the quality of recovery will depend on (1) the nature of the lesion: (2) the incidence of axonotomy: (3) the incidence of interruption of the basal lamina: (4) the obstacles posed to correct targeting by the complexity of functional representation in the nerve and the consequent liability to axon/target confusion; (5) the distance of the lesion from the target organs. We have already indicated that there are in effect no “purely motor” or “purely sensory” nerves, but the liability to axon/target confusion is plainly greater in a nerve such as the median with an extensive motor and cutaneous sensory distribution than it is in the posterior interosseous nerve, with no cutaneous sensory distribution. It is not doubted that full maturation of the regenerating nerve fibre depends on the establishment of the connection with its end-organ. The quality of recovery after neurotmesis and repair depends chiefly on the number of axons reaching their correct targets and on the later development and myelination of those axons. Factors influencing such re-innervation are: (1) the promptitude of repair; (2) the quality and viability of the opposed nerve ends; (3)The quality and accuracy of fascicular matching; (4) the degree of damage to the nerve ends
138
during suture; (5) the length of gap after resection; (6) the number of channels provided by the interposed graft for the regenerating columns; (7)the extent of fibroblastic infiltration of the stumps and of the interposed grafts.
4.5.1 Methods The methods used for measurement of power, of cutaneous sensibility and sympathetic function are set out in the following Chap. 5. We have made only slender progress towards measuremt of recovery in the deep afferent pathways. The documents which have been developed to record progress are illustrated in Chaps. 5 and 10. Recall. Recovery of nerves after suture is a prolonged business. Omer (1998) described recovery of the small muscles of the hand 8 years after nerve repair and he pointed out: “these well documented late returns of neurological function to the intrinsic muscles of the hand are contrary to all results previously reported.” Case report. Primary repair of median and ulnar nerves was performed in a 22 year old staff nurse. Recovery was quite good, but marred by cold sensitivity, so that her result was considered only fair at 3 years after operation. She came to see us 7 years after operation to report a striking improvement during the previous few months. Over-sensitivity had disappeared, the final result for the ulnar nerve was considered excellent – and it was very good for the median. Case report. A 39 year old nursing sister wounded the digital nerves to little and ring fingers. Recovery was marred by cold sensitivity, and overreaction to light touch, until there was quite sudden resolution, over a few weeks, of these difficulties 6 years after operation. Zachary (1954) analyzed the duration of followup in patients with nerves repaired in the five Medical Research Council Nerve Injury Centres. More than 70% of them were seen at 60 months. Most of the failed repairs of the radial nerve were subsequently treated by flexor to extensor muscle transfer and after excluding these the rate of recall at 5 years was close to 80%. It has always been difficult to match this in civilian practice, indeed, clinicians, now, may find it impossible to do so as they find themselves working in an over politicized, management led and target driven health service. We must do what we can. Adequate assessment of outcome after severe injuries to the brachial and lumbosacral plexus in the adult requires at least 5 years of study. In some patients even this is inadequate (Chap. 9). The length of the follow up must be longer still in children with birth lesions of the brachial plexus (BLBP) or in those with severe injuries to the main nerves of the lower limb because of the ever present risk of secondary deformity from muscular imbalance or impaired
Surgical Disorders of the Peripheral Nerves
growth and because of the high incidence of posterior dislocation at the shoulder in BLPP. As it is our follow up is adequate in 1,060 repairs of closed traction lesion of the supraclavicular brachial plexus in the adult, in 1,128 (250 repairs) cases of BLPP and in around 1,800 repairs of main nerves. The results are related in later chapters. Grading of results. The results for nerves as a whole are graded by systems based on those developed by Seddon (1975e) which were drawn from those used in the five Specialist Nerve Injury Units (Zachary 1954). In some nerves muscular function is a good deal more important than recovery of sensation and for the circumflex, musculocutaneous, radial, femoral and common peroneal nerves, little significance is attributed to return of cutaneous sensation except when recovery is complicated by significant pain when the result is considered poor. Recovery of sensation has been given equal importance to muscle function in describing results of repairs of median and ulnar nerves and of the tibial division of the sciatic nerve. One could argue that sensibility is the most important function of the median and tibial nerves. Three grades are used (Birch, Bonney 1998): 1. Good, which means that recovery of function is substantial, that it enables the patient to use the part and to lead a normal life without difficulty and that they are not troubled by pain or over reaction. 2. Fair, implies recovery of function which is useful but which falls far short of normal. The patient is sufficiently aware of the hand or the foot that they do not damage it. Muscle power of MRC grade 2 or 3 restores some balance across a joint and it may be possible to supplement this with the appropriate muscle transfer to enhance, for example, flexion of the elbow or extension of the wrist. 3. Poor, means no or insignificant recovery and this grade is used in all cases with persisting spontaneous neuropathic pain, hypersensitivity or over reaction. The criteria used for grading results in individual nerves are set out in Chap 8. All systems of measurement have their defects and it seems the more closely one looks at the result of a nerve repair the more defects are revealed. Birch and Raji (1991) and Kline and Hudson (1995) introduced an “excellent” result, a result which we have given to a small number of repairs of the median and ulnar nerves nerve in the forearm and to some cases of repair of the spinal accessory nerve. The outcome is set out according to the cause of injury: the “tidy” wound from knife, glass or scalpel; the “untidy” wound from axes, saw, penetrating missile, open fracture or dislocation or burn; and the closed traction lesion. The effects of associated arterial injury are severe: such cases are considered separately in each of the three main groups. This reflects our view that the two factors of overwhelming significance on outcome are first, the violence of the injury and the extent
Regeneration and Recovery
139
of damage to the nerve and adjacent tissues and second, delay before repair. Other factors which are important in prognosis were clearly defined in the Medical Research Council Special Report Series No. 282 (1954), by Seddon (1975c), Sunderland (1978), and Kline and Hudson (1995), Kline and Hudson (2008, 2008) amongst others. We shall examine these.
4.6 Factors in Prognosis Age. Recovery of function after repair of a nerve in a child is, on the whole, better than in the adult, but it is not as good as has sometimes been assumed. This comes as no surprise if we take into consideration the increased vulnerability of the immature nervous system to axonotomy. Some of the most difficult problems in reconstruction follow failure of recovery of either of the divisions of the sciatic nerve in the growing child. The unpredictability and the limitations of outcome after repair of the plexus in birth lesions are related in Chap. 10. The recovery of skeletal muscle and of sympathetic efferent function after injuries to the brachial or lumbosacral plexus in the growing child is generally inferior to that seen following urgent repair in adults. There is often severe shortening of the limb and atrophy of the hand or foot. On the other hand the recovery of cutaneous sensation in the infant or child is often remarkably good. The results after repair of digital nerves provides one example of the difference between the adult and the child (Table 4.1). Goldie et al. (1992) made a detailed study of 27 adult patients. Thirty nerves were repaired, all but one by primary repair using sutures for we did not have access to fibrin clot glue at that time. The digital artery was repaired if that had been damaged. Examination for sensation was thorough and it was helped by using weighted pins designed for us by Professor Ruth Bowden. The results were surprisingly disappointing, an indication perhaps that the more scrupulously that one examines outcome from a nerve repair the more one will find wrong. Thirty-seven per cent of fingers regained normal two-point discrimination but
only 27% of patients graded their overall result as “excellent” and 40% complained of persistent hyperaesthesia up to 2 years. Goldie and Coates concluded: “following repair of the divided digital nerve, normal sensation will never be regained. Hyperaesthesia may be present for months or years but will ultimately resolve. The final result will take two or three years to achieve.” Fortunately results do seem to be a good deal better in children. It does seem odd that recovery of sensation after a well executed primary repair of the median nerve at the wrist is regularly better than that from repair of digital nerves more distally. It is only fair to add that a number of patients reported striking improvement in sensation, with sudden loss of cold sensitivity and hypersensitivity to touch, some years after repair. There seems a trend towards neglecting nerve injuries in patients of a certain age, a policy for which we find a remarkable absence of supporting evidence. Indeed, the changes in the peripheral nerves with aging (Chap. 2) suggest that the older patient may in fact be more likely to develop severe pain after injury to a nerve. Some striking results have been seen after repair of the brachial plexus in patients aged over 65 years. In several of these the relief of pain and with it the improvement in the patient’s mental well being, were remarkable. It may be administratively convenient to ignore the elderly but they are, after all, still human beings and the risk of losing independence calls for a far more rational and vigorous approach than is often, too often, seen. Level of injury. This is undoubtedly important for the median, ulnar and radial nerves but it seems to be much less important for the sciatic nerve and its divisions. Only rarely is extension of the digits regained after injuries to the brachial plexus, the posterior cord or the radial nerve in the axilla. Urgent repair of lesions of C5 and C6 is often followed by recovery to a level far higher than that seen after repair of combined lesions of the circumflex, the suprascapular and the musculocutaneous nerves. There seems little doubt, however, that repair of the recurrent motor branch of the median nerve at the wrist, of the posterior interosseous nerve and of the deep branch of the ulnar nerve is followed by a return of function which is scarcely ever seen after high repairs of main nerves. On the other hand lesions of the distal
Table 4.1 Recovery of sensation after repair of 129 digital nerves. Grade Adults Repair within 48 h of injury Repair 2 weeks or more after injury Excellent
1
0
Children (aged 1 years or less) Repair at varying intervals after injury 17
18
Good
33
9
8
50
Fair
24
11
2
37
Poor
16
74 17 digital arteries repaired at primary operation. 20 flexor tendons repaired at primary operation.
8
0
24
28
27
129
140
branches of such cutaneous nerves as the sural, the medial cutaneous of forearm and the superficial radial nerve have deservedly gained an ill reputation. The nerve. Whilst there is no such thing as a purely “motor” or “sensory” nerve in the peripheral nervous system, it appears that recovery after repair of nerves which have no cutaneous distribution is generally better than after repair of main nerves. The spinal accessory and the nerve to serratus anterior generally recover well after repair. The suprascapular fares better than the circumflex. It is more than 50 years since Zachary established that recovery for the common peroneal nerve is far inferior than for the tibial nerve, an observation which has been repeatedly confirmed. The blood supply to the most proximal segment of the radial nerve and to the common peroneal nerve is rather poor, so that transection of these nerves is likely to enhance ischaemic changes in the distal segment.
4.7 Severity of Injury Useful recovery in function cannot be expected if the fundamental principles of treatment of a wound are neglected. Arteries must be repaired, the tissues of the limb must be perfused, the skeleton must be stabilized and repaired nerves and vessels must be covered by full thickness skin. Michael Glasby and his colleagues used the freeze-thawed muscle graft (FTMG) in a series of experiments which showed that recovery is worsened by delay, by arterial injury, by fractures, by cavitation and by haematoma (Fullerton et al. 1998, Glasby et al. 1997, Glasby et al. (1998).
Surgical Disorders of the Peripheral Nerves
inevitable.” Zachary and Holmes (1946) compared 55 cases of primary sutures of nerves referred to the Peripheral Nerve Injury Centre, at the Wingfield-Morris Hospital, Oxford, during the years 1940–1944 with 36 “early” secondary sutures. The results of secondary suture were distinctly better. The primary repair was resected and the nerve re-sutured in 16 patients. Histological examination of the resected material revealed a number of causes for the failure of the first operation: poor matching of the proximal and distal stumps; coarse suture material lodged between the stumps; separation of the stumps; and dense scar between or within the stumps. Zachary and Holmes concluded that “Formal nerve suture should be undertaken at the earliest moment when it is possible to recognise the extent of damage to the nerve, excise the injured segment, and bring together the mobilised nerve ends without the prospect of undue post operative tension.” These are painful but important findings: the lesson is that the results of operating on peripheral nerves depend to an extent hardly matched in any other branch of surgery on the skill of the surgeon and the quality of the technique. The clinician must always bear in mind that the sooner the distal segment is connected to the cell body and proximal segment the better the result will be. The quality of treatment of the wound at first operation and the timing of repair of the nerves are factors within the control of clinicians. The last 20 years has seen extensive laboratory work which confirms the view, long held by many clinicians, that the central nervous system suffers after interruption of a peripheral nerve. The rapidity and the severity of that response to violent proximal nerve injury is the over arching biological imperative which must guide action.
4.8 Delay References Sherren (1908) examined 50 cases of primary suture performed at the London Hospital. Pain sensibility recovered before touch; tactile localisation continued to improve for more than 2 years. Recovery of power was rather slower. There was no recovery in only one patient, whose wound had become infected. Sherren recommended primary suture “because the prognosis after secondary suture is more unfavourable.” Platt and Bristow (1924) reviewed the late results of nerve injuries treated in the First World War, and noted the “extreme perfection attained after so-called primary suture,” acknowledging that the more severe gunshot wounds of nerves were treated by secondary suture. Platt (1937) wrote “in primary sutures performed under ideal conditions, complete recovery of motor power and recovery of protopathic sensibility at least, is to be expected.....however, in more extensive wounds with widespread bruising and multiple tendon injuries, and in wounds in which infection had already secured a hold, partial or complete failure after primary suture is almost
Aitken JT, Sharman M, Young JZ (1947) Maturation of regenerating nerve fibres with various peripheral connexions. J Anat 81:1–22 Allieu Y (1989) Les Neurotisations par le nerf spinal dans les avulsions du plexus brachial de l’adulte. In: Alnot JY, Narakas A (eds) Les paralysies du plexus brachiale, monographies du groupe d’étude de la main. Expansion Scientifique, Paris, pp 173–179 Alnot JY, Narakas A (1989) Les Paralysies du plexus brachiale monographies du groupe d’étude de la main. Expansion Scientifique, Paris Alnot JY, Narakas A (eds) (1996) Traumatic brachial plexus injuries. Expansion Scientifique Française, Paris Anand P, Birch R (2002) Restoration of sensory function and lack of long-term chronic pain syndromes after brachial plexus injury in human neonates. Brain 125:113–122 Anand U, Otto WR, Casula MA, Day NC, Davis JB, Bountra C, Birch R, Anand P (2006) The effect of neurotrophic factors in morphology Trpvi expression and capsaicin responses of cultured human drug sensory neurons. Neurosci Lett 399:51–56 Anand U, Otto WR, Sanchez-Herrera D, Facer P, Yiangou Y, Korchev Y, Birch R, Benham C, Bountra C, Chessell IP, Aband P (2008) Canabinoid receptor CB2 localisation and agonist-mediated inhibition of capsaicin responses in human sensory neurons. Pain 138:667–680
Regeneration and Recovery Ballance C, Ballance H (1903) Remarks on the operative treatment of chronic facial nerve palsy of peripheral origin. Br Med J 2: 1009–1013 Ballance SC, Duel AB (1932) The operative treatment of facial palsy by the introduction of nerve grafts. Arch Otolaryngol 15:1–70 Banks RW (2005) The muscle spindle. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier, Philadelphia, pp 131–150 Barker D, Scott JJA, Stacey MJ (1986) Re-innervation and recovery of cat muscle receptors after long term denervation. Exp Neurol 94:184–202 Barker D, Berry RB, Scott JJ (1990) The sensory reinnervation of muscles following immediate and delayed repair in the cat. Br J Plast Surg 43:107–111 Barnes R, Bacsich P, Wyburn GM, Kerr AS (1945) A study of the fate of nerve homografts in man. Br J Surg 34:34–41 Barton NJ (2005) Calf nerve used as allograft for nerves in the arm in 1917. pers com: letter Peake J In: Borderlines, Country and Border Life Issue, 21 September 2005, Border, Oswestry SY10 7WN UK, p 4 Battiston B, Blanzetta M (1999) Repair of proximal ulnar by distal transfer- reconstruction of high nerve lesions by distal double median to ulnar nerve transfer. J Hand Surg Am 24:1185–1191 Beevor CE (1904) The Croonian lectures on muscular movements and their representation in the central nervous system. Adlard, London Benassy J, Robart J (1966) Un cas de transposition du nerf musculocutané sur le nerf median. Ann Chir Plast 11:187–191 Bennett MR, Ho S, Labisdis NA (1986) Competition between segmental nerves at end plates in rat gastrocnemius muscle during loss of polyneural innervation. J Physiol 381:351–376 Bergerot A, Shortland PJ, Anand P et al (2004) Co-treatment with riluzole and Gdnf is necessary for functional recovery after ventral root avulsion injury. Exp Neurol 187:359–366 Birch R (1979) Silicone rubber cuffs – a cause of nerve compression. Hand 11:211–212 Birch R (1989) The repair of peripheral nerves: experimental and clinical observations. Thesis for the degree of Master of Surgery, University of Cambridge Birch R, Bonney G (1998a) Regeneration. In: Birch R, Bonney G, Wynn Parry CB Surgical disorders of the peripheral nerves. Churchill Livingstone, London, pp 57–70 Birch R, Bonney G (1998b) Results. In: Birch R, Bonney G, Wynn Parry CB Surgical disorders of the peripheral nerves. Churchill Livingstone, London, pp 157–207 Birch R, Raji ARM (1991) Repair of median and ulnar nerves. J Bone Joint Surg 73B:154–157 Blaauw G, Sauter Y, Lacroix CL, Slooff AC (2006) Hypoglossal nerve transfer in obstetrical brachial plexus palsy. J Plast Rec Surg 59:474–478 Bonney G (1977) Some lesions of the brachial plexus. Ann Royal Coll Surg (Eng) 59:298–306 Bonney G, Jamieson A (1979) Communication au symposium sur le plexus brachial; reimplantation of C7 and C8. Int Microsurg 1:103–106 Bowden REM, Gutmann E (1944) Denervation and re-innervation of human voluntary muscle. Brain 67:273–313 Brunelli GA, Brunelli GR (1999) Restoration and walking in paraplegia by transferring the ulnar nerve to the hip: a report on the first cure. Microsurgery 19:223–226 Brunelli G, Monini L (1985) Direct muscular neurotisation. Proceedings of the second congress of the international federation of societies for surgery of the hand. J Hand Surg Am 10:993–994 Brushart TM (1988) Preferential reinnervation of motor nerves. J Neurosci 8:1026–1031 Brushart TM (1993) Motor axons preferentially reinnervate motor pathways. J Neurosci 13:2730–2738 Brushart TM, Seiler WA (1987) Selective reinnervation of distal motor stumps by peripheral motor axons. Exp Neurol 97:289–300
141 Brushart TM, Mather V, Sood R, Koschorke G-M (1995) Dispersion of regenerating axons across enclosed neural gaps. J Hand Surg Am 20:557–564 Bunnell S, Boyes JH (1939) Nerve grafts. Am J Surg 44:64–75 Bunting S, Di Silvio L, Deb S, Hall S (2005) Bioresorbable glass fibres facilitate peripheral nerve regeneration. J Hand Surg Br 30: 242–247 Cabaud HE, Rodky G, Nemeth TJ (1982) Progressive ultrastructural changes after peripheral nerve transection and repair. J Hand Surg Am 7:353–365 Caddick J, Kingham PJ, Gardiner NJ, Wyburg M, Terenghi G (2006) Phenotypic and functional characteristics of mesenchymal stem cells differentiated along a Schwann cell lineage. Glia 54:840–849 Cajal SRY (1928a) A regenerative process of the central stump. In: Degeneration and regeneration of the nervous system. Oxford University Press, London, Chapter 7, pp 141–166 (Trans ed: May RM) Cajal SRY (1928b) Late conditions shown by the nerve sprouts of the central stump and of the scar. In: Degeneration and regeneration of the nervous system. Oxford University Press, London, Chapter 9, pp 198–222 (Trans ed: May RM) Cajal SRY (1928c) Experiments dealing with the transplantation of nerves or their products, designed to prove especially an attractive or neurotropic action on nerve sprouts. In: Degeneration and regeneration of the nervous system. Oxford University Press, London, Chapter 15, pp 329–361 (Trans ed: May RM) Calder JS, Green CJ (1996) Nerve-muscle sandwich grafts: the importance of Schwann cells in peripheral nerve regeneration through muscle basal lamina conduits. J Hand Surg Br 20:423–428 Camp SJ, Allen D (2008) Cocontraction of swallowing musculature and trapezius following basal skull fracture. Injury Extra 39:95–97 Carlstedt T (1988) Reinnervation of the mammalian spinal cord after neonatal dorsal root crush. J Neurocytol 17:335–350 Carlstedt T (1991) Experimental studies on surgical treatment of avulsed spinal nerve roots in brachial plexus injury. J Hand Surg Br 16:477–482 Carlstedt T (1997) Nerve fibre regeneration across the peripheral-central transitional zone. J Anat 190:51–56 Carlstedt T (2007a) End to side or termino-lateral neurorrhaphy. In: Central nerve plexus injury. Imperial College Press, London, pp 70–71 Carlstedt T (2007b) Outcome of nerve transfer. In: Central nerve plexus injury. Imperial College Press, London, pp 71–74 Carlstedt T (2007c) Cocontraction. In: Central nerve plexus injury. Imperial College Press, London, pp 144–150 Carlstedt T (2007d) Basic science of experimental root injuries. In: Central nerve plexus injury, Chap 6. Imperial College Press, London, pp 78–111 Carlstedt T, Crane P, Hallin RG, Noren G (1995) Return of function after spinal cord implantation of avulsed spinal nerve roots. Lancet 346:1323–1325 Chen L, Gu YD (1994) An experimental study of contralateral C7 root transfer with vascularised nerve grafting to treat brachial plexus root avulsion. J Hand Surg Br 19:60–66 Chong MS, Woolf CJ, Haque NS, Anderson PM (1999) Axonal regeneration from injured dorsal roots into the spinal cord of adult rats. J Comp Neurol 410:42–54 Collier J (1940) Facial paralysis and its operative treatment. Lancet 1:91 Dellon AL, Munger BL (1983) Correlation of histology and sensibility after nerve repair. J Hand Surg Am 8:871–875 Durrenberger PF, Facer P, Gray RA, Chessell IP, Naylor A, Bountra C, Banati RB, Birch R, Anand P (2004) Cyclooxygenase – 2 (Cox-2) in injured human nerve and a rat model of nerve injury. J Peripher Nerv Syst 9:15–24 Dyck PJ, Thomas PK (2005) Peripheral neuropathy, 4th edn. Elsevier, Philadelphia
142 Dyck PJ, Dyck PJB, Engelstad J (2005) Pathological alterations of nerves. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier, Philadelphia, pp 733–829 Foerster O (1929) Die therapie der Schussverletzungen der peripheren nerven, resultate der plexus operationen. In: Foerster BO (ed) Handbuch der neurologie von lewandowski, vol 2. Springer, Berlin, pp 1676–1691 Forsmann J (1900) Zur kenntnis des neurotropismus. Ziegler’s Beiträge Zur Pathol Anat 27:407 Freeman LW (1952) Observations on spinal nerve root transplantation in the male guinea baboon. Ann Surg 136:206–210 Fullerton AC, Glasby MA, Lawson GM (1998) Immediate and delayed nerve repair using freeze-thawed muscle autografts in complex nerve injuries. Associated long bone fracture. J Hand Surg Br 23:360–364 Fullerton AC, Myles LM, Lenihan DV, Hems TEJ, Glasb M (2001) Obstetric brachial plexus palsy: a comparison of the degree of recovery after repair of 16 ventral root avulsions in newborn and adult sheep. Br J Plas Surg 54:697–704 Glasby MA, Gschmeissner SE, Huang CL, De Souza BA (1986) Degenerated muscle grafts used for peripheral nerve repair in primates. J Hand Surg Br 11:347–351 Glasby MA, Fullerton AC, Lawson GM (1997) Immediate and delayed nerve repair using freeze-thawed muscle autografts in complex nerve injuries. Cavitation, fibrosis and haematoma. J Hand Surg Br 22:479–485 Glasby MA, Fullerton AC, Lawson GM (1998) Immediate and delayed nerve repair using freeze-thawed muscle autografts in complex nerve injuries Associated arterial injury. J Hand Surg Br 23(3):354–359 Goldie BS, Coates CJ, Birch R (1992) The long term results of digital nerve repair in no man’s land. J Hand Surg Br 12:75–77 Gore RV (1978) A new method of nerve repair: repair of a lesion of the radial nerve with a branch to the triceps muscle. Br J Surg 65: 352–353 Griffin JW, Hoffman PN (1993) Degeneration and regeneration in the peripheral nervous system. In: Dyck PJ, Thomas PK, Lambert EH, Bunge R (eds) Peripheral neuropathy, 3rd edn. WB Saunders, Philadelphia, Chapter 22, pp 361–376 Gu YD, Zhang GM, Chen DS, Yan JG, Cheng XM, Chen L (1992) Seventh cervical nerve root transfer from the contralateral healthy side for treatment of brachial plexus root avulsion. J Hand Surg Br 17:518–521 Gutmann E, Sanders FK (1942) Functional recovery following nerve grafts and other types of nerve bridges. Brain 65:373–408 Gutmann E, Young JZ (1944) The reinnervation of muscle after various periods of atrophy. J Anat 78:15–43 Hall S (1997) Axonal regeneration through acellular muscle grafts. J Anat 190:57–71 Hall S (2005a) Mechanisms of repair after traumatic injury. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, Chapter 58, pp 1403–1434 Hall S (2005b) The response to injury in the peripheral nervous system. J Bone Joint Surg Br 87:1309–1319 Hall SM (2009) Biomaterials for the repair of peripheral nerves. In: di Silvio L (ed) Cellular response to biomaterials, Chap. 11. Woodhead, Cambridge, pp 252–290 Hallin R, Carlstedt T, Nilsson-Remahl IA, Risling M (1999) Spinal cord implantation of avulsed ventral roots in primates: correlation between restored motor function and morphology. Exp Brain Res 124(3):304–310 Harris W, Low VW (1903) On the importance of accurate muscular analysis in lesions of the brachial plexus and the treatment of Erb’s palsy and infantile paralysis of the upper extremity by cross-union of nerve roots. Br Med J 2:1035–1038 Hart AM, Terenghi G, Kellerth JO, Wiberg M (2004) Sensory neuroprotection, mitochondrial preservation and therapeutic potential of N-acetyl-cysteine after nerve injury. Neuroscience 125:91–101
Surgical Disorders of the Peripheral Nerves Healy C, Lequesne PM, Lynn B (1996) Collateral sprouting of cutaneous nerves in man. Brain 119:2063–2072 Highet WB, Sanders FK (1943) The effects of stretching nerves after suture. Br J Surg 30:355–369 Höke A, Redett R, Hameed H, Jari R, Zhou C, LI ZB, Griffin JW, Brushart TM (2006) Schwann cells express motor and sensory phenotypes that regulate axon regeneration. J Neurosci 26:9646–9655 Holmes W, Young JZ (1942) Nerve regeneration after immediate and delayed suture. J Anat 77:63–96 Htut M, Misra P, Anand P, Birch R, Carlstedt T (2006) Pain phenomena and sensory recovery following brachial plexus avulsion injury and surgical repair. J Hand Surg Br 31:596–605 Irwin MS, Abhilash J, Anand P, Manchahal J (2006) Free innervated sole of foot transfer for contralateral lower limb salvage. Plast Recon Surg 118:93e–97e Jabaley ME, Burns JE, Orcutt BS, Bryant WM (1976) Comparison of histologic and functional recovery after peripheral nerve transection and repair. J Hand Surg 1:119–130 Jamieson A, Eames RA (1980) Reimplantation of avulsed brachial plexus roots: an experimental study in dogs. Int J Microsurg 2:75–80 Joyce JL (1919) Nerve suture. Br J Surg 6:418 Kanamura A, Suzuki S, Sibuya M, Homma I, Sai K, Hara T (1993) Sensory reinnervation of muscle receptor in human. Neuroscience 161:27–29 Graham K (1904) On somepoints in the early development of motor nerve trunks, etc. in lepidosiren. Trans Roy Soc Edin 41:121–128 Kettle S (2003) The use of end to side repair of peripheral nerves for neurotization after loss of nerve tissue in a large tissue model. MD Thesis. University of Edinburgh Kilvington B (1907) An investigation on the regeneration of nerves, with regard to surgical treatment of certain paralysies. Br Med J 1:988–990 Kline DG, Hudson AR (1995) Nerve injuries. WB Saunders, Philadelphia Kline DG, Hudson AR (2008) Nerve injuries, 2nd edn. Saunders Elsevier, Philadelphia (with Kim DH, Midha R, Murovic JA, Spinner RJ) Kliot M, Smith GM, Siegal JD, Silver J (1990) Astrocyte-polymer implants promote regeneration of dorsal root fibres into the mammalian spinal cord. Exp Neurol 109:57–69 Kotani PT, Matsuda H, Suzuki T (1972) Trial surgical procedures of nerve transfers to avulsion injuries of the plexus brachialis. Excerpta Medica International Congress Series Number 291 (1973). Proceedings of the 12th Congress of S.I.C.O.T. Tel Aviv, October 9–12 1972, pp 348–351 Lago N, Ceballos D, Rodriguez FJ, Stieglitz T, Navarro X (2007) Effects of motor and sensory nerve transplants on amount and specificity of sciatic nerve regeneration. J Neurosci Res 85:2800–2812 Lang EM, Borges J, Carlstedt T (2004) Surgical treatment of lumbosacral plexus injuries. J Neurosurg Spine 1:64–71 Larsen M, Habermann TM, Bishop AT, Shin AY, Spinner RJ (2007) Epstein-Barr virus infection as a complication of transplantation of a nerve allograft from a living related donor. J Neurosurg 106: 924–928 Li H, Wigley C, Hall SM (1998) Chronically denervated rat Schwann cells respond to Ggf in vitro. Glia 21:1–14 Li Y, Carlstedt T, Berthold CH, Raisman G (2004) Interaction of transplanted olfactory ensheathing cells and host astrocyte processes provides a bridge for axons to regenerate across dorsal root entry zone. Exp Neurol 188:300–308 Lundborg G (1991) Neurotropism, frozen muscle grafts and other conduits. J Hand Surg Br 16:473–476 Lundborg G (2003) Nerve injury and repair. A challenge to the plastic brain. The Bunge Memorial Lecture. J Peripher Nerv Syst 8:209–226 Lundborg G (2004a) The growth cone. In: Nerve injury and repair, 2nd edn. Elsevier/Churchill Livingstone, Philadelphia, p 13
Regeneration and Recovery Lundborg G, Dahlin LB, Danielsen N, Nachemson AK (1986) Tissue specificity in nerve regeneration. Scand J Plast Reconstr Surg 6:265–281 Lundborg G, Dahlin L, Dohi D, Kanje M, Terad N (1997) A new type of “bioartificial” nerve graft for bridging extended defects in nerves. J Hand Surg Br 22:299–303 MacCarty CS (1951) Two-stage autograft for repair of extensive damage to sciatic nerve. J Neurosurg 8:319 MacKinnon SE (1996) Nerve allotransplantation following severe tibial nerve injury. J Neurosurg 84:671–676 MacKinnon S, Dellon L, O’brian J (1986a) Changes in nerve fibre numbers distal to a nerve repair in the rat sciatic model. Muscle Nerve 14:1116–1122 MacKinnon SE, Dellon AL, Lundborg G, Hudson AR, Hunter D (1986b) A study of neurotropism in a primate model. J Hand Surg Am 11:888–894 Malessy MJ, Thomeer RT, Van Dijk JG (1998) Changing central nervous system control following intercostal nerve transfer. J Neurosurg 89:568–574 Malessy MJ, Bakker D, Dekker AJ, Van Dijk JG, Thomeer RT (2003) Functional magnetic resonance imaging and control over the biceps muscle after intercostal-musculocutaneous nerve transfer. J Neurosurg 98:261–268 Mano Y, Nakamuro T, Tamara R, Takayang T, Kawanishi K, Tamai S, Mayer RF (1995) Central Motor re-organisation after anastamosis of the musculocutaneous and intercostal nerves following clinical root avulsion. Ann Neurol 38:15–20 Mayo-Robson AW (1917) Nerve grafting as a means of restoring function in limbs paralysed by gunshot or other injuries. Br Med J 1:117–118 Medical Research Council Special Report Series No. 282. (1954). Peripheral nerve injuries. Seddon HJ (ed). HSMO London. Millesi H (1973) Microsurgery of peripheral nerves. Hand 5:157–160 Mirsky R, Jessen KR (2005) Molecular signalling in Schwann cell development. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, Chapter 16, pp 341–376 Moberg E (1958) Objective methods for determining the functional value of sensibility in the hand. J Bone Joint Surg Br 40:454–476 Morris JH, Hudson AH, Weddell G (1972a) A study of degeneration and regeneration in the divided rat sciatic nerve based on electron microscopy ii the development of the “regenerating unit”. Zeitschrift für Zellforschumg 124:103–130 Morris JH, Hudson AH, Weddell G (1972b) Changes in the axons of the proximal stump. Zeitschrift für Zellforschumg 124:131–164 Morris JH, Hudson AH, Weddell G (1972c) Changes in fascicular microtopography. The perineurium and endoneurial fibroblasts. Zeitschrift für Zellforschumg 124:165–203 Mott FW, Halliburton WD (1901) The chemistry of nerve-degeneration. Phil Trans R Soc Lond B 194:437–466 Myles LM, Gilmour JA, Glasby M (1992) Effects of different methods of peripheral nerve repair in the number and distribution of muscle afferent neurons in the rat dorsal root ganglion. J Neurosurg 77:457–462 Nagano A, Tsuyama N, Ochiai N, Hara T (1989) Direct nerve crossing with the intercostal nerve to treat avulsion injuries of the brachial plexus. J Hand Surg Am 14(6):980–985 Narakas AO (1988) Neurotization or nerve transfer in traumatic brachial plexus lesions. In: Tubiana R (ed) The hand III. WB Saunders, Philadelphia, Chapter 62, pp 656–683 Nathaniel EJM, Nathaniel DR (1973) Regeneration of dorsal root fibers into the adult rat spinal cord. Exp Neurol 40:333–350 Navarro X, Calvet S, Rodriguez FJ, Stieglitz T, Blau T, Buti M, Valderrama E, Meyer JU (1998) Stimulation and recording from regenerated peripheral nerves through polyimide sieve electrodes. J Peripher Nerv Syst 3:91–101 Nichols CM, Brenner MJ, Fox IK, Tung TH, Hunter DA, Rickman SR, MacKinnon SE (2004) Effects of motor versus sensory nerve grafts on peripheral nerve regeneration. Exp Neurol 190:347–355
143 Oberlin C, Beal D, Leechavengvongs S, Salon A, Dauge MC, Sarly JJ (1994) Nerve transfers to biceps muscle using part of ulnar nerve for C5-C6 avulsion of the brachial plexus: anatomical study and report of four cases. J Hand Surg Am 19:232–237 Ochs S, Barnes CD (1969) Regeneration of ventral root fibers into dorsal roots shown by axoplasmic flow. Brain Res 15:600–603 Omer GE (1998) Peripheral nerve injuries and gunshot wounds. In: Omer GE, Spinner M, Van Beek AL (eds) Peripheral nerve problems, 2nd edn. WB Saunders, Philadelphia, Chapter 41, pp 398–405 Őzkan T, Őzer K, Gülgőnen A (2001) Restoration of sensibility in irreparable ulnar and median nerve lesions with use of sensory nerve transfer: long-term follow up of 20 cases. J Hand Surg Am 26:44–51 Pereira JH, Palande DD, Narayanakumar TS, Subramian AS, Gschmeissner S, Wilkinson M (2008) Nerve repair of denatured muscle autografts promotes sustained sensory recovery in leprosy. J Bone Joint Surg Br 90:220–224 Platt H (1937) Discussion on injuries to peripheral nerves. Proc R Soc Med Lond 30:863 Platt H, Bristow WR (1924) The remote results of operations for injuries of the peripheral nerves. J Br J Surg 11:535–567 Ramon-Cueto A, Nieto-Sampedro M (1994) Regeneration into the spinal cord of transected dorsal root axons is promoted by ensheathing glial transplants. Exp Neurol 127:232–244 De Ruiter GCW, Spinner RJ, Yaszmenski MJ, Windebank AJ, Malessy MJA (2009) Nerve tubes for peripheral nerve repair. Neurosurg Clin North Am 20:91–106 Sai K, Kanamura A, Sibuya M, Homma I, Hara T (1996) Reconstruction of tonic vibration reflex in the biceps brachii reinnervated by transferred intercostals nerves in patients with brachial plexus injury. Neurosci Lett 206:1–4 Sanders FK (1942) The repair of large gaps in the peripheral nerves. Brain 65:281–337 Sanders FK, Young JZ (1946) The influence of peripheral connections on the diameter of regenerating nerve fibres. J Exper Biol 22:203–212 Sanjuanbenito L, Esteban A, Gonzalez-Martinez E (1976) Regeneration of the spinal ventral roots. Acta Neurochir 34:203–214 Schuind F, Van Holder C, Mouraux D, Robert CH, Meyer A, Saliva P, Vermeylen N, Abramowicz D (2006) The first Belgian hand transplantation – 37 months results. J Hand Surg Br 31:371–376 Scott JJ (1995) The functional recovery of muscle proprioception after peripheral nerve lesions. J Peripher Nerv Syst 1:19–27 Scott JJA (2005) The Golgi tendon organ. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy (in two volumes), 4th edn. Elsevier Saunders, Philadelphia, Chapter 7, pp 151–161 Seddon HJ (1947) The use of autogenous grafts for the repair of large gaps in peripheral nerves. Br J Surg 35:151–167 Seddon HJ (1954) Nerve grafting. In: Peripheral nerve injuries by the Nerve Injury Committee of the Medical Research Council MRC Special Report Series 282. Hmso, London, pp 402–403 Seddon HJ (1975a) Degeneration and regeneration. In: Surgical disorders of peripheral nerves, Chap. 2, 2nd edn. Churchill Livingstone, Edinburgh, pp 9–31 Seddon HJ (1975b) Repair and tension. In: Surgical disorders of peripheral nerves, 2nd edn. Churchill Livingstone, Edinburgh, pp 298–299 Seddon HJ (1975c) Nerve Grafting and kindred procedures. In: Surgical disorders of peripheral nerves, 2nd edn. Churchill Livingstone, Edinburgh, pp 287–302 Seddon HJ (1975d) Homografts. In: Surgical disorders of peripheral nerves, 2nd edn. Churchill Livingstone, Edinburgh, p 299 Seddon HJ (1975e) Results of repair of the nerves. In: Surgical disorders of peripheral nerves, 2nd edn. Churchill Livingstone, Edinburgh, pp 303–314 Seddon HJ, Young JZ, Holmes W (1942) The histological condition of a nerve autograft in man. Br J Surg 29:378–384 Sherren J (1906) Some points in the surgery of the peripheral nerves. Edin Med J 20:217–231
144 Sherren J (1908) Injuries of nerves and their treatment. James Nisbet, London Simpson SA, Young JZ (1945) Regeneration of fibre diameter after cross-unions of visceral and somatic nerves. J Anat 79:48 Songcharoen P, Wongtrakul S, Mahrisavariya B, Spinner RJ (2001) Hemi-contralateral C7 transfer to median nerve in the treatment of root avulsion brachial plexus injury. J Hand Surg Am 26:1058–1064 Sorbie C, Porter TL (1969) Reinnervation of paralysed muscles by direct motor nerve implantation: an experimental study in a dog. J Bone Joint Surg Br 51:156–164 Strange FGS (1947) An operation for pedicle nerve grafting. Br J Surg 34:423–425 Sunderland S (1978) Nerve and nerve injuries, 2nd edn. Churchill Livingstone, Edinburgh Terenghi G, Calder JS, Birch R, Hall SM (1998) A morphological study of Schwann cells and axonal regeneration in chronically transected human peripheral nerve. J Hand Surg Br 23:583–587 Thomas M, Stirratt A, Birch R, Glasby M (1994) Freeze thawed muscle grafting for painful cutaneous neuromas. J Bone Joint Surg Br 76:474–476 Tinel J (1917) Nerve wounds (Revised and edited by Joll CA). Ballière Tindall and Cox, London Tower SS (1943) Regenerative capacity of ventral roots after avulsion from the spinal cord. Arch Neurol Psychiatry 49:1–12 Trachtenburg JT, Thompson WJ (1996) Schwann cell apoptosis at developing neuromuscular junctions is regulated by glial growth factor. Nature 379:174 Trumble TE (1991) Overcoming defects in peripheral nerves. In: Gelberman RH (ed) Operative nerve repair and reconstruction. JB Lippincott, Philadelphia, Chapter 36, pp 507–524 Tsao JW, George EB, Griffin W (1999) Temperature modulation reveals three distinct stages of Wallerian degeneration. J Neurosci 19: 4718–4726 Tsuyama N, Hara T (1973) Intercostal nerve transfer in the treatment of brachial plexus injury of root-avulsion type. Proceedings of the
Surgical Disorders of the Peripheral Nerves S.I.C.O.T. Tel Aviv Oct. 9–12 1972. Excerpta Medical International Congress Series No. 291: 348–351 Tuttle H (1913) Exposure of the brachial plexus with nerve transplantation. JAMA 61:15–17 Vizoso AD, Young JZ (1948) Internode length and fibre diameter in developing and regenerating nerves. J Anat 82:110–134 Waikakul S, Orapin S, Vanadurongwan V (1999) Clinical results and contralateral C7 root neurotization to the median nerve in brachial plexus injuries with total root avulsions. J Hand Surg Br 24:556–560 Wallman L, Zhang Y, Laurell T, Danielson N (2001) The geometric design of micromachined silicon sieve electrodes influences functional nerve regeneration. Biomaterials 22:1187–1193 Witte H, Bradke F (2005) Guidance of axons to targets in development and in disease. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, Chapter 21, pp 447–481 Witzel C, Brushart T (2003) Morphology of peripheral axon regeneration. J Peripher Nerv Syst 8:75–76 Yiangou Y, Birch R, Sangeswaram L, Eglen R, Anand P (2000) Sns/ PN3 and SNS2/NaN sodium channel like immunoreactivity in human adult and neonate injuries of sensory nerves. FEBS Lett 467:249–252 Young JZ (1942) Functional repair of nervous tissue. Physiol Rev 22:318–374 Young JZ (1993) First evidence of axonal transport. In: DyckPJ TPK, Griffin JW, Low PA, Poduslo JF (eds) Peripheral neuropathy, 3rd edn. WB Saunders, Philadelphia, p 2 Zachary RB (1954) Results of nerve sutures. In: Seddon HJ (ed) Peripheral nerve injuries. Medical Research Council. Special report series No. 282, pp 354–387 Zachary RB, Holmes W (1946) Primary suture of nerves. Surg Gynaecol Obstet 82:632–651 Zhao Q, Drott J, Laurell T, Wallman L, Lundstrom K, Bjursten LM, Lundborg G, Montelius L, Danielson N (1997) Rat sciatic nerve regeneration through a micromachined silicone chip. Biomaterials 18:75–80
5
Clinical Aspects of Nerve Injury
Clinical aspects: acute injury; penetrating missile wounds; symptoms and signs; importance of recognition of differential affection of fibres of different sizes; recognition of level and depth of lesion; significance of Tinel’s sign; examination of sensibility; quantitative sensory testing; the motor pathways; examination of muscles; late signs of nerve injury; records; signs of regeneration; aids to diagnosis. “Few surgeons see their patients from the beginning to the end. One could well wish that there was no evil in this. But there is much of evil, and not least is the fact that the orthopaedic surgeons work has enormously increased” (Keogh 1917).
In the acute injury the object of the clinician must be to recognise the fact of injury as soon as possible after the event, and later to go on to determine the nerve or nerves affected, the level or levels of injury and the extent and depth of the lesion or lesions. That this is not always easy nor always appreciated is apparent to anyone who has been able to study the records of the medical defense organisations. The history is important: high velocity injury, compound fracture and wounding, accidental, criminal, surgical or all three, are likely to mean that there has been a serious lesion. The use of a knife, often enough in the hand of a surgeon, is an indication that a nerve is likely to have been partly or completely severed. Advice from witnesses or emergency paramedical staff is always valuable. Potentially life or limb threatening injuries complicate closed traction lesion of the supraclavicular brachial plexus in at least 20% of cases. Even more patients with injuries to the lumbo-sacral plexus are so threatened. The subclavian artery is ruptured in 10% of complete lesions of the brachial plexus and in as many as 30% of cases of violent traction injury of the infraclavicular portion of the brachial plexus. The incidence of arterial lesion is high after fracture dislocations of the shoulder and elbow, higher still after fracture dislocations of the knee. It is important always to search with diligence for occult injuries to the head, the spine, the chest, the abdomen and pelvis before embarking upon treatment of the nerve lesion, both at the first hospital but also after transfer to another Unit (Fig. 5.1). The site and nature of the wound or wounds must be observed. In closed injuries the presence of swelling and bruising may give some indication of severity (Novak and Baratz 2006). In all cases of limb injury the adequacy of perfusion as judged by the state of the pulses, by colour and by temperature must be observed. Indications of associated
fracture must be sought (Fig. 5.2). In cases of serious injury, special attention has to be given to the patient’s general condition as shown by colour, pulse, blood pressure, respiration and other indicators. We follow Rank et al. (1973) in distinguishing between “tidy” wounds as from a knife and “untidy” wounds, such as those seen in open fracture. Soft tissue damage is worse in the latter; nerves and vessels are often
Fig. 5.1 This motor cyclist struck his shoulder against a traffic bollard. There is bruising and swelling of the left shoulder, neck and upper arm. Total avulsion.
R. Birch, Surgical Disorders of the Peripheral Nerves, DOI: 10.1007/978-1-84882-108-8_5, © Springer-Verlag London Limited 2011
145
146
Surgical Disorders of the Peripheral Nerves
Fig. 5.2 The Platt lesion. A 64 year old woman avulsed the fibular styloid standing up from a chair. The common peroneal nerve was ruptured.
Fig. 5.3 Bullet wound of divisions of the sciatic nerve in the thigh. The common peroneal nerve (above) was transected and repaired (neurotmesis). The tibial nerve (below, sling) recovered spontaneously (axonotmesis).
subjected to traction. In penetrating missile wounds it is important to distinguish between the shot gun, the hand gun or rifle and the fragment (Fig. 5.3–5.5). The immensely destructive effect of a close range shotgun injury is much more than that of wounds from more distant discharge. This fact was recognised by Stewart (Stewart and Birch 2001) who classed penetrating missile wounds of the brachial plexus into high energy transfer (HET) and massive energy
Fig. 5.4 Close range shot gun blast to the posterior triangle of the neck. There was rupture of the first part of the subclavian artery.
transfer (MET) injuries. Wounds from bomb fragments and shell splinters have some of the characteristics of shotgun injury, though on a larger scale and with a higher velocity. The International Committee of the Red Cross (ICRC) wound classification, described by Coupland (1993), has enhanced the understanding of the treatment of war wounds. Coupland writes “the surgical task presented by any wound
Clinical Aspects of Nerve Injury
147
Fig. 5.5 Penetrating missile wound. A military rifle bullet entered in the front of the shoulder, passed through the quadrilateral tunnel, destroying the circumflex nerve and posterior circumflex artery and exited posteriorly destroying the infraspinatus and much of the overlying skin.
depends on the wound severity i.e. the degree of tissue damage, and also the structure (s) that may have been injured. Recognition of this demands a clinical classification of wounds that is based on the features of the wound and not on the weaponry or the presumed velocity of the missile.” The system scores certain features of a wound: the maximum diameter, in centimeter, of the entry (E) and of the exit(X) wounds; the presence and the size of the cavity (C); the presence of a fracture (F) and the extent of comminution of that fracture; injury to a vital structure which may be the dura, the
pleura, the peritoneum, or a major vessel; and the retention of metallic fragments (Tables 5.1 and 5.2). The wounds are graded according to the amount of tissue damage by the E, X, C, and F scores into low energy transfer, high energy transfer and massive wounds, and then typed according to the structures injured. The wound is then placed in 1 of 12 categories by grade and type. We are indebted to Colonel Michael Stewart FRCS and his colleagues in the RAMC, Major K Brown, Major W Eardley FRCS, Major A Ramasamy FRCS, and for their extensive advice about the
Table 5.1 The red cross wound classification (Drawn from Coupland 1993). 1. The Field Wound Score
2. Subsequent Analysis. (A) Extent of Tissue Damage
E
Entry
diameter of entry wound, in centimeter
Grade 1
E + X < 10: CO, F0 or F1
Low energy transfer
X
Exit
diameter of exit wound, in centimeter
Grade 2
E + X < 10: c0 or F2
High energy transfer
C
Cavity
Grade 3
E + X > 10: c0 or F2
Massive energy transfer
F
V
M
Fracture
Vital Structure Metallic body
0
wound cavity too small to take two fingers
1
Wound cavity admits two or more fingers
0
No fracture
1
Simple fracture
2
Comminuted fracture
0
No injury to brain, vessels or viscera
1
Injury to the above
0
No metal fragment seen on radiograph
(B) Type Of Wound According To Structure Injured
ST
F0 and V0
F
F1 or F2 and V0
1
One fragment
V
F0 and V1
2
More than one fragment
VF
F1 or F2 and V1
148
Surgical Disorders of the Peripheral Nerves
Table 5.2 The 12 categories of wounds (Drawn from Coupland 1993). Grade I Grade II Grade III ST
Small simple wounds
2ST
3ST
F
1F
2F
3F
V
1V
2V
3V
VF
1VF
2VF
3VF Large wounds threatening life or limb
Fig. 5.6 Military rifle bullet, before debridement (above) and after debridement (below). Type 3F wound: E3, X8, C2, F2. The common peroneal nerve recovered (axonotmesis) (Courtesy of Major K Brown RAMC).
use of the IRCR classification in battle casualties and also for their providing numerous examples. They point out that the classification is somewhat defective in the analysis of severe blast injuries (Figs. 5.6 and 5.7).
5.1 Associated Symptoms and Signs The early symptoms of acute nerve injury are abnormal sensations, alteration or loss of sensibility, weakness, motor paralysis, impairment of function and sometimes pain.
Sometimes the patient is aware of warming and dryness of all or part of an extremity. Abnormal sensations or paraesthesiae, most often “pins and needles,” are usually associated with continuing noxious process. They are of course commonly associated with recovery of a transient conduction block. Their mechanism in that process was ingeniously explored by Merrington and Nathan (1949). Increasing pain and deepening of the nerve lesion signifies continuing noxious process. The patient’s failure to observe warming and anhidrosis is, regrettably, often shared by the examining clinician (Fig. 5.8). Pain, though not always present, is a most important symptom. It may be immediate or delayed; episodic or continuous; of all grades of severity. Its occurrence after injury often means that the noxious process is continuing, as when a nerve is stretched over a bony projection, compressed by a hard object or constricted by, for instance, a suture. It is a regular feature of injury caused by critical ischaemia. Such pain is neurostenalgia. It indicates not only that the noxious process is continuing but also that the lesion of the nerve may be deepening and it is an important indication for operation. Neuropathic pain is never easy to recognise in the injured patient who is probably confused, distressed and in pain. It can be distinguished from the pain of fracture or dislocation by loss of sensation, by painful, spontaneous sensory symptoms, (dysaesthesiae), expressed in the territory of the nerve, and by lancinating or shooting pain radiating into the distribution of that nerve. In some patients neuropathic pain is so severe that it overwhelms the pain from a fracture. Mothers may advise the clinician that the pain is worse than that of child birth. A constant crushing, bursting or burning pain in the otherwise undamaged hand or foot indicates serious and continuing injury to major trunk nerves. Progression of sensory loss with a deep bursting or crushing pain within the muscles of the limb, often accompanied by allodynia indicates impending critical ischaemia until proven otherwise. Two rare and easily recognizable forms of severe pain quite often begin soon after injury: they are (1) the pain associated with intradural damage to nerve roots, particularly in traction lesions of the brachial plexus, and (2) causalgia. These, and related matters are discussed in Chaps. 9 and 12. Examination should enable the clinician to extend the knowledge afforded by the history and the narrative of symptoms to permit accurate diagnosis to be made. All findings should be recorded in such a manner that the record will be intelligible later not only to the examiner but also to others. Unfortunately, the signs of acute nerve injury have to be sought at a time when the patient may be the least able to co-operate in an examination; soon after wounding, when there is likely to be distress and when the general condition may be affected by loss of blood and other injuries. The examination often has to be done in the often unfavorable
Clinical Aspects of Nerve Injury Fig. 5.7 A massive blast injury to the buttock and thigh. E more than 20, X0, C1, F0, V0. The profunda femoris artery was intact, the sciatic nerve recovered (axonotmesis) (Courtesy of Major K Brown RAMC).
Fig. 5.8 Sympathetic paralysis seen within a few days of transection of (left) the median nerve at the elbow and (right) the tibial nerve in the thigh.
149
150
Surgical Disorders of the Peripheral Nerves
Figures 5.9–5.32 show the sensory loss after transection, rupture, or avulsion of spinal and peripheral nerves Fig. 5.9 Complete avulsion of C4–T1 in two patients.
surroundings of an accident department. The patient may be a distressed child; an older child, an adolescent or an adult patient who may be affected by drink or drugs or by both. When the lesion has been inflicted by a surgeon or anesthetist, the patient’s response is likely to be distorted by post operative pain, by the effects of recent general anaesthesia or by sedative or analgesic drugs. The patient’s ignorance of medical process and even his or her trust in “the doctor” may lead him or her to think that pain and paralysis after operation is just something to be expected. That faith or even deference may inhibit him or her from voicing a complaint. These are no conditions for a quiet and comprehensive “neurological examination,” yet this is the time when the fact of nerve injury must at least be recognised if the best result is to be obtained from treatment. The examiner should at all times bear in mind that if there is a wound over the line of a main nerve and if there is any suggestion of loss of sensibility or impairment of motor function in the distribution of that nerve, it must be regarded as having been cut until and unless it is proved otherwise. Sensory loss is determined by response to light touch and pin prick and if circumstances permit, the patient outlines the area of sensory loss which is marked by a black skin marker pen. The surrounding zone of incomplete sensory loss can be similarly marked in red, and the limb then photographed (Figs. 5.9–5.32). Selected muscles are examined. The patient lying supine is usually able to demonstrate
Fig. 5.10 Sensory loss in avulsion of C5-T1. C4 innervates the skin of the outer aspect of the shoulder; T2 innervates the skin of the inner aspect of the arm.
Clinical Aspects of Nerve Injury
151
Fig. 5.11 Avulsion of C4, C5, C6, C7.
Fig. 5.13 Avulsion of C5. The area of sensory loss is usually larger than that seen after rupture of the circumflex nerve (see Fig. 5.19).
Fig. 5.12 In this patient with T6 spinal cord lesion and on clinical examination, C5 and C6 lesion, operation showed avulsion of C5, C6 and C7.
activity in serratus anterior by lifting the shoulders away from the couch, by “forward shrugging.” A radiograph of the chest will detect, inter alia, elevation of the hemi diaphragm. It is usually possible to observe the presence of flexion and abduction at the shoulder, flexion and extension of the elbow and wrist and flexion and extension of the fingers. The radial, median and ulnar nerves are tested by asking the patient to form an “O” between the thumb and little finger, to give the “thumbs up,” and to open and close the fingers like a fan. It should be possible, by gentle persuasion, to observe active flexion and abduction at the hip,
extension at the knee, and extension and flexion at the heel and toes. The palmar and plantar skin is scrutinised for changes in colour and in sweating. Although this may be more difficult in pigmented skin such changes are detectable. The standard tendon reflexes are examined. A more detailed examination is possible when the patient’s condition is stable, and when pain has been controlled. Limb dominance, occupation, marital status, underlying disease or continuing medication are recorded if this has not already been done. Neuropathic pain is by now somewhat easier to recognise, for this is less responsive to analgesics than is pain from skeletal injury.
5.2 Recognition of the Level and the Depth of Injury In the absence of wounding clinicians should be able to arrive at an accurate diagnosis of the level and depth of a lesion by clinical examination. A sound grasp of the level of the branches of the trunk nerves and of the contribution to those nerves coming from individual spinal nerves is a prerequisite. “Aids to Examination of the Peripheral Nervous System,” originally produced by the Medical Research Council and now in its
152
Surgical Disorders of the Peripheral Nerves
Fig. 5.14 Rupture of C5 and C6. Sensory loss does not extend to the thumb and the index finger.
Fig. 5.16 The area of sensory abnormality after section of the lateral cord in the axilla. As in Fig. 5.15, there was no complete loss of sensation. Fig. 5.15 Transection of C8 and T1. The area of sensory disturbance extends into the arm (medial cutaneous nerve of arm).
fourth edition under the direction of Michael O’Brian (2000), is essential reading. This slim volume should be in the possession of all doctors engaged in injury work. It easily fits into a pocket, but now that white coats have been abolished perhaps nurses and therapists who, of course, continue to wear their uniforms, might be invited to carry the volume.
To take one example, the level of injury to the posterior cord and the radial nerve can be determined by examining teres major (inferior subscapular nerve), latissimus dorsi (thoraco dorsal nerve), and deltoid (circumflex nerve). The nerves to the long head of triceps leave the main trunk proximal to the spiral groove. Those innervating the medial head of triceps pass away from the radial nerve at the entrance to and in the first part of the spiral groove whereas those innervating the lateral head of the muscle leave the main nerve
Clinical Aspects of Nerve Injury
still more distally. Paradoxically, the contribution from the spinal nerves is in reverse order: the medial head is usually innervated by the eighth cervical nerve, the long head by the seventh cervical nerve and the lateral head by the sixth cervical nerve. The nerve to brachioradialis consistently passes away from the trunk about three finger breadths above the lateral epicondyle; the nerve to extensor carpi radialis longus
Fig. 5.17 Rupture of the posterior divisions of the trunks of the brachial plexus deep to the clavicle.
Fig. 5.18 Rupture of the medial cord in the axilla. Loss of sensation was confined to the forearm and hand. There is sympathetic paralysis in the ulnar three fingers.
153
comes off about a centimeter more distally. One nerve to extensor carpi radialis brevis leaves the main nerve about 1 cm above the lateral epicondyle and another at the level of the branching into superficial radial and posterior interosseous nerves. Lesions of the sciatic nerve are often, incorrectly, placed at the knee, to the common peroneal nerve. These errors are prevented by examining gluteus medius, gluteus maximus and biceps femoris. Some of the most serious mistakes in the diagnosis and treatment of patients with injured nerves are made because the examiner fails accurately to assess the depth of injury, failing to distinguish between degenerative and non-degenerative injury and to estimate the extent in the nerve of each type of lesion. Some atavistic urge seems to cause clinicians to play down the severity of nerve injury. Perhaps beneath this urge there is a feeling that if there is a serious injury, much hard and possibly unrewarding work is going to be required. The tendency is of course particularly marked in cases of closed injury and of injury during operation. Too often the mantra “Neuropraxia” is pronounced: too often the soothing words “just some bruising of the nerve” are uttered. The diagnosis of the depth of the injury depends on the history and signs and on the simplest electrical examination. Serious injuries are likely to cause serious lesions of nerves. Severance of a nerve with a cutaneous sensory component will lead to well-defined loss of sensibility and to complete motor, sudomotor and vasomotor paralysis in the distribution of the nerve. Simple conduction block is likely to produce a patchy loss of sensibility and a patchy motor loss.
154
Surgical Disorders of the Peripheral Nerves
Fig. 5.20 Rupture of circumflex and musculocutaneous nerves. Fig. 5.19 Rupture of the circumflex nerve.
Further, it is likely to bear more heavily on the large axons than on the small ones: vibration sense and sensibility to light touch are likely to be impaired, whereas pain sensibility may be unaffected. If the axons are damaged (degenerative lesion), stimulation of the nerve below the level of the lesion 6 days after injury will not elicit a motor response. (Landau 1953, Gilliatt and Taylor 1959). If the axons are intact (conduction block) stimulation will evoke a motor response. It will, equally, be possible to stimulate and record from the nerve below the level of the lesion. This simple but rarely used test can permit the early recognition of the depth of the lesion. A firm diagnosis of neurapraxia should never be made unless 1 week after the injury stimulation of the nerve below the level of the lesion produces a motor response.
5.3 Signs The early signs of nerve injury are alteration or loss of sensibility, weakness or paralysis of muscles; vasomotor and sudomotor paralysis in the distribution of the affected nerve or nerves, and abnormal sensitivity over the nerve at the point of injury. For the reasons given, testing of sensibility is often difficult soon after wounding, or when nerve injury is associated with fracture of a long bone. In addition, it may even be
that for a few hours there is conduction across a clean transection without retraction of nerve ends (Smith and Mott 1986). The actions of some muscles can be simulated by the actions of others, so that the fact of paralysis can be missed in the early stages after nerve injury. However, one almost infallible sign is always present in the first 48 h after deep injury of a nerve with a cutaneous sensory component: because of the affection of small as well as of large fibres, the skin in the distribution of the affected nerve is warm and dry. (Bonney 1983, Birch 1986) In the small child, there may be an abnormal posture of the denervated digits (Fig. 5.33). Canale et al. (2000) describe a sign for anterior interosseous palsy in children. When the child is asked to bend the index finger they do it with the other hand. Other early signs which indicate a deep injury to a nerve include changes in texture of the skin rather like “goose pimples” development of a skin rash, and hypersensitivity surrounding the area of anaesthesia. Another good test for nerve injury in small infants is the “immersion test”: the injured hand or foot is placed, for a few minutes, in warm water. The skin of the denervated digits fails to wrinkle. When there is no breach of the skin and the injury of the nerve is caused by pressure or distortion, there is usually differential affection of fibres. Szabo et al. (1984), working with volunteer human subjects on the effects of an acute rise of pressure on the median nerve in the carpal tunnel, plainly showed that vibratory thresholds were the earliest measure of a decrease in nerve function. The largest fibres were the most susceptible to pressure. The use of vibratory, or
Clinical Aspects of Nerve Injury
155
Fig. 5.21 Sensory loss in two cases of rupture of the musculocutaneous nerve. Both patients were able to supinate the forearm fully; the power of elbow flexion was around 30%.
Fig. 5.22 Two examples of high lesion of the radial nerve. Right: there is early recovery into the wrist extensors after repair.
pressure stimuli as a non-invasive diagnostic test may, as the authors suggest, have a significant role in clinical work and in research. The smaller ones are spared, so that there is rarely any sudomotor or vasomotor paralysis, and delayed pain sensibility is preserved. We have however seen cases in which pressure paralysis was deep enough to affect the autonomic fibres. In one young girl whose attempt at suicide with narcotic drugs led to prolonged pressure on the tibial and
common peroneal nerves, there was prolonged vasomotor and sudomotor paralysis of the foot. In the case of intraneural hemorrhage or injection the situation is different: small fibres are early affected, sometimes, as with injection of local anaesthetic, before the large fibres. When the injected fluid is itself noxious there is early affection of all sizes of fibre. Peripheral ischaemia is usually signalled by pain, but in cases in which the vascular injury is
156
Surgical Disorders of the Peripheral Nerves
Fig. 5.25 Section of the median and the palmar cutaneous nerve at wrist. Fig. 5.23 Transection of median, ulnar, medial cutaneous nerve of forearm, and brachial artery in the arm of a 14 year old boy. Note the extent of skin innervation provided by the superficial radial and lateral cutaneous nerves of forearm. The intact radial nerve permits a sort of grasp.
associated with fracture, the significance of that pain may not be recognised. Ischaemia affects first the large fibres: discriminative sensibility and vibration sense are first affected (MacKinnon and Dellon 1988) It is not easy to test these modalities when ischaemia is developing because of damage to a main vessel associated with a fracture of a long bone, but if action is not taken until superficial sensibility is lost, it will come too late.
5.4 Tinel’s Sign
Fig. 5.24 High median nerve injury: no active flexion of the index finger and thumb.
Tinel (1915, 1917) was probably the first to draw attention to the indication of the “growing point” of the regenerating axons signalled by the production of paraesthesiae by tapping over the course of the nerve. Buck-Gramcko and Lubahn (1993) remind us that Tinel made the observation at about the same time that Hoffman working on the other side of the Western Front, made a similar discovery (Hoffman 1915a, b). Properly, the sign should be called the “Hoffman-Tinel sign.” Tapping the skin over the course of a recovering nerve reveals the presence of regenerating axons by the production of paraesthesiae in the sensory distribution of the nerve. Rothwell’s translation of Tinel (1917) says: “When compression or percussion is lightly applied to the injured nerve trunk, we often find, in the cutaneous region of the nerve, a creeping sensation usually compared by the patient to that caused by
Clinical Aspects of Nerve Injury
157
Fig. 5.26 A typical area of loss of sensibility after division of the median nerve at the wrist, sparing the palmar cutaneous branch.
Fig. 5.27 The area of sensory loss and “clawing” of little and ring fingers after division of ulnar nerve in the forearm.
electricity.... The “formication sign” is thus of supreme importance since it enables us to see whether the nerve is interrupted, or in the course of regeneration; whether a nerve suture has succeeded or failed, or whether regeneration is rapid and satisfactory, or reduced to a few insignificant fibres. All clinicians will recognise this sign. The element of unreliability is introduced by the fact that some of those regenerating axons are not on their way to any target. However, these
points can be stated: (1) a strongly positive Tinel sign over a lesion soon after injury indicates rupture of axons or severance of the nerve; (2) in favorable degenerative lesions (axonotmesis) or after repair which is going to be successful, the centrifugally moving Tinel sign is persistently stronger than that at the suture line; (3) after repair which is going to fail, the Tinel sign at the suture line remains stronger than that at the growing point; (4) failure of distal progression of
158
Surgical Disorders of the Peripheral Nerves
Fig. 5.28 The area of loss of sensibility in two cases of injury to the femoral nerve at, or proximal to the groin crease.
Fig. 5.29 Cutaneous distribution of the sacral plexus. Left showing area of sensory loss after closed fracture/dislocation of sacro-iliac joint. Right showing area of loss of sensibility after open fracture/dislocation of the pelvis. The muscles of the buttock are wasted.
Fig. 5.30 The area of sensory loss after transection of sciatic nerve in the thigh is confined to the leg and foot.
Clinical Aspects of Nerve Injury
159
Fig. 5.33 Infant’s hand 24 h after section of the palmar nerves to the index finger and thumb. Note that the anaesthetic digits are held out from the others. There was no tendon injury.
Fig. 5.31 Area of sensory loss after high division of common peroneal nerve. Note the pressure sore caused by a conventional ankle-foot orthosis.
conduction block for, at least, a significant number of axons within the nerve. It is important always to remember Tinel’s own advice that a positive sign should not be confused with the “hypersensitivity seen in some cases of neuralgia” (Tinel 1917). The Tinel-like sign elicited by percussion over schwannoma or over nerves in the early stages of entrapment neuropathy, such as the ulnar nerve at the elbow or the median nerve at the wrist, does not indicate that axons have been ruptured, rather that nerve fibres have become sensitised because of focal demyelination and changes in the expression of voltage gated ion channels at the level of lesion.
5.5 Eliciting the Tinel Sign in Closed Lesions
Fig. 5.32 Area of sensory loss after interruption of the deep division of the common peroneal nerve. Left, showing the leg of a 29 year old man in whom severe “compartment syndrome” was overlooked after intramedullary nailing of closed fracture of tibial shaft. The anterior compartment was infarcted and it was excised. Right, showing the area of sensory loss after transection of the deep division of the common peroneal nerve by a knife.
the Tinel sign in a closed lesion indicates rupture or other injury not susceptible of recovery by natural process; (5) a positive Tinel sign means the lesion is degenerative, not a
The examiner’s finger percusses along the course of the nerve from distal to proximal starting well below the presumed level of lesion. The patient is asked to say when the advancing finger elicits a wave or a surge of pins and needles or abnormal sensations, which may be painful, into the distribution of the nerve which must be clearly indicated by the examiner. A positive Tinel sign indicates that axons are ruptured at that level. The sign can be regularly elicited on the day of injury in a conscious patient. The level of the sign should be measured from a fixed point and the distance entered into the records. In the upper limb, we use the tip of the coracoid, the medial or lateral epicondyle and the styloid process of the radius and in the lower limb, the tip of the greater trochanter, the tip of the styloid process of the fibula and the lower points of the medial and the lateral malleoli, whichever is appropriate. At times the examination is painful and patients need to be warned about that. Percussion over the swollen posterior triangle of the neck in cases of multiple
160
Surgical Disorders of the Peripheral Nerves
Fig. 5.34 Static and progressing Tinel’s signs. A 43 year old woman sustained a complete and very painful, lesion of the common peroneal nerve from the kick of a horse. There was a strong, painful, Tinel sign at the level of lesion at the time of exploration 10 weeks after injury. The nerve was deeply compressed by scar from which it was removed. Her pain was abolished. The rate of progress of the Tinel sign for the superficial and the deep divisions of the nerve was about 2 mm a day. There was complete recovery.
avulsion usually elicits painful sensory phenomena which do not radiate into the dermatomes of the injured nerves. A strongly positive Tinel sign indicates to the clinician the level of lesion, and also the depth of lesion. Axons are ruptured where the percussing finger evokes radiation of sensations into the distribution of the nerve under examination. Subsequent examinations may provide evidence that the lesion is regenerating by demonstrating centrifugal progression of the sign which becomes progressively stronger at the distal rather than at the proximal level of sign (Fig. 5.34). Kato (Kato and Birch 2006) examined the evolution of Tinel’s sign in 137 nerves injured by closed fracture or dislocation. Recovery was good in 69 of the 86 nerves where the Tinel sign progressed. or where no Tinel sign was present. A static Tinel sign was one of the indications for operation in 43 of 51 nerves. Thirty-two were repaired and the other 19, which were embedded in callus or caught within a fracture or joint, could not have recovered spontaneously. Table 5.3 shows the
value of a static or advancing Tinel sign in predicting recovery in degenerative lesions after closed injury to the common peroneal, the radial and tibial nerves. An advancing sign proved misleading in 18 nerves. In most of these the distal muscles had been damaged by ischaemia so that the regenerating axons arrived at target organs which were irredeemably fibrosed. One case, of a radial nerve damaged by closed fracture of the humeral shaft was initially seen by one of us (RB) who predicted recovery because the Tinel sign was advancing. Marco Sinisi noted that there was discrepancy between the advancing sign and the failure of recovery of brachio radialis and when he exposed the nerve he found that about three quarters of it was embedded in the fracture; the surviving portion was destined for the superficial radial nerve. Tinel’s sign is particularly valuable in the diagnosis of post ganglionic rupture of the spinal nerves of the brachial plexus (Landi and Copeland 1979). If percussion in the posterior triangle induces radiation as far as the elbow then rupture of C5 is likely; rupture of C6 is anticipated when radiation extends to the lateral forearm and thumb and when radiation extends to the whole hand, especially to the dorsum, then rupture of C7 is expected. Table 5.4 summarises the role of Tinel’s sign in the early diagnosis of rupture of the spinal nerves in 100 consecutive cases of closed traction lesions of the supraclavicular brachial plexus. Most patients were examined within 14 days of injury: a positive sign was found in nine cases of lesion to C5 and C6 yet these nerves were found avulsed. Irradiation into the territory of the supraclavicular nerve (C4) was sometimes wrongly ascribed to a rupture of C5 and irradiation from percussion over a rupture of C5 sometimes extends to the forearm suggesting rupture of C6. Nineteen spinal nerves were found to be ruptured even though no Tinel sign was detected. These stumps were concealed by torn muscle or haematoma, and the level of rupture was close to the foramen. Absence of the sign over a peripheral nerve which is not working, suggests conduction block. The absence of the sign in closed traction lesion of the supraclavicular brachial plexus indicates one of two things: that the nerve has been avulsed or the lesion is conduction block. The distinction between these two is usually straight forward.
Table 5.3 Tinel’s sign as a guide to prognosis in 339 consecutive cases of degenerative lesions in closed injuries to the common peroneal (171 cases), the radial (139 cases) and the tibial (29 cases) nerves examined 2000–2007. Tinel Sign – Progressing Tinel Sign – Static Spontaneous Misleading – no, or poor, Spontaneous recovery Rupture or other lesion not suscepRecovery spontaneous recovery tible to recovery by natural process Common peroneal nerve with divisions
84
12
0
75
Radial nerve
103
5
0
31
Tibial nerve
16
1
0
12
TOTAL
203
18
0
118
Clinical Aspects of Nerve Injury
161
Table 5.4 Tinel’s sign in closed traction lesions of the brachial plexus in 100 adult patients examined and operated 2004–2005. Tinel sign absent Spinal Tinel sign present (358 nerves). Findings nerves (142 nerves). Findings at operation at operation Intact
Rupture
Avulsion Intact
Rupture Avulsion
C5
1
58
4
0
10
27
C6
0
41
5
10
6
38
C7
0
24
0
40
2
34
C8
0
4
0
45
0
51
T1
0
TOTAL 1
4
1
52
1
42
131
10
147
19
192
5.6 Tinel’s Sign and Recovery Henderson (1948) made serial observations on Tinel’s sign in 400 cases of nerve injury in his fellow prisoners of war. Henderson thought that the behaviour of the sign provided certain indications of satisfactory, or unsatisfactory, spontaneous regeneration by 4 months after the injury so that “a strongly positive sign at the level of the lesion, which gradually diminished as the sensitivity of the distal part of the nerve increased, progress peripherally, and faded centrally, was a certain indication that satisfactory spontaneous regeneration was in progressed. Poor recovery, attributed to great axonal confusion at the site of injury, was observed in cases where the sciatic lesion remained sensitive and a positive sign developed in the most proximal muscles within the distribution of the damaged nerve.” The situation is much more straightforward for the clinician examining main nerves damaged by closed fractures or dislocations in civilian practice and it is usually possible to distinguish between axonotmesis and neurotmesis by between 4 and 6 weeks from the day of the injury in closed lesions of the radial, median, ulnar, common peroneal and tibial nerves. Ruth Bowden found that clinically detectable recovery was evident by 64 days in 14 cases of axonotmesis of the radial nerve (Bowden and Scholl 1954). An advancing Tinel sign may be found when only a few nerve fibres are regenerating as in cases where trunk nerves are entrapped within a fracture or joint. Useful function will not return if the target organs are damaged by fracture or sepsis or, above all, by ischaemia. An advancing Tinel sign should be treated with caution in lesions of the common peroneal nerve, where some degree of post ischaemic fibrosis of the anterior compartment is found in about one-fifth of cases. Centrifugal progress of the sign is often unreliable in predicting recovery of lesions of the sciatic nerve incurred during arthroplasty of the hip. Most of these are mixed lesions, some nerve fibres are intact, others sustain conduction block whilst many more have sustained degenerative lesions which may or may not
be naturally favorable. Tinel’s sign can be detected over such “motor” nerves, as the posterior interosseous, but it is more difficult to elicit the sign over deep seated nerves such as the circumflex, the eighth cervical or first thoracic nerves. The sign advances more rapidly in degenerative lesions of naturally favorable prognosis (axonotmesis) than it does after repair of a nerve. A rate of progression of about 2 mm a day is usual. Tinel’s sign progresses more rapidly after repair of proximal lesions, particularly so when the repair is done urgently. Progression at the rate of 3 mm a day is by no means unusual after urgent repair of the supraclavicular brachial plexus in the segment from the neck to the upper arm. Thereafter, progress slows. Similar rates of advance are seen after urgent repairs of the sciatic nerve wounded in the buttock. These rates are close to those recorded by Bowden and Scholl (1954). Bowden and Scholl offered a caveat: “the choice of time for exploration of an injured nerve cannot be based on estimates of rates of recovery alone; in each individual case, all clinical factors must be taken into consideration. Where there is any reasonable doubt about the state of the nerve or the integrity of a suture line, operation should not be delayed since in experienced hands exploration is without harmful effect.” The distance between the sign and a suitable fixed bony point is recorded in the case notes at every examination. The rate of progress of the sign in a healing nerve matches that of the slow anterograde axonal transport system.
5.7 Examination of Sensibility Highet proposed a system for the examination of sensibility and muscle power in a memorandum submitted to the Nerve Injuries Committee of the Medical Research Council in 1941 and this, with some modifications, was adopted by that Committee. The Medical Research Council (MRC) method of recording sensibility offers a reasonable method for recording and measuring progress. As is the case with all such schemes it has obvious disadvantages, but no comprehensive method has yet been devised that does not have the overwhelming disadvantage of extreme complication (Table 5.5). So far as possible, sensation to light touch and pin prick, vibration sense and position sense should be tested, and the area of skin affected should be recorded. The timing of the response to pin prick and the ability to localise to the area of stimulation should if possible be noted. Anhidrosis is easily perceptible; vasomotor paralysis is shown by warming of the skin and, in the finger tips, by capillary pulsation. The examiner should use the pulp skin of his or her fingers, not back of the hand, which is, of course, less densely innervated. Our methods: The modalities routinely tested are light touch, temperature, position sense, pain, two point discrimination,
162
Surgical Disorders of the Peripheral Nerves
Table 5.5 Sensory recovery. The original grading by Highet (1941) Stage 0
Absence of sensibility in the autonomous zone of the nerve
Stage 1
Recovery of deep cutaneous pain sensibility within the autonomous zone
Stage 2
Return of some degree of superficial pain and tactile sensibility within the autonomous zone
Stage 3
Return of superficial pain and tactile sensibility throughout the autonomous zone with the disappearance of over-response
Stage 4
Return of sensibility as in Stage 3 with the addition that there is recovery of two-point discrimination within the autonomous zone
The Medical Research Council System by Seddon (1954) S0
Absence of sensibility in the autonomous area
S1
Recovery of deep cutaneous pain sensibility within the Autonomous area of the nerve
S2
Return of some degree of superficial cutaneous pain and tactile sensibility within the Autonomous area of the nerve
S3
Return of some degree of superficial cutaneous pain and tactile sensibility within the Autonomous area with disappearance of any previous over-reaction
S3+
Return of sensibility as in Stage 3 with the addition that there is some recovery of two-point discrimination within the Autonomous area
S4
Complete recovery
localisation and pressure. The stereognosis test has been used since 1975 (Wynn Parry and Salter 1976). Light touch: A wisp of cotton wool is moved lightly across the area under test. Temperature: Metal tubes are used: one contains cold water, the other, warm at about 35°C. These are applied alternately to the area under test. Position sense: All nearby joints, other than the one being tested, must be stabilised. The patient is first shown the direction in which the joint is being moved. Then, with eyes closed, he or she is asked to indicate the direction in which the joint is being moved. Pin prick: A bluntened pin is lightly applied to the skin and the patient is asked to say whether it feels sharp or blunt. Two point discrimination: This is done with the blunted points of a compass or the ends of a paper clip or with a special device. The patient is first instructed: “I shall touch your finger now with one point; now with two. If you feel one, say “one”; if you feel two, say “two”; if you are in doubt, say “one.”” Then, with closed eyes, the patient attempts distinction between one and two points. Two point discrimination is valuable because it indicates the degree of reinnervation of slowly adapting receptors. Of course, Highet (1941) used the test in his system of sensory grading. It is subject to limitations. The patient easily gets confused; it is difficult or impossible to ensure that the same
pressure is used throughout the test. Brand (1985) showed that even skilled clinicians used, during the test, pressures varying from 4 to 40 g/unit area. Such variability leads, of course, to error when results of different observers are compared. Bell and Burford (1982) using transducers and oscilloscopes, found that the difference between the pressure applied to one point and that applied to two easily exceeded the resolution threshold for normal sensitivity. They concluded that two point discrimination had poor validity. Refinements of the method were introduced by von Greulich (1976) and by MacKinnon and Dellon (1985). These were based on the use of as ring with circumferential prongs set at varying intervals. Useful instruments had, of course, been developed previously by Mannerfeldt. Dellon (1978) introduced the concept of moving two-point discrimination. The patient is asked to move his or her finger across a number of ridges separated by varying intervals. He or she then identifies the shortest interval appreciated, which is invariably less than that recorded by the static test. We think that this ingenious method is, to a certain extent, artificial since normal sensibility requires handling of an object between thumb and finger – a process that excites a far greater range of sensory receptors. However, Novak and MacKinnon (1999) found that a high degree of inter observer reliability between moving two point discrimination, static two point discrimination, and Semmes Weinstein filament tests. Localisation: We use the chart devised by Wynn Parry and Salter (1976) in which the hand is divided into different numbered areas. The blindfolded patient is asked to point to the area being touched, and this is recorded on the card. Thus, if a touch on the tip of the index finger is felt as a touch on the base, the number of the former area is recorded on the latter area on the chart. The localisation and texture and shape recognition charts were brought to the Peripheral Nerve Injury Unit by Christopher Wynn Parry when he joined the staff at the Royal National Orthopaedic Hospital as Director of Rehabilitation in 1975 (Figs. 5.35 and 5.36). Pressure: The padded blunt end of a pencil is used to indent the skin only lightly. We also use Von Frey hairs and the weighted pins (5 and 10 g) developed by Ruth Bowden. Recognition of textures and shapes: Riddoch (1940) asked patients to recognise different coins by touch. Porter (1966) asked patients to recognise steel type letters, in dimension 1.0 by 0.8 cm. The letters must be identical in reverse and H, O, U, V, Y are used. The “pick up” test (Moberg 1958) is, of course, a measurement of function as well as of sensibility. So is the assessment of ability to recognise objects, the importance of which we believe was first recognised by Wynn Parry and Salter (1976) and used at RAF Chessington and then subsequently in the PNI Unit at the Royal National Orthopaedic Hospital(Wynn Parry 1981). The blindfolded patient is presented with a series of objects of differing shape, texture and surface character, and asked to distinguish them. The number correctly identified and the
Clinical Aspects of Nerve Injury
163
Fig. 5.35 Chart recording improvement of the measurement of ability to recognise and to localise at intervals after repair of median nerve at wrist.
Fig. 5.36 Another example of recovery of recognition and localisation after repair of median nerve.
time taken are recorded. Later, common textures and small and large objects in daily use are presented for recognition. Rosen and Jerosch-Herold (2000), Jerosch-Herold (2000, 2005) thought that the shape and texture identification test may provide a better record of regeneration than those for two point discrimination.
5.8 Quantitative Sensory Testing Dyck et al. (2005) classified methods of measurement of different modalities of sensation into three groups. Semi quantitative sensory tests include the usual techniques which are useful in traumatic neuropathy where there is a
164
normal limb available for control. Quantitative sensory testing (QST) is the use of stimuli which are more precisely quantitative and more rigorously tested but the environment is not controlled. In laboratory QST all aspects of the examination are predetermined, standardised and controlled. Our experience is drawn from the work of Praveen Anand who has examined many hundreds of patients at our joint clinics over the last 20 years, first at the London Hospital and later at Imperial College, London (Anand et al. 1994, 1996) These investigations confirm the presence and the extent of axonopathy which cannot be detected by standard neurophysiological investigation. Focal, or generalised, neuropathy has been demonstrated in patients wrongly labelled as “complex regional pain syndrome Type 1” (CRPS Type 1) or even as malingerers. Our findings in some patients diagnosed with “CRPS Type 1” provide support for the critique from Oaklander et al. (2006). The differential susceptibility to the lesion between different groups of nerve fibres have been confirmed. The rates of recovery of different populations of nerve fibres have been analyzed in many cases of favorable, degenerative lesion (axonotmesis) and after repair. Profound differences between the infant and the adult have been revealed after repair of lesions of the brachial plexus (Anand and Birch 2002, Htut et al. 2006). The general assumption that the smallest fibres regenerate more rapidly has been called into question and we have shown that reinnervation of the skin lags far behind reinnervation of skeletal muscle after repair of the brachial plexus in the adult. The methods used are now outlined. Thermal threshold: A thermal threshold testing system (Somedic, Stockholm, Sweden) with a rate of rise in temperature of 1°C/s, is used. The small thermode (15 × 25 mm) is placed at the appropriate site, and the base line temperature is set at a neutral point between 30°C and 32°C. Thermal threshold is determined for cool and warm sensation by four separate consecutive tests for reach modality. The mean difference from base line temperature is recorded as threshold. The patient indicates their recognition of the stimulus by pressing a button. No significant difference has been seen by using either verbal or manual indication of change. “Abnormal” (more than two SD above the mean) values in children are: warm sensation greater than 3.8, cool sensation greater than 2.3; for adults aged between 20 and 30 years, abnormal values are: warm sensation greater than 3.9 and cool sensation greater than 2.6 (Fig. 5.37). Light touch: Thresholds are determined using Semmes Weinstein hairs which are made by A. Ainsworth, University College, London, UK. The number of the hair reliably detected (three or more out of five trials) with the lowest force is recorded and values are then transferred into respective gram values. A threshold to light touch greater than that elicited by the number three filament (0.0479g) is abnormal (Fig. 5.38).
Surgical Disorders of the Peripheral Nerves
Fig. 5.37 Testing of warm and cool thermal thresholds (Courtesy of Dr. Peter Misra).
Fig. 5.38 Monofilament perception to assess punctate touch (Courtesy of Dr. Peter Misra).
Vibration sense: The threshold is measured with a biothesiometer (Biomedical Instrument Company, Newbury, Ohio, USA) placed at the distal interphalangeal joint of a
Clinical Aspects of Nerve Injury
Fig. 5.39 Measurement of vibration perception thresholds (Courtesy of Dr. Peter Misra).
digit or over bony prominence in more proximal joints. Three ascending and three descending trials are carried out and values averaged. In children, an abnormal value is more than 8 V; in adults, it is more than 10 V (Fig. 5.39). These are subjective tests, they depend on the patient’s own perception of stimulus. Other methods are used to measure function in the post ganglionic sympathetic efferent fibres and the histamine induced flare response mediated by the axon reflex. Sweating: This is measured in the palm of the hand using an evaporimeter (Servomed, Stockholm, Sweden) in grams per square meter per hour. The instrument has two sensors which measure the relative humidity in an open cylinder at different distances from the skin surface and signals derived from these transducers are computed to provide first, the partial pressure of the water saturation gradient and then the evaporation rate. A value less than 50% of the contralateral hand is considered abnormal (Fig. 5.40). The Histamine induced flare response: The flare induced by the intradermal injection of 0.03 ml of histamine 1 mg/ml is measured by laser Doppler fluxmetry (Fig. 5.41). We have investigated conduction in smaller nerve fibres and in the motor pathways by techniques which supplement classical neurophysiological investigative methods. Contact heat evoked potential stimulator (CHEPS) enables studies of conduction within small sensory fibres (Ad and C fibres). The CHEPS machine (Medoc Ltd., Ramat Yishai, Israel) rapidly stimulates cutaneous small nerve fibres, and resulting evoked potentials can be recorded from the scalp. The machine can be put into a MR scanner so that functional MR images are available. Patients with symptoms of sensory neuropathy have been studied using CHEPS, and the findings compared to other objective measures of small nerve fibres such as the histamine-induced skin flare response and intra-epidermal fibres counts (IEF). CHEPS has been compared with the results of quantitative sensory testing (QST)
165
Fig. 5.40 Quantitative sweat testing (Courtesy of Dr. Peter Misra).
Fig. 5.41 Histamine induced flair assessment using laser Doppler scanning (Courtesy of Dr. Peter Misra).
(Atherton et al. 2007). Amplitudes of Ad evoked potentials (mV) following face, arm or leg stimulation were reduced in such patients. The reduced leg skin flare responses correlated
166
Surgical Disorders of the Peripheral Nerves
the level of brachial plexus or spinal cord injury, and in the analysis of the variable susceptibility of different populations of nerve fibres (Figs. 5.42 and 5.43). Conduction in somatic efferent pathways: Transcranial magnetic stimulation (TMS) or transcranial electromagnetic evoked potentials (TCEMEP) is a technique in which the motor cortex is stimulated by a brief electromagnetic stimulus delivered through a hand held magnetic stimulating circular coil. We use single pulse TMS with the Magstim®200 Monophasic stimulator (Novametrix Medical Systems Ltd., Whitland, UK) to assess motor pathways (see Htut et al. 2007). A high power 90 mm coil is placed over the head at the area over the motor cortex or over the dorsal skin of the neck. The evoked motor response can be detected by electrodes placed over the relevant upper limb muscles and the motor action potentials recorded on an EMG machine. Facilitation is used whenever possible; the patient is asked to attempt to contract the muscle from which the recording is made, as this can improve the amplitude of the recorded motor response. Responses from the injured and the contra-lateral (intact) limb can be compared (Fig. 5.44).
with leg Ad amplitudes. In leg skin biopsies, IEF counts were reduced and these also correlated with Ad amplitudes. CHEPS appears to be a sensitive measure, with abnormalities observed in some symptomatic patients who did not have significant IEF loss and/or QST abnormalities. CHEPS thus provides a clinically practical, non-invasive and objective measure and it has proved to be a useful tool for the diagnosis of sensory small fibre neuropathy, in the assessment of
5.9 Examination of Muscles 5.9.1 Some Pitfalls
Fig. 5.42 Contact heat evoked potential studies. The skin is stimulated by a thermode (in this case placed over the forearm) and the evoked potentials are recorded and averaged using an electroencephalographic (EEG) system with appropriate electro-placements and filtering (Courtesy of Professor Praveen Anand and Dr. Peter Misra).
Trick movements. Accuracy of assessment may be impaired by failure to recognise substitution of the action of paralyzed
Cz [uV]
Cz [uV]
−35
−30
−30
−25
−25
−20
−20
−15
−15
−10
−10
−5
−5
0
0 5 5 10 10
15
15
20
20
25
Fig. 5.43 Findings from contact heat evoked potential stimulation. The evoked potentials recorded from electrodes placed on the scalp.
25
30
30
35 40 0.0
0.2
0.4
0.6
0.8
1.0
(s)
35 0.1
0.2
0.4
0.6
0.8
1.0
(s)
Clinical Aspects of Nerve Injury
Fig. 5.44 Magnetic stimulation of, in this case, the brain demonstrates central and peripheral motor integrity and conduction times. Stimulation can also be applied to the neck so that the motor cortex and the motor roots are magnetically stimulated and recordings are made from limb muscles (Courtesy of Dr. Peter Misra).
muscle by the “trick” movement. Such movements are, of course, useful for the patient in providing function when the primary movers are permanently paralyzed. They may be produced by direct substitution, by an accessory insertion, by the “tenodesis effect” or by rebound. It is important for the examiner to palpate the belly of the muscle under examination and at the same time, to palpate the tendon of that muscle. Direct substitution: Most patients with paralysis of the deltoid muscle can, when the spinati and the rotator cuff are preserved, abduct the limb at the gleno-humeral joint. Even the clavicular part of pectoralis major and the long head of triceps can affect abduction when the humerus is rotated laterally by the action of the infraspinatus. When the biceps brachii and the brachialis muscles are paralyzed, the action of the brachioradialis alone often suffices to flex the elbow to, at least, power MRC Grade 4. Of course, the power of gravity suffices to extend the elbow in the absence of triceps. We have seen how a kind of opposition of the thumb can be produced by action of the flexor brevis and abductor longus when the thenar muscles supplied by the median nerve are paralyzed. Strong extension of the fingers can give an impression of an abducting action in the interossei, whilst strong flexion can give the appearance of an adducting action. The interosseous muscles extend the proximal interphalangeal joints whilst the metacarpophalangeal joints are flexed. Unfortunately, there is no substitute for the abductors of the hip. Lesions of the gluteal nerve leave the patient with serious defects in stability of the pelvis and in their gait. Accessory insertion: The abductor and flexor brevis pollicis have insertions to the extensor expansion, so that abduction of the thumb extends the interphalangeal joint, even when the extensor muscles are paralyzed.
167
“Tenodesis” action: When the long flexors of the fingers are paralyzed, extension of the wrist produces in them sufficient tension to cause flexion of the interphalangeal joints. It is this action that is exploited in the tenodesis of paralyzed finger flexors to give a grip when the wrist is extended. Rebound: When the antagonist to a paralyzed muscle contracts strongly and relaxes quickly it may appear as a contraction of paralyzed muscles. In paralysis of the common peroneal nerve, the patient can mimic active extension of the toes or active extension at the ankle by strong contraction and sudden relaxation of the flexors. Anatomical variations: Some of these have been described in Chap. 1. T1 provides functional extension of the digits in at least 10% of patients. Powerful extension of the digits is seen in more than one-third of patients in whom C5, C6 and C7 have been ruptured or avulsed. The ulnar nerve frequently sends a branch to the distal medial head of triceps, the radial nerve consistently innervates part of the brachialis and there are commonly variations in the innervation of the small muscles of the hand. Measurement of muscle power: No system for recording of motor power has really superseded that proposed in 1941 by Highet (1954) to the Nerve Injuries Committee of the Medical Research Council (Table 5.6). The scale is non linear. Sharrard (1953) estimated that a lower limb muscle graded as MRC 4, possessed about 40% of normal strength and that a grade 3 muscle was about 15% as strong as normal. Trumble et al. (1995) made objective measurements of muscle power by the use of force transducers; they used the results to “quantitate” the MRC Grades. Thus, they reckon Table 5.6 Motor recovery. The original grading proposed by Highet (1941) Stage 0
No contraction
Stage 1
Return of perceptible contraction in proximal muscles
Stage 2
Return of perceptible contraction in both proximal and distal muscles
Stage 3
Return of function in both proximal and distal muscles to such an extent that all important muscles are of sufficient power to act against resistance
Stage 4
Return of function as in Stage 3 with the addition that all synergic and isolated movements are possible
Stage 5
Complete recovery
The Medical Research Council System by Seddon (1954) M0
No contraction
M1
Return of perceptible contraction in the proximal muscles
M2
Return of perceptible contraction in both proximal and distal muscles
M3
Return of perceptible contraction in both proximal and distal muscles of such degree that all important muscles are sufficiently powerful to act against resistance
M4
Return of function as in Stage 3 with the addition that all synergic and independent movements are possible
M5
Complete recovery
168
that Grade M3 represented 17–42% of the function of the healthy muscle, and that Grade M4 represented 66–79%. The use of this technique is, unfortunately, too time consuming for use in most clinical work. Mills (1997) studied the first dorsal interosseous muscle in cases of amyotrophic lateral sclerosis. He found a linear relationship between the degree of wasting and the amplitude of the compound motor action potential (CMAP). That relation was exponential between the MRC Grade and the CMAP: “in effect, a reduction of one point on the scale is associated with, on average, halving of the CMAP amplitude.” Kaufmann (2005) comments:“the proportion of maximum strength required to overcome gravity is markedly different between muscle groups. Therefore manual muscle testing has limited value, particularly in the lower extremity.” Muscles work at about 25% of normal strength during walking, which is more or less equivalent to an MRC Grade of 3+ (Perry et al. 1986). The considerable reserve permits more strenuous activity and provides endurance. The assessment of motor function is extended by methods used to determine the stamina of muscles. The simplest method is to record with a stamina gauge or spring balance the length of time during which the subject can maintain contraction as a standard percentage of his or her maximal strength. Or, repeated contractions are made at maximal strength and the number that can be made before amplitude falls to 50% of the original, is recorded. A standard dynamometer or a sphygmomanometer with an aneroid or mercury gauge may be used. It is preferable to use a size of bulb to fit the size of the hand.
5.9.2 Our Methods We use the instruments devised by Mannerfeldt and made by HC Ulrich (Ulm) (Birch 1989; Birch and Raji 1991, Dunnet et al. 1995). The power of pinch grip is reduced by about one-third in low median palsy, and by nearly three quarters in low ulnar palsy. Power grip is reduced by about one-half in high ulnar palsy and it is as low as 20% in radial palsy, such is the importance of extension of the wrist (Fig. 5.45). A myometer (model D60107MK1. Penny and Giles Transducers, Christchurch, Hampshire) is used for the examination of more proximal muscles. For the shoulder and arm, the patient is seated comfortably with their back erect against the upright of a chair, both upper limbs are held in the same position. The examiner applies force against the arm using the appropriate cup. The amount of force required to overcome the patient’s resistance is noted and recorded as a percentage of the opposite limb. For hip flexion, the patient lies supine, on their side for abduction and for extension prone. Power of extension of the knee is best
Surgical Disorders of the Peripheral Nerves
measured with the patient sitting with the legs over the side of the couch (Fig. 5.46). Although many patients with isolated paralysis of deltoid show a complete range of active movement at the shoulder, the power of forward flexion and abduction is reduced to about 40% of the uninjured side. The power of extension of the shoulder, measured at 90° of abduction, is reduced to as little as 5%. The power of abduction after a “good” result of repair of the circumflex nerve reaches about 60% of the uninjured side. The power of elbow flexion after musculocutaneous palsy is reduced to between 20 and 40% of the uninjured side. It approaches 60–80% of normal after successful repair of the nerve. The power of dorsiflexion of the ankle after “good” results of repair of the common peroneal nerve is around 50% of normal, that of extension of the knee after successful repair of the femoral nerve, about 60% of normal. Although these figures fall short of normal, they are, of course, far superior to the power restored by muscle transfers.
5.9.3 Clinical Examination It is with the large proximal muscles, about the shoulder girdle and about the hip, that serious mistakes are most common. Delay before diagnosis of nerve injury is, in many cases, quite alarming despite the reliability of precise but elementary clinical examination. Thoraco scapular, thoraco humeral, and scapulo humeral muscles. Narakas (1993) described a simple clinical measure, the inferior scapulo-humeral angle (ISHA), which is very helpful in the analysis of injuries to the nerves to these muscles. The inferior scapulo-humeral angle is subtended by the long axis of the humerus and the lateral border of the scapula. The tip of that angle is centred over the glenohumeral joint. It is measured at rest and then with the arm in full active elevation. It measures the range of elevation at the glenohumeral joint (Fig. 5.47). Narakas acknowledged that Inman et al. (1944) had earlier arrived at this concept. Narakas(1993) measured the inferior SHA in 170 patients with nerve injuries affecting the shoulder girdle, and also in patients with rupture of the rotator cuff. The active inferior SHA was reduced to 30° or less in patients with isolated rupture of the rotator cuff, in cases of suprascapular nerve palsy, or in combined lesions of these nerves with or without concomitant rupture of the cuff. The angle was substantially reduced in lesions of the spinal accessory nerve; it was least reduced in cases of serratus anterior palsy. We have used the method of Narakas since 1993. Its role in the analysis of shoulder deformities in more than 1,500 cases of birth lesion of the brachial plexus is described in Chap. 10. It has been applied in over 800 adult cases of nerve lesions affecting the thoraco-scapular girdle. This simple investigation measures the respective contributions to elevation provided by the
Clinical Aspects of Nerve Injury
169
Fig. 5.45 Measurement of grip and pinch. Top left measuring of pinch grip between thumb and index finger, top right showing the “median power grip” with use of thumb; bottom left showing “ulnar power grip” with thumb excluded; bottom right another method for measuring grip strength.
thoraco-scapular and the gleno-humeral joints. Stiffness of the joints is detected by measuring the passive range which, in the normal limb, lies between 170 and 180°. This simple investigation should be more widely used. Combined lesions of the accessory nerve and the nerve to serratus anterior are, fortunately, rare, for these are crippling and painful (Fig. 5.48). The trapezius muscle is tested by asking the patient to elevate the limb or to press forward against resistance. The lower fibres can be tested by asking the patient to put a hand
behind the back and press it against the trunk. The rhomboids are tested by pressing in the opposite direction (Fig. 5.49). The action of the levator scapulae muscle is best felt at the beginning of abduction of the shoulder from a resting position of 30° of abduction. In her study of iatrogenous lesions of the spinal accessory nerves, Camp (2010) found that the diagnosis was made by the responsible surgeon in only three cases and the average delay before diagnosis was no less than 18 months (Fig. 5.50). This delay is inexplicable. Most patients experienced
170
Surgical Disorders of the Peripheral Nerves
Fig. 5.46 Measuring power of muscles at the shoulder. The patient is seated.
Fig. 5.47 The active inferior scapula-humeral angle (ISHA) in a normal shoulder lies between 150° and 170°.
immediate severe pain and demonstrated remarkable loss of function. The posture of the scapula is characteristic. It drops downwards, and away from the spine. The average ISHA, in Camp’s cases, was about 50°. The “winging” is often wrongly attributed to paralysis of serratus anterior muscle (Fig. 5.51). The nerve to serratus anterior is a frequent victim of the attentions of surgeons and transection is associated with pain
and loss of function only slightly less than that seen after accessory palsy. The usual way to test the serratus anterior muscle is to ask the patient to press against the wall with both hands. Weakness is shown by winging of the scapula on the affected side. The method is, unfortunately, inapplicable just when it is most needed: in paralysis of the upper limb from a complete lesion of the brachial plexus. The examiner
Clinical Aspects of Nerve Injury
171
Fig. 5.48 Two examples of combined palsy of the spinal accessory and the nerve to serratus anterior.
Fig. 5.49 The lower border of trapezius, 1, is demonstrated by forcing the left hand against the spine and the rhomboids, 2, are shown by pressing the right hand away from the spine against resistance. “T” towards, trapezius; “R” rearwards, rhomboids.
supports the paralyzed upper limb with one arm; the index finger and thumb of the examiner’s other hand grip below a pole of the scapula; the patient is asked to push forwards. Protraction of the scapula shows that serratus anterior is not paralyzed (Fig. 5.52). The nerve is particularly susceptible to involvement in neuralgic amyotrophy. The ISHA was, on average, 130° in 45 patients with lesions of the nerve. The scapula is elevated and approaches the spine (Fig. 5.53). The circumflex and suprascapular nerves – the rotator cuff. Inman et al. (1944) studied the joints of the shoulder girdle by inserting pins into the skin of volunteers and recording movement, by radiographs and by measurement of
current action potentials, from the muscles. They observed: “it should be clearly recognized that the standard text book teaching on these motions is entirely incorrect. These state that gleno humeral motion occurs up to a right angle and that thereafter further elevation is brought about by rotation of the scapula. Roentgenography and examination of the living proved beyond any doubt that scapula and humeral motion are simultaneously continuous” (Fig. 5.54). Recognition of rupture of the circumflex nerve can be very difficult. One reason for this is the widely held (and erroneous) view that the deltoid muscle is the abductor of the gleno humeral joint (Figs. 5.55–5.57). Wynn Parry (1981) examined 145 patients with paralysis confined to the deltoid muscle. He found that the range of abduction was full, or nearly so, and described a system of training compensatory movements which enabled most of his patients to return to full military duties: “it must be stressed that these movements providing full abduction and elevation are not trick actions in the sense usually associated with this word; all the muscles involved normally help to abduct the shoulder. The scapulo-humeral rhythm is quite normal and in the later stages of re-education the patient does not even need to rotate the humerus externally to initiate the movement.” Seddon (1975) was a little more cautious: “this perfect abductor action of the supraspinatus is rare; it is more usual to find abduction to about 155°, with the arm a little in front of the coronal plane of the body.” Curiously, the loss of abduction caused by lesions of the suprascapular nerve and/or of rupture of the rotator cuff is frequently and wrongly attributed to a lesion of the circumflex nerve (Figs. 5.58–5.60). In 63 cases we found that the active ISHA was diminished by about 20° in uncomplicated ruptures of the circumflex when there was no stiffness of the shoulder. The
172
Surgical Disorders of the Peripheral Nerves
Fig. 5.50 Right spinal accessory palsy. The scapula drops down and away from the spine. The active ISHA is 30°.
Fig. 5.51 Left spinal accessory palsy. Scapular winging, without prominence of the lower fibres of the trapezius, in a case where there is some early recovery into the upper fibres after repair of the spinal accessory nerve. At rest, the scapula is displaced downwards and away from the spine.
Clinical Aspects of Nerve Injury
Fig. 5.52 A method of examining serratus anterior. The examiner grasps the lower pole of the scapular between finger and thumb and the patient is asked to thrust the upper limb forwards. The examiner will be unable to block protraction of the scapula if serratus anterior is normal.
angle is reduced to less than 30° in most cases of suprascapular palsy or in complete ruptures of the rotator cuff (78 cases). Perhaps the most reliable sign of circumflex palsy is weakness of extension. The power of extension at the shoulder abducted to 90° is as little as 5–10% of normal when the deltoid is
173
paralyzed. The diagnosis of rupture of the circumflex nerve is only easy when it is too late to do anything about it, that is, when the atrophy of the muscle is all too plain. It is a very hard matter for the clinician treating a patient with fracture/dislocation of the shoulder to examine function in the muscles. The area of loss of sensation is inconsistent and some patients will describe sensation of the skin over the muscle as abnormal rather than absent. It is probably helpful for the treating clinician to bear the possibility of nerve injury in mind and to plan for examination of the muscles about the shoulder at 1 week after urgent treatment by which time sensory loss and paralysis can be readily detected. Examination of the adductors and medial rotators of the shoulder is usually straight forward. The subscapularis, the most powerful of the muscles comprising the rotator cuff, is more difficult to test. Placing the hand behind the back is a complex movement and requires integrity of the thoracoscapular muscles. It is usually impaired in patients with lesions of the spinal accessory, the nerve to serratus anterior or dorsal scapular nerves. Isolated paralysis of the subscapularis is rare. If the tendon is ruptured many patients are still able to maintain a full range of medial rotation although power is reduced in the last 20° or 30° (Figs. 5.61–5.63).
Fig. 5.53 Scapular winging in nerve to serratus anterior lesion. This is easily distinguishable from the winging provoked by accessory palsy by the position of the scapula which is drawn upwards and towards the spine by the unopposed action of trapezius, levator scapulae, and rhomboids.
174
Fig. 5.54 The deltoid. The anterior, middle, 1, and posterior, 2, components of the muscle should be observed and palpated during resisted abduction and extension.
All three heads of triceps should be examined for this provides information about which spinal nerves have been injured and offers an indication of the level of injury of the radial nerve (Fig. 5.64). Ruptures of the musculocutaneous nerve are frequently missed at first examination because brachioradialis is such a powerful flexor and also because the radial nerve often innervates the lateral portion of brachialis (Figs. 5.65 and 5.66). The power of supination and pronation of the forearm should be tested from the neutral position: it is important to remember that both biceps and supinator contribute to supination, and that the long flexor muscles of the forearm can often give useful pronation when the pronator teres muscle is paralyzed or weak. Paralysis of the extensor and flexor muscles of the wrist may be masked by the extensor and flexor muscles to the digits. Extensor pollicis longus is capable of extending the wrist, abductor pollicis longus can mimic this action if the wrist is partially pronated. It is important to palpate both the muscle and the tendon under examination (Fig. 5.67). The long flexors of the fingers: The flexor superficialis, usually innervated by the median nerve, acts on the proximal interphalangeal joints of the medial four digits. Its action on individual fingers is best tested by holding three of the digits almost straight and asking the patient to flex the free finger. Because the tendons of the superficialis are largely independent of each other and those of profundus are not, the former muscle will act to flex the proximal interphalangeal joint (Fig. 5.68).
Surgical Disorders of the Peripheral Nerves
The medial half of the flexor digitorum profundus (FDP) is usually innervated by the ulnar nerve; the lateral part, by the median nerve. With the exception of that to the index finger, the tendons are closely associated with each other, and independent movement is, in most cases, hardly possible. The test for the integrity of the profundus requires the subject to flex the distal interphalangeal joint of one finger with the proximal joints of that finger and all joints of the other fingers stabilised. One quick way to test the FDP simultaneously is to ask the patient to curl the fingers into the palm with the metacarpophalangeal joints extended. The examiner’s finger cannot extend the distal interphalangeal joint if the muscle is normal. The division of the supply of the intrinsic muscles of the hand between the median and ulnar nerves varies from subject to subject (Rowntree 1949). The most common arrangement is for the ulnar nerve to supply all muscles with the exception of the first lumbrical, the abductor brevis and opponens pollicis, and part of the flexor brevis pollicis. Most of the intrinsic muscles act on the metacarpo-phalangeal joints to flex, adduct and abduct. The particular action of the thenar muscles is to oppose the thumb to the little finger, and the muscle particularly responsible for rotating the metacarpal bone during that movement is the opponens. When the median nerve has been cut the opposing action of the thenar muscles can be mimicked by the combined action of an ulnar-innervated flexor brevis and the abductor longus muscle. Comparison with the intact side will usually show that this combined action does not reproduce the rotational action of the opponens. Similarly, the abducting action of the abductor brevis can in the absence of median nerve function be imitated by the action of the abductor longus muscle. These points are important in the early stages when there is no wasting to guide the examiner (Figs. 5.69 and 5.70). The action of the thumb in pinching against the forefinger is modified and weakened by the loss of the stabilising action of the flexor brevis pollicis on the first metacarpophalangeal joint. Without that action, pinch has to be mediated by the action of the flexor longus pollicis on the interphalangeal joint. The power of the ulnarinnervated muscles of the hand is best tested by examining the power of abduction and adduction of the fingers. The ease with which a sheet of paper may be pulled from between two adducted fingers gives some indication of this power.
5.9.4 The Lower Limb There is no great difficulty in testing the muscles connecting the pelvis to the femur in the healthy subject, but things are
Clinical Aspects of Nerve Injury
175
Fig. 5.55 Elevation of the upper limb in full medial rotation by the supraspinatus in the absence of deltoid in four cases of proven rupture of the circumflex nerve. Bottom left and right: note the activity in the clavicular head of pectoralis major.
different when this has to be done soon after replacement arthroplasty. Our observations suggest that in this situation there is also quite often a certain reluctance to look. In one case, a “drop foot” was observed soon after arthroplasty, but it was not until a year later that another examiner found paralysis of most of the muscles of the buttock. Superior gluteal palsy is crippling, yet delay in diagnosis is common. Much can be learnt from watching the patient walk. The integrity of the smaller glutei is tested with the patient standing; that of
the rotators of the hip with the patient seated, and that of the gluteus maximus with the patient prone (Fig. 5.71). It was common experience at times when poliomyelitis was common to see children and young adults walking quite well even though their quadriceps muscles were paralyzed (Fig. 5.72). Of course, they did this by a form of adaptation, a trick movement, in which the tensor fascia lata was responsible for stabilisation of the knee. In many cases there was the added factor of a hyperextension deformity of the knee. It is, however,
176
Surgical Disorders of the Peripheral Nerves
Fig. 5.56 The movement of the scapula in another case of rupture of the right circumflex nerve. The active ISHA on the side of the injury(bottom right) is reduced by 20° showing that 20° of the range of elevation is provided by extra movement at the thoraco-scapular joint.
Fig. 5.57 Combined injuries to the suprascapular and circumflex nerves. Left: showing the range of elevation in a patient with irreparable injury to the right suprascapular nerve but with a good result after repair of the circumflex nerve. Right: showing the elevation in another patient in whom repair of the left suprascapular nerve was successful but whose circumflex nerve injury was irreparable.
Clinical Aspects of Nerve Injury
Fig. 5.58 The infraspinatus, 1, is examined by resisting lateral rotation with the warm against the side.
177
the diagnosis was recognised only after the patients fell and damaged themselves. A lesion of the femoral nerve high enough to paralyze both hip flexors and extensor muscles of the knee is crippling. Examination of the power of flexion of the hip is best done with the patient lying supine. The power of extension of the knee is best measured with the patient sitting and the examiner should attempt to assess bulk and power in the three superficial components of the muscle, the vastus medialis, the rectus femoris, and the vastus lateralis. The flexors of the knee are best examined with the patient prone: it is usually possible to distinguish the action of the biceps femoris from that of the medial hamstrings. It is necessary, in testing the power of the flexors of the ankle, to make sure that the action is in fact effected by the soleus and gastrocnemius: the tibialis posterior and the flexors of the toes can produce ankle flexion, though they cannot sustain the patient’s weight. Both tibialis muscles effect inversion of the foot; the anterior muscle is the principal extensor of the ankle. It is not always easy to measure the strength in the evertor muscles of the ankle, (peroneus longus and brevis). It may be easier to detect early recovery with the patient lying on their side and the foot held plantigrade. The muscle belly and the tendon are observed and palpated (Fig. 5.73). We know of no really satisfactory way of testing the integrity of the intrinsic muscles of the foot. Some individuals can use these muscles independently of each other, most cannot (Fig. 5.74).
5.9.5 Late Signs of Nerve Injury
Fig. 5.59 Initiation of abduction, with opening of the active ISHA was the first sign of recovery into supraspinatus after repair of the suprascapular nerve. The lesion of the circumflex nerve was irreparable.
quite wrong to assume that an adult with a deep femoral nerve lesion could walk comfortably and without risk. In six of our cases of femoral palsies incurred during total hip arthroplasty,
Two weeks after a complete degenerative lesion, the area of loss of sensibility is well defined; the beginning of wasting indicates the extent of the motor affection. Anhidrosis is still present, but with the degeneration of peripheral fibres the warm isothermia of the skin gives way to poikilothermia and later to cold isothermia (Figs. 5.75 and 5.76). As time goes by, the changes of disuse appear: thinning of the skin; even ulceration from accidental injury; loss of substance in the tips of the digits; loss of skin markings; constant coldness and cyanosis; stiffness of joints; contractures; unmistakable wasting. Nails become brittle and discoloured and are prone to infection. Hair growth is disturbed, hairs are often coarse (Figs. 5.77–5.79). These changes occur rapidly in the ischaemic limb. At a late stage in partial denervation or with incomplete recovery after a deep lesion there may be muscle spasms and spontaneous fasciculation. Prolonged denervation of a growing limb leads to defective growth: this is of course well seen after birth injury of the brachial plexus. Lewis and Pickering (1936) thought that the changes in denervated limbs were simply the result of disuse rather than that of the loss of a “trophic” function of the nerves. However, it is a fact that in cases of greatly prolonged conduction block, the changes are always far less than they are in
178
Surgical Disorders of the Peripheral Nerves
Fig. 5.60 Rupture of rotator cuff with lesions of the suprascapular and circumflex nerves from fracture/dislocation of the shoulder. Above showing the range of elevation at the right shoulder in 74 year old ex-paratrooper in whom there was clear evidence of recovery for both of the nerves. Below: this shipwright held onto a cable to rescue a man from the Thames. The weight of the man and the force of the current was such that he felt the muscles tearing in his right shoulder, then he felt the head of the humerus pulling out from the socket and then his arm went dead. Mr. Sait (Dartford) confirmed rupture of the rotator cuff by MR scan. Electromyography (Dr. Cordivari, Queen Square) showed that the suprascapular nerve was intact and that there was, at 8 weeks, reinnervation of the posterior deltoid. A Tinel sign was detectable at the posterior aspect of the shoulder. A subsequent repair of the rotator cuff, performed by Mr. Sait, was successful.
Fig. 5.61 The clavicular head, 1, of pectoralis major is best examined by resisting adduction and medial rotation with the arm elevated.
Fig. 5.62 The sternal head, 1, of pectoralis major is tested with the arm in adduction.
Clinical Aspects of Nerve Injury
179
Fig. 5.65 Brachio radialis is a powerful flexor of the elbow. The biceps muscle and its tendon should be observed and palpated. 1. Biceps brachii, 2. Biciptal aponeurosis, 3. Medial head of triceps, 4. Brachioradialis, 5. Biceps tendon.
Fig. 5.63 Teres major, 2, and latissimus dorsi, 3, are tested by resisting adduction. The muscles are seen and palpated. The posterior deltoid, 1, is seen.
Fig. 5.64 The lateral, 1, long, 2, and medial, 3, heads of triceps are examined by observing and palpating the muscle.
degenerative lesions. We have seen that in degenerative lesions profound changes take place in both motor and sensory end-organs. The distal axon normally maintains a dense population of end-organs in skin and sweat glands and in the muscular component of arterioles. It is hard to resist the conclusion that the changes of “disuse” are at least in part due to the loss of distal axons and of their end-organs, and to the effects of that loss on superficial tissues. An extreme form of severe trophic changes in long standing “conversion paralysis” is described in Chap. 14.
Fig. 5.66 Elbow flexion. 1. Brachioradialis, 2. Brachialis, 3. Biceps, 4. Bicipital aponeurosis.
By the time the changes of degeneration are present, the patient is a better candidate for the examination halls than for restorative treatment. The object of the clinician must be to make the diagnosis before the signs of peripheral degeneration have appeared; before the best time for intervention has passed. Unfortunately, the peripheral neurologist is still likely to be presented with cases in which delay in diagnosis has permitted the development of these signs. The last are at this stage well marked; their absence in association with
180
Surgical Disorders of the Peripheral Nerves
Fig. 5.67 Activity in extensor carpi radialis brevis, 1, and longus, 2, is noted by observing and palpating the muscle bellies and their tendons.
Fig. 5.69 Below: the first dorsal interosseous, 1, and adductor pollicis, 2, muscles (ulnar nerve) are examined by thumb to index pinch grip. Above: abductor pollicis brevis, 3, (median nerve) is tested by resisting abduction of the thumb at right angles to the palm of the hand.
Fig. 5.68 Flexor digitorum superficialis is examined by immobilising the other digits.
persistent partial motor and sensory paralysis almost certainly means that the lesion was partly or wholly a conduction block.
5.9.6 Signs of Reinnervation The detection of initial and continuing reinnervation is often very important in aiding a decision about intervention. After
such a procedure, it is important as an indicator of success or failure, and consequently as a guide to a second intervention. In cases in which long spatial and temporal distances separate a lesion from the nerve’s target, the Tinel sign is useful but can mislead, especially when the target tissues have been damaged by ischaemia or sepsis. After a degenerative lesion, recovery proceeds centrifugally, so that the first sign of reinnervation is voluntary contracture in the muscle most proximally innervated. Sensibility too recovers centrifugally. As Head et al. (1905) observed, its quality improves over successive months, doubtless with re-myelination of the larger axons Recovery of sensibility may be preceded by recovery of sweating; indeed, sweating may be restored without any later recovery of sensibility; light touch sensibility may return in the absence of nociception and sweating In non degenerative lesions, recovery is not necessarily nor even generally centrifugal: distal muscles may well recover before proximal ones; loss or alteration of sensibility may anyway be patchy; vasomotor and sudomotor paralysis is rare. Because the lesion depends so much on demyelination, and in particular on demyelination of large fibres, a dissociated loss of position sense and vibration sense may be a good indicator of its type.
Clinical Aspects of Nerve Injury
181
Fig. 5.71 Above: the power of the abductors of the hip is tested with the patient in a lateral position, resisting abduction on the lower limb. Below: gluteus maximus is tested with the patient prone resisting extension at the hip.
Fig. 5.70 The small muscles of the hand in lesions of the ulnar nerve. Above: “Froment’s sign” is positive in the patient’s left hand. Below: clawing is corrected by passive flexion at the metacarpophalangeal joints.
Very often, in recovery after both degenerative and nondegenerative injuries, the patient experiences radiated pain and paraesthesiae and asks the clinician whether these feelings indicate progressive recovery. We have seen that Merrington and Nathan (1949) explored the nature and origin of paraesthesiae after conduction block: certainly, such feelings indicate recovery in that condition. They do not necessarily, nor even often, have that significance in the case of degenerative lesions. Features of imperfect recovery: Imperfect recovery, common in the adult after repair of a degenerative lesion, is apparent to the patient through: (1) Weakness and sometimes wasting; (2) alteration of sensibility;(3) dysfunction; (4) (sometimes) pain; (5) (sometimes) sensitivity over the site of the lesion; (6) exaggerated reaction of the affected part to external cold; (7) stiffness and even deformity of the joints of the affected part. The exaggerated reaction to cold requires some further comment; all other aspects are treated elsewhere. Cold affects healthy nerves: it affects even more severely those in which the number of functioning axons is
reduced, So, conditions of cold which would not seriously affect healthy nerves are liable badly to affect damaged nerves in which regeneration has been imperfect. Both motor and sensory function are affected. The cold affects not only nerve function; it affects too the skin and deep tissues of the affected limb, which is quick to cool and slow to warm. Sensitivity to cold may prevent return to work in the cold room of a butcher or fishmonger, or to outdoor work with machinery, even when recovery after repair of a median nerve at the wrist has been otherwise good. The response to cold stimulus was analyzed, in normal hands, by Davis and Pope (2002). Aching pain was experienced at temperatures of 15°C or less and prickling sensations were induced by even colder stimuli. Cold related symptoms after injury to peripheral nerves seem to be an exaggeration of this response. Irwin et al. (1997) noted immediate onset of symptoms in one-third of their patients. Symptoms had developed within 3 months in the remaining two-thirds. The problem was worse after untidy injuries and after injury to the axial artery. Kay (1985) “found no relationship between arterial inflow and cold intolerance” but there was a tendency for the symptoms to be worse with poorer nerve function. Vasospasm may contribute to the problem (Backman et al. 1993). An unpleasant sensation of intense coldness was associated with blanching or discolouration of
182
injured fingers in more than 70% of the patients described by Lithell et al. (1997). Malenfant et al. (1998) studied cold intolerance in patients with burns and demonstrated changes in sensation in normal skin in the contra lateral uninjured limb.
Surgical Disorders of the Peripheral Nerves
5.10 Records Recovery after severe injuries to nerves is prolonged, and persisting pain and loss of function interfere with rehabilitation. We have developed a number of systems which record the
Fig. 5.74 The small muscles of the foot, 1. The patient is asked to “make a fist” with their toes.
Fig. 5.72 Extension of the knee. The muscle bellies of vastus medialis, rectus femoris, and vastus lateralis can be seen and felt. Note the action of the tensor fascia lata which can, in some patients, extend the knee against gravity. Note the fascia lata, 1, the vastus lateralis, 2, the rectus femoris, 3, and vastus medialis, 4.
Fig. 5.73 Eversion at the ankle. The muscle bellies of peroneus brevis, 1, and longus, 2, and their respective tendons are observed and palpated.
Fig. 5.75 Late changes after nerve injury. Severe wasting of the hand with contractures, 1 year after stab wound to the lower part of the brachial plexus.
Clinical Aspects of Nerve Injury Fig. 5.76 Late changes after nerve injury. Left: there is sympathetic paralysis and unnoted burns after high lesion of the median nerve. Right: wasting and ulceration of the skin of middle finger and accidental injury to index finger after median nerve injury. The patient was working as a stone mason.
Fig. 5.77 Post ischaemic contracture in the hand. A 63 year old man lay in a coma for 18 h. There was compression of the axillary neurovascular bundle. The range of extension (left) and of flexion (right) is shown 1 year later.
183
184
Fig. 5.78 Post ischaemic fibrosis. The lateral geniculate artery was lacerated during arthroscopic meniscectomy. The false aneurysm which ensued remained undetected for 4 days in spite of his intense causalgia. The position of the leg, ankle and foot 3 years after injury.
progress after nerve injury, associated injuries and their consequences, and after “reconstruction” operations. Information is collected about the effects on daily life, on work or training or study. The evolution of pain, its response to treatment or to recovery is recorded. The progress for the nerve is recorded
Fig. 5.79 Late skin changes after nerve injury. Left: a skin rash in the distribution of C5, 6 months after rupture. Right: skin rash in the distribution of the common peroneal nerve 1 year after rupture at the knee.
Surgical Disorders of the Peripheral Nerves
sequentially for motor, sensory, and sympathetic function and a final grade is offered for the whole nerve. Suitably modified forms are used for injuries to the shoulder girdle and gleno-humeral joint, for injuries in the lower limb and for penetrating missile wounds (Fig. 5.80). Other forms are used for traumatic lesions of the brachial plexus (Fig. 5.81a and b). The documents used for birth lesions of the brachial plexus are illustrated in Chap. 10. These records are placed in the case notes at the patient’s first attendance, and remain there until treatment is completed. Information is transferred from the form onto appropriate data bases throughout treatment. These computerised systems certainly ease analysis but never replace the hand written entry made at every visit. All systems fail to a greater or lesser extent in (1) failing adequately to record the stamina of muscles; (2) failing adequately to record function of the sensory end-organs in muscles, tendons and particular capsules and ligaments and (3) failing adequately to record co-contraction. Co-contraction is particularly common after injuries to the brachial plexus in adults and in infants. Special apparatus is indeed available for testing stamina of muscles, in particular those of the hand. The amount of time involved is, however, too great to permit the routine use of these methods. Evidently, tests such as Moberg’s “pick-up” test to some extent examine deep as well as superficial sensibility, but there is no good measure of the extent of re-innervation of muscle spindles. The examination of tendon reflexes is, evidently, a coarse method of testing the function of the Golgi organs and muscle spindles, but more precise methods are lacking. We do not know how much the function of a muscle recovered after nerve repair is impaired by defect of input from it. It is probable that impairment of the
Clinical Aspects of Nerve Injury
185
Fig. 5.80 The document used for recording data in patients with compound nerve injury.
deep afferent pathway is one factor underlying co-contraction; such impairment is certainly a major cause of the regular failure of muscle transfers using reinnervated muscles.
5.11 Aids to Diagnosis Electrophysiological examination: We owe much to our colleagues, Dr. Nicholas Murray, Dr. Shelagh Smith, and Dr. Peter Misra, Dr. Carla Cordivari and their colleagues at the National Hospital for Nervous Diseases with whom joint electrodiagnostic clinics have been held for more than
20 years. More than 1,000 patients have been seen at these clinics, which provide prompt and valuable information about diagnosis and prognosis. This collaborative approach is also useful in teaching and in training. The cases are presented by the surgeon who offers a diagnosis of the level and of the depth of lesion and of the likely prognosis. That diagnosis is then confirmed, expanded or refuted by studies of sensory and motor conduction and by electromyography of selected muscles. More than 2,000 other patients have been seen by our colleagues at the National Hospital whose work represents a contribution in the analysis of injuries of profound importance.
186
Electrophysiological examination (Wynn Parry 1953, Smith 1996, 1998, 1999) is certainly the foremost aid to diagnosis, though in the acute stage the process is often hampered by pain and by local conditions. It must be done properly and results must expertly be interpreted. It is no substitute for clinical observation; it must not be used as device for deferring decision and delaying action. It can help in diagnosis of
Fig. 5.81 The document used for collecting data about patients with injuries to the brachial plexus.
Surgical Disorders of the Peripheral Nerves
the site and depth of a lesion and in the recognition and measurement of recovery. The extension of the process to the detection and measurement of potentials evoked from the cortex provides valuable evidence in the case of suspected avulsion of the roots of the brachial plexus. Perhaps the simplest, yet often neglected, technique of electrophysiological examination is that of stimulating the nerve below the level of
Clinical Aspects of Nerve Injury
187
Fig. 5.81 (continued)
the lesion and observing the motor response. If, 3 days after injury, stimulation below the level of the lesion produces a normal response in the muscles supplied by the nerve, the odds are that the lesion is a conduction block. If there is no motor response or if the response is much subdued, then the lesion is degenerative. The introduction of electrophysiological process during operation has brought massive advantages, in particular in (1) determining neural continuity across a lesion in continuity;
(2) determining the site of a conduction block; (3) determining which part of a nerve has suffered axonal interruption; (4) determining whether an apparently intact component of the brachial plexus has intact central connections. We are profoundly glad that the whole matter is considered at length in the next Chapter by an expert in the field. Magnetic resonance imaging: Whilst magnetic resonance neurography is uniquely informative in the diagnosis and accurate measurement of the extent of tumors in or near
188
nerves, in the more difficult entrapment syndromes, and in closed traction injuries of the brachial and lumbo-sacral plexus, the full potential of the method in the diagnosis of closed injuries to peripheral nerves has yet to be fully exploited. Ultrasonography seems more promising High resolution ultrasonography: Dr. Susan Krone (2005) (Maidstone) read an exceptional paper at the Royal National Orthopaedic Hospital in 2005 in which she outlined the application of this method for the display of major nerves and of their constituent bundles during regional anaesthesia. She pointed out that this investigation could prove valuable in the first days after closed injury before fracture haematoma is transformed into scar. Krone further suggested that the method might have a place in the examination of the neonatal brachial plexus in cases of birth lesion at a time when ossification of the cervical spine is limited. Martinoli et al. (1996) provided an important review: When examined with high-frequency probes, peripheral nerves show a peculiar arrangement of their internal structures. Using 15 MHz transducers the nerve appears to be composed of multiple hypo-echoic parallel discontinuous linear areas separated by hyper-echoic bands. The hypo-echoic areas are arranged in series and appear well defined and elongated according to the longitudinal axis of the nerve. On transverse sections, the hypoechoic areas become rounded and are embedded in a homogenous hyper-echoic background. Side by side comparison between ultrasound scans and respective histologic slides demonstrated that the hypo-echoic areas coincide with fascicles of neuronal fibres, whereas the hyper-echoic background correlates with the epineurium and passes into the spaces between the individual fascicles of nerve substance.
These workers go on to point out that the arrangement of fascicles is related to the ultrasound frequency employed. 15 MHz transducers are optimal but tissue penetration of these is poor so that their use is limited to the examination of nerves running within 2 cm from the skin surface. Deeper nerves can be examined using lower frequency probes but the fascicular pattern is less well defined with such transducers. The arrangement of bundles can be detected using a frequency as low as 7.5 MHz. Chan (2003), and Perlas and his colleagues (Perlas et al. 2003) show that the brachial plexus, in the posterior triangle of the neck, can be closely examined using high resolution ultrasonography. The potential for diagnosing nerve transection by ultrasound was examined by Cartwright et al. (2007), from the Wake Forest school of Medicine in 2007. The median, ulnar and radial nerves were transected in the arms of fresh cadavers and sham skin incisions were performed throughout the arm. The nerves were then scanned by an ultrasonographer who was unaware of the sites of transection. High resolution ultrasound was able to identify transected nerves with 89% sensitivity and 95% specificity. These workers conclude that: “this proof of concept study shows that ultrasound can accurately identify a nerve transection, which should lead to further ultrasound studies in patients with traumatic peripheral nerve injuries.” Two such studies are
Surgical Disorders of the Peripheral Nerves
provided by Cokluk and Aydin (2007 a, b) of the Department of Neurosurgery in Samsun, Turkey, who published two important papers in 2007. In all, 58 patients were examined The nerves investigated in the upper limb included the brachial plexus (4 cases) the ulnar nerve (9 cases) the radial nerve (6 cases) the median nerve (17 cases). In the lower limb, the femoral nerve(5 cases) and the sciatic nerve (17 cases) were examined. The examination was performed using a Tosbee ultrasound (Toshiba Inc., Tokyo) with a 5–7.5 MHz linear probe. The patients with injuries in the upper limb were placed supine. Ultrasound gel was plastered on the probe surface and the skin to enhance visualisation of peripheral nerves and the musculo-skeletal structures. The examination commenced about 10 cm proximal to the suspected region and continued 10 cm distally. Bone, muscles, tendons, vascular structures, and peripheral nerves were identified and distinguished. “Continuity, architecture, shape, calibration and integrity of the involved nerve and peripheral tissues were examined in the perpendicular and transverse planes.” The femoral nerve was examined with the patient supine, the sciatic nerve was examined with the patient placed prone. Sixteen of these patients were examined within 3 days of injury. In most cases the diagnosis made by ultrasound was matched with the findings at subsequent operation. The investigation proved reliable in identifying the nerve, in localising the level of injury and in the recognition of the nature of that injury. Toros et al. (2009) provide further valuable information about the technique. It seems likely that ultrasonography, in skilled hands, has enormous potential in the early detection of ruptures or other serious injuries to nerves. A number of orthopaedic and fracture surgeons are already well versed in the technique and we hope that the very real difficulty of recognition of rupture of the nerve trunk in a closed fracture will be overcome by the widespread use of this method by interested clinicians.
References Anand P, Birch R (2002) Restoration of sensory function and lack of long-term chronic pain syndromes after brachial plexus injury in human neonates. Brain 125:113–122 Anand P, Warner G, Parrett A, Terenghi G, Kopelman P, Ghatei MA, Bloom SR, Sinicropi D, Williams-Chestnut RE (1994) Nerve growth factor related small fibre dysfunction in human diabetic neuropathy. Ann Neurol 36:284A Anand P, Terenghi G, Warner G, Kopelman P, Sinicropi DV, WilliamsChestnut RE (1996) The role of nerve growth factor in human diabetic neuropathy. Nat Med 2:703–707 Atherton DD, Facer P, Roberts KM, Misra VP, Chizh BA, Bountra C, Anand P (2007) Use of the novel contact heat evoked potential stimulator (CHEPS) for the assessment of small fibre neuropathy: correlations with skin flare responses and intra-epidermal nerve fibre counts. BMC Neurol 7:21
Clinical Aspects of Nerve Injury Backman C, Nystr m A, Balkman A, Bjerle P (1993) Arterial spasticity and cold intolerance after digital replantation. J Hand Surg Br 18:551–555 Bell J, Burford W (1982) The force time relationship of clinically used sensory testing instruments. Presented at the American Society for Surgery of the Hand, 37th Annual Meeting, New Orleans Birch R (1986) Lesions of peripheral nerves: the present situation. J Bone Joint Surg Br 68:2–8 Birch R (1989) The repair of peripheral nerves: experimental and clinical observations. Thesis for the degree of Master of Surgery, University of Cambridge Birch R, Raji ARM (1991) Repair of median and ulnar nerves. J Bone Joint Surg Br 73:154–157 Bonney G (1983) Peripheral nerve lesions. In: Harris NH (ed) Clinical orthopaedics, Wright, Bristol, pp 726 Bowden REM, Scholl DA (1954) Rates of regeneration. In: Seddon HJ (ed) Peripheral nerve injuries, vol 282, Medical Research Council Special Report Series. HMSO, London, pp 16–24 Brand PW (1985) Clinical mechanics of the hand. Mosby, St. Louis, pp 61–191 Buck-Gramcko D, Lubahn JD (1993) The Hoffman-Tinel sign. J Hand Surg Br 18:800–805 Camp SJ (2010) Lesions of the spinal accessory nerve. J Bone Joint Surg (in print) Canale ST, Beaty JH, Warner WC, Sharp RS (2000) Passive flexion sign: a simple tool for diagnosis of anterior interosseous nerve injury in children. J Bone Joint Surg 82:1354–1355 Cartwright MS, Chloros GD, Walker FO, Wiesler ER, Campbell WW (2007) Diagnostic ultrasound for nerve transection. Muscle Nerve 35(6):796–799 Chan V (2003) Applying ultrasound imaging to Interscalene brachial plexus block. Reg Anesth Pain Med 28:340–343 Cokluk C, Aydin K (2007a) Ultrasound examination in the surgical treatment for upper extremity peripheral nerve injuries: part I. Turk Neurosurg 17(4):277–282 Cokluk C, Aydin K (2007b) Ultrasound examination in the surgical treatment for upper extremity peripheral nerve injuries: part II. Turk Neurosurg 17(3):197–201 Coupland RM (1993) War wounds of limbs. Butterworth Heinemann, Oxford, London Davis KD, Pope GE (2002) Noxious cold evokes multiple sensations with distinct time cause. Pain 98:179–185 Dellon AL (1978) The moving two-point discrimination test: clinical evaluation of the quickly-adapting fibre receptor system. J Hand Surg 3:474–481 Dunnet W, Housden P, Birch R (1995) Results of flexor to extensor tendon transfer. J Hand Surg Br 20(1):26–28 Dyck PJ, O’Brien PC, Johnson DM, Klein CJ, Dyck PJB (2005) Quantitative sensory testing. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier, Philadelphia Gilliatt RW, Taylor JC (1959) Electrical changes following section of the facial nerve. Proc R Soc Med 52:1080–1083 Head H, Rivers WHR, Sherren J (1905) The afferent nervous system from a new aspect. Brain 28:99–116 Henderson WR (1948) Clinical assessment of peripheral nerve injuries. Tinel’s test. Lancet 2:801–806 Highet WB (1941) Memorandum to the nerve injuries committee of the Medical Research Council and published in the Special Report Series No. 282 1954. HMSO, London Highet WB (1954) Quoted by Zachary RB. In: Seddon HJ (ed) Peripheral nerve injuries Medical Research Council Special Report Series No 282. HMSO, London, p 355 Hoffman P (1915a) Über eine Methode, den Erfolg einer Nervennant zu beurteilen. Med Klin 11:359–360 Hoffman P (1915b) Weiterres über das Verhalten frisch regenerierter Nerven über die Methode, den Erfolg einer Nervennaht Frühzeitig zu beurteilen. Med Klin 11:856–858
189 Htut M, Misra P, Anand P, Birch R, Carlstedt T (2006) Pain phenomena and sensory recovery following brachial plexus avulsion injury and surgical repair. J Hand Surg Br 31:596–605 Htut M, Misra VP, Anand P, Brich R, Carlstedt T (2007) Motor recovery and the breathing arm after brachial plexus surgical repair including re-implantation of avulsed spinal nerves into the spinal cord. J Hand Surg 32E:170–178 Inman VT, Saunders JB, Abbott LC (1944) Observation on the function of the shoulder joint. J Bone Joint Surg 26:1–31 Irwin MS, Gilbert SEA, Terenghi G, Smith RW, Green CJ (1997) Cold intolerance following peripheral nerve injury. Br J Hand Surg 22:308–316 Jerosch-Herold C (2000) Should sensory function after median nerve injury and repair be quantified using two point discrimination as a critical measure? J Plast Reconstr Surg 34:339–343 Jerosch-Herold C (2005) Assessment of sensibility after nerve injury and repair: a systematic review of evidence for validity, reliability and responsiveness of tests. J Hand Surg Br 30:252–264 Kato N, Birch R (2006) Peripheral nerve palsies associated with closed fractures and dislocations. Injury 37:507–512 Kaufmann KR (2005) Quantitative muscle strength assessment. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier, Philadelphia, pp 1095–1102 Kay S (1985) Venous occlusion plethysmography in patients with cold related symptoms after digital salvage procedures. J Hand Surg Br 10:151–154 Keogh Sir Alred (1917) Introduction to: notes on military orthopaedics (Author. Robert Jones), Cassell, London Krone S (2005) Ultrasound guided regional anaesthesia. Invited paper to advanced orthopaedic anaesthesia and critical care. Royal National Orthopaedic Hospital, Stanmore, Middlesex, 16 September 2005 Landau WM (1953) The duration of neuromuscular function after nerve section. Neurosurgery 10:64–68 Landi A, Copeland S (1979) Value of the tinel sign in brachial plexus lesions. Ann R Coll Surg Engl 61:470–471 Lewis T, Pickering GW (1936) Circulatory changes in the fingers in some diseases of the nervous system, with special reference to the digital atrophy of peripheral nerve lesions. Clin Sci 2:149–183 Lithell M, Backman C, Nystr m Å (1997) Pattern recognition in posttraumatic cold intolerance. J Hand Surg 22:783–787 MacKinnon SE, Dellon AL (1985) Two point discrimination tester. J Hand Surg Am 10:906–907 MacKinnon SE, Dellon AL (1988) Ischaemia of nerve: loss of vibration sensibility. In: Surgery of the peripheral nerve. Thieme Medical Publishers/Georg Thieme Verlag, Stuttgart, New York, p 57 Malenfant A, Forget R, Amsel R, Papillon J, Frigon J-Y, Choinière M (1998) Tactile thermal and pain sensibility in burned patients with and without chronic pain and paraesthesia problems. Pain 77: 241–251 Martinoli C, Serafini G, Bianchi S, Bertoletto M, Gandolfo N, Derchi LE (1996) Ultrasonography of peripheral nerves. J Peripher Nerv Syst 1:169–178 Merrington WR, Nathan PW (1949) A study of post-ischaemic paraesthesiae. J Neurol Neurosurg Psychiatry 12:1–18 Mills K (1997) Wasting, weakness and the MRC scale in the first dorsal interosseous muscle. J Neurol Neurosurg Psychiatry 62:541–542 Moberg E (1958) Objective methods for determining the functional value of sensibility in the hand. J Bone Joint Surg Br 40:454–476 Narakas AO (1993) Paralytic disorders of the shoulder girdle. In: Tubiana R (ed) The hand, vol. 4. Saunders, Philadelphia, pp 112–125, Chapter 9 Novak VP, Baratz ME (2006) Antero-medial ecchymosis about the elbow in an adult with a distal humerus fracture. J Hand Surg Am 31:860–862 Novak CB, MacKinnon SE (1999) Letter to editor. J Hand Surg Am 24: 869–870
190 O’Brian M (ed) (2000) Aids to the examination of the peripheral nervous system, 4th edn. Elsevier, London Oaklander AL, Rismiller JG, Gelman LB, Zheng L, Chang Y, Gott R (2006) Evidence of focal small-fibre axonal degeneration in complex regional pain syndrome-1 (reflex sympathetic dystrophy). Pain 120:235–243 Perlas A, Chan V, Simons M (2003) Brachial plexus examination and localization using ultrasound and electrical stimulation. Anaes thesiology 99:429–435 Perry J, Ireland ML, Gronley J, Hoffer MM (1986) Predictive value of manual muscle testing and gait analysis in normal ankles by dynamic electromyography. Foot Ankle Int 6:254 Porter RW (1966) New test for finger sensation. BMJ 2:927 Rank BK, Wakefield AR, Hueston JT (1973) Surgery of repair as applied to hand injuries, 4th edn. Churchill Livingstone, Edinburgh Riddoch G (1940) The coin test. In: Surgical disorders of the peripheral nerves (Attributed: Seddon HJ 1975), Churchill Livingstone, Edinburgh, p 53 Rosén B, Jerosch-Herold C (2000) Comparing the responsiveness over time of two tactile gnosis test: two-point discrimination and the STI test. J Hand Surg 5:114–119 Rowntree T (1949) Anomalous innervation of the hand muscles. J Bone Joint Surg Br 31:505–510 Seddon HJ (1975) Surgical disorders of the peripheral nerves, 2nd edn. Churchill Livingstone, Edinburgh, London, New York, p 177 Seddon HJ (ed) (1954) Peripheral nerve injuries. Medical Research Council Special Report Series No 282. HMSO, London Sharrard WJW (1953) Correlations between the changes in the spinal cord and muscular paralysis in poliomyelitis. Proc R Soc Lon 40:346 Smith SJM (1996) The role of neurophysiological investigation in traumatic brachial plexus injuries in adults and children. J Hand Surg Br 21:145–148
Surgical Disorders of the Peripheral Nerves Smith SJM (1998) Electrodiagnosis. In: Birch R, Bonney G, Wynn Parry C (eds) Surgical disorders of the peripheral nerves, 1st edn. Churchill Livingstone, Edinburgh, London, pp 467–490, Chapter 19 Smith SJM (1999) Neurophysiological interpretation after nerve injury in the upper limb. Curr Orthopaed 13:27–32 Smith PJ, Mott G (1986) Sensory threshold and conductance testing in nerve injuries. J Hand Surg Br 11:157–162 Stewart M, Birch R (2001) Penetrating missile injuries. J Bone Joint Surg Br 83:517–524 Szabo RM, Gelberman RH, Williamson RV, Dellon AL, Yavu NC, Dimick NP (1984) Vibratory sensory testing in acute peripheral nerve compression. J Hand Surg Am 9:104–109 Tinel J (1915) Le signe du “fourmillement” dans les lésions des nerfs periphériques. Presse Méd 47:388–399 Tinel J (1917) Nerve wounds. Ballière Tindall & Cox, London. Authorised translation by Rothwell F, revised and edited by Joll CA Toros J, Karabay N, Őzaksar K, Sugun TS, Kayalar M, Bal E (2009) Evaluation of peripheral nerves of the upper limb of the upper limb with ultrasonography. J Bone Joint Surg Br 91:762–766 Trumble TE, Kahn U, Vanderhooft E, Bach AW (1995) A technique to quantitate motor recovery following nerve grafting. J Hand Surg Am 20:367–372 von Greulich M (1976) Der zweipunkte-Stern. Handchirurgie 8:97–99 Wynn Parry CB (1953) Electrical methods in diagnosis and prognosis of peripheral nerve injuries. Brain 76:229–265 Wynn Parry CB, Salter RM (1976) Sensory re-education after median nerve lesions. Br J Hand Surg 8:250–257 Wynn Parry CB (1981) Rehabilitation of the hand, 4th edn. Butterworth, London
6
Clinical Neurophysiology in Peripheral Nerve Injuries Shelagh Smith and Ravi Knight
Development of clinical neurophysiology as a diagnostic aid, applications and purpose of clinical neurophysiology, electrodiagnostic techniques and new methodologies, limitations and caveats, safety issues.
6.1 Introduction The Clinical Neurophysiologist participates in clinical evaluation of patients with known or suspected injury to a peripheral nerve, plexus or spinal root, and the information gained from neurophysiological evaluation often contributes directly to therapeutic decision making in such patients. Conduction along sensory and motor nerve axons is essentially an electrical process, and electrophysiological methods are a premier investigation for acquired disorders of peripheral nerves. Clinical neurophysiology alone does not always provide complete diagnostic information; nerve conduction studies and electromyography should be viewed as diagnostic aids, which extend the clinical examination and support other investigative tools, as outlined in Chap. 5. Electrodiagnostic techniques verify or exclude the clinical suspicion of a neuropathological process, and can offer a precise definition of the site, type and degree of a neural lesion, or reveal abnormalities that were clinically uncertain, silent or unsuspected. Electrodiagnostic study in neural injury is principally aimed at: • • • •
Localisation of the lesion Determination of pathophysiology Establishing severity Identification of reinnervation
The action potential in the peripheral nervous system was first described in the nineteenth century by Du Bois-Reymond and Bernstein (Seyffarth 2006; Pearce 2001) and is the fundamental unit of neural activity at the cellular level. The earliest application of electrophysiology in the study of human peripheral nerve disease was the demonstration that conduction velocity is slowed in injured regenerating nerve (Hodes et al. 1948). Considerable advances occurred during the 1940s and 1950s. Simpson reported delayed conduction in the median nerve in carpal tunnel syndrome (Simpson 1956), and Dawson in London developed an averaging technique for sensory nerve
recording using superimposed photographic traces of potentials (Dawson and Scott 1949). Further work with Dawson’s averaging method was carried out by Gilliatt and Sears at the Institute of Neurology in London, who applied it to the clinical study of nerve lesions (Gilliatt and Sears 1958), and P.K. Thomas, who described abnormalities of nerve conduction in ulnar neuropathies at the elbow and in the hand. An important contribution to the understanding of the basis of nerve conduction abnormalities was the experimental work on diphtheritic neuropathy by Lambert’s group at the Mayo Clinic, which identified segmental demyelination as the cause of slow nerve conduction velocity (Kaeser and Lambert 1962). Investigation of electrical properties of muscle fibres, notably by Lord Adrian and colleagues in Cambridge during the 1920s (Adrian 1925), and later by Buchthal in Sweden, led to the introduction of electromyography into clinical practice. Buchthal’s group went on to describe the characteristics of normal motor unit potentials, and the abnormalities seen on EMG in a wide range of myopathic and neurogenic disorders in man (Buchthal et al. 1957). Following these pioneering developments, a range of techniques for clinical study of peripheral and proximal nerves has been introduced and the pathophysiological basis of nerve conduction abnormalities clarified through correlation with morphological changes seen on nerve biopsy. The diagnostic role of electrophysiological investigation is further extended through introduction of methods for evaluation of previously inaccessible parts of the nervous system, including central motor and sensory pathways, and small myelinated and unmyelinated nerve fibres.
6.2 Electrodiagnostic Techniques In the clinical setting, summated responses from supramaximal stimulation of a sensory nerve (sensory nerve action potential) and motor nerve (compound muscle action potential) are
R. Birch, Surgical Disorders of the Peripheral Nerves, DOI: 10.1007/978-1-84882-108-8_6, © Springer-Verlag London Limited 2011
191
192
respectively recorded over the cutaneous course of the sensory nerve, and from the belly of the muscle, using surface electrodes applied to the skin. Needle electromyography (EMG) records electrical activity from within muscle, and describes the integrity of the motor unit – which comprises a single alpha-motor neurone and all of the muscle fibres it innervates. Action potentials of single motor and sensory nerve fibres are summated to produce compound responses or potentials. These have a triphasic character, and parameters such as amplitude, duration, latencies and conduction velocities are determined efficiently by computer software integral to modern neurophysiological equipment. Normative data are available for a range of nerves and muscles, and enable quantification of any deviation from the norm. Conventional or standard nerve conduction studies assess function and conduction in large fast conducting nerve fibres – principally the A beta nerve fibres that mediate fine touch, with velocities in the range of 40–60 m/s. Slower conducting small myelinated A-delta and unmyelinated B fibres, mediating pain and temperature sensation, with typical conduction velocities of 3–10 m/s, cannot be assessed with standard nerve conduction tests. Evaluation of small fibre function requires specialised techniques previously accessible in a few neuroscience centres, but now more widely available (detailed application of these techniques in clinical practice is further discussed in Chap. 5). Somatosensory evoked potentials are of especial value in establishing presence of a pre-ganglionic lesion, proximal to the sensory or dorsal root ganglion, through measurement of the passage of near-field potentials resulting from stimulation of a mixed nerve or dermatome, at the level of the lumbar or cervical spine, and further cranially at the somatosensory cortex. They may indicate the presence of clinically unsuspected disease, or a second level injury in the central nervous system. Case Report: A 37 year old man had sustained multiple shrapnel injuries including one penetrating the neck anteriorly, with a fragment still lodged in the spinal canal. He developed proximal weakness in the right upper limb and paraesthesia in a C5 and C6 distribution; he also complained of numbness in the left leg. The right brachial plexus was explored approximately 1 week after injury. The plexus was considered intact by direct visual inspection; the subclavian vein was repaired. The patient was referred for neurophysiological investigation because of persisting sensory symptoms and arm weakness. The nerves studied were bilateral median, ulnar and radial sensory; right median motor, right ulnar motor, bilateral musculocutaneous and axillary motor. Extensive EMG sampling was undertaken in the deltoid, infraspinatus, biceps, triceps, brachoradialis and first DIO muscles of the right arm, and in mid and lower cervical paraspinal muscles bilaterally. Lower limb SSEPs were performed with tibial nerve stimulation to evaluate sensory symptoms in the left leg. The abnormal findings were a relatively small median sensory potential from the right thumb (12 uV compared
Surgical Disorders of the Peripheral Nerves
with 35 uV in the left arm), and evidence of denervation (large or polyphasic MUPs, reduced recruitment) in right deltoid, infraspinatus and biceps. Fibrillations were present in infraspinatus > biceps. Cervical paraspinal muscles were normal. Cortical SSEPs from the left tibial nerve were poorly formed compared with the right leg; latency was normal but amplitude was reduced. Responses from the cervical cord were normal. Neurophysiological interpretation: evidence of a partial proximal lesion of the right brachial plexus affecting the upper trunk. The SSEP findings are suggestive of an axonal lesion in somatosensory pathways between the upper cervical spine and primary sensory cortex.
6.2.1 Glossary of Electrodiagnostic Procedures Compound sensory action potential (SNAP): is recorded following synchronous stimulation of a sensory nerve, or the sensory fibres within a mixed nerve. Stimulation and recording along the course of the sensory nerve are undertaken using electrodes placed on the skin surface. Needle electrodes may be used if the nerve is relatively inaccessible or deep to the skin surface (e.g., lateral femoral cutaneous nerve). SNAP amplitudes are small, in the order of microvolts; averaging of 10–20 responses is required to optimise signal to noise ratio (Fig. 6.1). SNAP onset and peak latencies are measured as the time interval to the initial deflection, and the next positive or negative deflection. Conduction velocity is calculated by dividing latency by the distance from the stimulating cathode (Fig. 6.2). Mixed nerve action potential (NAP): is obtained by percutaneous stimulation over nerve trunk, which contains sensory as well as motor fibres, such as the median nerve from
Fig. 6.1 Median sensory action potential recording (orthodromic): stimulation of digit 2, skin recording electrodes placed over median nerve at wrist.
Clinical Neurophysiology in Peripheral Nerve Injuries
193
Fig. 6.2 Median sensory action potential: A2 response from digit 2, A3 response from palmar stimulation (recorded on Nicolet Viking™ EMG system). Markers are set to onset and peak latency; amplitude measures are shown in far right column of numerical data.
the palm, or ulnar nerve action potential recorded from ulnar groove; evoked responses recorded over the skin surface. Amplitude of the NAP is similar to that of SNAPs. Compound muscle action potential (CMAP): is the summated response from the synchronous stimulation of motor nerve fibres evoked by per-cutaneous stimulation over the trunk of a motor nerve. The CMAP response is recorded through skin or needle electrodes placed near to the motor point of a muscle or muscle belly, with a reference electrode placed over an inactive site such as the bony point of muscle insertion (Figs. 6.3 and 6.4). CMAP responses typically have amplitude of several millivolts: they are much larger than SNAP or NAP potentials, and averaging of summated responses is generally not required. CMAP amplitude is influenced by the size, duration, and temporal dispersion of single muscle fibre action potentials, and number of functional motor units within the muscle. Amplitude measurements are made to negative peak, or peak to peak; distal latency is the time interval between stimulation of the nerve and the onset of the response. Motor conduction velocity or speed or propagation of action potential along the nerve is calculated as conduction time/conduction distance (Fig. 6.5).
Fig. 6.3 Median nerve motor study: CMAP recorded with electrodes placed over the APB muscle, with stimulation of the median nerve at the wrist above the carpal tunnel.
H reflex (after Hoffman) : a late CMAP response of consistent latency obtained by anterograde stimulation of alphamotor neurones, which activates Ia fibres innervating muscle
194
Fig. 6.4 Tibial nerve motor study: CMAP recorded with electrodes placed over the adductor hallucis muscle, with stimulation of the tibial nerve trunk at the medial malleolus.
spindles forming the afferent arc of this monosynaptic reflex. H reflexes are most readily obtained from the gastrocnemiussoleus muscles; the responses are analogous to the ankle jerk, and may be useful in suspected S1 radiculopathy. F response (from foot, where first obtained): is a late CMAP response of varying latency, produced by non-synaptic, retrograde activation of the anterior horn cell by electrical stimulation of the motor nerve. Activation of different motor units, and collision of action potentials accounts for
Surgical Disorders of the Peripheral Nerves
the variability of latency. F responses are a useful measure of conduction in proximal parts of motor nerves, which may be inaccessible to direct stimulation. Central motor conduction time (CMCT): measures the latency of the cortical motor evoked potential (MEP) to a muscle, from which is subtracted the peripheral conduction time from the anterior horn cell to the muscle. The former is usually determined by transcranial magnetic stimulation over the scalp, the latter estimated from F response latency (Figs. 6.6 and 6.7). CMCT is a particularly useful test of the gross integrity of the corticospinal or pyramidal pathway to a muscle. Tests of neuromuscular transmission: include repetitive stimulation at low (3 Hz) and high (20–50 Hz) frequencies, to assess muscle fibre fatiguability, and single fibre EMG (SFEMG). In repetitive nerve stimulation, a decrement of CMAP amplitude and area of >10% is found in disordered post-synaptic transmission, as in acquired and congenital myasthenia gravis. Disorders of neuromuscular transmission cause increased temporal variability in the firing patterns of individual muscle fibres within the same motor unit (“jitter”). This is measured using specialised single fibre EMG needles to record pairs of potentials; SFEMG is a more sensitive (95–98%) test of disordered neuromuscular transmission than repetitive stimulation (Fig. 6.8). Electrotonus: is a generic term for nerve excitability studies, which provide information about the membrane properties of the axon at the site of stimulation. In contrast to
Fig. 6.5 Normal compound muscle action potential recorded from APB. Arrow marks distal motor latency; amplitudes of the distal (wrist) and proximal (elbow) responses are measured from baseline to negative peak.
Clinical Neurophysiology in Peripheral Nerve Injuries
195
axon reflex loop, evoking a sweat response that can be quantified (QSART). This thermoregulatory sweat test is reliable and reproducible. 3. Thermal thresholds – a psychophysical test measuring threshold for perception of temperature sensation, useful in the interrogation of small myelinated A delta (cold threshold) and unmyelinated C (heat threshold) fibre function, with the method of limits used as the most common algorithm (Fig. 6.9).
Fig. 6.6 One of the authors (RK) demonstrating transcutaneous magnetic brain stimulation to measure central motor conduction time in a normal volunteer.
conventional nerve conduction techniques, electronus determines change in membrane potential caused by activation and deactivation of electrogenic ion pumps and channels (Kiernan et al. 2000, 2001a; Nodera and Kaji 2006). Nerve excitability studies are being used to delineate patterns and mechanisms of axonal injury, although there is to date very limited application in traumatic nerve injury. Quantitative sensory testing: small nerve fibre function can be assessed using 1. Sympathetic skin response – a galvanic response of the eccrine sweat glands to electrical (sometimes other sudden noxious) stimulation, and a useful screen of integrity of the sudomotor cholinergic system. The SSR is evoked by passing a current between two points on the skin, measuring electrical skin resistance between these points. 2. Sudomotor axon reflex – post-ganglionic sudomotor cholinergic axon terminals are antidromically stimulated with iontophoresed acetylcholine. This produces orthodromic release of acetyl choline at the nerve endings through an
Sensory evoked potentials (SSEPs) – external (electrical) stimulation of a peripheral nerve trunk in upper or lower limbs evokes responses in peripheral nerve pathways and the central nervous system. Short latency somatosensory evoked potentials (SSEPs) refer to that part of the waveform that occurs within 25 ms (N20) of median nerve stimulation, and within 40–50 ms of tibial nerve stimulation (P40). The best waveform is identified with multiple scalp recordings. Bilateral stimulation, and recording over multiple sites (peripheral nerve, plexus, spinal cord, somatosensory cortex) provides internal control and helps localise the anatomical level of the lesion (Fig. 6.10). Cutaneous stimulation in various dermatomal areas produces dermatomal SEPs (DSEPs). Sensory evoked potentials are used in evaluation of spinal cord disorders and radiculopathies, and in assessment of central sensory pathways. They are not disease specific, but reduction in amplitude generally indicates axonal loss, and prolonged latencies suggest demyelinating pathology, which may be subclinical. Electromyography (EMG): the recording and analysis of spontaneous, insertional and volitional electrical activity of muscle, usually by insertion of a concentric needle electrode. (The term “EMG” is often used colloquially to refer to electrodiagnostic studies that incorporate nerve conduction, with or without electromyography and other investigative techniques. This is misleading and should be avoided. Nerve conduction studies and electromyography, whilst inter-related, are distinct procedures).
• Insertional activity: describes electrical activity generated during needle electrode insertion. • Spontaneous activity following nerve injury leading to denervation typically includes fibrillation potentials and positive sharp waves (spontaneously generated action potentials of individual muscle fibres), fasciculations (spontaneous discharges of individual motor units), and complex repetitive discharges (polyphasic action potentials of uniform morphology and frequency [5–100 Hz], with abrupt onset and offset). EMG records motor unit action potentials (MUPs), generated from single discrete motor units, these being the anatomic and functional unit of an anterior horn cell and its connections (axons, neuromuscular junction) to the level of the muscle fibres. In EMG, MUPs represent the compound
196
Surgical Disorders of the Peripheral Nerves
Fig. 6.7 Motor evoked potentials in abductor digiti minimi in the right hand, transcranial magnetic stimulation applied to scalp overlying the left motor cortex. Sequential cortical evoked responses are displayed; central motor conduction time is calculated using shortest latency responses (A5 & A6).
Fig. 6.8 Compound muscle action potential (CMAPs) evoked by repetitive stimulation at 3 Hz. In this normal subject, there is no significant drop in amplitude or change in area of the evoked potentials in the eight sequential responses.
Clinical Neurophysiology in Peripheral Nerve Injuries
197
Fig. 6.9 thermal thresholds measured in foot of subject with Fabry’s disease (Morgan et al. 1990) in which there is preferential involvement of A-delta fibres, causing elevation of threshold to cooling.
firing characteristics or recruitment and interference pattern. Recruitment refers to the successive activation of motor units following volitional muscle contraction, and interference pattern, the electrical activity recorded during maximal contraction of the muscle (Fig. 6.11). Quantitative EMG (QEMG): uses descriptive as well as automated quantitative methods such as turns/amplitude analysis, frequency analysis to describe morphology and firing characteristics of MUPs. Surface EMG: this records grouped MUPs from muscles using surface electrodes applied to the skin surface, and is useful for multi-level muscle recording in neurogenic disorders to detect fasciculations, and in the study of neurological disorders such as tremor and dystonia (movement disorders in which there is sustained focal, segmental or generalised muscle contraction or tone).
6.2.2 Intra-operative Neurophysiological Procedures
Fig. 6.10 SSEP recording, median nerve stimulation at wrist, evoked response in plexus (N9), spinal cord (N13) and sensory cortex (N20).
action potentials of muscle fibres within the recording territory of the needle electrode, and are described in terms of their duration, amplitude, number of phases, and by the
Electrophysiological monitoring is used in a wide range of surgical procedures to assist the surgeon and maximise functional return for the patient. Techniques have been developed to guide dissection, identify the region of injury, monitor function in sensory and motor nerves and pathways, and protect against iatrogenic injury. A medico-legal role is secondary to these clinical purposes. Monitoring techniques must have the fundamental ability to provide rapid and reliable information about function, at an early and reversible stage
198
Surgical Disorders of the Peripheral Nerves
Fig. 6.11 Normal MUPs recorded from concentric EMG needle placed in deltoid. Upper trace shows full recruitment pattern of motor units of varying size and firing rates.
of potential neural injury. Choice of technique is determined by surgical procedure and neural structures at risk; careful planning between clinical neurophysiologist, surgeon and anaesthetist is required ahead of surgery to select the most appropriate monitoring procedure, enable correct positioning of electrodes used for recording and stimulation, and optimise anaesthesia. Inhalation anaesthetics abolish or markedly attenuate SSEPs, and full neuromuscular blockade will prevent acquisition of EMG; for monitoring of motor nerve or root function using free running or stimulated EMG, partial neuromuscular blockade (up to 75%) (Holland 1998) or short acting neuromuscular blocking agents are required. Free running and stimulated EMG: is used to monitor function in muscles innervated by specific motor nerves or motor roots considered to be at risk during surgery; needle EMG electrodes are placed into the belly of muscles to record motor unit activity evoked by neural stimulation, or to record abnormal high frequency EMG activity (neurotonic discharges) which appear when blunt trauma is applied to motor nerves (Beatty et al. 1995). For stimulated EMG, bipolar neural stimulation is preferable, as this minimises current
spread to closely lying or adjacent nerves – particularly problematic in plexus surgery. Current spread may also occur through liquid pools in the surgical field; in this situation, hook electrodes can be used to lift the nerve clear for stimulation. EMG signals cannot be recorded during diathermy, because of high frequency artefact. Nerve action potentials: are recorded direct from a nerve to assess continuity across the site of a nerve injury or lesion. In the setting of a post-ganglionic nerve injury, the presence of a nerve action potential indicates at least several thousand intact or regenerating axons, and correlates with functional recovery (Kline and Happel 1993), potentially obviating the need for nerve graft or repair. Somatosensory, dermatomal and root evoked potentials: are used to assess afferent pathways from peripheral nerve or dermatome through plexus and dorsal columns of the spinal cord to somatosensory cortex. Intra-operative SSEPs are the most commonly used modality in spinal surgery to reduce iatrogenic neurological injury, but give limited or no information about individual roots and motor pathways. Recording electrodes may be placed epidurally, on the surface of the
Clinical Neurophysiology in Peripheral Nerve Injuries
cervical cord, or over the cortex. Predictors of post-operative deficit are a 50% reduction in SSEP amplitude; the operation should stop if this occurs, and resume only when there is recovery of the potential. Evidence that SSEP monitoring may protect against spinal cord injury is drawn principally from retrospective reviews of outcome, with and without monitoring; there are no randomised controlled trials of efficacy – such trials now viewed as being unethical. Motor evoked potentials (MEPs): are increasingly used to assess efferent motor pathways and function in corticospinal tracts during spinal surgery and neurosurgery (Sutter et al. 2007). Current methodology includes ‘D’ wave monitoring and direct recording of EMG in individual muscles. The ‘D’ wave recorded epidurally, represents descending asynaptic volleys in the fast conducting neurones in the corticospinal tract, evoked by single pulse transcranial stimulation. Direct EMG/MUP recording requires multi-pulse stimulation, with multiple motor tracts being involved in generation of responses. Long term motor deficit has been reported to occur post-operatively when there is >50% reduction in the cord D wave in combination with bilateral loss of muscle EMG (Deletis and Sala 2007), whereas motor function was only transiently impaired following surgery in those cases showing unilateral or bilateral loss of muscle EMG, but an unchanged or lesser reduction (<50%) in D wave response (Sala et al. 2001). Additional intra-operative techniques: include cranial nerve monitoring in skull base surgery: recording of facial muscle EMG and facial nerve stimulation to evoke CMAPs; brain stem acoustic evoked responses to monitor function in the VIII cranial nerve; and free running EMG to assess function of the trigeminal nerve, spinal accessory nerve, and the cranial nerves innervating extra-ocular muscles.
6.2.3 Electrodiagnostic Techniques and Localisation Clinical neurophysiology offers a powerful menu of techniques to assess neural integrity along the entire length of peripheral and central nervous systems. Sensory nerve recording is used to determine whether a lesion lies in postganglionic or pre-ganglionic fibres; focal nerve lesions can be precisely identified by demonstration of focal conduction block and inching techniques, with stimulation of a nerve at different levels along its course, and needle EMG sampling of the muscles and myotomes it supplies. Delineation of the site of damage in the brachial plexus or lumbo-sacral plexus is established by measuring evoked responses in sensory and motor nerve branches arising from plexus elements, in combination with EMG sampling of multiple muscles. The electrodiagnostic triad of root avulsion is (1) normal SNAP (first recognised in seminal work by Bonney and Gilliat 1958) and
199
a preserved histamine response (2) denervation in paraspinal muscles and (3) loss of the somatosensory evoked potential. Together with evidence of fibrillations and absence of voluntary motor unit activity in peripheral muscles supplied by the root, this localising triad is of great value in distinguishing a proximal root lesion from more distal pathology, when there is clinical suspicion of root avulsion in traumatic injuries affecting the brachial plexus. Anatomical localisation using a combination of nerve conduction studies and EMG is illustrated by the patient who presents with unilateral foot drop. Loss of the superficial peroneal SNAP and denervation in tibialis anterior, peroneus longus and peroneus tertius muscles indicates a lesion of the common peroneal nerve at the knee, and slowing of motor conduction and reduction in evoked muscle action potential (conduction block) on stimulation of the nerve as it traverses the lateral knee localises a compressive lesion to the fibula head. If there are denervation changes on EMG in short head of biceps femoris, this would signify a more proximal lesion involving the peroneal component of the sciatic nerve. A normal superficial peroneal sensory response and denervation in tibialis posterior as well as muscles in the anterior and lateral compartments of the lower leg are indicative of L5 radiculopathy.
6.3 Limitations and Pitfalls of Electrodiagnostic Investigation There are pitfalls that may trap the unwary in both the performance and interpretation of nerve conduction studies. Biological variables can affect nerve conduction, and these must be taken into account and controlled for carefully. Age is an important factor (Hyllienmark et al. 1995): in children below the age of 5 years, conduction velocities are approximately one-third to one-half those in older children and adults, due to age related progressive myelination of peripheral nerves. Amplitudes of sensory nerve action potentials decline in the elderly; transcutaneous sural nerve responses and certain other sensory potentials may not be recordable above the age of 65 years. Normative data used in the neurophysiology laboratory must be appropriate for the age range of the patient population. Height can also affect peripheral nerve conduction velocity, more so in lower limb nerves: conduction velocity in sural, tibial and peroneal nerves is inversely correlated with height, whereas median nerve conduction velocity shows no relationship to height in normal subjects (Rivner et al. 1990). One possible explanation for the height effect is axonal size – distal segments of longer length axons have smaller diameter, less myelin and shorter inter-nodal distance. However, the most important biological variable in clinical practice is limb temperature. Skin temperature below 30°C will reduce nerve conduction velocity and enhance neuromuscular transmission
200
(Denys 1991). Amplitudes of nerve action potentials are increased in cool limbs, possibly as a result of temperaturerelated lengthening of the opening time of sodium channels in the nerve membrane. It should be standard practice to measure skin temperature, and to maintain skin surface temperature above 30°C for the duration of the test. If measured, skin temperature should then be documented in the neurophysiological report. Technical factors influence nerve conduction measurement and validity of data. Waveform morphology and amplitude is dependant on the position of the indifferent or reference electrode. Stimulus artefact may affect signal quality, and limit the ability to accurately measure latency and amplitude of potentials. Normative data are referenced to maximal stimulation of nerves; insufficient or sub-maximal stimulation may result in spuriously low amplitudes or slow velocities. A nerve response that is apparently absent may be so because of erroneous placement of recording electrodes – this is particularly challenging when assessing unfamiliar nerves, notwithstanding the wealth of excellent reference texts of surface anatomy. Inaccurate measurement of nerve length, either technical or the result of incorrect assumption as to the course of a nerve, is a common source of error when calculating nerve conduction velocity – and is especially problematic in short nerve segments, such as the ulnar nerve as it circumvents the elbow through the ulnar groove. Least error will made in an ulnar nerve conduction study if the elbow is flexed to >90° for stimulation. Knowledge of normal anatomical variants or anomalous innervation is essential (Sonck et al. 1991), and some anomalies are relatively common. The Martin–Gruber anastomosis between the median and ulnar nerves in the forearm is estimated to occur in about 20% of the population. In this anastomosis, median nerve fibres leave the main nerve trunk or the anterior interosseus branch in the forearm, cross over to the ulnar nerve and terminate in intrinsic hand muscles normally supplied by the ulnar nerve. This manifests electrophysiologically as a median evoked CMAP at the wrist being larger than that evoked by stimulation at the wrist – a reversal of the usual pattern of CMAP amplitude – and/or anomalous morphology of the CMAP. A Martin-Gruber anastomosis can present diagnostic difficulties in carpal tunnel syndrome, because some median fibres will be spared from compression at the wrist, and the anatomical variant may also affect electrodiagnosis of ulnar neuropathies at the elbow, where a severe lesion may appear less so because of anomalous innervation of intrinsic hand muscles by the crossed median nerve fibres. Other anomalous innervations include the “all median” and “all ulnar” hands, variations in sensory innervation on the dorsal side of the hand, and the accessory deep peroneal nerve, a branch of the deep peroneal nerve which runs around the lateral malleolus to innervate part of the extensor digitorum muscles in the foot.
Surgical Disorders of the Peripheral Nerves
Errors can be led by the examiner, who may fail to test enough nerves for identification of a more diffuse problem such as a generalised neuropathy underlying an apparently focal entrapment, or who may omit stimulation of nerves in proximal sites for detection of high level conduction block, or who applies the wrong procedure for the clinical problem, such as standard nerve conduction studies in a patient whose clinical symptoms suggest a small fibre neuropathic process. In EMG examination of muscles, several sites must be sampled, since the recording area of a concentric needle is quite small, with the potential to miss a focal area of denervation, reinnervation or myopathic change. Reinnervation proceeds along time as well as space, and assessment of its rate and degree will require serial EMG recordings. Whilst needle EMG may confirm reinnervation, clinical return of movement across the joint may not occur as the number of reinnervated muscle fibres may not be sufficient. Unlike certain other laboratory tests which have absolute meaning if abnormal, clinical neurophysiological data must always be placed in clinical context, and numerical results should not be over-interpreted. For example, sub-clinical slowing of median nerve conduction across the wrist does not equate in itself to carpal tunnel syndrome. Over simplification of findings must be avoided: some patients will have multiple diagnoses, such as a co-existing neuropathy in a patient with obvious primary muscle disease. Crucially, the test strategy should be based on the patient’s symptoms and signs, rather than adhering to fixed protocols of investigation. An important consideration when using serial nerve conduction studies to evaluate progression of a neuropathy, or recovery of a focal nerve lesion, is inter- and intra-examiner variability. Intra-examiner consistency of results is high, whereas amplitudes of SNAPs and CMAPs can vary quite considerably when tests are performed by different examiners (Bleasel and Tuck 1991). Hence, longitudinal nerve conduction studies are best undertaken by the same examiner. Most will be gained from clinical neurophysiological testing if the questions posed by the referring clinician are specific, explicit and answerable. Referrals should be considered in terms of (1) what knowledge will be gained about the patient from the investigation (2) how and whether the result may alter management (3) the degree of urgency of the request. The latter has some practical importance in the UK where clinical neurophysiology is a relatively scarce resource. Regular collaboration and communication between clinicians and clinical neurophysiologists is highly advantageous; the joint peripheral nerve trauma clinics run between the Royal National Orthopaedic Hospitals and the Clinical Neurophysiology department at the National Hospital, Queen Square, since the mid 1980s are a testament to the clinical value of such interaction. The report of the investigation should contain numerical data for all nerves and muscles included in the test, a
Clinical Neurophysiology in Peripheral Nerve Injuries
summary of the findings, and a clinical interpretation. Styles vary according to local laboratory practice, and to some extent, automated report generating systems embedded in modern neurophysiology recording equipment.
6.4 Safety Aspects Nerve conduction studies and needle EMG are generally well tolerated, although patients will experience some discomfort from transcutaneous nerve stimulation and EMG needle insertion. Test procedures are relatively safe, being largely non-invasive (American Association of Electro diagnostic Medicine Professional Practice Committee 1999), but there are some risks associated with procedures (Al Shekhlee et al. 2003). EMG can be hazardous in patients with coagulation disorders or who are taking anti-coagulants, such as warfarin, that prolong prothrombin or partial thromboplastin time. Patients on intensive care often have a higher risk of clotting disorders, or may be on high dose heparin; a clotting screen is prudent before EMG investigation is performed. Needle EMG is safe in patients on low dose aspirin. Peritonitis is a potential complication of needle examination of intercostal and abdominal muscles, and pneumothorax is a risk of needle insertion into intercostal, serratus anterior, suprapinatus and paraspinal muscles. Nerve injury may occur on needle EMG examination of gluteal muscles (sciatic), flexor pollicis longus (superficial radial), pronator quadratus (ulnar) and pronator teres (median). These hazards can be avoided by appropriate positioning of the patient and careful EMG technique. There is a small risk that the electrical field generated by nerve stimulation may alter function of implanted cardiac pacemakers and cardioverter/defibrillator devices, and cardiological advice should be obtained before proceeding. Electrodes must be placed at least 15 cm from the device; stimulus duration should be less than 0.2 ms, and repetition rate no more than 1 Hz. Repetitive nerve stimulation techniques, including those for evaluation of recovery cycles and nerve excitability should not be performed in patient with pacemakers in situ. Appropriate infection control measures are used to minimise risk of transmission of diseases such as Hepatitis B and C, HIV and transmissible spongiform encephalopathies. Infection risk is increased when performing EMG in intrinsic foot muscles in patients with peripheral vascular disease, or in limbs of patients who have undergone extensive lymph node dissection, such as in breast cancer surgery (Al Shekhlee et al. 2003), and should only be performed in such cases if the clinical value of diagnostic information obtained from EMG outweighs infection hazard.
201
6.5 Pathophysiological Correlates Neurophysiological correlates of morphological and functional changes following different types of neural injury, distinction between neurapraxia, axonotmesis, neurotmesis, time course of neurophysiological changes after injury, neurophysiological markers of regeneration and reinnervation
The pathophysiological response of large fibre peripheral nerves to a wide range of injurious processes and pathologies is limited either to loss of axons or demyelination, or a combination of these two processes. The repertoire of response in small diameter unmyelinated nerve fibres is even more restricted – these will either survive or show axon loss (Table 6.1). The principal electrophysiological markers of demyelination are reduction in conduction velocity and occurrence of conduction block. Axonal loss manifests electrophysiologically as attenuation of amplitude of evoked sensory or motor response, with relatively little change in conduction velocity, unless axonal degeneration is of severe degree. Up to 30% reduction in conduction velocity may be seen in less extensive axonal atrophy, due probably to axonal shrinkage or secondary demyelination (Baba et al. 1982). Needle EMG has particular value in assessing the degree and duration of motor nerve axonal loss. Fibrillation potentials and other markers of acute denervation appear first in proximal muscles; the timing of their appearance is dependant on the distance between the site of the nerve lesion and muscle, and is also influenced to some extent by age (see later section on birth lesions of the brachial plexus). There may be an interval of up to 10–40 days before fibrillations appear, limiting the value of early EMG in assessment of axonal damage after nerve trauma, and quantification of fibrillations does not give Table 6.1 electrophysiological consequences of demyelination. Electrophysiological change Pathophysiological basis Slowing of conduction
Caused by delays in regeneration of action potentials at the nodes of Ranvier, or by a transformation of saltatory to linear conduction in a demyelinated segment (Bostock and Sears 1976)
Conduction block
Failure of transmission of action potentials
Decrement in CMAP
Impaired transmission of trains of action potentials manifest as decrement of the CMAP to repetitive stimulation at high frequencies, as a result of greater internal longitudinal resistance and unavailability of current for impulse propagation in the demyelinated segment (Davis 1972). CMAP compound muscle action potential.
202
a direct measure of the degree of axonal loss. As the motor unit undergoes remodelling during denervation and reinnervation, motor unit potentials undergo morphological change from highly polyphasic, complex unstable units seen in the earlier stages to more stable, large units of longer duration that signal axonal reinnervation and re-growth.
6.5.1 Types of Nerve Lesion: the Electrophysiological Consequences In severe focal nerve injury which results in axonal damage or death, conduction failure or block occurs at the site of injury. Distal to this, axons undergo a process of Wallerian degeneration but continue to conduct impulses until axonal degeneration is sufficiently advanced to cause conduction failure. With lesser degrees of neural injury, myelinated nerve fibres may undergo focal demyelination, resulting in either conduction slowing or conduction block. Conduction slowing per se does not produce clinical symptoms, whereas the clinical effects of conduction block are similar to those of axon loss, resulting in weakness or sensory deficit. Demyelinating conduction block may co-exist with axonal degeneration, and the CMAP can aid distinction of focal demyelination causing conduction block from axonal degeneration – a near normal CMAP at 7 days after injury in a clinically weak muscle indicates that significant axonal loss or denervation has not occurred. Both conduction block and conduction failure affect amplitudes of evoked responses from motor and sensory nerves. However, in conduction block, amplitudes are affected when stimulating above or at the site of block, but not below the lesion, whereas conduction failure affects amplitudes when stimulating at any point along the nerve length, distal or proximal to the lesion. The optimal timing for neurophysiological assessment of degree and type of lesion after the initial injury is usually between days 8 and 10, but the length of the distal stump is important in determining the time course of pathophysiological change, with short distal stumps showing the earliest onset of SNAP or CMAP abnormality. In general terms, the CMAP response declines by day 3–5 after nerve injury, and is absent by day 7–9 (Chaudhry and Cornblath 1992). Changes in the sensory response tend to lag behind CMAP diminution by 2–3 days. Immediately after nerve transection, no changes are evident at the level of the neuromuscular junction and the sarcolemmal membrane. Depending upon the length of the distal stump, there is failure of neuromuscular transmission 8–20 h after nerve section (Miledi and Slater 1970). Other length dependent changes of note are a reduction in resting membrane potential and increased sensitivity to acetyl choline (denervation hypersensitivity). The latter occurs as a result of a proliferation of acetyl choline receptors at sites
Surgical Disorders of the Peripheral Nerves
other than the endplate. In motor nerves, abnormal spontaneous activity (fibrillations, sharp waves) occurs 1–4 weeks after the injury, according to distance between the site of neural injury and the motor units in the target muscle. Fibrillation potentials appear when there is spontaneous and asynchronous depolarisation of the muscle membrane. At the ultrastructual level, fibrillations are initiated at the endplates as result of increased sodium conductance, and propagate along the muscle fibre (Smith and Thesleff 1976). The intensity of fibrillation activity approximates to the severity of axonal damage, and the size of the potentials and frequency of firing diminish over time (Fig. 6.12). Nerve lesions due to compression, which are characterised by neurapraxia or axonotmesis, result in focal slowing of conduction as a consequence of localised narrowing, with nodal and paranodal demyelination. Distal to the point of compression, axonal narrowing may produce slowing along the length of the nerve. The earliest electrophysiological abnormality in compressive neuropathies producing conduction block is a significant reduction of the amplitude of the CMAP proximal to the site of the compression. If Wallerian degeneration occurs, there is disruption of axonal conduction to the muscle fibre, which then fires spontaneously, manifesting electrophysiologically as fibrillations and positive sharp waves. Compressive neuropathies can be primarily Schwann cell mediated, rather than a mild form of Wallerian degeneration. Animal studies have shown early Schwann cell apoptosis and proliferation following nerve compression, with down-regulation of myelin
25
20
15
10
5
0
1
2
3
4
5
6
7
8
9
10
Number of days after nerve injury CMAP amplitude (mV) SNAP amplitude (microvolts) Fibrillations (frequency)
Fig. 6.12 Temporal sequence of electrophysiological changes after nerve injury.
Clinical Neurophysiology in Peripheral Nerve Injuries
203
proteins, myelin associated glycoprotein in particular. Axonal integrity is generally unimpaired, allowing for local de-and remyelination, as well as axonal sprouting. Seddon in his classic review (Seddon 1943), recounts the historical classification of nerve injuries into (1) complete anatomical division, (2) a lesion in continuity, and (3) transient block. He credits Prof Henry Cohen with creation of the now standard terms of neurotmesis, where there is loss of anatomical and functional continuity, although the epineurium may be intact, axonotmesis, in which complete peripheral (Wallerian) degeneration occurs but recovery takes place because of the preservation of axon sheaths and the internal architecture, and neurapraxia, where invariably complete recovery of function occurs without peripheral degeneration. Sunderland extended classification of neural injury by emphasising histological changes that occur in a nerve trunk in his categorisation of five different types or degrees of nerve injury (Sunderland 1990). The relationship between these classification systems, and the electrophysiological changes that may be seen in different types of nerve injury, are summarised in (Table 6.2). Neurapraxia: blunt trauma or stretch produces neurapraxia, which is characterised pathologically by focal demyelination. Loss of the myelin sheath impairs the transmission of impulses by changing the biophysical properties of the paranodal and internodal membranes (Bostock and Sears 1976). There is an increase in membrane capacitance and a reduction in the transverse resistance in paranodal and internodal regions, with a resultant deficiency in the current available to depolarise the next node of Ranvier. Demyelination of the paranodal segment in particular markedly influences membrane capacitance, with exposure of the potassium channels leading to persistent hyperpolarisation. In addition, the Na+/K+ ATPase electrogenic pump further drives the membrane towards the
K+ equilibrium potential, which in turn reduces the safety factor for transmission of the action potential (Sears and Bostock 1981). Conduction block thus ensues, characterised clinically by acute weakness in the affected limb, and motor nerve conduction studies typically demonstrate a significant drop in amplitude across the demyelinated nerve segment. Early experimental work confirmed the earliest involvement of large myelinated fibres (Aguayo et al. 1971), with the most superficial fibres being most severely affected. In an animal model of traumatic conduction block in the baboon (Papio papio), in which neurapraxia was induced by pneumatic pressure, the spatial extent or length of the compressive lesion correlated directly with the severity and duration of conduction block (Ochoa et al. 1971). Since axons remain intact in a neurapraxic lesion, function is restored by focal remyelination within days to weeks. Nerve dysfunction may be even more shortlived in transient conduction block following deforming mechanical injury such as compression and traction. The contribution of vascular or ischaemic injury to conduction block was raised by Denny-Brown and others (Denny Brown and Brenner 1944), with subsequent researchers concluding that both biomechanical and vascular factors were responsible. However histological evidence from the work of Gilliatt and associates (Gilliatt 1981), in particular the finding of nodal intussusception, reinforces the role of mechanical deformation as the principal mechanism in the pathogenesis of conduction block in compressive nerve lesions. Neurophysiologically, the neurapraxic lesion manifests with a small/absent CMAP stimulating at and above the site of the lesion, whereas nerve stimulation below the lesion evokes a normal response. Focal slowing of conduction may be demonstrable across the lesion. There are similar changes in the evoked responses of sensory nerves, but demonstration of conduction block in sensory nerves is more difficult, since
Table 6.2 Nerve injury classification and their electrophysiological correlates. Classsification Seddon Sunderland Pathology
Electrophysiological Correlate
Neurapraxia
Grade 1
Myelin injury or ischaemia
Conduction block, with or without conduction slowing
Axonotmesis
Grade 2
Axon loss
Fibrillations
Stromal derangement
Mild diminution to complete absence of SNAP and CMAP responses, in proportion to degree of axonal loss
Endoneurium, perineurium and epineurium intact
± varing degrees of conduction block and slowing associated with myelin injury
Grade 3
Endoneurium disrupted
Fibrillations
Grade 4
Perineurium disrupted
Grade 5
Epineurium disrupted
Nerve distal to lesion shows normal conduction
Absent SNAP and CMAP responses Fibrillations Absent SNAP and CMAP responses Neurotmesis
Fibrillations
Absent SNAPs and CMAPs CV conduction velocity, SNAP sensory nerve action potential, CMAP compound muscle action potential.
204
there is a normal drop in SNAP amplitude when the distance between recording and stimulation sites is 25 cm or more. Fibrillations and other evidence of denervation do not occur in neurapraxic injury. However, since these changes may not appear for up to 2 weeks, neurapraxic lesions cannot be conclusively distinguished from axonal degenerative lesions in the first few days after injury. The term neurapraxia, which typically guarantees rapid and complete recovery, is best reserved for those situations where electrodiagnosis has conclusively shown that demyelinating conduction block is solely responsible for the neural lesion. Axonotmesis: In axonal injuries that do not involve the supporting perineurium, the perineurial sheath provides a channel for axonal regeneration from the cell body, facilitating recovery. Although axonal function is disrupted immediately after the injury, the disconnected distal segment survives for the next 4–7 days. As Wallerian degeneration proceeds centrifugally, the distal segment becomes progressively inexcitable. Axonotmesis and neurotmesis show similar electrophysiological features in the first few days after nerve injury. Initially, the CMAP is small or absent proximal to the lesion, and normal below the site of injury, prior to Wallerian degeneration. In the second week after injury, axonotmetic lesions show progressive reduction in CMAP and SNAP amplitude distal to the site of injury, and fibrillations may appear, depending on distance between site of injury and muscle. Regeneration occurs from the intact nerve cell body, at an average rate of 1 mm/day. The extent to which recovery is possible depends on the degree of internal disorganisation within the nerve, and the distance between the lesion site and the end organs. Neurotmesis In neurotmesis, the nerve is severed or axonal disorganisation is so severe that axonal regrowth is impossible. Axons in the nerve segment below the site of injury remain viable, and therefore excitable, for a few days. If in the second week of injury, stimulation of the distal segment evokes no response, the term neurotmesis can be applied. However, even up to 2 weeks after injury, axonotmesis or lesser degrees of axonal damage cannot be reliably distinguished from neurotmesis using electrophysiologic testing. Neurotmesis is established with certainty only by direct observation of the nerve, but will be suspected in open wounds where significant axonal loss is likely to have occurred through penetrating injury. Since electrodiagnostic testing cannot reliably distinguish neurotmesis and lesser degrees of axonal injury in the first few days after nerve injury, early surgical exploration is recommended if there is suspicion that the nerve is transected or fully sectioned. However early (day 2–3) EMG examination is advocated in the context of iatrogenous nerve injury, since demonstration of even a few volitional motor unit action potentials is evidence for neural continuity although this may not be clinically apparent (Aminoff 2004). Demonstration of conduction across the site of the nerve injury using direct nerve stimulation is also valuable in this situation.
Surgical Disorders of the Peripheral Nerves
6.5.1.1 Special Types of Neural Injury Cold thermal nerve injury: significant nerve injury is found at temperatures below 10°C. Action potentials are preserved between 5°C and 15°C (Paintal 1965). Depending upon the degree of cooling and the duration of exposure, the continuum of pathology includes mild reversible conduction block at the milder end, and axonal swelling and Wallerian degeneration at the severe extremity. The epineurium and perineurium are relatively resistant to injury and are therefore preserved (Denny-Brown et al. 1945). Electrophysiologically, there is an increase in the nerve action potential amplitude and duration (area under the curve) initially, with significant reduction of amplitude and area beginning at 16°C. The temperature is independent of nerve diameter, but lower in unmyelinated nerve fibres (Franz and Iggo 1968). There is also a reduction in conduction velocity, likely related to altered function at the sodium channels (Kiernan et al. 2001b), although experimental studies on rat sciatic nerve suggest that passive properties of the axon may contribute to the reduction in conduction velocity seen at low temperatures (Stecker and Baylor 2009). In the clinical situation, trench foot caused by non-freezing cold, is a typical example of cold induced nerve injury. Although the literature is scarce, case reports suggest damage to myelinated and unmyelinated nerve fibres, with ischaemia and reperfusion contributing to the injury (Irwin et al. 1997). Electrical neural injury: these represent 3–6% of admissions to burns units (Sances et al. 1979). The mechanism of injury is somewhat contentious, although generated heat is generally regarded as being the principal causative agent (Lee and Astumian 1996). Electroporation, a transient opening of ion channels in the cell membrane and ion fluxes leading to cell death, is sometimes invoked as a mechanism in tissue injury (Jellinek 1960). A direct correlation between the strength of current and duration of exposure, with a critical level of irreversible injury (30 mA/3 mm nerve diameter/5 s shock) was noted in experimental studies (Alexander 1941), but scarring and concomitant vascular injury significantly influence the severity and ultimate outcome of neural injury. Nerve injuries may be caused by direct heat (contact burn), and range in severity from reversible conduction block to severe axonal loss. Typically, isolated or multiple mononeuropathies in the vicinity of the electrical discharge occur acutely in about a third of electrical injuries (Duncan 2001) although delayed motor neuronopathies, suggesting involvement of the anterior spinal cord have been described (Ratnayake et al. 1996; Fu et al. 2008), and may relate to ischaemia of the anterior spinal artery (Ko et al. 2004). Radiation injury to nerves: ionising radiation as a cause of extrinsic nerve damage has been recognised for several decades. The pattern of the response is independent of the type of ionising radiation, and shows an initial period of enhanced activity, reflected in increased conduction velocity
Clinical Neurophysiology in Peripheral Nerve Injuries
and spike amplitude, and reduction in threshold, likely caused by a physico-chemical change in the nerve membrane itself. There is then progressive decline of activity, with the greatest vulnerability seen in small diameter sensory fibres. The total dose of irradiation, rather than the frequency, determines the severity of damage, which may range from myelin fragmentation to Wallerian degeneration (Love 1983). Secondary neural damage is caused by fibrotic and ischaemic change. The most common clinical situation is iatropathic, with brachial and lumbosacral plexus lesions following radiotherapy for adenocarcinomas of the breast, ovary, and lymphomas accounting for most cases. Rare mononeuropathies of the sciatic and femoral nerve have been described (Gikas et al. 2008). Radiation induced brachial plexopathy is more common in younger patients and in those receiving concomitant chemotherapy, and is directly correlated with fraction size, with >2 Gy doses producing more extensive and severe lesions (Olsen et al. 1993). Sensory disturbance is universal. Electromyography is useful in differentiating neoplastic from radiation plexopathy, as myokymias and fasciculations are characteristic of the latter (Lederman and Wilbourn 1984).
6.5.2 Regeneration and Reinnervation Rates of axon regeneration appear to be similar in all somatic nerve fibres (Sunderland 1991). The rate of maturation correlates inversely with nerve thickness, and is influenced by local factors such as level, duration and severity of the lesion, as well as systemic factors such as age and temperature. Nerve repair occurs by two processes – axonal sprouting, with new growth of axons making functional contact with end organs or muscle fibres, and axonal regeneration along the nerve fibres. Histologically, nodal sprouting from the nodes of Ranvier, as well as terminal sprouting from intact nerve terminals, can be identified (Edds 1949). Nodal sprouting is principally determined by the distance between the node and the endplate zone, through the vacant perineurial sheath (Hoffman 1950; Slack and Williams 1981). Factors influencing nodal sprouting include age, type of injury (crush versus complete section), and type of muscle fibre (fast versus slow twitch). When nerve injury is complete, recovery can only take place through axonal regeneration, and reliable electrophysiological monitoring of motor nerve axonal growth is difficult until neuromuscular junctions begin to be reestablished. Reinnervation is heralded by reduction in the amount of fibrillations and other spontaneous activity in a muscle, but this is not easily quantified. The earliest definite electrophysiological evidence of reinnervation is recruitment of a few small unstable MUPs on volitional effort, or evoked by stimulation of the motor nerve distal to the site of injury (nascent units) (Table 6.3).
205 Table 6.3 Electromyography (EMG) findings in denervation and reinnervation. Denervation
Reinnervation
Spontaneous activity – fibrillations, positive sharp waves in acute denervation; fasciculations and complex repetitive discharges in chronic denervation Early
Normal MUPs with increased duration because of late potentials or satellite fibres incorporated through collateral sprouting
On-going
Moderate amplitude polyphasic MUPs of long duration, unstable firing due to variable conduction along unmyelinated sprouts and low safety margin of neuromuscular transmission
Late
Large amplitude increased duration MUPs with stable transmission
MUP motor unit potential.
Immature nerves are typically relatively inexcitable and require higher levels of stimulus intensity to evoke a response. Regenerating nerves also show slower conduction speeds, before remyelination is complete, which manifests as slow velocity and increased distal motor latencies. As maturation and regeneration proceeds, the number of MUPs and size of evoked CMAP increases. On needle EMG, MUPs show increased duration reflecting the enlarged motor unit, and increased polyphasia reflecting reinnervation and increased temporal dispersion of the individual muscle fibre potentials. Recruitment is reduced, and MUP firing rates increase disproportionately in relation to the number of motor units activated. Evidence of recovery of sensory nerves is usually found around the same time as the small unstable MUPs begin to be detected. Regenerating sensory potentials are initially low amplitude and often show dispersion. Recordable nerve action potential responses indicate the presence of at least several thousand moderate-diameter regenerating fibres, and this number of fibres correlates with clinical recovery of injured nerves in experimental animal models (Kline and DeJonge 1968). Whilst clinical tests such as an advancing Hoffman-Tinel sign provide useful indication of recovery in sensory nerves, quantitative sensory tests can help to establish the degree and type of small fibre involvement, although agreement between the two methods may not be evident (Leffler and Hansson 2008). In partial nerve injury, recovery incorporates regeneration of severely affected axons, and reinnervation through collateral sprouting from nerve fibres spared by the injury. Collateral reinnervation usually precedes axonal regeneration, and results in motor units that have increased amplitude, fibre density, territory and duration. The degree of motor axon regeneration may be quantifiable through measures such as automated analysis of MUP shape and firing
206
rates (Dorfman et al. 1989), estimation of motor unit number (McMomas et al. 1971) and estimation of axon number by approximation of conduction velocity (Dorfman 1984). Electrophysiological evaluation can play only a very limited role in the early stages of neural regeneration, before sensory or muscle end organs are reinnervated – although this is arguably the stage at which assessment of potential for recovery would be most valuable for the clinician. Even the detection of a few motor units in previously paralyzed muscles does not necessarily imply that recovery will be complete enough to result in return of useful clinical function.
6.6 Clinical Applications Neurophysiological findings in focal nerve lesions of the upper and lower limbs, plexopathies, radiculopathies, diffuse or multilevel disorders; clinical applications and uses of specific electrodiagnostic techniques
6.6.1 Upper Limb Neuropathies 6.6.1.1 Median Nerve Median neuropathy at the wrist Carpal tunnel syndrome, with a corrected average annual incidence of 60–120/100,000 is probably the commonest surgical referral to the neurophysiology department. Dysfunction of the median nerve is caused by increased pressure between the flexor retinaculum and the floor of the carpal tunnel. Clinical symptomatology includes weakness in the hand, and sensory disturbance that may involve the forearm and upper arm, as well as the hand, with nocturnal and positional provocation of sensory symptoms as well recognised features. There is a bimodal distribution of frequency with a peak at 50–59 years, and another at 75–84 years. Aetiology is multifactorial, with factors such as age, sex, body mass index (BMI) and pregnancy, as well as diseases such as diabetes and rheumatoid arthritis contributing to the disorder (Bland 2005; Becker et al. 2002). Some controversy surrounds the aetio-pathogenesis of the nerve damage, with both deranged myelination and ischaemic axonal injury invoked as primary mechanisms (Kiernan et al. 1999). Correlation of clinical severity with degree of neurophysiological abnormality has been validated in several studies. Bland, using one of the largest databases of 8,501 patients with carpal tunnel syndrome, has demonstrated a strong linear correlation of clinical with neurophysiological severity using a 7 point scoring system of electro physiological abnormality (Bland 2000) (Table 6.4).
Surgical Disorders of the Peripheral Nerves Table 6.4 Neurophysiological grading of median nerve compression in carpal tunnel syndrome, Canterbury system (Adapted from Bland 2000). Grade Degree of Neurophysiological abnormality severity 0
Nil
Nil
1
Very mild CTS
Abnormality detected only in two sensitive tests (e.g., inching, palm-wrist median/ulnar comparison, ring finger “double peak”)
2
Mild CTS
Orthodromic sensory conduction velocity from index finger to wrist < 40 m/s with motor terminal latency from wrist to abductor pollicis brevis [APB] < 4.5 ms
3
Moderately severe CTS
Motor terminal latency > 4.5 ms and < 6.5 ms; preserved index finger SNAP
4
Severe CTS
Motor terminal latency > 4.5 ms and < 6.5 ms, with absent SNAP
5
Very severe CTS
Motor terminal latency > 6.5 ms
6
Extremely severe CTS
Surface motor potential from APB < 0.2 mV, peak-to-peak
Median and sensory nerve conduction studies (NCS) provide accurate and reproducible measures that allow diagnosis or confirmation of carpal tunnel syndrome with a high degree of sensitivity (66–82%) and specificity (82–97%) (Jablecki et al. 2002). Using a clinical questionnaire as a criterion standard, standard electrophysiological tests have moderate sensitivity and specificity, with ulnar-median sensory latency comparison having the highest diagnostic accuracy. Electrodiagnostic yield is increased by combining standard NCS with additional sensory and motor nerve tests (Chang et al. 2002), including short nerve segment testing such as palm-wrist sensory conduction (Mills 1985), comparison of conduction measures in the median nerve with those in other unaffected nerves, such as the ulnar or radial nerve, second lumbrical/first dorsal interosseus motor latency comparison (Walters and Murray 2001) and median terminal latency comparison (Lee et al. 2009). The clinical value of new electrodiagnostic methods, which bring very small incremental improvement in diagnostic yield, may be debatable: additional new tests must be robust and efficient in application, with no undue prolongation of overall procedure time for quality of information gained. The British Society for Clinical Neurophysiology has developed good practice guidelines for electrodiagnostic verification of suspected carpal tunnel syndrome, that are suitable for multi-disciplinary use (www.BSCN.org.uk) (Table 6.5). Needle EMG of abductor pollicus brevis is indicated in suspected carpal tunnel syndrome when there is clinically
Clinical Neurophysiology in Peripheral Nerve Injuries
207
Table 6.5 Standards, guidelines and options for electrodiagnostic procedures in suspected carpal tunnel syndrome (Adapted from practice guidelines published by the British Society for Clinical Neurophysiology, www.BSCN.org.uk). Neurophysiological test Additional procedure Standards
Hand temperature is measured and maintained above 30°C
Report includes numerical data and statement of abnormality detected
SNC in median digital sensory nerve in most affected hand using surface electrodes, and measuring response amp. & latency/velocity
Professional status of practitioner performing the investigation and report is identified
Comparative test of conduction in a digital nerve not innervated by the median nerve is performed in the same hand
The report is signed by the person taking medico-legal responsibility for it
Median MNC across wrist is measured in affected hand, using surface electrode and measuring response amp. & latency/velocity Guidelines
Digital SNC is performed in the contra-lateral hand
Referrals are screened to assess appropriateness
A second test of median SNC performed. This may include: median palmar sensory study; median/ulnar palmar ratio; median/radial sensory latency comparison to thumb; median/ulnar sensory latency comparison to ring finger
Clinical history and examination performed to identify co-existing disease
MNC in the ulnar nerve is performed in the affected limb using surface electrodes and measuring response ampl. & latency/conduction velocity Median MNC is performed in the contra-lateral limb Options
A second test of median MNC is performed, such as median/ulnar motor latency comparison to second lumbrical and second interosseous
The report contains illustrations of recorded waveforms
Needle EMG recording of median innervated arm muscles is performed by a medically qualified practitioner. This may include recording of the abductor pollicis brevis CMAP during median nerve stimulation, but not as a substitute for MNC measured as per standard above F wave latencies are recorded SNC sensory nerve conduction, ampl amplitude, MNC motor nerve conduction, EMG electromyography, CMAP compound muscle action potential.
evident muscle wasting, or carpal tunnel syndrome graded as severe or above, and in acute presentation of traumatic/atraumatic median nerve compression to detect axonal damage. More extensive nerve and muscle testing will be required if there is suspicion of co-existing disease, or disorders encompassing mononeuritis multiplex and generalised polyneuropathy to motor radiculopathy and anterior horn cell disease. Whilst symptoms usually begin and are more severe in the dominant hand, population based studies have demonstrated bilateral electrophysiological abnormalities in over half of the sample surveyed (Bland and Rudolfer 2003). The incidence of bilateral carpal tunnel syndrome rises to 87% in a hospital based prospective series (Padua et al. 1998) and underlines the importance of examining both hands. The degree of conduction abnormality seen pre-operatively is some guide to the rate and amount of recovery of neural function following surgery. When pre-operative nerve conduction studies demonstrate conduction block as the predominant abnormality, full and rapid response to carpal tunnel decompression is expected. Recovery will be slower if axonal damage has intervened, and probably incomplete. Goodman and Gilliat’s studies of the effect of treatment in what was
then characterised as severe carpal tunnel syndrome (distal motor latency to abductor pollicis brevis >10 ms) showed that median motor latency took up to 18 months to recover, and residual abnormalities were common. In their cases deemed moderate, with latency of 7–10 ms, motor conduction took 8–12 months to recover; mild cases (latency <5 ms) recovered within 6 months (Goodman and Gilliatt 1961). Sensitive techniques such as analysis of stimulus response curves show transient reduction in amplitude of the CMAP as well as the input-output curve at approximately 1 month after surgical decompression (Ginanneschi et al. 2008) . These parameters were significantly improved at 6 months, drawing attention to the early apparent electrophysiological deterioration, likely caused by impaired efficiency of axonal recruitment immediately following surgery. High median neuropathies: These unusual lesions occur at the level of the shoulder, elbow or within the forearm, with direct trauma accounting for the majority of cases. Extrinsic compression, such as caused by axillary crutches, may also produce a high median neuropathy, in isolation or combination with radial and ulnar neuropathies. A lesion of the median nerve in the shoulder region is associated with
208
weakness of forearm pronation and wrist flexion, and flexion of the lateral two digits. Flexor digitorum superficialis involvement also produces weakness of flexion involving the proximal interphalangeal joint of the ring and little fingers. Sensory loss involves the thenar eminence and the thumb, index and the lateral side of ring finger. At the elbow, the nerve may be compressed at three discrete sites: proximally by an aberrant ligament of Struther’s above the medial epicondyle, at or just below the elbow by the pronator teres muscle, or more distally when the trunk continues in the forearm as the anterior interosseous nerve. A distinct pronator syndrome was described by Seyffarth in 1951, where compression is attributable to the lacertus fibrosus, a hypertrophied pronator teres, or a fibrous band of the flexor digitorum superficialis (Seyffarth 1951). Proximal median entrapment neuropathies may mimic symptoms of carpal tunnel syndrome (Lee and Lastayo 2004), or produce nonspecific pain and sensory disturbance in the forearm, provoked by repetitive forearm movement. Needle EMG sampling is the single most useful electrophysiogical examination in high median neuropathies, as sensory and motor studies in the hand do not discriminate between distal and proximal nerve lesions. Electromyography also helps exclude differential diagnoses that include cervical radiculopathy and anterior horn cell disease. Case Report: This 25-year old serving soldier suffered open fractures of his left radius and ulna upon exposure to 105 mm artillery round (shrapnel): multiple entry wounds were seen over the dorsal and volar surfaces of the forearm. He underwent fasciotomies and a left sural cable graft to the median nerve in the forearm within the first week of his injuries. Post-operatively, extensive soft tissue scarring made clinical assessment of muscle function difficult. There was no discernible movement in thenar muscles. The patient was thus referred for neurophysiological assessment, to evaluate post-operative neural continuity, and reinnervation of more distal muscles. Electrophysiology, by short segment inching stimulation studies along the median nerve below the elbow to the level of flexor pollicis evoked low amplitude CMAPs from APB. EMG examination of APB and FPL revealed nascent type unstable motor unit potentials, indicative of axonal regeneration and functional continuity of the nerve 4 months after surgical repair. It was possible to offer a favorable prognosis for recovery (Fig. 6.13). Anterior interosseous nerve syndrome: This purely motor branch arises about 5 cm distal to the lateral epicondyle, and innervates the flexor pollicis longus, flexor digitorum superficialis to the index and middle finger and pronator quadratus. Isolated lesions are associated with trauma, and occur as a feature of brachial neuritis. Whilst needle EMG sampling remains the electrodiagnostic investigation of choice, distal latency to the pronator quadratus muscle may show prolongation, and is a useful adjunct (Nakano 1977).
Surgical Disorders of the Peripheral Nerves
Fig. 6.13 Post-operative appearance of forearm in case 2, following fasciotomy and sural cable graft to median nerve.
6.6.1.2 Ulnar Nerve Ulnar neuropathy at the elbow: Simpson in 1956 was the first to demonstrate the utility of nerve conduction studies in the localisation of ulnar neuropathy at the elbow (Simpson 1956). The importance of electrophysiology in localisation of the nerve lesion is underscored by a recent study reporting poor sensitivity (28–60%) and specificity (53–80%) of the clinical utility of provocative tests such as Tinel’s sign in the diagnosis of ulnar entrapment neuropathies at the elbow (Beekman et al. 2004). A standard ulnar nerve electrodiagnostic study should include recording of an ulnar digital SNAP, ulnar NAP at elbow, CMAP from an ulnar-innervated hand muscle, ulnar nerve conduction velocity measured elbow to wrist, above elbow to wrist, and above elbow to below elbow, and needle EMG examination of the first dorsal interosseous (the most frequently abnormal muscle, Stewart 1987) and ulnar innervated forearm flexor muscles. Ulnar nerve conduction studies have been beset by controversy regarding the optimal elbow position in which to test the nerve, given the discrepancy between true length of the nerve and measured skin distance in different elbow positions, potentially resulting in spuriously slow conduction around the elbow. The recommended elbow position is flexion between 70° and 90° from horizontal, as this minimises errors in measurement of nerve length (Campbell et al. 1999). Nerve conduction studies can localise a lesion to the elbow either by demonstration of slowing of the conduction velocity in the across-elbow segment compared to normal, or slowing of conduction in the elbow segment in relation to the below elbow to wrist segment. Published practice parameters (Campbell et al. 1999) recommend an across elbow conduction velocity of less than 50 m/s, or slowing of more than 10 m/s in the above elbow to below elbow segment compared to the below elbow to wrist segment as being indicative of abnormal conduction. A reduction in CMAP amplitude of more than 20% or a clear change in compound muscle
Clinical Neurophysiology in Peripheral Nerve Injuries
action potential morphology between the below elbow and above elbow sites are also suggestive of ulnar neuropathy at the elbow. Electrodiagnosis in mild lesions of the ulnar nerve, manifesting with sensory symptoms alone, can be difficult as nerve conduction studies are often normal or equivocal in these cases. Abnormalities of the ulnar sensory potential are sensitive for detection of ulnar nerve pathology, but do not usually help to localise the lesion. Motor studies have been shown to be more sensitive than the ulnar mixed sensory study across the elbow in localising ulnar neuropathy of the elbow (Kothari et al. 1998a). Considerable methodological variability contributes to the wide confidence intervals of electrodiagnostic accuracy in different published studies, but overall, the literature reports sensitivities of 25–100% for slowing of motor conduction, and up to 68% for conduction block. If routine motor studies recording from adductor digiti minimi are inconclusive, conduction studies should then be made using the first dorsal interosseus muscle, or vice versa, as nerve fibre bundles to individual muscles may be differentially affected in compressive lesions of the ulnar nerve at the elbow. Peripheral nerves have a
209
complex internal arrangement, comprising fascicles in which nerve fibres bound for a particular target are organised together. In a partial nerve lesion, fascicular involvement will be determined by proximity of a specific fascicle to the injurious lesion. Short segment conduction studies (“inching”) significantly increase detection of ulnar nerve lesions at the elbow (Gooch et al. 2003), and may also provide precise localisation, such as to the cubital tunnel region where the nerve is vulnerable to compression by the fascia connecting the two heads of the flexor carpi ulnaris muscle. The technique of inching involves movement of the stimulating electrode over small distances, looking for abrupt change in CMAP amplitude, morphology or latency (Campbell and Geiringer 1998); it can be utilised for the localisation of many focal neuropathies in upper and lower limbs. Whilst the typical electrophysiological findings in ulnar neuropathy at the elbow are attenuation or loss of the ulnar digital sensory response, and conduction slowing around the elbow, other features of demyelination, such as temporal dispersion and conduction block may also be present (Figs. 6.14–6.16).
Fig. 6.14 Slowing of the evoked motor response from adductor digiti minimi, maximal across the elbow.
210
Surgical Disorders of the Peripheral Nerves
Fig. 6.15 Temporal dispersion of the compound muscle action potential (CMAP) from first dorsal interosseous at the elbow.
Intra-operative electroneurography at the time of ulnar nerve exploration can assist precise localisation of the site of ulnar nerve compression – Campbell and colleagues have employed peri-operative stimulation for this purpose, and to identify residual compression in patients undergoing reexploration (Campbell et al. 1988). In addition to defining the level of the lesion, electrodiagnosis can help determine prognosis. Electrodiagnostic signs of demyelination – presence of motor conduction block and motor velocity slowing across the elbow – were found to be associated with better outcome, independent of other factors in 74 ulnar neuropathies at the elbow (Beekman et al. 2004). In a recent review of a large series (n = 244 arms) of ulnar neuropathy at the elbow, severe motor conduction block signified by conduction block of greater than 50% was associated with a more favorable long term prognosis as assessed by clinical examination and patient assessment at a median follow up period of 25 months (Dunselman and Visser 2008). In those with severe motor conduction block, outcome as assessed by clinical examination was good in 81%, and 84% when judged by patient assessment. These outcomes were significantly better than those in the patient group showing motor conduction block of 20% or less (44% and 41% respectively).
As with other apparent mononeuropathies, an underlying or predisposing neuropathy may be present in patients with ulnar nerve lesions at the elbow, and electrodiagnostic assessment should include examination of at least one other upper limb nerve. If denervation in hand muscles is found in suspected ulnar neuropathy at the elbow, needle EMG should be performed in non-ulnar muscles innervated by the C8 root, medial cord and lower trunk, to exclude brachial plexopathy or a radicular lesion. Lesions of the ulnar nerve in the upper arm above the elbow such as due to compression by the arcade of Struthers or external compression in the axilla by arm crutches are difficult to distinguish reliably from more distal lesions using EMG. In these cases, percutaneous serial short segment stimulation or inching studies may help in localisation through demonstration of highly focal conduction block or other CMAP abnormality. Distal ulnar nerve lesions: Ulnar nerve lesions at or below the wrist are relatively uncommon; the clinical neurophysiologist is most likely to encounter a distal lesion in Guyon’s canal, or a lesion of the deep palmar branch to first dorsal interosseus. Distal ulnar neuropathies can manifest with a pure motor, pure sensory or mixed clinical picture, according
Clinical Neurophysiology in Peripheral Nerve Injuries
211
Fig. 6.16 Conduction block below the elbow, with significant (>50%) drop in amplitude in the proximal segment.
to whether the lesion is sited in the main nerve trunk, sensory branch or deep motor branches. Wu and colleagues have developed a precise classification of distal ulnar lesion, by categorising 55 published cases of ulnar neuropathy at the wrist into five groups according to clinical and anatomic presentation (Wu et al 1985). The type I lesion (commonest) is proximal to or within Guyon’s canal, involves both the superficial and deep branches, and causes mixed motor and sensory deficit. The neurophysiological findings will be reduction or loss of the ulnar digital SNAP, denervation of all ulnar innervated hand muscles, and abnormalities of conduction dependant on degree of demyelination and axonal damage in the nerve. The dorsal ulnar cutaneous branch, which exits the main trunk of the ulnar above the wrist, is spared, and this SNAP is a useful discriminator between distal ulnar neuropathy and ulnar nerve lesions at the elbow. The type II lesion, usually located in Guyon’s canal or in the palmaris brevis, involves the superficial branch and will manifest neurophyiologically with a small or absent ulnar digital SNAP. Types III through V are pure motor syndromes. Type III is caused by a lesion proximal to the branch to the hypothenar muscles, and causes global ulnar hand weakness. Type IV is caused by a lesion distal to the hypothenar branches, resulting in
weakness of all muscles except the hypothenars. Type V (least common) is caused by a lesion just proximal to the termination of the palmar branch, and produces weakness limited to the first dorsal interosseous and adductor pollicis, sparing other interossei and the ulnar innervated lumbricals. Neurophysiological findings in the pure motor types are typically denervation in relevant ulnar hand muscles. Needle EMG examination of palmaris brevis, innervated by the superficial branch of the ulnar nerve arising near Guyon’s canal, is helpful, as this muscle is spared in deep motor branch lesions (Iyer 1998). When the lesion is primarily compressive, such as in damage to the deep branch of ulnar in keen cyclists, there may be prolongation of distal motor latency to first dorsal interosseus (Ebeling et al. 1960).
6.6.1.3 Radial Nerve The radial nerve can be compressed or damaged in the upper arm, or around the elbow; occasionally a lesion may affect the superficial sensory branch in the forearm or at the wrist. Radial or posterior interosseus nerve involvement occurs in multifocal motor neuropathy with conduction
212
block, and the radial nerve is also selectively affected in leprosy and lead poisoning. A frequently encountered radial neuropathy is “Saturday night palsy,” when the nerve is compressed against the humeral shaft within the spiral groove of the humerus bone. Although the lesion is primarily neurapraxic in type (Watson and Brown 1993), additional mild denervation is often seen on EMG in extensor digitorum communis, extensor pollicis longus, supinator and extensor carpi radialis. The lateral head of triceps may also be involved, depending on the exact point of nerve compression. It is sometimes possible to locate the site of conduction block in the upper arm by measuring CMAPs from extensor pollicis longus or extensor indicis proprius, stimulating the radial nerve above and below the spiral groove. Recovery may be protracted and eventually incomplete if amplitudes of the radial SAP and CMAPs from radially innervated muscles are reduced by more than 50% distal to the site of compression (Watson and Brown 1993). Lesions of the C7 root, or a brachial plexopathy involving middle trunk or posterior cord, are the main differential diagnoses in suspected lesions of the radial nerve above the elbow, and electrodiagnostic assessment should include needle EMG of other muscles innervated by the C7 root (serratus anterior, latissimus dorsi, lower cervical paraspinal muscles). Posterior cutaneous sensory conduction studies may confirm a postganglionic C7 lesion localisation or exclude a posterior interosseous neuropathy; when abnormal, they may predict a poor outcome (Lo et al. 2004). Pure lesions of the superficial radial nerve are rare, and then usually due to direct compression such as from handcuffs, tight casts or a watch band, or a superficial radial neuritis (Wartenberg syndrome). Symptoms can be misleading, and some cases are misdiagnosed as tenosynovitis in the wrist. Diagnosis is established by showing an abnormal superficial radial sensory potential, typically a reduction in amplitude and perhaps slowing of sensory conduction, in comparison with the unaffected arm or the lateral antebrachial cutaneous nerve on the same side (Spindler and Dellon 1990). Posterior interosseus neuropathy. This motor branch of the radial nerve exits the main trunk just below the elbow, to enter the supinator muscle under the arcade of Frohse. Compression at the arcade or within the supinator muscle will result in wrist and finger drop, but weakness of wrist extension is partial due to sparing of extensor carpi radialis. Nerve conduction studies in posterior interosseous neuropathy are of limited use (Fisher 1993). Denervation on EMG may be seen in muscles innervated by the posterior interosseus nerve; normal EMG findings in supinator, extensor carpi radialis and other proximal muscles distinguishes a posterior interosseus neuropathy from a high radial nerve lesion. The superficial radial sensory response is unaffected, or shows minimal reduction in amplitude. Conduction block or focal
Surgical Disorders of the Peripheral Nerves
slowing in the region of the arcade of Frohse may be demonstrable in neurapraxic posterior interosseus lesions, but is technically difficult. A radial or posterior interosseus neuropathy is sometimes considered in patients with treatment resistant “tennis elbow” or pain on repeated supination/ pronation of the elbow – and often classified as “radial tunnel syndrome.” Variable neurophysiological abnormalities have been reported in such cases. A series of 28 patients found that routine radial motor conduction studies were of no value, but an increase in motor latency of more than 0.5 ms was seen when conduction studies were repeated during forced supination of the arm. Inconclusive abnormalities were found on EMG of muscles innervated by the posterior interosseus nerve (Rosen and Werner 1980). Whether these cases represent a true compression neuropathy is uncertain.
6.6.1.4 Suprascapular Nerve Suprascapular nerve entrapment should be considered in patients with unexplained shoulder pain or weakness. The nerve is most commonly compressed within the suprascapular notch, resulting in denervation of both infraspinatus and supraspinatus muscles. Less often, entrapment is located distally at the spinoglenoid notch, and infraspinatus alone is affected (Kiss and Komar 1990, Bouzaidi et al. 2005). The suprascapular nerve appears to be vulnerable to injury at any point along its length due to relative fixation of the nerve at its origin in the brachial plexus and termination in the infraspinatus muscle, with potential for angulation at the suprascapular notch (Rengachary et al. 1979). Studies by Sever before current research ethical standards showed that if the head and shoulder of an infant at post-mortem were pulled in opposite directions with great force, the suprascapular nerve was the first to snap (Sever 1916). Electro diagnosis is principally by needle EMG, to reveal denervation in supraspinatus and/or infraspinatus, and distinguish suprascapular nerve lesions from more widespread processes, brachial plexopathy or C5 root pathology. Neurogenic changes on EMG are not uncommonly chronic in type as diagnosis of suprascapular entrapment is often a late clinical consideration. Additionally, suprascapular nerve conduction can be measured (Kraft 1972), stimulating at Erb’s point and measuring CMAPs in the supraspinatus and infraspinatus muscles using either surface or needle electrodes; normative data for amplitudes and latencies have been published (Casazza et al. 1998). Latency and CMAP values may have predictive value for outcome of treatment: in a retrospective survey of patient rated motor outcome, 12 of 16 patients with good or excellent outcome who did not undergo surgery had normal motor latency but significant reduction (greater than 50%) in CMAP amplitude, implying reversible conduction block not requiring surgical decompression (Smith 2004).
Clinical Neurophysiology in Peripheral Nerve Injuries
6.6.1.5 Spinal Accessory Nerve Whilst a rare neuropathy overall, this nerve is not infrequently damaged during lymph node dissection in the neck. Lesions typically produce weakness of the trapezius, with scapular winging – which in our experience is not uncommonly mis-diagnosed clinically as a long thoracic nerve palsy (Fig. 6.17a and b). Very proximal lesions, such as due to infiltration by neoplastic processes, may also produce weakness of lateral neck rotation, with involvement of the sternomastoid. Electromyography helps to exclude other causes of scapular winging such as lesions of the long thoracic or dorsal scapular nerves. Intra-operative studies have been described as helpful in revealing lesions in continuity, with nerve action potentials recorded beyond the lesion or demonstration of trapezius contraction on direct stimulation of the spinal accessory nerve (Donner and Kline 1993). Case Report: This 55 year old woman presented with shoulder weakness and stiffness following surgical removal of a cervical lymph node. Scapular winging was evident. She was seen in the joint Royal National Orthopaedic
a
b
213
Hospital/National Hospital for Neurology and Neurosurgery peripheral nerve trauma clinic, where clinical and electromyographic examination confirmed an isolated right spinal accessory nerve lesion with fibrillations and a reduced interference pattern of wide duration motor units in the trapezius muscle, with sparing of the sternomastoid, allowing for uncomplicated surgical reconstruction of the nerve, and near complete recovery.
6.6.1.6 Axillary (Circumflex) Nerve The axillary nerve is the smaller division of the posterior cord of the brachial plexus, and derives from the C5–6 roots. From its origin in the axilla, the nerve passes deep and posterior to the subscapularis, and enters the quadrilateral space, where the branch to teres minor separates. It then passes deep to the posterior part of the deltoid muscle, gives off the superior lateral brachial cutaneous branch, and winds around the neck of the humerus, innervating the lateral and anterior fibres of the deltoid. The nerve is most frequently damaged by traction injury to the shoulder, and is clinically characterised by weakness of shoulder abduction and sensory loss in the chevron distribution. Anterior shoulder dislocation and fractures of the proximal humerus account for a significant proportion of the traumatic lesions (de Laat et al. 1994). Unreduced dislocations and older age are significantly associated with nerve damage (Pasila et al. 1978). Neuralgic amyotrophy not infrequently affects the axillary nerve – 6% as a mononeuropathy in Parsonage and Turner’s classic series, and 42% in combination with the suprascapular nerve (Parsonage and Turner 1948). Nerve conduction studies are of limited value in axillary nerve injuries, as the single proximal (or supraclavicular) site of stimulation at Erb’s point precludes distinction between axonal and demyelinating lesions. The key electrodiagnostic procedure is needle EMG to examine deltoid and teres minor, with sampling of other muscles as appropriate to exclude a brachial plexopathy or accompanying trauma of other nerves such as the suprascapular. Needle EMG is also helpful in assessing extent and timing of recovery as teres minor and the posterior fibres of the deltoid recover earliest after the initial insult.
6.6.1.7 Long Thoracic Nerve
Fig. 6.17 (a) and (b) showing trapezius muscle wasting in a case of spinal accessory neuropathy (case 3 above).
The long thoracic nerve arises from the anterior rami of the C5, C6 and (usually) C7 roots, passes lateral to the brachial plexus behind the clavicle, and down the anterolateral chest wall to supply the serratus anterior muscle. Blunt, as in backpack use, or penetrating trauma are often implicated in isolated lesions of this nerve (Makela et al. 2006). The long thoracic nerve is involved in as much as 70% of cases of
214
Surgical Disorders of the Peripheral Nerves
idiopathic brachial neuritis (van Alfen and van Engelen 2006). Presentation of a long thoracic nerve lesion is with muscular pain, and weakness of shoulder abduction caused by a failure of anterior rotation, and scapular winging exacerbated by shoulder abduction is noted on clinical examination. Needle EMG will demonstrate neurogenic changes in serratus anterior, and helps to exclude more extensive neural involvement such as in brachial neuritis, or a muscle disorder such as facio-scapular muscular dystrophy, which may present to the orthopaedic surgeon with isolated unilateral scapular winging.
6.6.1.8 Brachial Plexus Lesions Diverse pathologies affect the brachial plexus – trauma, neoplastic infiltration, radiation induced damage, inflammatory processes including brachial neuritis, hereditary pressure palsies, and compression due to cervical ribs or bands. In a large series (n = 203) of brachial plexus lesions from the Johns Hopkins hospital (Moghekar et al. 2007), of median age 37 years ranging from birth to 79 years, the three commonest pathologies were brachial neuritis, birth injury and trauma. The lesion was supraclavicular in 90%, and infraclavicular in 10%, with the upper trunk being most often affected (27%), followed by the lower trunk (11%), upper and middle trunks 11%, lower and middle trunk 7%, and upper, middle and lower trunk in 25%. In general, electrophysiological testing is primarily focused on establishing the level of pathology through the combination of sensory nerve studies (amplitudes of SNAPs for median digits 1–3, ulnar, radial, lateral antebrachial cutaneous, medial antebrachial cutaneous) and needle EMG, to delineate lesions affecting roots, trunks, cords and peripheral nerves and distinguish a pre-ganglionic lesion from a process affecting neural elements distal to the dorsal root ganglion. Electrodiagnostic evaluation must be comprehensive, even exhaustive, to identify clinically unsuspected involvement within the plexus (Table 6.6).
Table 6.6 Electrodiagnostic findings in pre and post-ganglionic lesions. Post-Ganglionic Pre-Ganglionic SNAPs
Absent/small
Normal/intact
CMAPs
Absent/small
Absent
F responses
Absent or delayed
Absent
Distal muscles
Denervation
Denervation
Paraspinal muscles
Normal
Denervation
Plexus SSEP
Absent
Intact
Cortical SSEP Absent Absent SNAP sensory nerve action potential, SSEP somatosensory evoked potential.
Ferrante and Wilbourn (Ferrante and Wilbourn 1995) assessed utility of sensory nerve conduction responses in 53 cases of brachial plexopathy with a single truncal element involved. In upper trunk lesions, almost all had abnormal median digit 1 and lateral antebrachial cutaneous SNAPs; 58% showed small or absent radial sensory responses. In patients with lower trunk lesions, the ulnar (digit 5) SNAP was abnormal in almost all, and the medial antebrachial cutaneous SNAP in 65%. The single case with a middle trunk lesion showed abnormality of the median digit 3 SNAP. Evidence of denervation on needle EMG in spinatii, biceps, deltoid, triceps, pronator teres and ECR indicates an upper trunk lesion; in ECR, ECU, EDC and EIP, a middle trunk lesion, and in FCU and intrinsic hand muscles, a lower trunk lesion. Reduction in CMAP responses from median and ulnar innervated muscles provides additional evidence of a lesion affecting the lower trunk or medial cord. Cervical root and brachial plexus lesions can be distinguished by measuring SNAPs, median and ulnar SSEPs, and needle EMG of axial, limb and paraspinal muscles. Needle sampling of rhomboids, innervated by the dorsal scapular nerve which exits directly from the C5 root, differentiates an upper trunk plexopathy from C5 radiculopathy. EMG of serratus anterior is useful for identification of a C7 lesion, as the long thoracic nerve leaves the plexus above the upper and medial trunks (Table 6.7). Determining the type of neural injury is more difficult, particularly in acute plexus lesions, as evidence of axonal damage may not manifest for several days after injury, and conduction block is less readily demonstrated in plexus elements compared with more accessible peripheral nerve trunks. Proximal conduction is amenable to measurement by use of F responses and percutaneous high intensity electrical stimulation of cervical roots in the neck and the plexus at Erb’s point; median and ulnar SSEPs may be used to evaluate conduction in sensory pathways traversing the brachial plexus and cervical cord. Adult traumatic brachial plexopathy: The need for, and timing, of surgical exploration in adults with acute traumatic injury of the brachial plexus will often be determined by associated injury to bony and vascular elements. Moreover, operative repair should be undertaken as promptly as permitted by the clinical condition of the patient, since outcome of surgery to the brachial plexus is adversely influenced by delay in operation. Thus, the contribution of neurophysiological evaluation in acute plexus trauma is limited before surgery, as the evidence of denervation in distal muscles and features which allows confident electrodiagnostic distinction of partial from complete neural injury is not apparent in the first few days after nerve trauma. Indeed, the optimal time for neurophysiological evaluation is somewhere between 3 and 12 weeks after plexus injury – electrodiagnostic testing is likely to be falsely negative for denervation, if EMG is undertaken within 3 weeks of trauma.
Clinical Neurophysiology in Peripheral Nerve Injuries
215
Table 6.7 Brachial plexus elements – innervation of muscles and origin of upper limb sensory nerves. Level Nerves Muscles Snap Roots Trunk
C5
Dorsal scapular
Rhomboids
C5,6,7
Long thoracic
Serratus anterior
Upper
Suprascapular
Supraspinatus Infraspinatus
Middle
Lower
Cords
Lateral Posterior
Medial
Lateral pectoral
upper part of pect. major
Musculocutaneous
Biceps, brachialis
Median (lateral)
Pronator teres, FCR
Part of radial
Brachoradialis
Axillary
Deltoid, teres minor
Thoracodorsal
Latissimus dorsi
Subscapular
Teres major
Radial
all except brachoradialis
Median (lateral)
Pronator teres, FCR
Medial pectoral
lower part of pect. major
Ulnar
all ulnar muscles
Median (medial)
FDS, FDP (2,3), FPL, median hand muscles
Radial (C8 fibres)
EIP, EPB, (EDC,ECU)
Musculocutaneous
Biceps, brachialis
Median (lateral)
Pron. Teres, FCR
thoracodorsal
Latissimus dorsi
Subscapular
Teres major
Axillary
Deltoid, teres minor
Radial
All radial innervated
Ulnar
All ulnar innervated
Median (medial)
FDS, FDP, FPL, median
Median digit 1, lateral antebrachial cutaneous, some radial
Median digit 2 & 3, some radial
Ulnar digit 5, dorsal ulnar cutaneous, med. antebrachial cutaneous
Hand muscles
At the time of surgery, nerve continuity can be assessed by direct intraoperative nerve stimulation to record evoked nerve responses or EMG, and by perioperative SEP recording (Hetreed et al. 1994). Following surgical repair, serial neurophysiological measurement of temporal rate and magnitude of regeneration in central and peripheral neural pathways provides supportive evidence of clinical motor and sensory functional recovery. Misra and colleagues at Imperial College and the Royal National Orthopaedic Hospitals, London, have used EMG studies, transcranial magnetic cortical stimulation and quantitative sensory tests (axon reflex vasodilatation, thresholds for thermal and light touch sensation) to evaluate regeneration and functional plasticity in patients who have undergone spinal root repair and/or reimplantation of avulsed ventral roots into the spinal cord after brachial plexus injury. Reinnervation on EMG, first evident at 9–15 months after
surgery, normal latency and conduction properties of transcranial evoked motor potentials, together with clinical signs of recovery of sensation and muscle strength, demonstrates the potential for axonal regrowth and reinstatement of local segmental spinal cord circuits in human spinal cord after repair of avulsed spinal roots (Carlsted et al. 2000). The “stinger injury”: Stingers are well known amongst players of American football and ice hockey. The injury, which follows a neck or shoulder blow, is characterised by an electric shock sensation that travels down the arm, sometimes accompanied by paralysis of the arm. Stingers are typically transient, lasting minutes to days, but some cases may develop a prolonged syndrome with muscle wasting and sensory loss. Possible mechanisms include traction of the brachial plexus with lateral flexion of the neck, a percussive injury to the plexus at Erb’s point, and neck hyperextension and ipsilateral rotation (Kuhlman and McKeag 1999). Whilst damage is
216
generally considered to occur at brachial plexus level, the potential mechanisms have led to proposals that traumatic injury to cervical roots is more likely. Stingers amongst UK rugby players have not yet been reported in the medical literature, but sporting bodies and players are familiar with the injury, and incidence may be rising as a result of changing body habitus and tackling style. Gareth Payne and two colleagues from Cardiff and Bristol have undertaken electrodiagnostic studies in 14 cases of stinger injury in professional and amateur rugby football players (Payne 2008, personal communication). In seven players, the pattern of denervation in infraspinatus, supraspinatus, deltoid, biceps implicated the upper trunk of the brachial plexus. Five had findings suggesting an isolated axillary nerve lesion, or an upper trunk lesion restricted to the axillary nerve fascicles. One player, with a history of repeated stingers, had more widespread weakness and wasting of the arm suggesting multiple brachial plexus injuries. True neurogenic thoracic outlet syndrome: Electrophys iological findings in thoracic outlet syndrome (TOS) were first described by Roger Gilliatt and others in their seminal paper published in the Journal of Neurology, Neurosurgery and Psychiatry in 1970 (Gilliatt et al. 1970), and in a later publication from the same group, nerve conduction studies were shown to localise the lesion to the lower trunk of the brachial plexus affecting predominantly the C8-T1 roots (Gilliatt et al. 1978). Compression of the lower trunk of the brachial plexus produced by a fibrous band from an elongated C7 transverse process, or a rudimentary cervical rib was demonstrated in all patients in Gilliatt’s series. Neural and vascular elements traversing the thoracic outlet are variably affected by these vestigial structures, and the condition is best considered as being either one which is purely vascular, purely neurogenic (“true neurogenic thoracic outlet syndrome”), or involving both neural and vascular elements to a greater or lesser extent. True neurogenic TOS is very rare. The salient presentation is in a female patient who describes upper limb pain, often poorly localised, sensory deficit in the inner forearm and hand, and unilateral atrophy of thenar muscles (“the Gilliatt hand”). Widely accepted diagnostic neurophysiological criteria for true neurogenic TOS, indicative of chronic axonal loss in the lower trunk of the brachial plexus, are a small or absent ulnar sensory response, small or absent medial antebrachial cutaneous sensory response, normal median sensory response, attenuated motor CMAP amplitude from the abductor pollicis brevis, and low amplitude ulnar motor response (Le Forestier et al. 1998; Kothari et al. 1998b; CruzMartinez and Arpa 2001). Side to side comparison of nerve conduction studies significantly increases the sensitivity of electrodiagnostic testing (Tsao et al. 2008). Diagnosis of true neurogenic syndrome is challenging, not least because clinical symptoms are frequently non-specific, and overlap with carpal tunnel syndrome and ulnar nerve
Surgical Disorders of the Peripheral Nerves
entrapment neuropathy. Many techniques have been proposed over the years to enhance diagnosis and detect cases at an early stage before the development of usually irreversible hand muscle atrophy. Wulff and Gilliatt measured F waves recorded from hypothenar muscles evoked by stimulation of the ulnar nerve at the wrist and found prolongation of latency in the affected hand compared with the opposite limb (Wulff and Gilliatt 1979). In a majority of cases, the latency remained abnormally long after surgery, suggesting permanent degeneration of faster conducting fibres. SSEPs have been used to show abnormalities of proximal sensory conduction through the brachial plexus (Chodoroff et al. 1985); the N9 plexus response evoked by stimulation of the ulnar nerve may be of reduced size and latency, whereas median nerve N9 responses are normal. Cervical root stimulation using magnetic or high intensity electrical stimulation has been used to assess proximal conduction; Felice and colleagues describe a patient with unilateral proximal conduction block on root stimulation, who showed modest improvement in hand function post-operatively (Felice et al. 1999). For cases of mild lower brachial plexus lesions, Seror has suggested a new electrodiagnostic pattern of comparatively low or low amplitude medial antebrachial cutaneous sensory response, normal median and ulnar sensory and motor action potential amplitudes and normal or slightly reduced interference pattern in some C8-T1 innervated muscles (Seror 2004). Neurophysiological evaluation of the patient presenting with suspected neurogenic TOS should include exclusion of median nerve compression at the wrist and an ulnar neuropathy at the elbow. When SNAP abnormalities are equivocal, the differential diagnosis of apparent unilateral hand muscle wasting includes anterior horn cell diseases, radicular pathologies and multifocal motor neuropathy. Neuralgic amyotrophy: Neuralgic amyotrophy is a disorder characterised by an acute attack of neuropathic type pain followed by patchy weakness in a limb. The usual clinical course is with acute onset of pain in the neck or shoulder region, initially continuous and severe (and typically impairing sleep), lasting up to 4 weeks, with subsequent paresis. Sensory involvement occurs in three quarters of cases, but is mild relative to the neuropathic pain and muscle atrophy. Neuralgic amyotrophy usually occurs sporadically; rare hereditary forms exist, with predisposition to recurrent painful peripheral nerve lesions. A characteristic presentation is the young adult who experiences an acute and very painful brachial neuropathy with subsequent winging of the scapula. However, since the earliest description of cases by Parsonage and Turner, it has become apparent that neuralgic amyotrophy is a complex disorder which typically, but not always, affects the upper limbs, with a highly variable phenotype. The complexity is possibly due to the pathophysiological process being that of a mononeuritis multiplex, comprising multifascicular inflammatory
Clinical Neurophysiology in Peripheral Nerve Injuries
and constrictive lesions, as argued by Austin Sumner (Sumner 2007). The variability in location and extent of symptoms suggests that the alternative diagnostic label of brachial neuritis is a misnomer, as the disorder can affect the lumbo-sacral plexus as well as many peripheral nerves, albeit with a predilection for involvement of the cervical roots, brachial plexus and neural branches thereof. The most commonly affected nerves in a series of 40 cases (Cruz-Martinez et al. 2002) were in rank order, the suprascapular, axillary, musculocutaneous, long thoracic and radial nerves; a small number of patients had lesions of the anterior interosseus, lateral antebrachial cutaneous or phrenic nerves. In a large series of neuralgic amyotrophy (n = 246, 199 sporadic cases, and 47 with hereditary form) reported from the Netherlands (van Alfen and van Engelen 2006), damage occurred most frequently (71.1%) in the upper and middle trunk distribution with involvement of the long thoracic and/or suprascapular nerve. The characteristic prodrome of severe neuropathic pain and patchiness of neural involvement will usually prompt diagnosis, and a further clue is the relatively common occurrence of an antecedent event, such as infection, strenuous exercise, surgery, pregnancy/puerperium or vaccination. Recognition is more challenging in those cases where pain is mild or absent: 3.7% in the Dutch series, but considerably higher at 47% in the cohort reported from the Johns Hopkins Hospital (Moghekar et al. 2007). Electrodiagnosis in neuralgic amyotrophy serves to supplement clinical examination by confirming or excluding involvement of specific nerves or plexus elements, and it may contribute to an estimation of recovery, before this is apparent to the patient. It also aids differentiation of neuralgic amyotrophy from hereditary neuropathy with liability to develop pressure palsies (HNPP), in which neurophysiological pointers to a more diffuse neuropathic disorder will be evident through abnormalities of motor and sensory conduction in the clinically unaffected arm and lower limbs. Birth lesion of the brachial plexus: Traumatic injury to the neonatal brachial plexus is uncommon, but it carries the potential to cause severe long term functional disability. There is a spectrum of neural damage ranging through transient conduction block, a lesion in continuity, severe neural rupture to root avulsion and the extent of injury can vary from a single nerve element to all components of the brachial plexus. We have found that traumatic plexus injury in neonates is most likely to produce partial intradural damage and to result in some residual conducting tissue across lesions because of the anatomical arrangement of spinal cord and roots within the spinal canal of the newborn, and the less abrupt nature of traction compared with adult traumatic plexopathy (Birch et al. 1998; Bisinella et al. 2003). Neurophysiological assessment of traumatic birth lesions of the brachial plexus (BLBP) has as its main purpose the
217
identification of children with poor prognosis for recovery – those with severe degenerative axonal injuries or root avulsion – in whom surgical repair offers the best opportunity for a good functional outcome. Electrodiagnostic and radiological investigation of traumatic BLBP should supplement clinical examination by clearly indicating the severity of neural injury at each level of the plexus. Ideally, ancillary investigations should also offer clues as to aetiology, allowing separation of rare palsies that occur during the intrauterine period from those resulting from events at the time of birth. Despite the opportunities for electrodiagnostic testing to answer questions regarding prognosis and causation, few centres use EMG in traumatic BLBP. This appears to be due to perception that EMG studies are insufficiently reliable and accurate, with high false/positive rates or findings that are unduly optimistic or pessimistic when measured against functional outcome (Slooff 1993; Yulmaz et al. 1999). Royden Jones has reviewed the limitations of neurophysiological assessement of brachial plexus injuries in children (Jones 1996). Demonstration of preserved SNAPs and fibrillations in cervical paraspinal muscles – the hallmarks of root avulsion – may be difficult or impossible to achieve in small neonates for technical and safety reasons. The spectrum of sensory nerve responses available for measurement is more limited than in older children or adults, particularly for digital nerves, and those nerves with relatively short distance between stimulating and recording electrodes (radial, cutaneous antebrachial nerves) since stimulator artifact may affect SNAP quality. Normative values for SNAPs amplitudes show wide variation in infants and judging normality or otherwise may be difficult especially in the instance of bilateral brachial plexus injury. Amplitude of the CMAP is age dependant, with a threefold increase in size in the arm through maturity. Infants also have smaller muscle fibres (17 mm at 3 months of age compared with approximately 60 mm in adults, Brooke and Engel 1969), and their motor unit potentials are both smaller and shorter in duration. Neurophysiological distinction between intra-uterine plexopathy and birth related traumatic damage to the brachial plexus is reliant on accurate knowledge of the time course of change in CMAPs, SNAPs and motor unit potentials in the neonate, but there is very little information on this time course, as so few EMG studies are undertaken in the immediate few days after birth. Theoretically, fibrillations should occur much earlier in neonates than in adults, since babies have nerves which are shorter in length and smaller in diameter. The time that fibrillations appear is inversely correlated to the volume of the denervated segment of the nerve below the level of section; calculations based on neonatal nerve length and size suggest that fibrillations would be expected to appear within a few days of injury (Van Dijk et al. 2001). Experimental work in piglets with brachial plexus section supports presence of fibrillations in denervated muscles within 24 h of injury (Gonick
218
Surgical Disorders of the Peripheral Nerves
et al. 1998), and implies maximal reduction in CMAPs within 24–48 h, and SNAPs being reduced 1–2 days later. Fibrillations and other evidence of denervation have indeed been demonstrated within 4–5 days of birth in infants with traumatic BLBP days of birth (Colon and Vredeveld 2001; Mancias et al. 1994). There are strong arguments for advocating very early and sequential EMG studies to detect first appearance of fibrillations, and thereby enable accurate timing of injury, especially in cases where there is circumstantial evidence of birth related trauma (Jones et al. 2003; Pitt and Vredefeld 2005). Intrauterine injury may be suspected if EMG reveals chronic denervation within the first 1–2 weeks of birth: Paradiso and colleagues described an infant born by difficult breech delivery, in whom C5 and C6 muscles showed wasting and chronic denervation on EMG at 18 days after birth, findings which clearly placed the lesion as occurring some weeks before birth (Paradiso et al. 1997). The intra-uterine environment does not protect against pressure palsy of nerves– focal mononeuropathies of radial (Ross et al. 1983) and peroneal (Jones et al. 1996) have been identified in neonates, and were shown neurophysiologically to pre-date birth. In addition to needle EMG, techniques applied in neurophysiological evaluation of traumatic BLBP include axillary and musculocutaneous motor latency, upper limb motor and sensory conduction, ulnar H reflex (Eng et al. 1996,), SSEP for detection of root avulsion (Hashimoto et al. 1991), and intra-operative CMAP, NAP and SSEP recording to determine neural continuity (Papazian et al. 2000). Heise and colleagues used CMAP measures to estimate degree of axonal damage or loss; they found axillary and radial nerve CMAP amplitudes below 10% of the unaffected arm were accurate in predicting outcome for C5 and C6, and ulnar nerve CMAPs were useful for C8 and T1 prognosis (Heise et al. 2004). Our preferred method for neurophysiological evaluation in BLBP is to combine median and ulnar mixed nerve action potentials in the forearm (NAP) with EMG. In 73 infants aged 3 months or older with slowly recovering brachial plexus injury who were managed conservatively (Bisinella
et al. 2003), neurophysiological prediction of a favourable long term outcome (NAP amplitude 50% or above of unaffected arm, presence of significant amounts of reinnervation or normal motor units on EMG) was confirmed at long-term follow up with mean of 4.3 years in 93% of C6 lesions and 97% of C7 lesions. Predictions for C5 were confirmed in a smaller proportion (78%), possibly because of inability to record an NAP for C5 and the high frequency of secondary shoulder pathology contributing to functional disability. An unexpected finding in this series of slowly recovering traumatic BLBP was the neurophysiological evidence suggesting neurapraxic injury in a significant number of roots – normal NAP responses, reduced recruitment of normal MUPs and absence of denervation/reinnervation change on EMG. The mean age at which neurophysiological studies were performed was 5 months, and one of these infants with apparent prolonged conduction block was age 10 months at the time of evaluation. None of the infants whose neurophysiological assessment implied neurapraxic injury had clinical function in biceps by age 3 months, and they were clinically indistinguishable from cases with more severe lesions, prompting referral to the tertiary centre. All had a very good outcome with conservative management. The explanation for long lasting conduction block is unclear, but identification is important as there is a very good chance of spontaneous recovery. Eng and colleagues identified similar cases of prolonged conduction block (n = 8) in their large series of 135 children with obstetric palsy (Eng et al. 1978). Our group has also found strong correlation between neurophysiological prediction of lesion type, based on NAP and EMG measures, and nerve injury type identified at operation (Birch et al. 2005), suggesting that more severe degrees of axonal injury can be discriminated from those with potential for recovery using pre-operative electrodiagnosis (Table 6.8). Whilst needle EMG is the most important measure of which roots are involved, and whether there is discontinuity in nerve supply, reliance on EMG alone may produce inaccurate information – the so-called “overly optimistic EMG” (Vredeveld
Table 6.8 Birth lesions of the brachial plexus – proposed correlation between neurophysiological diagnosis and type of nerve lesion, adapted from Birch et al. 2005. NAP/SAP EMG Neurophysiological Pathophysiology or IOP SEP at Foramen/Stimulation diagnosis Lesion type distal to lesion Normal or near normal
NSA
50% or above of normal
Fibrillations +/−
Absent or less than 50% normal
Fibrillations +
Present in some, more often absent
Fibrillations ++
Conduction block
Neurapraxia
Not applicable
Favourable rupture
Axonotmesis
Normal/>50%
Unfavourable rupture
Neurotmesis
Reduced/<50%
Avulsion and/or severe rupture
Pre-ganglionic or total neurotmesis
Both reduced/absent
Normal units Copious MUP activity Limited MUP activity
Very limited MUP activity NSA no spontaneous activity; MUP motor unit potential.
Clinical Neurophysiology in Peripheral Nerve Injuries
et al. 1996). Van Dijk and colleagues have suggested possible explanations for an unduly optimistic or pessimistic EMG picture in their thoughtful review of obstetric lesions of the brachial plexus (Van Dijk et al. 2001). These include an incorrect clinical examination, as is more likely in the severe or complete lesion, or overestimation of MUP activity by the neurophysiologist, because of the small size of neonatal muscle fibres relative to recording needle. There may be luxury or polyneuronal innervation at birth, possibly derived from multiple segments or roots; distribution of individual spinal nerve roots is more diverse than traditionally indicated in anatomical textbooks (Colon et al. 2003). Central nervous system involvement could play a role in modifying recruitment patterns following deafferentation injury in neonatal brachial plexus injury. Aberrant regeneration or misdirected axons can result in MUPs appearing in the wrong muscle, as in “the breathing hand,” and synkinesis may be difficult to assess in the less cooperative infant or child. Finally, whilst neurophysiology can identify continuity of neural conduction, this is not the only important factor in regaining useful arm function following brachial plexus injury in the neonate.
6.6.2 Lower Limb Neuropathies 6.6.2.1 Sciatic Nerve The sciatic nerve arises from the lumbosacral plexus and is formed by the L4/5 and S1 roots. The long anatomical course from the ischial notch through the thigh renders this nerve prone to traumatic injury. The peroneal division, by virtue of the fascicular arrangement of fibres that are encased in loose areolar tissue, and tethered at the sciatic notch and the fibular neck, is particularly vulnerable and is usually disproportionately affected. The most common pathological mechanisms for sciatic nerve injury are compression, traction or direct injury by mechanical disruption as is seen in intragluteal muscle injection, still a significant factor in developing countries. In an extensive review of the published literature, hip surgery was identified as the most common cause by far (Plewnia et al. 1999). Iatrogenic lesions have also been reported in cardiac surgery (Kempster et al. 1991; McManis 1994), gynaecological procedures (Burkhart and Daly 1966), femoral nailing (Britton and Dunkerley 1990) and craniotomy in the sitting postion (Gozal and Pomeranz 1994; Keykhah and Rosenberg 1979). Other traumatic causes include direct injury to the nerve from gun-shot or knife wounds, and compartment syndromes caused by prolonged immobility (Hynes and Jackson 1994). Traction related sciatic neuropathies have been reported following assumption of unusual limb positions as in Yoga (Vogel et al.1991) and after martial arts (M Pitt, 2009, Great Ormond Street Hospital, London, personal communication). Rarely, recurrent sciatic neuropathy
219
raises the possibility of an underlying liability to pressure palsies (Ichikawa and Nezu 2005). The typical electrophysiological features of a sciatic nerve lesion are attenuation of sensory and motor nerve responses in both peroneal and tibial divisions. In an electrophysiological review of 100 consecutive patients with sciatic neuropathy (Yuen et al. 1995), the peroneal division was more severely involved in most cases (66%), with 12% demonstrating sole involvement of this division. Evidence of axonal change was the predominant electrophysiological finding. However, electrodiagnostic testing revealed involvement of the tibial division in some patients, not apparent on careful clinical testing. It may be difficult to differentiate sciatic neuropathy with predominant involvement of the peroneal division from a common peroneal nerve lesion. In such cases, needle EMG of the short head of biceps femoris is useful, as this muscle is supplied by the lateral trunk of the sciatic nerve (Katirji and Wilbourn 1994; Katirji 2002). The differential diagnosis of sciatic neuropathy includes lumbosacral plexopathy, which may be excluded by needle EMG sampling the gluteal and anterior femoral muscles, and root lesions at lower lumbar and sacral levels. Piriformis syndrome: Sciatic nerve compression in the piriformis syndrome is purported to be caused by hypertrophy of the piriformis muscle, or an anatomical variation in the course of the sciatic nerve as it crosses the piriformis muscle. This anomaly, which usually involves the peroneal division, has been identified in 5% of cadavers (Peccina 1979), with the implication that the syndrome may be under-recognised. Most patients described with piriformis syndrome present with buttock or leg pain or other sciatic symptoms, typically developing on exercise or in certain postures such as sitting. There are no specific electrodiagnostic features other than those indicative of a sciatic nerve lesion. The FAIR (flexion, adduction, internal rotation) provocative H reflex test has been proposed as increasing sensitivity for diagnosis (Fishman and Zybert 1992; Fishman et al. 2002). However, Holder at University College Hospital in London found a positive FAIR test in only two of 35 cases with suspected piriformis syndrome (Holder, 2009, Neurophysiological Diagnosis and Intervention in Piriformis Syndrome, personal communication) – and the pathophysiological basis of rapid increase in H reflex latency would be difficult to explain on the basis of a process presumed to be compressive with resultant conduction block. Needle EMG may be of most value in cases of so-called piriformis syndrome to guide therapeutic intra-muscular injection of local anaesthetic, steroid or botulinum toxin.
6.6.2.2 Common Peroneal Nerve The lateral division of the sciatic nerve continues as the common peroneal nerve below the knee, winding around the neck of the fibula. This renders it susceptible to compression: common
220
peroneal nerve lesions at the knee are the commonest type of lower limb mononeuropathy. An Italian multicentre study of peroneal neuropathy identified a traumatic causation in 32%, with a preponderance of direct and indirect iatrogenous injury (25%). Surgery at remote sites such as the thyroid and abdomen accounted for most of the iatrogenic cases. In 16%, no cause was identified (Aprile et al. 2005). Other rare sites of a common peroneal nerve lesion include the lower leg and under the extensor retinaculum of the ankle. Electrodiagnosis of a common peroneal neuropathy is based on the demonstration of an attenuated SNAP in the superficial peroneal nerve, and reduction in the CMAP amplitudes from the peroneal muscles such as extensor digitorum brevis and tibialis anterior. Slowing of motor conduction across the fibular head, and/or conduction block will suggest a demyelinating lesion. As in ulnar entrapments at the elbow, inching or short segment studies may aid diagnosis of a demyelinating lesion across the fibular head (Raudino 2004). In cases where the principal pathological process is demyelination, as indicated by focal conduction block, prognosis should be favorable: an absent or severely diminished motor response showed strong inverse correlation with recovery (Cruz-Martinez et al. 2000). In a large case series review of 116 lesions (Katirji and Wilbourn 1988), axonal loss was the most common electrophysiological finding (55%), conduction block was identified in 17% of patients, and mixed axonal/demyelinating features were seen in 25%. Motor conduction studies to the tibialis anterior muscle identified conduction block more consistently than extensor digitorum brevis. In the series reported by Kang and colleagues, a diminished superficial peroneal nerve response was seen in slightly less than half (48%) of cases, suggesting fascicular selectivity Kang et al. 2005. A generalised underlying polyneuropathy must always be excluded in cases presenting with a common peroneal nerve lesion. L5 radiculopathy is an important consideration in the differential diagnosis, particularly as lateral disc herniation may produce compression distal to the dorsal root ganglion affecting the superficial peroneal sensory response (Levin 1998). In the patient presenting with foot drop, EMG evidence of denervation in tibialis posterior muscle will usually reliably distinguish a common peroneal neuropathy from an L5 root lesion.
6.6.2.3 Tibial Nerve and its Branches Isolated lesions of the proximal part of the tibial nerve are relatively uncommon, as the nerve is well protected in the knee. Nerve damage may however occur following direct knee injury, in cases of Baker’s cyst (DiRisio et al. 1994), and iatrogenically such as during placement of fixation screws or endoscopic vein surgery (Geselschap et al. 2001).
Surgical Disorders of the Peripheral Nerves
Neurophysiological studies will usually reveal evidence of axonal loss – reduced amplitude of the abductor hallucis CMAP and neurogenic change in calf muscles and intrinsic foot muscles innervated by the tibial nerve. The H reflex recorded from soleus may be small or absent. Lesions above the popliteal fossa result in reduction or loss of the sural sensory response. In the lower leg, the tibial nerve passes behind and below the medial malleolus through the tarsal tunnel, underneath the flexor retinaculum. The calcaneal sensory branch may be given off proximal to the tunnel, but the medial and lateral plantar nerves arise within the tarsal tunnel. Any of these nerves may be affected in the (posterior) tarsal tunnel syndrome, which is the pedal counterpart of the carpal tunnel syndrome. This rare entrapment neuropathy is frequently misdiagnosed as plantar fasciitis. Trauma such as ankle fracture or dislocation is most commonly implicated in the pathogenesis, but medical conditions such as rheumatoid arthritis may also be causative, as is joint hypermobility (Francis et al. 1987). Pain in the sole of the foot, particularly prominent at night, is the most common clinical symptom in reported cases. Weakness of the intrinsic foot muscles may be present, but is difficult to define clinically. Prolongation of the distal motor latency to abductor hallucis and/or abductor digiti quinti is a very suggestive electrodiagnostic finding in suspected tarsal tunnel syndrome. However absent or diminished sensory responses from the sole, and the great toe, may be a more sensitive feature (Guiloff and Sherratt 1977), except in patients over 65 years in whom lower limb SNAPs may be difficult to record even in the absence of pathology. Motor recording from the first dorsal interosseous muscle supplied by the lateral plantar nerve, may also be useful (Galloway and Greathouse 2006). Needle EMG assists exclusion of first sacral radiculopathy, which may produce similar symptoms, and the examination should include at least one muscle innervated by the common peroneal nerve to rule out a sciatic neuropathy. Most clinical neurophysiologists will share our experience that many of the cases of possible tarsal tunnel syndrome referred for electrodiagnosis show normal nerve conduction studies and EMG findings. The anterior tarsal tunnel syndrome, whilst very uncommon overall, appears to be more frequent in persons frequently adopting the kneeling (or Namaz) position at prayer (Akyuz et al. 2000). Pain in the dorsal forefoot, worse at night and relieved by a change in position, are the principal clinical features. Electrophysiological confirmation is by demonstration of disproportionately prolonged distal motor latency to extensor digitorum brevis, with denervation in the muscle on needle EMG sampling. Morton’s metatarsalgia was probably the earliest focal neuropathy to be clearly described in the literature, but the actual pathology, a traumatic neuroma of the lateral digital nerve, was only recognised in the 1940s (King 1946). The
Clinical Neurophysiology in Peripheral Nerve Injuries
clinical feature of burning pain over the head of the fourth metatarsal which radiates forward to the toe, is sufficiently characteristic, but near nerve recording from the interdigital nerves from toes four and five may help in confirmation of the diagnosis. Demyelination is not a prominent feature. Other rare pedal entrapment neuropathies include Joplin’s neuroma, caused by traumatic fibrosis of the perineurium of the medial plantar digitial nerve of the great toe. Pain and variable sensory loss over the medial aspect of the great toe are noted clinically. Electrophysiological confirmation is by near needle recording from the tibial nerve at the ankle, stimulating the medial plantar digital nerve in the great toe (Cichy et al. 1995; Oh 2007). Sural neuropathy: The sural nerve emerges from the fascial layer of the postero-lateral leg about 20 cm above the lateral malleolus, passing behind the lateral malleolus and supplying the skin of the lateral ankle and foot up to the little toe. The most common cause of an isolated sural neuropathy currently is iatrogenic viz. nerve grafting or biopsy. However, the sural nerve can also be injured by a Baker cyst in the popliteal fossa, and during surgical vein stripping or compression. Orthopaedic procedures at the ankle, tumors and direct traumatic injury also contribute. The lesion is usually axonal, and a side to side reduction in amplitude of less than 45% in comparison with the normal leg confirms the diagnosis (Seror 2002).
6.6.2.4 Femoral Nerve Femoral nerve lesions are usually due to nerve compression – during pelvic surgery, from a retroperitoneal haematoma, and as a result of prolonged lithotomy positioning. The nerve is also vulnerable during hip surgery, particularly hip revision or replacement. Clinical manifestations are weakness of hip flexion if the lesion is above the inguinal ligament proximal to the motor branch to psoas, weakness of knee flexion, and sensory loss in the antero-medial thigh. Electrodiagnostic findings in femoral neuropathy are determined by whether the lesion is primarily axonal, when there will be loss of the saphenous SAP and denervation on EMG, or of demyelinating type, consequent to external compression or stretch injury. Demyelinating lesions may show a reduced CMAP of rectus femoris or vastus medialis, stimulating the femoral nerve at the groin, an EMG pattern suggesting conduction block (reduced recruitment of normal MUPs) and preservation of the saphenous SAP. The CMAP size has prognostic value in axonal femoral neuropathy – if the response is less than 50% of the unaffected side, fewer than half of patients had recovered by 1 year, whereas all patients with a CMAP exceeding 50% showed full recovery (Kuntzer et al. 1997). EMG is the most useful procedure in a suspected femoral neuropathy, to aid localisation and to exclude extra-femoral involvement. Neurogenic EMG abnormalities will be seen in
221
the psoas muscle when the lesion is sited within the pelvis above the inguinal ligament. In lumbar plexopathy, extra- femoral denervation is found, involving all or some of tibialis anterior, peroneal muscles and thigh adductor muscles, and the saphenous and superfical peroneal SAPs may be reduced or absent. Denervation will also be more extensive in lumbar radiculopathy, including paraspinal muscles. The entity of “diabetic amyotrophy” was previously and erroneously considered to be a femoral neuropathy based on presentation with acute severe pain in the upper leg, lower back or buttock with weakness and wasting of proximal leg muscles. However, electrodiagnosis reveals denervation in several lower limb muscles in different myotomes, and often evidence of a generalised sensori-motor axonal neuropathy. Although the precise level of pathology is uncertain (Chokroverty 1987), it should be considered as a diffuse disorder affecting lumbosacral plexus, lumbo-sacral roots and peripheral nerves (BrunsGarland syndrome).
6.6.2.5 Other Lower Limb Nerve Disorders Meralgia paraesthetica (Bernhardt-Roth syndrome): This entrapment neuropathy of the lateral femoral cutaneous nerve, which arises from the upper lumbar plexus, is not uncommon, and deserves especial consideration. Unpleasant paraesthesias affect the upper outer thigh, with a small area of well circumscribed sensory loss. The focus of entrapment is at the level of the inguinal ligament, through or under which traverses the lateral femoral cutaneous nerve. Pregnancy, obesity or rapid weight gain and diabetes are significantly associated with this condition. Although the diagnosis is primarily clinical, electrodiagnosis is useful to establish the lesion, and to rule out a femoral neuropathy, L4 radiculopathy or lumbo-sacral plexopathy, principally by needle EMG in lower limb muscles. Other diagnostic techniques include lateral femoral cutaneous SNAP amplitude and sensory conduction velocity measurement, but the nerve can be difficult to study with surface or skin electrodes and near nerve needle recording may be required. Side to side comparison, rather than the absolute amplitude of the sensory response, is more sensitive in identifying abnormality (Lagueny et al. 1991; Seror and Seror 2006). Obstetric nerve injuries: The medical literature reports a frequency of postpartum lumbosacral spinal and lower extremity nerve injury between 0.8 and 50/100 000. A large survey (6,057 women) of post-partum lower extremity injuries in the United States estimates the incidence of post-partum nerve injury at 0.92% (Wong et al. 2003). The mechanisms of nerve injury in labour are diverse, and include transection, stretch, compression as well as ischaemic/vascular injury. The most common nerve lesion (60%) is that of unilateral or bilateral injury of the lateral cutaneous nerve
222
(meralgia paraesthetica), with the hyperlordosis of pregnancy associated with compression of the nerve by the posterior fascicle of the inguinal ligament. Stretching by retractors during Caesarian section, as well as actual severing during a wide Pfannenstiel incision are other aetiopathogenic mechanisms. Femoral neuropathy is the second most common injury, with most women demonstrating hip flexion as well as knee extension weakness. Interestingly, none of the women reported in the study by Wong had caesarian delivery. Excessive hip abduction and external rotation is considered to stretch the intra-pelvic portion of the femoral nerve, and stretch–induced ischaemia may be aggravated by poor blood supply. Other peripartum neural complications described in the literature include sciatic (Roy et al. 2002), common peroneal, obturator (Lindner et al. 1997) and lumbosacral plexus lesions. Saphenous nerve lesions: Saphenous neuropathy is most frequently caused by saphenous vein harvesting for coronary artery bypass grafts (Chauhan et al. 1981). Occasional causes include compression by the femoral artery, schwannomas, damage to the nerve at knee level during arthroscopy and other orthopaedic procedures (Mochida and Kikuchi 1995), and during varicose vein surgery (Cox et al. 1974). Focal lesions can be confirmed by sensory nerve recording, with needle EMG if there is suspicion of lumbar radiculopathy (L4) or a partial femoral nerve lesion.
6.6.2.6 Lumbosacral plexopathy The lumbosacral plexus has two structurally and functionally discrete divisions- the lumbar plexus, formed by the anterior rami of the L1 to (most of ) L4 roots, and the sacral plexus which is formed by the lumbosacral trunk (part of L4, L5) and the anterior rami of the S1-S4 roots. The lumbar plexus lies within the posterior part of the psoas muscle. Sensory innervation includes the skin of the pubis and (partly) the external genitalia. It also supplies the front, inner and posteromedial thigh via the genito-femoral and obturator nerves, and the inner and anteromedial leg to the ankle (saphenous-femoral nerves). Muscular contribution includes the iliacus and psoas muscles, as well as the anterior and medial femoral muscles through the femoral and obturator nerves. The sacral plexus is located over the anterior surface of the piriformis muscle, in the posterior pelvis, and is therefore in close proximity to the pelvic organs such as colon, bladder, uterus and rectum. Of the divisions of the sacral plexus, the sciatic nerve is the largest, formed by the fusion of anterior (tibial) and the upper four posterior (peroneal) divisions. Other divisions are the gluteal and posterior femoral cutaneous nerves. Lumbosacral plexopathies are much less frequent than brachial plexus lesions, with non-traumatic causes accounting
Surgical Disorders of the Peripheral Nerves
for a majority of cases reported in the literature. Neoplastic infiltration was found to be eight times more frequent than trauma, with infiltration by malignancies of the pelvic organs (uterus, bladder, rectum) being more common than primary tumours of the lumbosacral plexus such as neurofibromas (Jaeckle 2004). Pelvic fractures resulting from motor vehicle accidents, falls, and industrial accidents are the usual injuries associated with traumatic lumbosacral plexopathies; transverse sacral fractures are particularly liable to damage nerve roots (Wilbourn 2007). Motor vehicle accidents more frequently involve the upper plexus, whereas gunshot wounds tend to produce discrete, multifocal lesions of the lumbosacral plexus, predominantly affecting the upper part of the plexus (Chiou-Tan et al. 2001). The co-existence of skeleto-muscular trauma makes precise localisation difficult clinically, and electrodiagnostic evaluation is complex and challenging in such cases. Our own experience from the joint Peripheral Nerve Injury and National Orthopaedic and National Hospital clinic is that a much higher proportion of the lumbosacral plexopathy cases we see are due to pelvic fracture and penetrating missile injuries. Accurate localisation to the lumbar or sacral plexus is more difficult than in brachial plexus lesions for a number of reasons. There is a limited repertoire of sensory nerve responses (sural, saphenous, superficial peroneal) that can be reliably studied in the lower limbs, especially in elderly subjects; moreover, motor involvement is often the predominant feature of a lumbo-sacral plexopathy. In a large series, only a third of 171 cases could be localised to the sacral plexus, using the criteria of reduced/absent SNAPs, denervation of plexus innervated muscles, and absence of paraspinal denervation (Tavee et al. 2007). Age related loss of sural SNAPs, prior lumbar surgery and duration of the lesion contribute to the imprecise localisation (Wilbourn 2007). Differential diagnosis includes traumatic sciatic and common peroneal nerve neuropathies. A peroneal pattern of sensory loss and weakness was seen most frequently in a retrospective series of neurological injuries associated with pelvic trauma (Chiodo 2007). This is explained by the anatomical proximity of the peroneal division to the sacroiliac joint. Prognosis for gait outcome was significantly associated with abnormalities of peroneal nerve conduction, and denervation in the tibialis muscle; gluteal involvement tended to be mild and did not contribute to gait abnormality. Neuropathic pain was seen in 37%, and was not correlated to the type or level of injury. As in brachial plexopathy, evidence of myokymia on EMG can be specifically helpful in differential diagnosis of tumor and radiation induced lumbosacral plexopathy. Myokymic discharges were seen in 57% of patients with radiation damage, but in none of the cases with neoplastic infiltration (Thomas et al. 1985).
Clinical Neurophysiology in Peripheral Nerve Injuries
6.6.2.7 Electrodiagnosis in Root Lesions Radiculopathies are one of the most common reasons for referral to the Clinical Neurophysiologist, and electrodiagnostic testing has been used in the evaluation of nerve root compression for many years as a complement to neuroimaging. In occasional cases, EMG studies may confirm diagnosis even when radiological examination is negative (Wilbourn and Aminoff 1988). The approach of combining nerve conduction studies and needle EMG sampling has recently shown comparative diagnostic efficiency with neuroradiological examination in lumbosacral radiculopathies (Fisher et al. 2008). In general, with the exception of an L5 root compression (Levin 1998; Ho et al. 2004), sensory nerve conduction studies are normal in radiculopathy since the lesion lies in the preganglionic segment of sensory fibres, proximal to the dorsal root ganglion. Motor nerve conduction studies may reveal abnormalities of CMAP amplitude, as well as aberrations of F waves and the H response. However, nerve conduction studies play the conjunctive role of exclusion of a co-incidental focal peripheral nerve lesion in suspected radicular disease. In those patients with radiculopathy who have normal EMG findings, one explanation for a false negative study is the greater variability in cervical and lumbosacral myotomes than is usually indicated in standard texts. Levin and colleagues compared surgical and electromyographic localisation of single cervical root lesions, and found a fairly stereotyped pattern of muscle involvement in C5, C7 and C8 lesions, whereas it varied in C6 radiculopathies. In half of these cases, muscles affected were similar to those in C5 root lesions with additional involvement of pronator teres; in the remainder, the pattern resembled that of a C7 radiculopathy (Levin et al. 1996). Further challenges are uncertainties as to precise segmental level in paraspinal muscles at both cervical and lumbosacral regions, and the persistence of EMG abnormalities following previous spinal surgery. The value of SSEPs in radicular disease is limited. Relatively short segments of demyelination in sensory roots may not manifest with significant prolongation of latency, and amplitude of SSEP components is overall too variable to allow reliable identification of focal conduction block. Several peripheral nerves derive from multiple nerve roots – the median nerve for example contains fibres from C6, C7, C8 and T1. Sensitivity of dermatomal SEPs varies from 25% of cases showing abnormality (Aminoff et al. 1985) to 95% (Katifi and Sedgwick 1987); the difference in study findings may be explained by use of wider criteria for abnormality. Use of magnetic stimulation of roots to measure motor evoked responses (Chokroverty et al. 1989) is limited by uncertainty as to exact site of stimulation, and to a lesser extent by discomfort of the procedure. The “double crush” syndrome This syndrome was first proposed by Upton and McComas based on their observation of patients with a combination of cervical root lesions and upper
223
limb entrapment neuropathies, such as carpal tunnel syndrome or ulnar neuropathy at the elbow (Upton and McComas 1973). The double crush concept is based on the occurrence of focal compression at more than one site along the length of a nerve fibre. Upton and McComas proposed that a sub-clinical distal focal neuropathy could be converted into a clinical lesion, probably through serial constraint of axonal flow, and thus a distal compressive lesion had a greater effect when in the presence of a proximal lesion. The double crush syndrome became a widely accepted concept, and was further expanded to include the reverse double crush, whereby a distal lesion such as affecting the ulnar nerve at the wrist could trigger an ulnar neuropathy at the elbow by comprising retrograde axonal transport (Dahlin and Lundborg 1990), and even multiple-crush syndromes. However Wilbourn and Gilliatt have provided a compelling critique of the double crush concept, arguing that it is far more widely cited than is justifiable on anatomical or physiological grounds (Wilbourn and Gilliatt 1997). The principle most often breached in clinical examples of the double crush syndrome is the most basic – the requirement for anatomical continuity of nerve fibres between the two (or more) lesion sites. This is especially so when the distal lesion affects a peripheral nerve that has extensive proximal origin such as the median nerve at the wrist, which derives from four cervical roots, and then traverses multiple components of the brachial plexus. Hence, a patient who presents with median nerve compression at the wrist and clinico-radiological evidence of a C6 cervical radiculopathy is likely to have two neurological lesions which are not directly related. The other context in which the double crush concept poses difficulty is when the proximal lesion involves sensory radicular fibres within the intra-spinal canal, and thus proximal to the dorsal root ganglion. Pre- and post-ganglionic fibres are not in anatomical continuity, and injury to one set of fibres will not affect the other unless the dorsal root ganglion is also damaged. Morgan and Wilbourn analyzed a large series (n = 12,736 limbs) of carpal tunnel syndrome or ulnar neuropathy at the elbow (Morgan and Wilbourn 1998). In 435 limbs (3.4%), there was a coexisting cervical radiculopathy, but only 98 limbs (0.8%) had an association that was anatomically appropriate. Cumulatively, only 69 of the 12,736 cases of carpal tunnel syndrome or ulnar neuropathy at the elbow satisfied pathophysiologic and one of the anatomic requirements of the double crush hypothesis, suggesting that a cervical radiculopathy could only rarely serve as the proximal lesion with these entrapment neuropathies in the double crush syndrome.
6.6.3 Diffuse Problems It is important to remember that an isolated, apparently compressive or traumatic lesion of a peripheral nerve or nerve
224
root can be the presentation of a more diffuse process. This may be an hereditary polyneuropathy such as hereditary neuropathy with liability to pressure palsies (HNPP), a myopathy such as facio-scapulo-humeral dystrophy, or even a disorder of the anterior horn cell such as the amyotrophic lateral sclerosis form of motor neurone disease. The role of the electromyographer therefore extends beyond localisation and prognostication of nerve or nerve root injury to recognition and diagnosis of diffuse neuromuscular disorders, or a condition unsuspected by the referrer. Electrodiagnostic evaluation of a generalised polyneuropathy is directed towards assessing the degree of involvement of sensory and/or motor fibres, and identifying whether the neuropathy has pathophysiological features of an axonopathy or a demyelinating neuropathy. The distinction between these two types of neuropathy is determined principally by conduction velocity – if there is slowing of nerve conduction velocity, the neuropathy is demyelinating in type. In axonal neuropathies, there is reduction in size or loss of sensory and/or motor nerves responses; conduction velocities are relatively preserved, depending on the degree of axonal loss. The distribution of abnormality reveals whether the neuropathy is symmetrical or multifocal, as in mononeuritis multiplex syn. multifocal neuropathy, or a chronic demyelinating polyneuropathy. Distal dying back axonopathies of metabolic aetiology such as diabetes, or due to toxins such as alcohol or chemotherapeutic agents, initially involve the longest i.e., lower limb nerve fibres. Neurophysiological characterisation of the type and pattern of a polyneuropathy is of substantial help when considering possible underlying causes. Case Report: A 62 year old man presenting with symptoms suggesting median nerve compression in the right wrist underwent carpal tunnel decompression. There was no clinical improvement following surgery. The finding of a positive Tinel’s sign over the median nerve at the elbow led to operation at this site, again without any improvement in sensory symptoms. The patient was then referred for nerve conduction studies. These revealed absent median and ulnar sensory responses in both hands, a small superficial radial sensory response of 6 mV, and absent lower limb sensory responses. Conduction velocities and F responses were abnormally slow in several motor nerves of the arms and legs, with patchy involvement of different nerve segments. Conduction block was seen in both ulnar nerves at the elbow and in the left common peroneal nerve. Mild degrees of chronic partial denervation were seen in distal limb muscles. The nerve conduction studies thus revealed a diffuse polyneuropathy, of demyelinating type in view of slow conduction velocities. The patchy nerve involvement suggested an inflammatory cause i.e., chronic inflammatory demyelinating polyneuropathy, later confirmed on nerve biopsy, and successfully treated with immunosuppressive therapy.
Surgical Disorders of the Peripheral Nerves
Muscle weakness and atrophy are cardinal features of anterior horn cell disorders – which comprise motor neurone diseases, spinal muscular atrophies, poliomyelitis and other neurotrophic viral infections. Bulbar as well as spinal muscles may be affected. Fasciculations are often a prominent physical sign; reflexes may be brisk or lost, depending on the degree of involvement of upper and lower motor neurones. Anterior horn cell disorders are diagnosed by demonstrating EMG features of acute and chronic partial denervation; central motor conduction studies can be used to reveal pathological involvement of upper motor neurones in the cortico-spinal tracts. Electromyography has a critical role in showing the extent of segmental involvement of denervation, and in identifying subclinical neurogenic abnormality. The peripheral nerve disorder, motor neuropathy with multifocal conduction block (Chaudhry et al. 1994) can mimic motor neurone disease clinically. This immunologically mediated and treatable condition is distinguished by electrophysiological demonstration of conduction block in peripheral motor nerves. Detailed and careful electrodiagnostic assessment is critical, since the conduction block may be in atypical sites such as the mid forearm, or in the very proximal parts of motor nerves. Muscle or myopathic disorders include genetically determined conditions such as limb girdle dystrophy and Duchenne muscular dystrophy, and acquired disorders such as polymyositis and inclusion body myositis. Muscle diseases present with weakness, fatiguability and variable muscle atrophy. Myopathic motor units have a characteristic low amplitude short duration appearance on EMG, although the pattern is not specific to primary muscle disease, being seen in disorders affecting the neuromuscular junction and occasionally in neurogenic processes. The pattern of muscle involvement is an important clue to diagnosis – facial muscles, sternomastoid, spinatii and upper arm muscles are affected in the genetic disorder facio-scapulohumeral muscular dystrophy, whereas quadriceps, tibialis anterior and forearm deep flexor are preferentially involved in inclusion body myositis (IBM). IBM usually presents in older men and may be associated with dysphagia. Its cause is uncertain. Additional EMG findings can help in differential diagnosis of primary muscle disease, although are not specific for aetiology. Thus, fibrillations occur in inflammatory myopathies such as polymyositis; and myotonic discharges are a feature of myotonic dystrophies. There are no electromyographic findings which reliably distinguish between relapse of myositis and drug induced myopathy in patients with inflammatory muscles disease treated with steroids. Case Report: This 21 year old man presented with acute shoulder weakness and scapular winging following an accident while waterskiing. The clinical suspicion was of an acute long thoracic nerve lesion, and this was the referral question when the patient was sent for neurophysiological studies. Needle electromyographic examination of serratus anterior and trapezius was normal, confirming the integrity
Clinical Neurophysiology in Peripheral Nerve Injuries
of the long thoracic and spinal accessory nerves. Instead, EMG revealed myopathic changes (short duration low amplitude and highly polyphasic motor units) in deltoid, brachioradialis, tibialis anterior, peroneus longus, and infraspinatus, leading to a diagnosis of facio-scapulo-humeral dystrophy. This is a primary muscular dystrophy of genetic origin, with autosomal dominant inheritance. The patient was adopted, and had no knowledge of his family history.
References Adrian ED (1925) Interpretation of the electromyogram: Oliver-Sharpey Lecture. Lancet 208:1229–1233 Aguayo A, Nair CP, Midgley R (1971) Experimental progressive compression neuropathy in the rabbit. Histologic and electrophysiologic studies. Arch Neurol 24:358–36 Akyuz G, Us O, Turan B, Kayhan O, Canbulat N, Yilmar IT (2000) Anteriortarsal tunnel syndrome. Electromyogr Clin Neurophysiol 40(2):123–128 Alexander L (1941) Electrical injuries of the nervous system. J Nerv Ment Dis 94:622 Al Shekhlee A, Shapiro BE, Preston DC (2003) Iatrogenic complications and risks of nerve conduction studies and needle electromyography. Muscle Nerve 27:517–526 American Association of Electrodiagnostic Medicine Professional Practice Committee (1999) Risks in electrodiagnostic medicine. Muscle Nerve 22:S53–S58 Aminoff MJ (2004) Electrophysiologic testing for the diagnosis of peripheral nerve injuries. Anesthesiology 100:1298–1303 Aminoff MJ, Goodin DS, Parry GJ, Barbaro NM, Weinstein PR, Rosenblum ML (1985) Electrophysiological evaluation of lumbosacral radiculopathies: electromyography, late responses and somatosensory evoked potentials. Neurology 35:1514–1518 Aprile I, Caliandro P, La Torre G, Tonali P, Foschini M, Mondelli M, Bertolini C, Piazzini DB, Padua LJ (2005) Multicenter study of peroneal mononeuropathy: clinical, neurophysiologic, and quality of life assessment. Peripher Nerv Syst 10(3):259–268 Baba M, Fowler CJ, Jacobs JM, Gilliatt RW (1982) Changes in peripheral nerve fibres distal to a constriction. J Neurol Sci 54:197–208 Beatty RM, McGuire P, Moroney JM, Holladay FP (1995) Continuous intraoperative electromyographic recording during spinal surgery. J Neurosurg 82:40 Becker J, Nora DB, Gomes I, Stringari FF, Seitensus R, Panosso JS, Ehlers JC (2002) An evaluation of gender, obesity, age and diabetes mellitus as risk factors for carpal tunnel syndrome. Clin Neurophysiol 113(9):1429–1434 Beekman R, Wokke JHJ, Schoemaker MC, Lee ML, Visser LH (2004) Ulnar neuropathy at the elbow: follow-up and prognostic factors determining outcome. Neurology 63:1675–1680 Birch R, Ahad N, Kono H, Smith SJ (2005) Repair of obstetric brachial plexus palsy: results in 100 children. J Bone Joint Surg Br 87:1089–1095 Birch R, Bonney G, Wynn Parry C (1998) Birth lesions of the brachial plexus. In: Birch R, Bonney G, Wynn Parry CB (eds) Surgical disorders of the peripheral nerves. Churchill Livingstone, Edinburgh, pp 209–233 Bisinella GL, Birch R, Smith SJM (2003) Neurophysiological predictions in slowly recovering obstetric lesions of the brachial plexus. J Hand Surg Br 28:148–152 Bland J (2005) The relationship of obesity, age, and carpal tunnel syndrome: more complex than was thought? Muscle Nerve 32(4):527–3
225 Bland J (2000) A neurophysiological grading scale for carpal tunnel syndrome. Muscle Nerve 23:1280–128 Bland JD, Rudolfer SM (2003) Clinical surveillance of carpal tunnel syndrome in two areas of the UnitedKingdom, 1991–2001. J Neurol Neurosurg Psychiatry 12:1674–1679 Bleasel AF, Tuck RR (1991) Variability of repeated nerve conduction studies. Electroencephalogr Clin Neurophysiol 81:417–420 Bonney G, Gilliat RW (1958) Sensory nerve conduction after traction lesion of the brachial plexus. Proc R Soc Med 51:365–367 Bostock H, Sears TA (1976) Continuous conduction in demyelinated mammalian nerve fibres. Nature 263(5580):786–787 Bouzaidi K, Ravard-Marsot C, Debroucker F (2005) Ganglion cyst at the spinoglenoid notch. A case report and literature review. Rev Méd Interne 26(4):335–338 Britton JM, Dunkerley DR (1990) Closed nailing of a femoral fracture followed by sciatic nerve palsy. J Bone Joint Surg Br 72:318 Brooke MH, Engel WK (1969) The histographic analysis of human muscle biopsies with regard to fiber types. 4. Children’s biopsies. Neurology 19:591–605 Buchthal F, Guld C, Rosenfalck A (1957) Multielectrodes study of the territory of a motor unit. Acta Physiologica Scandinavia 39:83–104 Burkhart FL, Daly JW (1966) Sciatic and peroneal nerve injury: a complication of vaginal operations. Obstet Gynecol 28:99–102 Carlsted T, Anand P, Hallin R, Misra PV, Norén G, Seferlis T (2000) Spinal nerve root repair and reimplantation of avulsed ventral roots into the spinal cord after brachial plexus injury. J Neurosurg (Spine 2) 93:237–247 Campbell WW, Geiringer SR (1998) The value of inching techniques in the diagnosis of focal nerve lesions. Muscle Nerve 21:1554–1561 Campbell DW, Sahni SK, Pridgeon RM, Riaz G, Leshner RT (1988) Intraoperative electroneurography: management of ulnar neuropathy at the elbow. Muscle Nerve 11:75–81 Campbell WW, Carroll DJ, Greenberg MK et al (1999) Practice parameter: electrodiagnostic studies in ulnar neuropathy at the elbow. Neurology 52:688–690 Casazza BA, Young JL, Press JP, Heinemann AW (1998) Suprascapular nerve conduction: a comparative analysis in normal subjects. Electromyogr Clin Neurophysiol 38:153–160 Chang M-H, Wei S-J, Chiang H-L, Wang H-M, Hsieh PF, Huang S-Y (2002) Comparison of motor conduction techniques in the diagnosis of carpal tunnel syndrome. Neurology 58:1603–1607 Chaudhry V, Cornblath DR (1992) Wallerian degeneration in human nerves: serial electrophysiological studies. Muscle Nerve 15: 687–69 Chaudhry V, Corse AM, Cornblath DR, Kuncl RW, Freimer MV, Griffin JW (1994) Multifocal motor neuropathy: electrodiagnostic features. Muscle Nerve 17:198–205 Chauhan BM, Kim DJ, Wainapel SF (1981) Saphenous neuropathy: following coronary artery bypass surgery. NY State J Med 81: 222–223 Chodoroff G, Lee DW, Honet JC (1985) Dynamic approach in the diagnosis of the thoracic outlet syndrome using somatosensory evoked responses. Arch Phys Med Rehabil 66:3–6 Chiodo A (2007) Neurologic injury associated with pelvic trauma: radiology and electrodiagnosis evaluation and their relationships to pain and gait outcome. Arch Phys Med Rehabil 88:1171–1176 Chiou-Tan FY, Kemp K Jr, Elfenbaum M, Chan KT, Song J (2001) Lumbosacral plexopathy in gunshot wounds and motor vehicle accidents: comparison of electrophysiologic findings. Am J Phys Med Rehabil 80(4):280–285 Chokroverty S (1987) Diabetic amyotrophy. Muscle Nerve 10:679–684 Chokroverty S, Sachdeo R, Dilullo J, Duvoisin RC (1989) Magnetic stimulation in the diagnosis of lumbosacral radiculopathy. J Neurol Neurosurg Psychiatry 52:767–772 Cichy SW, Claussen GC, Oh SJ (1995) Electrohysiological studies in Joplin’s neuroma. Muscle Nerve 18(6):671–672
226 Colon AJ, Vredeveld Jw (2001) Denervation potentials shortly after obstetric brachial plexus palsy: no indication of intrauterine injury. Muscle Nerve 24:140 Colon AJ, Vredeveld JW, Blaauw G, Slooff AC, Richards R (2003) Extensive somatosensory innervation in infants with obstetric brachial palsy. Clin Anat 16:25–29 Cox SJ, Wellwood JM, Martin A (1974) Saphenous nerve injury caused by stripping the long saphenous vein. Br Med J 1:415–417 Cruz-Martinez A, Arpa J, Palau F (2000) Peroneal neuropathy after weight loss. J Peripher Nerv Syst 5:101–105 Cruz-Martinez A, Arpa J (2001) Electrophysiological assessment in neurogenic thoracic outlet syndrome. Electromyogr Clin Neurophysiol 41:253–256 Cruz-Martinez A, Barrio M, Arpa J (2002) Neuralgic amyotrophy:variable expression in 40 patients. J Peripher Nerv Syst 7:198–204 Dahlin LB, Lundborg G (1990) The neurone and its response to peripheral nerve compression. J Hand Surg Br 15:5–10 Davies FA (1972) Impairment of repetitive impulse conduction in experimentally demyelinated and pressure-injured nerves. J Neurol Neurosurg Psychiatry 35:537–544. Dawson GD, Scott JW (1949) The recognition of nerve action potentials through the skin in man. J Neurol Neurosurg Psychiatry 12:259–267 de Laat EA, Visser CP, Coene LN, Pahlplatz PV, Tavy DL (1994) Nerve lesions in primary shoulder dislocations and humeral neck fractures. J Bone Joint Surg 76B:381–383 Deletis V, Sala F (2007) Intra-operative neurophysiological monitoring of the spinal cord during spinal cord and spine surgery. Clin Neurophysiol 119:248–264 Denny Brown D, Brenner C (1944) Paralysis of nerve induced by direct pressure and by tourniquet. Arch Neurol Psychiatry 51:1–26 Denny-Brown D, Adams RD, Brenner C, Doherty MM (1945) The pathology of injury to nerve induced by cold. J Neuropathol Exp Neurol 4:305 Denys EH (1991) The influence of temperature in clinical neurophysiology. Muscle Nerve 14:795–81 DiRisio D, Lazaro R, Popp AJ (1994) Nerve entrapment and calf atrophy caused by a Baker’s cyst: case report. Neurosurgery 35:333–334 Donner TR, Kline DG (1993) Extracranial spinal accessory nerve injury. Neurosurgery 32:907–911 Dorfman LJ (1984) The distribution of conduction velocities (DCV) in peripheral nerves: a review. Muscle Nerve 7:2–11 Duncan C (2001) High-tension electrical burns: special problems and their management. Trauma 3:181–185 Dorfman LJ, Howard JE, McGill KC (1989) Clinical studies using automatic decomposition electromyography in needle and surface EMG. In: Desmedt JE (ed) Computer aided electromyography and expert systems. Elsevier, Amsterdam, pp 189–204 Dunselman HH, Visser LH (2008) The clinical, electrophysiological and prognostic heterogeneity of ulnar neuropathy at the elbow. J Neurol Neurosurg Psychiatry 79(12):1364–1367 Ebeling P, Gilliatt RW, Thomas PK (1960) The clinical and electrical study of ulnar nerve lesion in the hand. J Neurol Neurosurg Psychiatry 23:1–6 Edds MV (1949) Experiments on partially deneurotized nerves. J Exp Zool 1:211–226 Eng GD, Binder H, Getson P, O’donnell R (1996) Obstetric brachial plexus palsy (OBPP) outcome with conservative management. Muscle Nerve 19:884–891 Eng GD, Koch B, Smokvina MD (1978) Brachial plexus palsy in neonates and children. Arch Phys Med Rehabil 59:458–463 Felice KJ, Butler KB, Druckemiller WH (1999) Cervical root stimulation in a case of classic neurogenic thoracic outlet syndrome. Muscle Nerve 22:1287–1292 Ferrante MA, Wilbourn AJ (1995) The utility of various sensory nerve conducion responses in assessing brachial plexopathies. Muscle Nerve 18:879–889
Surgical Disorders of the Peripheral Nerves Fishman LM, Dombi GW, Michaelsen C, Ringel S, Rozbruch J, Rosner B, Weber C (2002) Piriformis syndrome: diagnosis, treatment and outcome-a 10 year study. Arch Phys Med Rehabil 83:295–301 Fishman LM, Zybert PA (1992) Electrophysiological evidence of piriifomis syndrome. Arch Phys Med Rehabil 73:359–364 Fisher MA (1993) Other mononeuropathies of the upper extremity. In: Brown W, Bolton C (eds) Clinical electromyography, 2nd edn. Butterworth-Heineman, Boston, pp 271–322 Fisher MA, Bajwa R, Somashekar KN (2008) Routine electrodiagnosis and multiparameter technique in lumbosacral radiculopathies. Acta Neurol Scand 118:99–105 Francis H, March L, Terenty T, Webb J (1987) Benign joint hypermobility with neuropathy: documentation and mechanism of tarsal tunnel syndrome. J Rheumatol 14(3):577–581 Franz DN, Iggo A (1968) Conduction failure in myelinated and nonmyelinated axons at low temperatures. J Physiol (Lond) 199:319–345 Fu PK, Hsu HY, Wang PY (2008) Delayed reversible motor neuronopathy caused by electrical injury. J Chin Med Assoc 71(3):152–154 Galloway KM, Greathouse DG (2006) Tibial nerve motor conduction with recording from the first dorsal interosseous: a comparison with standard tibial studies. Neurol Neurophysiol Neurosci 10:2 Geselschap JH, van Gent WB, Wittens CH (2001) Complications in subfascial endoscopic perforating vein surgery: a report of two cases. J Vasc Surg 33(5):1108–1110 Gikas PD, Hanna SA, Aston W, Kalson NS, Tirabosco R, Saifuddin A, Cannon SR (2008) Post-radiation sciatic neuropathy: a case report and review of the literature. World J Surg Oncol 6:130 Gilliatt RW, Le Quesne PM, Logue V (1970) Wasting of the hand associated with a cervical rib or band. J Neurol Neurosurg Psychiatry 33:615–624 Gilliatt RW (1981) Physical injury to peripheral nerves. Mayo Clin Proc 56:361–370 Gilliatt RW, Sears TA (1958) Sensory nerve action potentials in patients with peripheral nerve lesions. J Neurol Neurosurg Psychiatry 21: 109–118 Gilliatt RW, Willison DM, Dietz V, Williams IR (1978) Peripheral nerve conduction in patients with a cervical rib or band. Ann Neurol 4:124–129 Ginanneschi F, Milani P, Reale F, Rossi A (2008) Short-term electrophysiological conduction change in median nerve fibres after carpal tunnel release. Clin Neurol Neurosurg 110(10):1025–1030 Gonick B, McCormick EM, Verweji BH, Rossman KM, Nigro MA (1998) The timing of congenital plexus injury: A study of electromyographic findings in the newborn piglet. Am J Obstet Gynecol 178:688–695 Gooch C, Azrieli Y, Weimer L, Lovelace R (2003) The utility of segmental nerve conduction studies in ulnar mononeuropathy at the elbow. Muscle Nerve 27(1):46–50 Goodman HV, Gilliatt RW (1961) The effect of treatment on median nerve conduction in patients with the carpal tunnel syndrome. Ann Phys Med 6:137–155 Gozal Y, Pomeranz S (1994) Sciatic nerve palsy as a complication after acoustic neurinoma resection in the sitting position. J Neurosurg Anesthesiol 6:40–42 Guiloff RJ, Sherratt RM (1977) Sensory conduction in medial plantar nerve. J Neurol Neurosurg Psychiatry 40:1168–1181 Hashimoto T, Mitomo M, Hirabuki N, Miura T, Kawai R, Nakamura H, Kawai H, Onon K, Kozuka T (1991) Nerve root avulsion of birth palsy: comparison of myelography with CT myelography and somatosensory evoked potential. Radiology 178:841–845 Heise CO, Lorenzetti L, Marchese AJT, Gherpelli JLD (2004) Motor conduction studies for prognostic assessment of obstetrical plexopathy. Muscle Nerve 30:451–455 Hetreed MA, Howard LA, Birch R (1994) Evaluation of preoperative SEP recorded from nerve roots to the cervical epidural space during brachial plexus surgery. In: Jones SJ, Boyd S, Hetreed M, Smith NJ (eds) Handbook of spinal cord monitoring. Kluwer Academic, Lancaster, pp 171–178
Clinical Neurophysiology in Peripheral Nerve Injuries Ho YH, Yan SH, Lin YT, Lo YK (2004) Sensory nerve conduction studies of the superficial peroneal nerve in L5 radiculopathy. Acta Neurol Taiwan 13(3):114–119 Hodes R, Larrabee MG, German W (1948) The human electromyogram in response to nerve stimulation and the conduction velocity of motor axons; studies on normal and on injured peripheral nerves. Arch Neurol Psychiatry 60(4):340–365 Hoffman H (1950) Local reinnervation in partially denervated muscle: a histophysiological study. Aust J Exp Biol Med Sci 28: 383–397 Holland NR (1998) Intraoperative electromyography during thoracolumbar spinal surgery. Spine 23:1915–1922 Hyllienmark L, Ludvigsson J, Brismar T (1995) Normal values of nerve conduction in children and adolescents. Electroencephalogr Clin Neurophysiol 97(5):208–214 Hynes JE, Jackson A (1994) Atraumatic gluteal compartment syndrome. Postgrad Med J 70:210–212 Ichikawa K, Nezu A (2005) Hereditary neuropathy with liability to pressure palsies in childhood: report of a case and a brief review. Brain Dev 27(2):152–154 Irwin MS, Sanders R, Green CJ, Terenghi G (1997) Neuropathy in nonfreezing cold injury (trench foot). J R Soc Med 90(8):433–438 Iyer VG (1998) Palmaris brevis sign in ulnar neuropathy. Muscle Nerve 21:675–677 Jaeckle KA (2004) Neurogical manifestations of neoplastic and radiation induced plexopathies. Semin Neurol 24:385–393 Jablecki CK, Andary MT (2002) Practice parameter: electrodiagnostic studies in carpal tunnel syndrome. Report of the American Association of Electrodiagnostic Medicine, American Academy of Neurology, and the American Academy of Physical Medicine and Rehabilitation. Neurology 58:1589–1592 Jellinek S (1960) The problem of electrical trauma placed on a new level as a result of basic research. Wien lin Wschr 72:501 Jones HR (1996) Plexus and nerve root lesions. In: Royden Jones H, Bolton CF, Harper CM (eds) Paediatric clinical electromyography. Lippincott-Raven, Philadelphia, pp 123–169 Jones HR, Herbison GJ, Jacobs SR, Kollross PR, Macones GA (1996) Intrauterine onset of a mononeuropathy in a newborn with electromyographic findings at day one compatible with prenatal onset. Muscle Nerve 19:88–91 Jones HR, Miller TA, Wilbourn A (2003) Brachial and lumbosacral plexopathies. The medico-legal debate. In: Jones HR, De Vivo DC, Darras BT (eds) Neuromuscular disorders of infancy, childhood and adolescence. A clinician’s approach. Butterworth-Heinemann, Philadelphia, p 61 Kaeser HE, Lambert EH (1962) Nerve function studies in experimental polyneuritis. Electroenceph Clin Neurophysiol 22:29–35 Kang PB, Preston DC, Raynor EM (2005) Involvement of superficial peroneal sensory nerve in common peroneal neuropathy. Muscle Nerve 31(6):725–729 Katifi HA, Sedgwick EM (1987) Evaluation of the dermatomal somatosensory evoked potential in the diagnosis of lumbo-sacral root compression. J Neurol Neurosurg Psychiatry 50:1204–1210 Katirji B, Wilbourn AJ (1988) Common peroneal mononeuropathy: a clinical and electrophysiologic study of 116 lesions. Neurology 38(11):1723–1728 Katirji B, Wilbourn AJ (1994) High sciatic lesion mimicking peroneal neuropathy at the fibular head. J Neurol Sci 121(2):172–175 Katirji B (2002) Electrodiagnostic approach to the patient with suspected mononeuropathy of the lower extremity. Neurol Clin 20: 479–501 Kempster P, Gates P, Byrne E, Wilson A (1991) Painful sciatic neuropathy following cardiac surgery. Aust NZJ Med 21:732–735 Keykhah MM, Rosenberg H (1979) Bilateral footdrop after craniotomy in the sitting position. Anesthesiology 51:163–164 Kiernan MC, Burke D, Andersen KV, Bostock H (2000) Multiple measures of axonal excitability: a new approach in clinical testing. Muscle Nerve 23:399–409
227 Kiernan MC, Cikurel K, Bostock H (2001a) Effects of temperature on the excitability properties of human motor axons. Brain 124: 816–825 Kiernan MC, Lin CSY, Andersen KV, Murray NM, Bostock H (2001b) Clinical evaluation of excitability measures in sensory nerve. Muscle Nerve 24:883–892 Kiernan MC, Mogyoros I, Burke D (1999) Conduction block in carpal tunnel syndrome. Brain 122:933–941 King AJ (1946) Note on the pathology of Morton’s metatarsalgia. Am J Clin Pathol 16:124 Kiss G, Komar J (1990) Suprascapular nerve compression at the spinoglenoid notch. Muscle Nerve 13:556–557 Kline DG, Dejonge BR (1968) Evoked potentials to evaluate peripheral nerve injuries. Surg Gynecol Obstet 127(6):1239–1248 Kline DG, Happel LT (1993) A quarter century’s experience with intraoperative nerve action potential recording. Can J Neurol Sci 20:3–10 Ko SH, Wook C, Hyun CK (2004) Delayed spinal cord injury following electrical burns: a 7-year experience. Burns 30:691–695 Kothari MJ, Heistand M, Rutkove SB (1998a) Three ulnar nerve conduction studies in patients with ulnar neuropathy at the elbow. Arch Phys Med Rehabil 79:87–89 Kothari MJ, Macintosh K, Heistand M, Logigian EL (1998b) Medial antebrachial cutaneous sensory studies in the evaluation of neurogenic thoracic outlet syndrome. Muscle Nerve 21:647–649 Kraft GH (1972) Axillary, musculocutaneous and suprascapular nerve latency studies. Arch Phys Med Rehabil 53:383–387 Kuhlman GS, McKeag DB (1999) The “Burner”: a common nerve injury in contact sports. Am Family Phys 60(7):2035–2040 Kuntzer T, Van Melle G, Regli F (1997) Clinical and prognostic features of femoral neuropathies. Muscle Nerve 20:205–211 Lagueny A, Deliac MM, Deliac P, Durandeau A (1991) Diagnostic and prognostic value of electrophysiological tests in meralgia paraesthetica. Muscle Nerve 14:51–56 Lederman RJ, Wilbourn AJ (1984) Brachial plexopathy: recurrent cancer or radiation? Neurology 34:1331 Lee K-Y, Lee Y-J, Koh S-H (2009) Usefulness of the median terminal latency ratio in the diagnosis of carpal tunnel syndrome. Clin Neurophysiol 120:765–769 Lee MJ, Lastayo PC (2004) Pronator syndrome and other nerve compressions that mimic carpal tunnel syndrome. J Orthop Sports Phys Ther 34(10):601–609 Lee RC, Astumian RD (1996) The physicochemical basis for thermal and non-thermal ‘burn’ injuries. Burns 22:509–519 Le Forestier N, Moulonguet A, Maisonobe T, Leger J-M, Bouche P (1998) True neurogenic thoracic outlet syndrome: electrophysiological diagnosis in 6 cases. Muscle Nerve 21:1129–1134 Levin KH (1998) L5 radiculopathy with reduced superficial peroneal sensory responses: intraspinal and extraspinal causes. Muscle Nerve 21(1):3–7 Levin KH, Maggiano HJ, Wilbourn AJ (1996) Cervical radiculopathies: comparison of surgical and EMG localisation of single-root lesions. Neurology 46:1022–1025 Leffler A-S, Hansson P (2008) Painful traumatic peripheral partial nerve injury-sensory dysfunction profiles comparing outcomes of bedside examination and quantitative sensory testing. Eur J Pain 12: 397–402 Lindner A, Schulte-Mattler W, Zierz S (1997) Postpartum obturator nerve syndrome: case report and review of the nervecompression syndrome during pregnancy and delivery. Zentralbl Gynäkol 119(3): 93–99 Lo YL, Prakash KM, Leoh TH (2004) Posterior antebrachial cutaneous nerve conduction study in radial neuropathy. J Neurol Sci 223(2): 199–202 Love S (1983) An experimental study of peripheral nerve regeneration after x-irradiation. Brain 106(Pt 1):39–54 Mancias P, Slopias JM, Curtis CG, Zuker RM, Seifu Y, Andrews DF (1994) The natural history of obstetrical brachial plexus palsy. Plast Reconstr Surg 93:675–680
228 Makela JP, Ramstad R, Mattila V, Pihlajamaki H (2006) Brachial plexus lesions after backpack carriage in young adults. Clin Orthop Relat Res 452:205–206 McManis PG (1994) Sciatic nerve lesions during cardiac surgery. Neurology 44:684–687 McMomas AJ, Fawcett PRW, Campbell MJ, Sica REP (1971) Electrophysiological estimation of the number of motor units within a human muscle. J Neurol Neurosurg Psychiatry 34:121–131 Miledi R, Slater CR (1970) On the degeneration of rat neuromuscular junctions after nerve section. J Physiol (Lond) 207:507–528 Mills KR (1985) Orthodromic sensory action potentials from palmar stimulation in the diagnosis of carpal tunnel syndrome. J Neurol Neurosurg Psychiatry 48:250–255 Mochida H, Kikuchi S (1995) Injury to infrapatellar branch of saphenous nerve in arthroscopic knee surgery. Clin Orthop Relat Res 320:88–94 Moghekar AR, Moghekar AR, Karli N, Chaudhry VJ (2007) Brachial Plexopathies: etiology, frequency, and electrodiagnostic localization. Clin Neuromusc Dis 9:243–24 Morgan G, Wilbourn AJ (1998) Cervical radiculopathy and coexisting distal entrapment neuropathies – Double-crush syndromes? Neurology 50:78–83 Morgan SH, Rudge P, Smith SJM, Bronstein AM, Kendall BE, Holley E, Young EP, D’A crawford M, Bannister R (1990) The neurological complications of Anderson-Fabry disease (alpha galactosidase A deficiency) – investigation of symptomatic and presymptomatic patients. QJ Med 75:491–508 Nakano KK (1977) Anterior interosseous syndrome. Arch Neurol 34:477 Nodera H, Kaji R (2006) Nerve excitability and its clinical application to neuromuscular diseases. Clin Neurophysiol 117:1902–1916 Ochoa J, Danta G, Fowler TJ, Gilliatt RW (1971) Nature of the nerve lesion caused by a pneumatic tourniquet. Nature 233(5317):265–266 Oh SJ (2007) Neuropathies of the foot. Clin Neurophysiol 118:954–980 Olsen NK, Pfeiffer P, Johannsen L, Schrøder H, Rose C (1993) Radiation-induced brachial plexopathy: neurological follow-up in 161 recurrence-free breast cancer patients. Int J Radiat Oncol Biol Phys 26(1):43–49 Padua L, Padua R, Nazzaro M, Tonali P (1998) Incidence of bilateral symptoms in carpal tunnel syndrome. J Hand Surg Br 23(5):603–606 Paintal AS (1965) Block of conduction in mammalian myelinated nerve fibres by low temperatures. J Physiol (Lond) 180:1–19 Papazian O, Alfonso I, Yalali I, Velez I, Jayakar P (2000) Neurophysiological evaluation of children with traumatic radiculopathy, plexopathy and peripheral neuropathy. Semin Pediatr Neurol 7:26–35 Paradiso G, Granana N, Maza E (1997) Prenatal brachial plexus injury. Neurology 49:261–262 Parsonage MJ, Turner JW (1948) Neuralgic amyotrophy; the shoulderGirdle syndrome. Lancet 251:973–978 Pasila M, Jaroma H, Kiviluoto O, Sundholm A (1978) Early complications of primary shoulder dislocations. Acta Orthop Scand 49(3):260–263 Payne G (pers comm) Burners and stingers: shoulder injuries in rugby players. Conference Abstract. ABN/BSCN Joint Meeting, Croke Park, Dublin, 2008 Peccina M (1979) Contribution to the aetiologic explanation of the piriformis syndrome. Acta Anat 105:181–187 Pearce JM (2001) Emil Heinrich Du Bois-Reymond (1818–96). J Neurol Neurosurg Psychiatry 71(5):620 Plewnia C, Wallace C, Zochodne D (1999) Traumatic sciatic neuropathy: a novel cause, local experience, and a review of the literature. J Trauma 47(5):986–989 Pitt M, Vredeveld JW (2005) The role of electromyography in the management of the brachial plexus palsy of the newborn. Clin Neurophysiol 116:1756–1761 Ratnayake B, Emmanuel ER, Walker CC (1996) Neurological sequelae following a high tension electrical burn. Burns 22(7):574–577
Surgical Disorders of the Peripheral Nerves Raudino F (2004) The value of inching technique in evaluating the peroneal nerve entrapment at the fibular head. Electromyogr Clin Neurophysiol 44(1):3–5 Rengachary SS, Neff JP, Singer PA, Brackett CE (1979) Suprascapular entrapment neuropathy: a clinical, anatomical, and comparative study. Part II: anatomical study. Neurosurgery 5:447–451 Rivner MH, Swift TR, Crout BO, Rhodes KP (1990) Toward more rational nerve conduction interpretations: the effect of height. Muscle Nerve 13:232–239 Rosen I, Werner CO (1980) Neurophysiological investigation of posterior interosseus nerve entrapment causing lateral elbow pain. Clin Neurophysiol 50:125–133 Ross D, Jones HR, Fisher J, Konkol RJ (1983) Isolated radial nerve lesion in the newborn. Neurology 33:1354–1356 Roy S, Levine AB, Herbison GJ, Jacobs SR (2002) Intraoperative positioning during caesarean section as a cause of sciatic neuropathy. Obstet Gynecol 99(4):652–653 Sances A, Larson SJ, Myklebust JB et al (1979) Electrical injuries. Surg Gynecol Obstet 149:97 Sala F, Niimi Y, Berenstein A, Deletis V (2001) Neuroprotective role of neurophysiological monitoring during endovascular procedure in the spinal cord. Ann NY Acad Sci 939:126–136 Sears TA, Bostock H (1981) Conduction failure in demyelination: is it inevitable? Adv Neurol 31:357–375 Seddon HJ (1943) Three types of nerve injury. Brain 66:237–288 Seror P (2004) Medial antebrachial cutaneous nerve conduction study, a new tool to demonstrate mild lower brachial plexus lesions. A report of 16 cases. Clin Neurophysiol 115:2316–2322 Seror P (2002) Sural nerve lesions: a report of 20 cases. Am J Phys Med Rehabil 81:876–880 Seror P, Seror R (2006) Meralgia paresthetica: clinical and electrophysiological diagnosis in 120 cases. Muscle Nerve 33(5):650–654 Sever JW (1916) Obstetric paralysis. Am J Dis Child 12(6):541 Seyffarth EA (2006) Julius Bernstein (1839-1917): pioneer neurobiologist and biophysicist. Biol Cybern 94(1):2–8 Seyffarth H (1951) Primary myoses in the pronator teres as a cause of the lesion of the median nerve. Acta Psychiatrica Scandinavia 74(S):251 Simpson JA (1956) Electrical signs in the diagnosis of carpal tunnel and related syndromes. J Neurol Neurosurg Psychiatry 19(4):275–280 Slack JR, Williams MN (1981) The absence of nodal sprouts from partially denervated nerve trunks. Brain Rsr 226:291–297 Slooff AC (1993) Obstetric brachial plexus lesions and their neurosurgical treatment. Clin Neurol Neurosurg 95(suppl):73–77 Smith JW, Thesleff S (1976) Spontaneous activity in denervated mouse diaphragm muscle. J Physiol (Lond) 257:171–186 Smith HR (2004) Nerve conduction study results and treatment outcome in suprascapular neuropathy. Neurology 62(Suppl 5):A376 Sonck WA, Francx MM, Engels HL (1991) Innervation anomalies in upper and lower extremeties: potential clinical implications. Electromyog Clin Neurophysiol 31:67–80 Spindler HA, Dellon A (1990) Nerve conduction studies in the superficial radial entrapment syndrome. Muscle Nerve 13:1–5 Stecker MM, Baylor K (2009) Peripheral nerve at extreme low temperatures 1: effects of temperature on the action potential. Cryobiology 59:1–11 Stewart JD (1987) The variable clinical manifestations of ulnar neuropathies at the elbow. J Neurol Neurosurg Psychiatry 50:252–258 Sumner AJ (2007) Parsonage Turner revisited. J Clin Neuromuscul Dis 8(4):237–239 Sunderland S (1990) The anatomy and physiology of nerve injury. Muscle Nerve 13:771–784 Sunderland S (1991) Nerve injuries and their repair – a critical appraisal. Churchill Livingstone, Edinburgh, p 103 Sutter M, Deletis V, Dvorak J, Eggspuehler A, Grob D, MacDonald D (2007) Current opinions and recommendations on multimodal intraoperative monitoring during spine surgeries. Eur Spine J 16: S232–S237
Clinical Neurophysiology in Peripheral Nerve Injuries Tavee J, Mays M, Wilbourn AJ (2007) Pitfalls in the electrodiagnostic studies of sacral plexopathies. Muscle Nerve 35(6):725–729 Thomas JE, Cascino TL, Earle JD (1985) Differential diagnosis between radiation and tumour plexopathy of the pelvis. Neurology 35:1–7 Tsao BE, Shields RW, Wilbourn AJ (2008) True neurogenic thoracic outlet syndrome: electrodiagnostic analysis of 42 cases. Clin Neurophysiol 119:e45 Upton RM, McComas AJ (1973) The double crush in nerve entrapment syndromes. Lancet 11:359–362 Van Alfen N, Van Engelen BG (2006) The clinical spectrum of neuralgic amyotrophy in 246 cases. Brain 129(Pt 2):43 Van Dijk J, Pondaag W, Malessy MJ (2001) Obsetric lesions of the brachial plexus. Muscle Nerve 24:1451–1461 Vredeveld JW, Blaauw G, Slooff AC, Richards R, Rozeman SC (1996) The findings in paediatric brachial palsy differ from those in older patients: a suggested explanation. Electromyogr Clin Neurophysiol 99:299 Vogel CM, Albin R, Alberts JW (1991) Lotus footdrop: sciatic neuropathy in the thigh. Neurology 41(4):605–606 Walters RJ, Murray NM (2001) Transcarpal motor conduction velocity in carpal tunnel syndrome. Muscle Nerve 24:966–968
229 Watson WF, Brown BV (1993) Acute retrohumeral radial neuropathies. Muscle Nerve 16:706–711 Wilbourn AJ (2007) Plexopathies. Neurol Clin 25(1):166–167 Wilbourn AJ, Gilliatt RW (1997) Double crush syndrome: a critical analysis. Neurology 49:21–29 Wilbourn AJ, Aminoff M (1988) AAEM Minimonograph #32. Muscle Nerve 11:1099–1114 Wong CA, Scavone BM, Dugan S, Smith JC, Prather H, Ganchiff JN, McCarthy RJ (2003) Incidence of postpartum lumbosacral spine and lower extremity nerve injuries. Obstet Gynecol 101(2):279–288 Wu JS, Morris JD, Hogan GR (1985) Ulnar neuropathy at the wrist: case report and review of literature. Arch Phys Med Rehabil 66:785–788 Wulff CH, Gilliatt RW (1979) F waves in patient with hand wasting caused by a cervical rib and band. Muscle Nerve 2:452–457 Yuen EC, So YT, Olney RK (1995) The electrophysiologic features of sciatic neuropathy in 100 patients. Muscle Nerve 18(4):414–420 Yulmaz K, Cahskan M, Oge E, Aydmh N, Tunaci M, Ozmen M (1999) Clinical assessment, MRI and EMG in congenital brachial plexus palsy. Eur J Paediatr Neurol 21:705–710
7
Operating on Peripheral Nerves
Operating on nerves: indications for operation; conditions and requirements; general principles of operation; intra-operative diagnosis; methods of repair; neurolysis; approaches to individual nerves; possibilities of intradural repair; hazards in entrapment neuropathy; pitfalls in operating on nerve tumours.
7.1 Indications and Objects of Intervention Petrie (2006) defines an experimental study as one: “in which the investigator intervenes in some way to affect the outcome. Such studies are longitudinal and prospective; the investigator applies the intervention and observes the outcome some time later”. The objects of intervention are (1) to confirm or establish diagnosis; (2) to restore continuity to a severed or ruptured nerve; (3) to remove a noxious agent compressing or distorting or occupying a nerve. It is difficult to overstate the significance of worsening of pain and deepening of nerve lesion caused by expanding haematoma or ischaemia. Clinicians must never forget that nerves compressed in a swollen ischaemic limb or in a tense compartment progress from conduction block to much deeper and much less favourable degenerative lesions. Pain persisting after a focal injury to a nerve is an indication for operation at almost any interval after injury. No one should forget the lesson of case 3 in Birch and Strange’s (1990) series. There, exploration and decompression of the sciatic nerve in the notch 3 years after injury was rapidly followed by relief of pain and recovery of a paralysis affecting the common peroneal component. That lesion was of course a conduction block prolonged by external causes. Montgomery et al. (2005) reported a similarly gratifying outcome after neurolysis of the sciatic nerve in a patient who had endured severe pain for a number of years. The lesion was predominantly a conduction block, prolonged and maintained by fibrosis tethering the nerve following total hip arthroplasty. Even more remarkable is the case described by Camp, Milano and Sinisi (2008). The patient suffered intractable and increasing pain for 18 years. Pain was abolished by neurolysis of the ulnar nerve which had become adherent to the pulsatile vein graft used to repair the brachial artery. That lesion was a conduction block, prolonged by external causes: the pain was neurostenalgia.
“It has all been done before” (Holmes 1881). The British Medical Journal of 15 May 1915 carried a report of an address by Inspector-General Delorme (1915) to the Académie de Médecine on the treatment of gunshot wounds of the nerves. In this address most of the procedures to be recommended here are canvassed, including the use of electrical stimulation during operation. Alas! The medical world then was evidently unprepared for Delorme’s proposals, many of which were sharply criticised by the great and good. Bland-Sutton (1930) relates that in the 1870s J. W. Hulke, then surgeon to the Middlesex Hospital in Mortimer Street London, repaired the cut median and ulnar nerves of a woman. Bland-Sutton continues: “The operation was successful and this was the first instance of a successful secondary suture of a divided nerve in London. Surgeons had been afraid to suture cut nerves in this way for fear of tetanus” The harmfulness of delay before repair and the limitations of the different methods of repair have been described in Chapters 3 and 4. Some important practical advantages of urgent exploration include the ease of recognising a rupture and the ease with which the stumps can be approximated. The best time to explore such injuries is before distal conduction is lost (Figs. 7.1–7.3). There is no need to explore the nerve if the clinician expects spontaneous recovery. Seddon (1975a) had this to say about nerve lesions complicating closed fractures of the shaft and the lower end of the humerus: “however, on balance, it would seem proper to await spontaneous recovery of the nerve, provided the two conditions are satisfied. The first is reasonable apposition of the bony fragments, and the other complete certainty that there is no threat of ischaemia of the forearm muscles”. If however, the clinician elects to convert a closed fracture to an open one by internal fixation, then it is wisest to expose nerves which are not working. The fracture surgeon who does not do this is asking for trouble (Fig. 7.4). We define primary repair as one performed within 5 days of injury and delayed primary repair
R. Birch, Surgical Disorders of the Peripheral Nerves, DOI: 10.1007/978-1-84882-108-8_7, © Springer-Verlag London Limited 2011
231
232
Surgical Disorders of the Peripheral Nerves
for one completed at intervals from 5 days to 3 weeks after injury. Secondary repair is performed at between 3 weeks and 3 months after injury. Late, or neglected, repair is reserved for those cases where delay exceeds 3 months.
7.1.1 Special Units: Their Role
Fig. 7.1 Ease of diagnosis at operation done soon after injury. Incomplete transection of the femoral nerve occurred during a difficult pelvic operation. The surgeon recognised the event and enabled display and repair 24 h after wounding.
From 1956 onwards primary suture of severed nerves was practised at St Mary’s Hospital whenever it was possible. This policy was extended to revascularisation of damaged limbs, to replantation of amputated hands and severe injuries of the brachial plexus. With growing experience it became obvious that Special Units would evolve to provide opinion and treatment without delay. These require the constant availability of adequately supported senior advice. It is essential that all members of staff understand the overriding purpose of such a Unit which is to provide urgent advice and
Fig. 7.2 Ischaemia and conduction. Traction lesion of the brachial plexus was accompanied by rupture of the subclavian artery. There was a weak pulse. At operation, 54 h after injury, stimulation of the avulsed ventral roots of C7, C8 and T1 evoked strong contraction in the relevant muscles distally. This showed that there was neither critical ischaemia
within the limb nor that there was a second, more distal, lesion. Strong SSEP’s were recorded from the stumps of C5 and C6 (1). The dorsal root ganglia of C7, C8 and T1 (2) and their ventral roots (3) are shown. An extensive repair was done (Case investigated and referred by Mr Tanaka and Mr Shandall, Royal Gwent Hospital).
Operating on Peripheral Nerves
Fig. 7.3 It is often very hard to determine from inspection alone the nature of the injury in delayed or neglected cases. Ten weeks after a severe lesion of C5, C6 and C7 it was not possible to distinguish between rupture and avulsion. The phrenic nerve (1) suprascapular nerve (2) and the neuroma (3) are seen.
Fig. 7.4 The explanation for persisting painful median palsy in a 7 year old child after fracture of distal humerus, which had been treated by “closed” wiring, was revealed when the nerve was found embedded within the bone 11 months later (Courtesy of Mr Roderick Dunn, Salisbury).
233
organism before admission and MRSA carriers are treated in isolation beds. Emergency admissions are admitted into single rooms until their MRSA status is known. Elective admissions are stopped if the number of positive MRSA cases exceeds the number of single rooms available. This admirable policy of segregation is not a new one. Rodger (2004) describes how the building of the Royal Naval Hospitals at Haslar (Portsmouth) and at Stonehouse (Plymouth) came about because of the perceived need to control epidemics of typhus: “At a cost of over £100,000…. Haslar provided for the highest standards of medical care, careful segregation of infectious diseases, and (not least) careful guard against the risk of desertion. Haslar, and even more the second naval Hospital, Stonehouse near Plymouth (finished in 1762), the first to be built on the “pavilion” system with separate blocks isolated from one another, became a model studied and copied all over Europe”. Those responsible for the control of hospital infection might learn something from recent Naval experience: “throughout the eighteenth century British naval officer’s fanatical attention to cleanliness of their ships and men aroused the astonishment of visiting foreigners, but it had unquestionably good results in limiting disease” (Rodger 2004). The difficulties and the obstacles faced by Special Units are increasingly severe but they must have an important role particularly in the treatment of severe injuries to the brachial and lumbosacral plexi and also in the treatment of iatrogenous injuries. Special Units must work closely with the referring hospital to ensure continuity of care through the process of rehabilitation and they should provide good records drawn from adequate follow up with thorough audit. One object of a Special Unit is to plan its own demise by training surgeons to take on the work in their own hospitals.
7.2 General Principles of Operation 7.2.1 Control of Bleeding
urgent treatment. Systems were developed to facilitate hospital to hospital transfer (Giddins et al. 1998). The work of such Units require commitment and support from hospital managers as it must from those senior to whom they must answer. Such support unfortunately appears to be less and less forthcoming as Bircher et al. (2006) point out when they relate the difficulties imposed upon their work with patients with fracture dislocations of the pelvis. Bircher et al. (2006) cite interference with referrals, the imposition of waiting time targets, and the failure to provide isolation beds to cope with the problem of hospital acquired infection. The last of these is a serious and growing problem. Nixon and his colleagues (2006) describe their policy for the control of methicillin resistant staphylococcus aureus (MRSA) infection in orthopaedic wards. All elective patients are screened for the
The development of methods to control bleeding is described, with graceful scholarship, by Kirkup (2007) in “A History of Limb Amputation”. Petit’s screw, invented in 1718 “dominated practise for more than two centuries, being still available during World War I”. Crumplin and Harrison (2005) relate how Nelson lost his right arm at Santa Cruz in 1795. He was struck on the right elbow by either a canister round or a musket ball: “he was bleeding, and Nisbet, after concealing this with his hat, removed his silk stock and bound the Admiral’s arm above the elbow to control the haemorrhage, so saving the Admiral’s life”. Blutleere (bloodless) surgery, introduced by Esmarch (1873), was advocated by Stromeyer (Smith 2006) who commented “in Edinburgh, some opposition was shown, as usually happens with everything that is admired in London”.
234
Crumplin (2007) describes how Guthrie, from 1808, controlled the subclavian artery by digital compression against the first rib. Hennen (1820) recorded that in one case of shoulder disarticulation at the Battle of Vitoria (1813) “the amount of blood loss from the principal artery was no more than the quantity contained between the point of pressure and the point of incision through the vessel”. Digital compression of the femoral artery at the groin was used to control bleeding in the first successful case of disarticulation at the hip. Hennen (1820) described a successful case operated 19 days after injury, at Waterloo and he observed that “the haemorrhage was very little more than in the common amputation with the tourniquet”. The emergency control of bleeding by pressure is an art to practice for we have seen cases where patients were close to exsanguination following stab wounds to the axillary or brachial, or to the superficial femoral arteries. Such extreme blood loss might have been reduced by firm pressure above and below the wound. The tourniquet is potentially dangerous. One of us, as an aspirant surgeon, was asked what was the correct course of action should a surgeon realise that a tourniquet had been left, inflated, overnight after an operation: the correct answer was that the tourniquet should be left inflated, and that the surgeon should speak with the Presidents of the Colleges of Surgeons before proceeding to amputation of the limb. Such a catastrophe is inconceivable nowadays but we are indebted to Professor Averil Mansfield for the observation that the use of a tourniquet in a limb in which an arterial prosthesis has been inserted, is absolutely contra-indicated. It seems that the implant is insufficiently elastic to dilate after release of the cuff; also, collateral circulation is likely to be defective. Tourniquet times must be reduced in patients with rheumatoid arthritis, diabetes mellitus, alcohol addiction or other possible causes of neuropathy. In many such patients it is best to avoid altogether the use of a tourniquet. Klenerman (2003) provides an important manual about tourniquet use. The superficially attractive idea of temporary release of the tourniquet is undermined by Concannon et al. (1992) who studied patterns of free radical production after tourniquet ischaemia. Post operative pain is worsened by the tourniquet by longer periods of ischaemia and in older patients (Ömeroglu et al. 1998). The duration of application, inflation pressure and site of application should be always recorded in an ordered operating note and the times of application and release should be written on board in the operating theatre and in the case notes. The duration of tourniquet ischaemia is reduced by inflating the cuff after preparation and towelling of the limb. We prefer Lister’s (1879) method of elevation of the limb before inflation of the cuff to the exsanguinating bandage. Because these operations are usually time-consuming, it is especially important adequately to protect the pressure points at the knee, the elbow and elsewhere with suitable padding. In operations on the nerves of the neck in particular, special care
Surgical Disorders of the Peripheral Nerves
must be taken to avoid air embolism, to recognise it if it does occur and to be prepared to deal effectively with any such occurrence. In the case of severe injuries to the brachial plexus, when there may be a sudden release of spinal fluid, it is necessary to be prepared quickly to alter the position of the patient to avoid “coning” of the medulla. If it seems likely that nerve or vein grafts are going to be needed, the donor sites should be prepared and monitoring equipment, arterial and venous lines and the the drapes should be placed accordingly.
7.3 Apparatus and Instruments Most operations on injured peripheral nerves are best done with general anaesthesia. If the anaesthetist uses a muscle relaxant must be prepared to reverse its effect when the nerve stimulator/recorder is being used. Special apparatus required includes bi-polar diathermy, stimulating and recording apparatus and instruments for magnification. Neurophysiological examination: “It is astonishing that in this age of technological idolatory, exploration of nerves is still being performed without the aid of this instrument” (Seddon 1975b). Seddon refers here to a simple nerve stimulator. Nerve stimulation and the recording of nerve conduction is used to (1) identify the nerve, (2) to identify individual bundles within a nerve or in a prepared stump, (3) to demonstrate conduction across a lesion and (4) to record continuity between the central and peripheral nervous systems. Intraoperative neurophysiological studies have been performed in more than 4,000 operations since 1953. For simple stimulation and observation of motor response only the simplest uni- or bi-polar stimulator is necessary. For stimulation and recording from muscle and nerve, more elaborate apparatus is required. We now use the Medelec synergy monitoring system (Vaisys Health care, Madison, Wisconsin, USA) (Fig. 7.5). The electrodes are provided by Ambu, Ballerup, Denmark. In measuring conduction across a lesion, hand held bi-polar stimulators are placed on either side of the lesion. The ground electrode can be placed in a convenient adjacent area. The interval between the electrodes and between each electrode and the lesion is measured. Kline et al. (2008a) relate further experience with this method in an extensive and authoritative manner, describing findings in nearly 2,300 patients operated between 1967 and 2001 (Fig. 7.6). For recording somatosensory evoked potentials (SSEP’s) the reference electrode is placed on the forehead, the ground electrode at the temple, and the recording electrode on the skin overlying the second or third cervical intervertebral space. The skin is prepared with abrasive paste and alcohol wipes to lower resistance with the object of balancing impedance between reference and recording electrodes, at a level
Operating on Peripheral Nerves
Fig. 7.5 The apparatus for recording sensory evoked potentials.
less than 2.0 kOhms. Once the surface electrodes are positioned they are secured with tape. A hand held stimulator is used to record signals from the median and ulnar nerves before preparing and towelling the upper limb, using the uninjured side as a control. A sterile hand held bi-polar stimulator is used to stimulate the nerve directly. Somato-sensory evoked potentials (SSEP) are recorded using a stimulus rate of 3–5 pulses per second, of duration 2.0 mS, and intensity 150–300 V. Signals are averaged from between 50 and 200 sweeps. The stimulus rate for the hand held bi-polar stimulator is 3–5 pulses per second. The quality of the traces may be adversely affected by a number of factors. Ambient noise interference from electrical equipment in the operating theatre was a particular problem in the operating theatres at St Mary’s Hospital. Difficulties are encountered when nerves are embedded in dense fibrosis
235
or the operating site is permitted to become too wet or too dry. Compression of nerves by haematoma causes anoxic conduction block. SSEP’s are relatively unaffected by most anaesthetic agents, but muscle relaxants block neuromuscular conduction (Fig. 7.7). We have no experience with transcranial electromagnetic evoked potentials (TCEMEP) during operation. Schwartz and colleagues (2007) observed that changes in TCEMEP appeared some minutes before alterations in somato-sensory evoked potentials (SSEP) during operations on the deformed spine. By this method, Lee and colleagues (2006) detected anterior haematoma in two cases of anterior fusion of the cervical spine. The first case was diagnosed by a sudden fall in amplitude of TCEMEP. Improvement in TCEMEP correlated with recovery of the cord in the second case. Lee et al. conclude that: “TCEMEP is… proving to be considerably more sensitive than conventional SSEP monitoring which carries an unacceptably high rate of false negative results in cervical spinal surgery”. Magnification: the operating microscope was used by ophthalmic and ENT surgeons for decades before its regular use in orthopaedic and plastic surgery (Nylen 1972). The discipline of microsurgery seems to have acquired a mystique which is not altogether justified. The elements of microsurgical technique are no more than the application of basic surgical skills. They are acquired by practice. (Birch 1987). For magnification we use loupes or the operating microscope. The microscopes are OPMI 6SD FC and OPMI 6 (both Carl Zeiss, Oberkochen): the stand is the universal S3B (Carl Zeiss, Oberkochen) Instruments: Stromeyer (Smith 2006) commented: “I used my stay in London to buy surgical instruments for the Freiburg clinic at Weiss and Fergusson. They were excellent. Professor Hecker, when he amputated for the first time with an English knife, was quite alarmed at its sharpness.” Our earliest micro-instruments were bought from John Weiss of Wigmore Street, and Dumont’s watch makers forceps were bought, ten at a time for £12, from the jewellers of Hatton Garden. Probably the best of the early microsutures was that developed by Owen (1975) who swaged a 4 mm needle onto a 10/0 perlon filament. The Joll’s thyroid retractor is excellent for reflection of skin flaps in the neck. The small Deaver’s retractor is useful in the neck. A set of malleable retractors are essential. Conventional toothed self retaining retractors are avoided, because of the risk to nerves and vessels. Three sizes of vascular clamps are used. The Satinsky clamp is especially useful in end to side anastamosis. DeBakey’s scissors and forceps are used for both arterial and nerve work. The range of sutures includes: 6/0, 8/0 nylon on a 6 or 8 mm vascular needle and 8/0, 9/0, 10/0 and 11/0 on a 4 mm or a 6 mm needle used with the appropriate needle holders. Howarth’s dental elevator and Lemperts raspatory are very good for fine bone work. A range of spinal punches,
236
Surgical Disorders of the Peripheral Nerves
Fig. 7.6 Severe pain and deep paralysis complicated closed dislocation of the gleno-humeral joint in a 78 year old woman. The nerves were exposed 3 months later; there were no ruptures. CNAP and SSEP were recorded traversing the lesions of median, ulnar and musculocutaneous nerves. They were absent across the radial nerve. Recovery through the radial nerve extended only to the triceps and flexor to extensor transfer was performed later. The other nerves recovered, there was early relief of pain.
angled bone nibblers and bone cutters are needed. Plasma “glue” (Young and Medawar 1940, Tarlov 1944) was regularly used in our hospitals until the early 1980s when the raw material became unavailable. A commercial product was reintroduced which is now used, (Tisseel (TM) Immuno Ltd, Arctic House, Rye Lane, Dunton Green, Sevenoaks, TN14 5HB). The aprotinin must be diluted with sterile water: otherwise there is a risk of inducing fibrosis. The undiluted preparation is reserved for haemostasis. The needle tip should be directed away from the repair to avoid disruption. A steady
gentle pressure is exerted so that a film of the fluid bathes the repair and seals it . Fibrin clot glue acts as an envelope around the nerve but offers no resistance to tension. Narakas (1988) found that the proportion of good results of his nerve repairs was increased by 15% after turning to fibrin clot glue. Prevention of Pain: Surgeons can and should do more to prevent post operative pain. The complications of blind nerve or regional blocks can be severe (Chapter 3). Henry (1975) describes his technique of thigh amputation without tourniquet or general anaesthetic by a method first described before
Operating on Peripheral Nerves
237
Fig. 7.7 Examples of normal and abnormal SSEP traces.
the Second World War: “so, after infiltrating areas of flap or cuff, and all the operative field – most thoroughly- with procaine (0.5%), we can in comfort see and block the great sciatic trunk”. Then follows ligation of main vessels and then: “during these activities, however, the great sciatic trunk will rest in peace – until the time comes to remove the limb. With procaine infiltration, not less than 20 min must elapse before injecting and dividing this capacious conduit of shock impulse. And that, indeed, is little time enough”. Surgeons can and should do a great deal more to diminish postoperative pain by the simple and safe expedient of infiltration of the line of incision and the skin on either side, with local anaesthetic. We use a preparation of Levobupivicaine 0.25%, with adrenalin 1:200,000. The maximum dose is 2 mg per Kgm body weight. It is a simple matter to infiltrate the tissues around the
supraclavicular nerves and to inject local anaesthetic into the joint itself for operations at the shoulder. If amputation of the lower limb is indicated because of a deformed and painful foot, a circumferential block of the skin of the mid thigh is made before exposing the sciatic nerve. This is bathed in local anaesthetic infused through an epidural catheter, with its tip adjacent to the nerve, before proceeding to amputation. The infusion is maintained for 48 h after operation. Other nerves are infiltrated with local anaesthetic before cutting them. The use of prolonged nerve block in the treatment of the painful nerve lesion is outlined in Chapter 12. Incisions need to be adequate and, where possible, extensile. There is no place for short incisions. Main nerves and major vessels are exposed first proximal to the lesion and then distal to it. The lesions are displayed by dissection from
238
Mandibular br. of facial n.
Surgical Disorders of the Peripheral Nerves
Lesser occipital n. Splenius capitis Great auricular n. Spinal accessory n.
Transverse cervical n.
Sternocleidomastoid m.
Supraclavicular n.n.
Trapezius m.
Fig. 7.8 Some of the more important superficial nerves in the neck. Note that the mandibular branch of the facial nerve comes at one point below the lower margin of the mandible. The cervical branch has been omitted.
above and below. All tissues must be treated gently. They are, after all, alive, and on their continued viability depend the healing of the wound without infection and the recovery of the nerve lesion. Indeed, avoidance of infection probably has more to do with tender handling of the tissues and accurate haemostasis than has the administration of antibiotics. The line of the proposed incision should be marked and cross-hatched and be planned with cutaneous innervation in mind. Lasting trouble can follow wounding of an apparently trivial cutaneous nerve; it is an ill matter for a surgeon purporting to assist healing of lesions of peripheral nerves to damage one set of nerves on the way to repairing another (Figs. 7.8 and 7.9). In the neck, the skin flaps should include
the platysma; elsewhere, they should be cut to full thickness. Skin flaps should be held with fine skin hooks; if the procedure is going to take a long time, the flaps should be sutured back to the surrounding skin. So far as possible, dissection should be done with the knife or with sharp blunt-ended scissors. Although in the limbs the pneumatic tourniquet can be used during dissection and exposure, it must be released for stimulation, repair and closure. The best possible haemostasis must be secured after the cuff has been deflated, by diathermy, ligation and haemostatic sponge. Cut bone surfaces should be sealed with wax or similar preparation. Once the flaps are raised, the field should be kept as free of blood as possible, but it must be kept moist by regular irrigation. Even well shielded operation lamps generate enough heat to accelerate desiccation of the tissues. When muscles have to be divided, their ends should be marked with sutures and if necessary labelled, so that when the procedure is completed they can accurately be re-united. Nerves should be handled with extreme care; retracted with very fine skin hooks in the epineurium or with plastic slings. Colour coding of the slings adds pleasing variety to the proceedings and, more importantly, permits the surgeon to identify to the assistant the nerve to be retracted. They should not be mobilised over such a length as to impair their blood supply. The wound should not be closed before bleeding points have been checked; even with good haemostasis it is wise to use a suction drain in most wounds. However, the business end of the drain should not be placed near the site of repair, for fear of damage to the anastomosis by the suction or by later withdrawal of the drain. Divided muscle layers should be repaired accurately and securely. In the neck the platysma should carefully be closed with interrupted sutures. Although with careful handling of tissues wound infection is rare, the exposure is often so much prolonged that prophylactic use of antibiotics is advisable. Such cover should always be used if there is any liability to ischaemia or if there is an associated fracture. We use antibiotic cover (cefuroxime 750 mg intravenously three times in 24 h starting just before operation) in all cases of major intervention
Extensor pollicis longus
Fig. 7.9 Radial side of the hand, with further warning about the terminal branches of the radial nerve and lateral cutaneous nerve of forearm (see also Fig. 1.29).
Extensor pollicis brevis
Dorsal carpal branch of radial a.
Abductor pollicis longus
Extensor carpi radialis longus
Extensor retinaculum
Lateral cutaneous n. of forearm
Superficial radial n.
Operating on Peripheral Nerves
on the brachial plexus, in all compound nerve injuries, in interventions on the proximal part of the sciatic nerve and in most second operations on main nerves. The record should carefully be maintained: it is best to follow a standard form and to supplement this with a diagram and photographs. Under no circumstances must descriptions of the lesion, of the state of the stumps after resection or of the gap after resection be omitted. The operation record should be written or dictated by the operating surgeon as soon as possible after completion of the procedure. The first copy is retained with the medical case notes, a copy is sent to the family practitioner, another to the referring clinician, and the final copy is stored with files of coded operating records (Figs. 7.10 and 7.11).
7.4 Methods of Repair There are remarkable similarities between the technique of repair of nerves and vessels. In arterial injury the first principle is rapid control of the vessels proximal and distal to the wound. Whilst the situation is less urgent in nerve repair, the nerve should be exposed first in healthy tissue above and below the level of lesion. Damaged vessel and nerve must be resected, the repair will fail unless healthy tissues are coapted. Undue tension guarantees failure. Adventitial tissue must be resected to expose the media and the epineurium (Fig. 7.12).
7.4.1 The Vascular Repair Of combined venous and arterial injury Barros d’Sa (1982) says: “ligation (of the vein) should be avoided at all costs. A vein tolerates lateral suturing much better than an artery. At least one major channel of satisfactory calibre must be restored so as to avoid a serious rise in peripheral venous resistance and pressure which reduces arterial flow to the limb and can lead to thrombosis at the site of the arterial repair with disastrous consequences. In combined venous and arterial injury, the vein should be repaired first”. Having secured proximal and distal control of the injured vessel any associated fracture or dislocation must be neutralised as rapidly as possible. A Rush nail, passed from proximal to distal provides rapid and adequate stability of the humerus before repair of the axillary or brachial artery. Direct suture is sometimes possible in the fresh stab wound; a vein patch is better than lateral suture at the mouth of a false aneurysm. Interrupted sutures are preferred to continuous suture to reduce the risk of stenosis at the suture line, ease co-aptation of vessels of different diameter, and facilitate end to side suture.
239
Division of the axillary and brachial sheath and of the deep fascia of the forearm is essential in all cases save those where a simple wound is successfully sutured within 3 h of injury. Subcutaneous fasciotomy of the forearm is adequate if there is no distal injury of elbow and forearm. The deep fascia of the forearm is exposed through a short incision on the medial side of, and parallel to, the biceps tendon. The skin is retracted and the plane between it and the deep fascia is gently developed by blunt dissection. The fascia is now incised and the plane deep to it opened out in a similar manner so that the fascia can be split safely using blunt tipped scissors. In more severe cases a fasciotomy should include the skin: this is usually susceptible to delayed primary closure at about 48 h. The indications for fasciotomy are more stringent in the lower limb. Barros d’Sa (1992) emphasised that decompression of all four compartments of the leg is essential in repair of shot gun wounds to the popliteal vessels. This principle should be followed in closed lesion complicated by fractures of the tibia or deep contusion of the muscles. We use two incisions. That over the fibula exposes the deep fascia enclosing the anterior and lateral compartments. The medial incision starts just above the midpoint between the medial malleolus and the Achilles tendon and extends to the upper leg. It is very important to identify and to open the fascia over the deep flexor compartment. Repair of Artery or Vein: Both proximal and distal vessels can be infused with heparinised saline but systemic anti-coagulation is not used. The proximal and distal stumps of the vessels are securely controlled by plastic slings and appropriate clamps. A skilful and patient assistant is charged with controlling the clamps and adjusting their position. Adventitia is removed after confirming back flow from the distal stump of the vessel. Repair is by interrupted sutures, 5/0 or 6/0 nylon for the subclavian, axillary, femoral or popliteal arteries, 7/0 or 8/0 for brachial, radial, ulnar and tibial arteries. The first two sutures are placed at the equator. It is often easier to repair the posterior wall first before turning the clamps to expose the anterior wall. The sutures pass through the media and the intima at intervals of about three quarters of a millimetre. The reversed vein graft. This is taken from an uninjured limb where possible. The long saphenous vein in the leg is best for larger arteries. The prepared stumps of the artery are gently drawn together and the gap between them measured, then the vein graft is prepared to match that gap. A very light touch must be used in handling the vein. Diathermy should not be used on the small branches. The proximal vessel is tied off leaving a long strand of suture to indicate that this must be placed distally during the repair. A flexible cannula, mounted on a syringe, is introduced into the distal stump of the vein and secured with another tie. The segment of the vein is distended with heparinised saline. This reduces spasm. End-to-side anastomosis is used when there is disproportion between the stumps of the
240 Fig. 7.10 An example of an operating record.
Surgical Disorders of the Peripheral Nerves
Operating on Peripheral Nerves
241
Fig. 7.11 A further example of operating record.
vessels (Fig. 7.13). Bleeding from the suture lines is best controlled by pressure for several minutes before inserting further sutures. The repaired vessel should be kept exposed for as short a time as possible and kept moist at all times. Nerve grafts must be elevated as swiftly as possible and much of this can be done whilst the arterial repair is underway. “Plasma” glue is invaluable in these cases, for its saves a great deal of time.
7.4.2 The Nerve Operations 7.4.2.1 Neurolysis A good deal of the argument about the value of “neurolysis” arises from imperfect definition of the term. We owe a great deal to Frykman and colleagues (1981) for their
clarification of this matter. They plainly showed the amount of fibrosis that follows so simple a procedure as interfascicular injection of saline; that which follows interfascicular neurolysis is certain to be more severe. That is not the only danger: their investigation was prompted by paralysis after a very careful interfascicular neurolysis of a radial nerve. External neurolysis is the freeing of the nerve from a constricting or distorting agent by dissection outside the epineurium. The first description of external neurolysis remains one of the best. Horsley (1899 ) operated on a cavalry officer who had dislocated his knee 8 months previously whilst putting his horse at a fence. The patient had returned to active service in spite of a persisting complete common peroneal palsy. There was extensive scarring in the popliteal fossa and the lateral aspect of the thigh. Horsley wrote: “I suggested that as a preliminary measure he could try massage, but that proved no good, and therefore I
242
Fig. 7.12 Rupture of axillary artery from fracture/dislocation in a 68 year old man. The atheromatous intima was fractured. A reversed vein graft was necessary after resection back to healthy intima.
Surgical Disorders of the Peripheral Nerves
determined to cut down upon it. I raised a flap of skin, and found the nerve was not only involved in a scar for about 7 in., but for about 3 in. of its course was greatly reduced in size. After dissecting it quite free from the scar I removed all the scar tissue I could and laid the flap of skin back again over the nerve, which by this time looked a great deal the worse for wear. To make a long story short, he has now recovered motion and sensation in the parts supplied by the external popliteal, after about an interval of a year or 13 months after the operation”. External neurolysis is especially valuable when the nerve is intact but tethered, strangled or immobilised by scar. Pain (neurostenalgia) is usual in such cases and relief of pain with improvement in function is regularly seen after external neurolysis and decompression. External Neurolysis after Repair or Amputation. Neurolysis is usually fruitless in a nerve which has been repaired. The decision to revise that repair is governed by failure to progress, persisting pain and a static Tinel sign (Fig. 7.14). However, it is not uncommon to find that the function regained after successful suture of the median or ulnar nerves at the wrist is marred by pain because the nerve has become adherent to the adjacent flexor tendons. Pain is worsened by movements of the digits and the neuroma can be seen moving up and down during flexion and extension. Liberating the nerve by incision of adhesions is often successful but a bed of healthy synovium must be restored. A
Fig. 7.13 Reversed vein grafting in the axilla: end-to-side (above) and end-to-end (below).
Operating on Peripheral Nerves
243
7.4.2.2 Biopsy
Fig. 7.14 Futile neurolysis. Poor recovery after primary suture of an ulnar nerve. The nerve was re-explored at 8 months. A CNAP was detectable and neurolysis was done. There was little recovery.
neuroma which has become adherent to the scar over an amputation stump is often extremely painful. Again, liberation of the nerve from scar, which may require cutting the nerve again so that the stump lies in healthy tissue usually relieves the pain. Epineurotomy: means the simple division of the epineurium in the line of the nerve and it probably has a limited place in cases of damage from injection of a noxious substance near a nerve or by localised pressure on a nerve, which has produced localised thickening and fibrosis of the epineurium. Unhappily, the fibrosis is unlikely to be limited to the perifascicular epineurium: there is likely to be fibrosis of the interfascicular epineurium too. The place of epineurotomy in radiation neuropathy is uncertain. The most that can at present be said for it is that when done carefully it is unlikely to do any harm.Internal neurolysis or inter fascicular neurolysis is the exposure of the bundles by epineurotomy and their separation by dissection between them or by the removal of interfascicular scar tissue. It is an essential part of three important procedures: (1) separation of intact from damaged fascicles in nerves which have suffered partial damage; (2) separation of a motor fascicle of a nerve for transfer; (3) separation of intact fascicles during removal of a benign but infiltrative tumour. Nagano and colleagues 1996 show that it may be required in at least one entrapment syndrome.The traction lesion in continuity is a common finding at exploration and it is very difficult. The nerve, which is usually in the axilla or at the knee, is exposed after a severe closed traction injury, and is found elongated by one third or more. The epineurial blood vessels are torn, but the perineurium and individual bundles appear intact. The damaged segment may exceed 25 cm. in length and resection and repair of such extensive lesions is scarcely feasible. Useful recovery occurs naturally in one third to one half of these injuries (Chapter 8). It seems that the damaged segment acts as an imperfect graft because the perineurium is in continuity and some, at least, of the Schwann cells survive.
Biopsy of a nerve requires removal of portions of conducting tissue and it necessitates at least epineurotomy and excision of one or more bundles so that the perineurium and its contents are made available for examination. Biopsy may extend to the entire nerve. The examination of teased single nerve fibres was reintroduced by JZ Young and PK Thomas and this method has greatly enhanced understanding of peripheral neuropathies (Dyck and Thomas 2005). Nerve biopsy is neither trivial nor without risk and it should never be a matter of unthinking routine. Dyck et al. (2005) provide a detailed account of biopsy of one fascicle from the sural nerve and the experience of one of them, who underwent the procedure, provides a salutary example of persisting pain and sensory disturbance after a meticulously performed biopsy. One of our patients, a 52 year old woman with chronic inflammatory peripheral neuropathy experienced increasing pain and a deepening sciatic neuropathy raising the possibility of plasmacytoma. One fascicle of the nerve was biopsied. On the following day she noted a virtually complete but painless, sciatic palsy. Haematoma was excluded by ultrasonography and by reexploration. The biopsy excluded plasmacytoma and it seems that an apparently relatively innocuous intervention in which a segment of one fascicle of the sciatic nerve was removed induced a dense ischaemic lesion of the whole trunk. It is in the investigation of nerve tumours that the greatest errors are seen. Case Report: A 43 year old woman presented with a 3 year history of intermittent abdominal symptoms; a mass was palpable. MRI showed a large tumour in the retroperitoneum, extending from L1 to the sacrum. A diagnosis of soft tissue sarcoma was made on the basis of a needle biopsy and the mass was excised, including a segment of the lumbo sacral plexus L4, L5 and S1. Chemo-and radiotherapy followed, complicated by a massive pulmonary embolism. We reviewed the histological material: the diagnosis was revised to a benign schwannoma. The patient now has profound weakness of the abductors and extensors of the hip, paralysis of the dorsi flexors of ankle and foot, weakness of quadriceps and other muscles of the lower limb. She is able to walk 50 yards with two sticks. Knight and her colleagues (2007) described the complications attending biopsy, usually by fine needle, in 53 patients with benign solitary schwannoma. 1. Eight biopsies failed to yield diagnostic material. 2. Ten biopsies removed portions of normal nerve. 3. An erroneous diagnosis of a soft tissue sarcoma was made in two core needle biopsies. 4. Most patients experienced significant or severe pain as a result of the biopsy and there was significant loss of function in 22 of them. 5. Fibrosis induced by the biopsy distorted the tissue planes of the nerve and added greatly to the difficulties of later enucleation of this benign tumour.
244
Surgical Disorders of the Peripheral Nerves
leaving the nerve well alone, closing the wound and referring the patient to an interested colleague. We emphasise the importance of frozen section biopsies during resection of MPNST which are essential in proving an adequate margin of resection. Santos-Pallazzi (1991) of Barcelona, has, for many years, examined specimens from the stumps of ruptured spinal nerves by frozen section which has enabled him to classify the potential for regeneration by the amount of fibrosis. Gschmeissner et al. (1991) offered a 2 min assessment of the quality of the stumps by examination at frozen sections.
7.5 The Nerve Repair Fig. 7.15 Biopsy of a benign lesion of the femoral nerve caused severe pain and complete paralysis of the extensor muscles of the knee. No action was taken for 10 months. The first surgeon did not examine the specimen. Histological findings: fascicles of a main nerve.
It is important for the surgeon who comes to treat such lesions to ask to see the sections taken from the material removed at the primary operation. Often enough, a cross section of a trunk nerve is included in the specimen (Fig. 7.15). Geisinger and Abdul-Karim (2001) emphasise the limitations of fine needle aspiration biopsy of soft tissue tumours. The sample may not be representative of the tumour and it may prove insufficient to make a diagnosis. There is no justification for biopsy in tumours of nerves which are clearly benign. The clinical features supported by the findings from MR and US scanning enable the accurate diagnosis of nearly all cases of schwannoma and intraneural ganglion. An incorrect diagnosis of a benign lesion in malignant peripheral nerve sheath tumour (MPNST) may cost the patient his or her life because of dissemination of tumour from the nerve into adjacent soft tissues or because of the continuing extension of the tumour during the months before the correct diagnosis becomes all too clear. We have seen ten such cases and in five of these the initial error, compounded by delay, lost the chance of adequate surgical excision. Whilst biopsy is essential in cases in which there is any doubt about diagnosis and in particular in those where there is a possibility of primitive neurectodermal tumour, neuro-epithelioma or extra osseous Ewing’s tumour, we think that such biopsy is best performed within the Institution where definitive treatment will take place. The possibility that the tumour is catecholamine secreting should be considered in adrenal and extra-adrenal retroperitoneal lesions (Sood et al. 2007). It is for the responsible surgeon to decide whether biopsy is needed and which method is used. It is the responsibility of that surgeon to examine the biopsy tissue with an experienced pathologist. Surgeons do not always get things right, neither do radiologists or pathologists. We believe that a surgeon encountering an unexpected tumour within a nerve, one which is not evidently a benign schwannoma will do no harm in
The object of nerve repair is the accurate coaptation of healthy conducting elements without tension. In practical terms this means accurate coaptation of the bundles (Birch and Raji 1991). This is done after skeletal injuries have been stabilised and vessels repaired, and after muscles, tendons, joint capsule and synovium have been drawn together so restoring gliding planes. Cover by healthy full thickness skin is essential. In delayed repair the incision is planned so that a skin flap is raised over the nerve (Fig. 7.16). Once the decision is made for or by the surgeon, the nerve ends are cut back progressively until healthy pouting bundles show in the cut surfaces (Fig. 7.17). Resection is less in urgent repairs. No more than 1–2 mm of nerve is removed in tidy wounds. Finding the right level of section in the closed traction rupture or in the untidy wound is much easier in urgent repairs where there is still conduction in the distal trunk. In ruptures of the spinal nerves, the stimulator is moved slowly from the rupture towards healthy tissue until an SSEP becomes detectable (proximal stump), or, until muscular activity returns (distal stump). This usually coincides with a healthy looking nerve face. In traction ruptures, the amount of tissue resected usually lies between 5 and 10 mm. More nerve must be resected in late
Fig. 7.16 A tidy wound. Primary repair of all divided structures is indicated.
Operating on Peripheral Nerves
245
2. It is impracticable to bridge with grafts gaps in lesions of the whole sciatic nerve. Sufficient graft material is not available. The gap has to be closed by flexion of the knee and extension of the hip and later maintenance of that position for the appropriate time. 3. Anterior transposition of ulnar or radial nerves gives at most 3 cm. 4. No gaps in the median nerve in the forearm can be closed by end to end suture. Ideally, it is best to match bundle to bundle, sensory fibres to sensory fibres and motor fibres to motor fibres. It is often easy to match bundle to bundle simply by looking at the nerve end under the microscope or with the loupe, though the changing architecture of the nerve along its length makes this difficult when a long gap has to be bridged. When operation is done soon after injury, it is easy, as Vandeput and colleagues (1969) suggested, to determine the sites of motor fascicles in the distal stump and to effect an “electrophysiological orientation”: Certainly Kato and his colleagues (1998) improved anatomical matching in repair of low median and ulnar nerve lesions by displaying individual bundles and orientating them by electrical stimulation. However, the late distinction between motor and sensory fibres requires more advanced methods. The possibilities of histochemical identification of nerve fibres are reviewed by Brushart (1999). Fig. 7.17 Good stumps after resection in traction injury of the radial nerve. Below: The rupture displayed: Above: resection is done back until clearly separated “pouting” bundles are visible.
cases. When the case has been complicated by infection as much as 4 cm of proximal and distal stumps are irretrievably fibrosed. Palpation of the nerve detects the difference between soft, healthy tissue from the firm or hard, scarred segment. Then the ends must be united, preferably by end to end suture. So long as the gap after resection is small, little mobilisation of the nerve is needed to close it, and the repaired nerve lies without tension, without excessive flexion of adjacent joints. One simple test as to the advisability of direct suture of a nerve trunk at the wrist or in the forearm involves passing an epineurial suture of 7.0 nylon, with the wrist flexed to no more than 30°. If this suture will draw the stumps together without tearing the epineurium and without causing blanching of the epineurial vessels, then suture is reasonable. Failing that, grafting is necessary. It will be seen that in many circumstances it is better to bridge a gap with a graft than to force direct suture; it is better to resect to healthy bundles and create a wide gap than to resect too little in order to facilitate direct suture. We take as guides the following principles: 1. End to end suture of the nerves of the brachial plexus above the clavicle or of the accessory nerve is rarely practicable.
7.5.1 Methods of Suture Orgell (1987) described a modified fascicular suture: “group fascicular suture” and he concluded that since there was little difference between the results of epineurial and perineurial suture, epineural suture was “the technique of choice for most acute nerve lacerations.” He pointed out that it was easier and faster and entailed less manipulation of the internal structure of the nerve than did fascicular suture. Spinner (2008) thinks that “fascicular” suture is useful in distal median and ulnar repairs and he emphasises that the most important cause for failure of suture is “inadequate resection of injured nerve back to healthy tissue”. During urgent or emergency operation the surgeon should indeed proceed to a neat epineurial suture using 6/0 or 7/0 sutures or to immediate grafting if circumstances permit. We have seen many excellent results following this approach in severe cases of injury or when the nerve has been inadvertently divided during an operation. Accurate matching is assisted by the orientation of epineurial vessels, and by making a sketch of the prepared faces indicating the size and the disposition of different bundles. We prefer bundle (fascicular) suture, combined with epineurial suture for most main nerves, with the exception of the sciatic. In the early days after division of a nerve bundles
246
Surgical Disorders of the Peripheral Nerves
are mobile within the epineurium, and epineurial suture increases the risk of malalignment. Dissection within the nerve is avoided as this surely leads to fibrosis. The needle is passed through the condensed inner epineurium and the perineurium to secure accurate coaptation of larger bundles. The repair is completed by epineurial suture. In delayed repair, fibrosis within the epineurium stabilises the bundles so that they cannot rotate within the epineurium. In these, epineurial suture may be adequate. The atrophy of the distal stump and the
extent of fibrosis in both proximal and distal stumps increases the difficulties in neglected cases (Figs. 7.18 and 7.19). In both primary and secondary suture, the areolar adventitial tissue is pushed back from each stump to expose the true epineurium. In fascicular suture, matched bundles, identified by size and by position in the nerve, are united by perineurial sutures of 9/0 or 10/0 nylon. Once these “key” bundles have been united, the union is completed by passing sutures of 8/0 or 9/0 nylon through perineurium and epineurium The nerve
Fig. 7.18 Primary suture of median nerve at the wrist. Top right: the pattern of the bundles is examined. Top left: The first sutures pass through the perineurium of the larger bundles then follows epi-
perineurial suture. Middle: the nerve is rolled on a swab for access to the posterior aspect. The completed repair is seen below.
Operating on Peripheral Nerves
Fig. 7.19 The opportunity for primary suture was lost in this transection of median nerve displayed 3 months after wounding.
can be rotated on a saline-soaked dental swab, first from one side and then from the other, so that the whole circumference is accessible. Between 18 and 25 sutures are used to repair the adult median nerve at wrist level. In epineurial repair, orientation of bundles is achieved as well as is possible, and the epineurium is united with two lateral sutures of 8/0 nylon, the ends of which are left long. The repair of the anterior face is completed with sutures of 8/0 or 9/0 nylon, and the nerve is then rotated by manipulation of the lateral sutures so that the posterior epineurium can be united. If plasma glue is available this reduces the number of sutures required. The glue is applied after suture. Heavier epineurial sutures (4/0, 5/0 or 6/0) are used for suture of the sciatic nerve.
7.5.2 Grafting Wherever possible we use cutaneous nerves from the damaged limb and prefer the medial cutaneous nerve of forearm (MCNF) if it is available. The superficial radial nerve (SRN) and the lateral cutaneous nerve of forearm (LCNF) provide valuable innervation to the skin of the hand and neither should be used unless the parent nerve is irretrievably damaged. The ipsi lateral sural nerve should never be used for the repair of a low lesion of the tibial nerve for this adds to the denervation of the skin of the heel. These patients are better off losing one of the medial cutaneous nerves of forearm. No graft should be elevated until the injured nerves have been exposed and the extent of lesion defined and the gap between the prepared faces measured. This is a good time for the surgeon to pause for a few minutes of reflection, so that the repair can be properly planned. This is particularly important when repairs of several main nerves are necessary. A map of the proximal and distal faces is prepared outlining the pattern of the bundles in each stump and measuring the length of the gap between the prepared stumps. Then, a
247
calculation is made of the number of grafts required for each nerve, and the disposition and length of the grafts which are cut to about 15% more than the measured gap. Elevation and preparation of the graft. The MCNF is taken from the arm through a straight incision. One anterior branch will be seen in the middle of the arm, lower down the nerve divides into two branches which straddle the main brachial vein. The SRN is delivered through separate incisions. The nerve is identified where it emerges from deep to the brachioradialis. The terminal branches are divided at the wrist. The radial nerve is displayed between brachio-radialis and brachialis, and the superficial branch identified by gentle traction. It is delivered into the elbow wound, a manoeuvre which has the advantage of stripping it of much of its adventitia. Up to 30 cm of nerve are available in the adult. The LCNF is found just lateral to the biceps tendon and it can be displayed, in the lower part of the arm, between the biceps and brachialis muscles. About 15 cm of graft is available. Reissis and his colleagues (1992) showed that the terminal 4 cm of the posterior interosseous nerve provides useful material for grafting palmar digital nerves (Fig. 7.20). The sural nerve(s) is almost always required in repair of lesions of the brachial plexus in the adult and in cases where more than one main nerve must be repaired. The patient is prone for repair of the sciatic nerve and its divisions, otherwise they are placed supine. The lower limb(s) is elevated by a stoical assistant or the knee is flexed to about 70° with the foot resting on the table. The nerve is exposed through a long midline incision which moves laterally in the distal one third of the leg to a point midway between the posterior aspect of the lateral malleolus and the lateral margin of the Achilles tendon. The incision may be extended into the popliteal fossa as a Z. Up to 50 cm of graft is available. The grafts, tenderly handled, are laid between saline soaked swabs and cut to appropriate length with a new blade or with vascular scissors. The nerve ends are prepared so that the fascicles protrude, and are laid in the prepared bed which must be healthy and unscarred. Fixation is usually by fibrin clot glue but the grafts can be sewn into place, two lateral sutures of 9/0 nylon being used for each. The sutures unite the fascicles of the stumps with the grafts. Because the individual strands of the grafts are likely to be around the same size as the individual bundles, the suture unites epineurium of graft to perineurium of bundle (Fig. 7.21). Millesi (1981) commences his dissection of the “fascicle groups” in the healthy nerve above and below the lesion, and makes the section at the point at which each group loses its healthy appearance. In the distal stump, for instance in a lesion of the median nerve at the wrist, the fascicle groups are traced back from the point of identification of their destination to the level just distal to the lesion. Fascicular patterns of distal and proximal stumps are matched, and individual grafts are used to unite the fascicular groups. Millesi resects fascicular groups at different levels, then uses 10/0 nylon through interfascicular
248
Surgical Disorders of the Peripheral Nerves
Fig. 7.20 Nerve grafts. Above left: The superficial radial nerve (1) and lateral cutaneous nerves of forearm (2) are displayed at the elbow. Above right: the radial nerve is drawn into this wound after section of the distal branches at the wrist. Below: the communicating branches of the sural nerve are variable. This nerve is best elevated through a long
incision. Below left: the sural nerve divides into its terminal branches close to the short saphenous vein (3) above the ankle. Below right: note the communicating branch from the common peroneal nerve in the upper leg (4).
epineurium or perineurium for coaptation. The result is to produce grafts interdigitating with the fascicle of the stumps. We prefer to graft the proximal stump of the nerve first, leaving the distal ends of the grafts laid out across the bed before proceeding to the distal stump, matching as far as is possible, fascicle to fascicle. It is very important, once both ends have been anastomosed, carefully to inspect the grafts and the suture lines to make certain that during the union of one end the union of the other end has not been disturbed. It is better to bridge the gap with grafts if intact fascicles have been separated from the lesion. All nerve repairs must be secluded from naked bone, tendon, or lacerated muscle by healthy tissue. Grafts are stabilised by closing healthy synovium or fat, or muscle, over them. The fat pad is a valuable shield in the posterior triangle of the neck and it should be carefully apposed over the repair. It is necessary to restate that repairs of nerves and vessels must be covered by healthy, full thickness skin. Split skin grafts induce severe fibrosis which strangles. Vascularised grafts: The evolution application results and limitations of the free vascularised ulnar nerve graft in repair of the brachial plexus is described in Chapter 9 (Fig. 7.22).
The method was a development of the ingenious operations described by Strange (1947) and MacCarty (1951). In Strange’s operation the ulnar nerve is used to repair the median nerve in cases where both are otherwise irreparable and in MacCarty’s technique the common peroneal nerve is used to bridge long gaps in the tibial nerve in otherwise hopeless injuries of the sciatic nerve. These techniques provide a full calibre graft which is not only vascularised but also predegenerate and remain valuable in the most severe injuries. The steps required are described using the sciatic nerve injured in the mid or lower part of the thigh as an example. 1. At the first operation the proximal and distal stumps are identified and the gap between them measured. 2. The proximal stumps of the tibial and common peroneal nerves are prepared and then sutured together. 3. The proximal segment of the common peroneal nerve is traced and then the bundles within the trunk are cut across at a distance from the suture line equivalent to the gap already measured. The blood vessels in the mesoneurium and in the epineurium are preserved. In effect a vascularised graft has been prepared.
Operating on Peripheral Nerves
249
Fig. 7.22 Free vascularised ulnar nerve graft. Above: the nerve is shown based on superior ulnar recurrent vessels (1). Below: shows the condition of a VUNG displayed close to the clavicle during an operation for non union 28 months after grafting.
Fig. 7.21 In grafting the suture unites epineurium of graft to perineurium of the bundle.
4. At the second operation, which is performed no sooner than 3 weeks after the first, the proximal segment of the common peroneal nerve is sectioned and brought down to the distal stump of the tibial nerve to which it is sutured. Bleeding, which is at times quite copious, is seen at the face of the common peroneal graft when it is apposed to the distal tibial nerve. Case Report: An 18 year old woman sustained open fracture of mid shaft of femur in a road traffic accident. Most of the skin of the lower part of the thigh was avulsed and there was much destruction of muscles. Extensive skin grafting proved necessary. The tibial nerve was repaired by the pedicle technique: a
24 cm long segment of the proximal common peroneal nerve was prepared and it proved a simple matter to separate this from the tibial division as far as the level of the neck of the femur. The proximal stumps of the tibial and of the common peroneal nerves were prepared and sutured. The bundles within the common peroneal nerve were sectioned 25 cm proximal to this suture line. Four weeks later the common peroneal graft was mobilised and sutured to the distal stump of the tibial nerve in the upper part of the leg. She regained flexion of the heel and of the toes to power MRC grade 4, warm and cool sensation in the sole of the foot and could accurately localise to the plantar skin without over reaction. There was recovery of sweating.
7.5.3 Indications For and Methods of Nerve Transfer Nerve transfers work best when a healthy nerve, or a portion of a healthy nerve, is transferred to a nerve to muscle of roughly equivalent size without any interposed nerve graft (Narakas 1987). They cannot be asked to do too much; a rivulet cannot feed the Nile. Addas and Midha (2009) provide a valuable review of this field. They say “nerve transfers tend
250
to take the surgeon away from exploring the injury site, the brachial plexus, which carries the potential for surgeons to not even offer an anatomic nerve reconstruction, even in situations when these are perfectly appropriate…. With the increasing use of transfers, newly trained peripheral nerve surgeons are less likely to have exposure to the brachial plexus and they will be increasingly unfamiliar with the detailed anatomy and intraoperative electrophysiology assessment of the lesion”. These are important warnings. We offer two more. The first is that no nerve of vital function should be used for the sake of a non vital function. A birth lesion of the phrenic nerve is a life threatening complication. Some adults with severe injuries to the brachial plexus experienced breathing difficulties because of the associated lesion of the phrenic nerve. The second is that reinnervation of muscle, to be successful, must involve reinnervation of the deep afferent pathway from the muscle spindles and the Golgi tendon organs. This fact may account for the disappointing results following transfer of the hypoglossal nerve because afferent pathway from the tongue passes through the lingual nerve. Some of the more common methods are now described.
7.5.3.1 The Intercostal Nerves The patient is placed in a semi-sedentary position with preparation including the forequarter and the chest wall from the mid line to the iliac crest. An incision is made below the fold of pectoralis major extending to the axilla as a “Z”. The serratus anterior is exposed, and the neurovascular pedicle to that muscle identified and protected. The lateral perforating branches of the intercostal nerves are identified. The upper four digitations of the serratus anterior are released from the ribs, leaving a sufficient cuff of muscle for later repair. The muscle is reflected so opening the plane between the muscle and the rib cage.. The cutaneous branch of the intercostal nerve is traced back to a narrow foramen in the external intercostal muscle. The muscle is released from the rib above as far as the posterior angle. The deep division of the intercostal nerve is found by dividing the attachment of the middle intercostal muscle from the rib above. Gentle traction with a phrenic hook, underneath the upper rib, reveals the deep branch which is divided anteriorly. A Gelpi knee retractor can now be inserted to spread the rib cage. The deep division is traced back to its junction with the perforating branch. Elevation of the nerve becomes progressively easier. Bi-polar diathermy must be used throughout (Fig. 7.23). The intercostal nerves from T3 to T6 are available by this method. These are brought to the anterior surface of serratus anterior through a tunnel in the upper part of the muscle. The serratus anterior is repaired.
Surgical Disorders of the Peripheral Nerves
Fig. 7.23 Nerve transfer. Upper left intercostal nerves T2, T3, T4 (1) raised and united to the lateral cord of the plexus (2).
If only one or two deep divisions of the intercostal nerves are being used for the nerve to serratus anterior, then a suitably placed transverse incision in that muscle is adequate. It is wise always to see a radiograph of the chest before the patient leaves the operating theatre. The Spinal Accessory Nerve: This is a powerful motor and it should be used with discrimination. THE INNERVATION OF THE UPPER ONE THIRD OF TRAPEZIUS MUST NOT BE COMPROMISED. Even when the nerve is divided deep to the clavicle, the loss of function is less but it is not insignificant. The nerve is found at the lateral end of the transverse supraclavicular incision in the plane between the fat pad and the deep face of trapezius. The nerve passes down it in a rather sinuous fashion and is accompanied by a longitudinal artery and vein. These can cause troublesome bleeding. The nerve is joined by a branch from the cervical plexus just above the level of the clavicle and intra-operative stimulation here only occasionally evokes a muscle twitch. The nerve is sectioned distal to that junction (Fig. 7.24). Ulnar to Biceps Transfer (Oberlin et al. 1994) is regularly effective in cases of avulsion of C5 and of C6. With care, it can be extended to cases where C5, C6, C7 or even C5, C6, C7 and C8 have been avulsed. Fastidious dissection of the bundles within the ulnar nerve is necessary. The nerves are exposed through an incision along the length of the brachial bundle. The musculocutaneous nerve is identified. It is not uncommon to find that the nerves to biceps and brachialis arise directly from the median nerve. The nerve to biceps is accompanied by a sizeable artery and vein. It is traced proximally and separated from the main nerve. It is sectioned here so that it drops down onto the ulnar nerve. The epineurium of the ulnar nerve is incised and the bundles exposed. Nerve stimulation used at very low intensity leads the surgeon to a bundle in the antero-lateral quadrant of the nerve passing to the flexor muscles of the forearm but not to the muscles of the hand. This is divided and an end-to-end suture is done.
Operating on Peripheral Nerves
251
This valuable principle has been extended widely, using bundles from the median or the ulnar nerve to nerves to triceps, to the nerve to extensor carpi radialis brevis or to reinnervate free functioning muscle grafts and it has proved particularly
valuable in the repair of injuries to the brachial plexus in which some spinal nerves are intact whilst others are avulsed. In these a bundle from an intact nerve may be used to reinnervate the suprascapular nerve or an avulsed ventral root (Figs. 7.25–7.28).
Fig. 7.24 Accessory (1) to suprascapular (2) transfer. Above, seen from the head: the spinal accessory nerve is displayed on the inner face of the trapezius at the base of the posterior triangle, Middle and below, the view from the shoulder: the suprascapular nerve is passed deep to the fat pad and united to the spinal accessory nerve.
Fig. 7.25 Avulsion of C5 and of C6 with dorsal scapular palsy. The suprascapular nerve was reinnervated by one bundle from C7. The suprascapular nerve (1) posterior division of upper trunk (2) anterior division of upper trunk (3) C7 (4) and the selected bundle from C7 (5) are displayed.
252
Surgical Disorders of the Peripheral Nerves
Fig. 7.26 Reinnervation of the avulsed ventral roots of C6 and C7 from the proximal stump of C5. The ventral root of C6 is apposed to the stump of C5. Grafts were passed to the ventral root of C7 and to suprascapular nerve. C5 (1), C6 (2), C7 (3), the ventral root of C6 (4) and the grafts (5) are seen.
7.5.3.2 Direct Muscular Neurotisation This technique is useful in lesions of the circumflex or musculocutaneous nerves in which the nerve has been avulsed directly from the muscle or, in the case of circumflex, the distal stump of the nerve is wrecked by fibrosis. The proximal stump is prepared in the usual way and two lengths of graft, usually the MCNF are united to the proximal stump and then implanted into the muscle through short incisions through its sheath. The grafts are passed subcutaneously, and the portal of entry is exposed by a short incision of the skin at those points. About ten portals of entry are fashioned using the terminal branches of the graft or by subdivision of the distal stump. The entry points of the implanted nerve are sealed with fibrin clot glue (Fig. 7.29). Freeze thawed muscle graft (FTMG). This is useful in the repair of cutaneous nerves or their branches which have formed a painful neuroma. The stumps of the nerve are prepared and the gap measured. A segment of adjacent muscle, of the equivalent calibre, is excised. This should
Fig. 7.27 Reinnervation of part of the circumflex nerve (1) by a nerve to long head of triceps (2) (Operation performed by Dr Andrew Yam RNOH).
be at least two and a half times the length of the measured defect in the nerve. The muscle is then enveloped by a piece of aluminium foil and immersed in liquid nitrogen for about 60 s. The packet is then placed in distilled water for a couple of minutes. The prepared muscle graft can then be trimmed to appropriate length and breadth and it may be secured to the nerve stumps by suture and then sealed by plasma clot glue. Fragmentation is usual with muscle grafts more than 3 cm long.
7.5.4 Immobilisation He or she labours in vain who repairs a nerve and leaves the junction or junctions open to the hazards of movement of fragment of a bone or of movement of a related joint. Although the antique methods of immobilisation and gradual
Operating on Peripheral Nerves
Fig. 7.28 Transfer of a bundle from the ulnar nerve (2) to the nerve to biceps (1) (Oberlin’s operation).
extension by “turnbuckle” or hinged plaster after its cessation are now rarely necessary, repaired nerves still need protection during at least the first 3 weeks after repair. In most cases, a simple plaster slab gives sufficient protection. Only in the case of extensive proximal lesions of the whole sciatic nerve is elaborate protection required. Narrowing of the gap to permit end to end suture or adequate repair by graft requires flexion of the knee and avoidance of flexion of the hip. The necessary position of immobilisation in a hip spica is awkward and uncomfortable; further, gradual extension of the knee after 3 weeks may be necessary in order to protect the line of the repair. Usually, it is sufficient when the splint is removed to warn the patient about the danger of excessive movement of related joints, and to rely on the patient steadily to restore movement of those joints. There is particular difficulty in protecting repairs of the brachial plexus, an even greater problem when it becomes necessary to protect roots re-united to the spinal cord. In these cases, Fournier (2005) applies a Minerva plaster of Paris jacket. At present we rely on a sling for the upper limb of the adult: the best in our
253
Fig. 7.29 Muscular “neurotisation”. Above: The nerve graft is prepared. Below: It is passed from the proximal stump of the circumflex nerve and implanted into the muscle.
experience is that designed by Evelyn Hunter RGN. Alternatively, slings with straps which secure the arm across the body, can be used (Fig. 7.30). We supplement the sling with a soft cervical collar after repairs of the supraclavicular brachial plexus. A plaster of Paris “cocoon” is applied after repair of the plexus in infants. The method used in the post operative care of severe wounds at the wrist is as follows (Fig. 7.31). The splint holds the elbow at 90 degrees of flexion, the wrist at between 30 and 40 degrees of flexion, the metacarpophalangeal (MCP) joints at about 70 degrees of flexion, and the proximal interphalangeal (PIP) joints at no more than 30 degrees of flexion. The dorsal splint extends to the tips of the fingers and the palmar splint to the PIP joints only. The splints are bandaged so that there is restriction but not rigid immobilisation. Gentle, active flexion of the fingers and the thumb is encouraged from the outset. The arm is supported in a sling, but there should be encouragement of gentle active lateral rotation and of elevation of the shoulder to 90° from the first
254
Surgical Disorders of the Peripheral Nerves
movements of the joints can be done by the patient or by his or her parents, but weekly supervision by a physiotherapist is useful in keeping the mind concentrated on the work during the long march towards recovery. In the case of the metacarpophalangeal joints of the fingers and the carpo-metacarpal joint of the thumb, “lively” splints are useful, but the problem of the stiff metacarpo-phalangeal joint in the paralysed hand persists. Accurate diagnosis is the foundation of rehabilitation and operations for repair of the nerves should be seen as but the first stage in that process. It should be superfluous to add that the sooner the diagnosis is established the better, and that urgent efforts must be made to improve the prognosis.
7.6 Approaches to Individual Nerves The Neck: cervical and brachial plexus; the subclavian and axillary vessels; the spinal nerves; spinal accessory nerve; suprascapular nerve.
7.6.1 The Transverse Supraclavicular Approach: (Anterior, or Anterolateral) Fig. 7.30 Evelyn Hunter’s sling for control of the fore quarter after repair of the brachial plexus. Note the check on lateral rotation at the gleno-humeral joint.
postoperative day. At 3 weeks the splints and the sutures are removed. The next splint does not restrict the elbow. The wrist is splinted to prevent extension beyond 20°. The dorsal hood, which again extends to the tips of the fingers, blocks the MCP joints to 30 degrees of flexion and the PIP joints to 30 degrees of flexion. Increasingly, vigorous active flexion of the fingers and the thumb is now permitted and gentle active flexion at the wrist is also encouraged within the confines of the splint, bandaged as it is to the forearm and hand. In direct sutures of nerves in the elbow region a hinged splint is applied at 3 weeks from operation. This permits active flexion but blocks extension. The range of permitted movement is increased at weekly intervals by adjusting the hinge. By 6 weeks, splints are discarded and vigorous active flexion work against resistance is introduced. Now, gentle passive stretching work can be introduced for the fingers and the thumb. Early gentle active movements within the confines of the splint and the bandages are encouraged. It is of course very important during the period of recovery to maintain the mobility of the joints some or all of whose governing muscles are paralysed and to warn the patients of the danger, especially in cold weather, of accidental damage to the anaesthetic skin. Most of the work on passive
This is used for exposure of the supraclavicular part of the brachial plexus. Its disadvantage is that the vertebral artery stands between the operator and the most proximal parts of the nerves. The risk of skin necrosis is negligible. The scar is reasonable. The length and the level of the incision are modified according to the lesion. In the urgent case the incision extends from beyond the mid line to beyond the anterior fold of the trapezius and is placed just above the clavicle. For lesions of the upper trunk of the brachial plexus the incision is shorter and is placed about two fingers breadth above the clavicle. The exposure of Fiolle and Delmas (1921) is achieved by adding a vertical limb to the transverse supraclavicular wound. Bonney’s transclavicular approach (Birch et al. 1990) is, in effect, a medial extension of the transverse supraclavicular approach and it gives excellent access to the first part of the subclavian artery, the first part of the vertebral artery and the whole of the brachial plexus. The Operation. The patient lies supine in a semi-sedentary position, with the head elevated to about 30° and the neck in extension. After placing recording electrodes to the scalp and the skin of the neck, the head is bandaged to a neurosurgical rest. The neck is extended but not rotated. The area of skin preparation includes the whole of the forequarter, extending to the jaw line and the ear, beyond the mid line to the inferior rib margin. The skin flap is developed deep to the platysma (Fig. 7.32). This is easier to identify at the posterior margin of the
Operating on Peripheral Nerves
255
Fig. 7.31 Protection after repair. Above: Plaster after repair of both nerves and flexor tendons at the wrist. Below: Hinged plaster with adjustable check.
sternocleidomastoid muscle. Below, the flap is developed to reveal the inferior margin of the clavicle; above as far as the greater auricular and transverse cervical nerves crossing the anterior face of sternocleidomastoid (SCM). The incision is deepened in the plane between external jugular vein and the SCM, displacing the supraclavicular nerves posteriorly. The lateral part of the insertion of the muscle can be elevated from the clavicle. The fat pad and the omohyoid muscle are
now seen. The omohyoid is divided between stay sutures and reflected. The fat pad may present as two leaves, split by branches of the transverse cervical vessels. These must be preserved in cases of rupture of the subclavian artery because of their contribution to the collateral circulation (Fig. 7.33). The scalenus anterior is exposed and the phrenic nerve is seen coursing obliquely across it. It is mobilised and held in nerve sling. Access to bony structures behind the plexus
256
Fig. 7.32 The transverse supraclavicular approach to the brachial plexus. Note the position of the patient and the line of incision.
should be achieved by separating the component nerves rather than by forward retraction of the plexus. Following the phrenic nerve cranially brings the surgeon to the fifth cervical nerve and it is sometimes helpful to work on the medial side of the phrenic nerve to expose a ruptured stump at C5 or C6. The upper trunk and the suprascapular nerve are traced. After traction injury it may be easier to identify the suprascapular nerve first and follow it back to the upper trunk. The seventh cervical nerve can now be seen behind the upper trunk. Division of scalenus anterior reveals the subclavian artery; the plane between the muscle and the subclavian artery requires careful definition. We have seen the subclavian artery crossing anterior to scalenus anterior in three cases and through it in seven more. The eighth cervical and first thoracic nerves are traced by following the plane between the artery and the lower trunk. The nerve to serratus anterior lies lateral and deep to the upper trunk, behind the suprascapular nerve and posterior to scalenus medius. The rami forming this nerve are displayed by dividing scalenus medius. There are usually three, usually that from C6 is the largest. The spinal accessory nerve, is identified running in a vertical
Surgical Disorders of the Peripheral Nerves
and sinuous manner on the deep face of trapezius. The dorsal scapular nerve can be found passing posteriorly away from C5. Kline (2008b) provides a good demonstration of the course and relations of this nerve (Fig. 7.34). Extension by the operation of Fiolle and Delmas. This affords full display of the supra, retro and infra clavicular plexus, of the second part of the subclavian to the terminal axillary artery and of the subclavian and axillary veins deep to the clavicle. The originators said “the exposure of the vessels can be achieved in three minutes” and this is possible in an emergency. There is no doubt that the operation secures rapid access to and control of major vessels from the scalenus anterior to the lowest part of the axilla. The plexus itself is exposed from the spinal nerves above to the terminal branches below. It is especially valuable in fresh cases of laceration or rupture of the great vessels deep to or below the clavicle and in later cases of false aneurysm, and is often necessary for the delayed exposure of ruptured nerves after primary vascular repair. It is the exposure of choice in the closed infraclavicular rupture of vessels and nerves. The incision is T-shaped, the vertical limb running in the deltopectoral groove, retracting the cephalic vein medially, then curving into the apex of the axilla underneath the fold of pectoralis major. In urgent cases a Gigli saw is used to divide the clavicle at the lateral edge of SCM: otherwise, a hole is drilled and prepared for a compression screw, inclined at 45° to the long axis of the clavicle. Subsequent fixation is made easier by contouring a plate to the bone and drilling and preparing one hole in the lateral fragment. The clavicle is cut at 90° to the axis of the hole for the compression screw. It is possible, in some cases, to avoid cutting the clavicle. The bone can be drawn upwards or downwards by a nylon tape (Figs. 7.35 and 7.36). The fascia at the root of the axilla is divided and the finger develops a plane between pectoralis major and minor in front, and the axillary bundle behind. Pectoralis major is detached from the humerus. Pectoralis minor is reflected taking care for the musculocutaneous nerve. The limb is rolled into lateral rotation so that the trunks, divisions and cords of the brachial plexus are displayed with their accompanying vessels. The exposure has been used in more than 250 cases. We know of nine cases of non union of the clavicle; infection complicated three of these.
7.6.2 The Transclavicular Exposure Bonney developed this operation in 1976 as an approach to the anterior aspect of the cervico-dorsal spine for excision of malignant cartilaginous tumour which had eroded the bodies of C7 and D1 and compressed the spinal cord. The approach was designed to give adequate exposure and control of the
Operating on Peripheral Nerves
257
Fig. 7.33 Above left: The posterior margin of sternocleidomastoid is defined, anterior to the external jugular vein and supraclavicular nerves (1). Above right: This is retracted to display the fat pad and transverse cervical vessels (2). Below left: The omohyoid is divided between stay
sutures. Below right: the nerve to serratus anterior (3) is displayed by section of scalenus medius deep to the upper trunk (5). The largest ramus comes from C6. There is often a communicating branch from the dorsal scapular nerve (4).
venous trifurcation, the first part of subclavian artery, the vertebral artery and the recurrent laryngeal nerve. Duval (1921) faced with the problems of deep wounds in the neck outlined an approach in which the sternoclavicular joint was turned downwards. He described it as: “this major and hazardous intervention which should be undertaken only by the most consummate surgeon”. Antibiotics, blood transfusion and advances in anaesthetics have greatly diminished some of these hazards but the intervention remains an extensive and difficult one calling for a high degree of anatomical knowledge and practical versatility. The exposure rests on the elevation of the osseo-muscular flap comprising the medial portion of clavicle with the sterno clavicular joint based on the sternocleidomastoid muscle (SCM). The whole of the brachial plexus can be displayed after some lateral work; the cervico dorsal spine can be seen from C3 to T3 by developing the plane between the carotid sheath and the visceral axis (Fig. 7.37).
The transverse limb of the incision runs from the fold of the trapezius on the side of operation to the mid point of the opposite posterior triangle. This limb can be extended to increase the exposure. The vertical limb runs to the sternal angle. The flaps are widely elevated to include platysma. The SCM is defined anteriorly and posteriorly from its insertion below to the uppermost limit of the wound above. The accessory nerve is identified. The clavicle is displayed subperiosteally at its middle point. The subclavius muscle and suprascapular vessels are divided. The omohyoid muscle is divided and reflected. Pectoralis major is detached from the inferior portion of the medial clavicle revealing the subclavian vein. The strap muscles are released from the notch of the manubrium and the plane deep to this developed carefully using a dental Howarth elevator and a curved Adson’s dissector. The plane is enlarged with the finger and a malleable retractor passed. The first costochondral junction is defined after detaching pectoralis major from the adjacent manubrium
258
Surgical Disorders of the Peripheral Nerves
Fig. 7.35 The Fiolle Delmas exposure. The incision at the shoulder was made for the purpose of inserting an intramedullary nail. The wound, caused by the fracture, has been left open.
Fig. 7.34 Rupture of C5 and C6 with avulsion of C7, C and T1 displayed at 6 days after injury. Above: the fat pad is split to display the transverse cervical vessels (1). Middle: the phrenic nerve (2) is seen deep to the pre vertebral fascia. Below: division of the fascia reveals the ruptured stumps of C5 and C6 (3) and the avulsions of C7, C8 and T1 (4).
and the interval between the first and second sternocostal junction is developed deep to the sternum by the means outlined above. A malleable retractor is passed to meet its fellow from above. The first costal cartilage is cut with a scalpel, a
fine osteotome is used to divide the manubrium with an L-shaped cut. The clavicle is now divided and the medial clavicle, sternoclavicular joint and upper corner of the manubrium are elevated on the SCM. The residual strap muscles are released. The internal jugular vein is now seen and traced to its junction with the subclavian vein. The brachiocephalic vein is traced and lightly held in a vascular sling. The phrenic nerve is elevated from scalenus anterior and the muscle divided. The first part of subclavian artery, the vertebral artery and the recurrent laryngeal and vagus nerves are seen. To display the whole of the plexus pectoralis major and pectoralis minor are detached from the humerus. When it comes to closure, the manubrio-clavicular fragment is reattached to the manubrium with wires, and to the clavicle with a plate and screws. The soft tissue layers are carefully closed. This exposure has been used on 80 occasions, chiefly for tumours of the brachial plexus. One patient, a 62 year old man, with a malignant neurofibroma complicating neurofibromatosis type 1(NF1), died of consumptive coagulopathy. Osteomyelitis of
Operating on Peripheral Nerves
259
Fig. 7.36 Two large neurofobromata exposed through the approach of Fiolle and Delmas. Top left: the skin incision. Bottom left: The tumour exposed and, Top right: removed. There was no loss of function
(Courtesy of Marco Sinisi). Bottom right: Another neurofibroma arising in the posterior triangle of the neck. It was not necessary to divide the clavicle in either of these cases.
the clavicle occurred in two cases, and non union in four more. The recurrent laryngeal nerve is more vulnerable on the right side. Damage to the thoracic duct can usually be avoided, except in a scarred field and is usually recognised because of the leak of milky fluid. It is best to identify and repair the lesion, because if the pleura has been opened, there is a real danger of formation of a chylothorax. If the hole cannot be repaired, it can be plugged with a muscle graft.
that the upper ribs are exposed. The first rib is defined and removed extra-periosteally, if necessary after the second rib has been removed sub-periosteally to facilitate exposure. The scalenus medius muscle is partly liberated during the removal of the first rib, and further mobilisation permits its upward retraction to display the brachial plexus. Kline and colleagues (1978) warn that it is necessary to avoid damage to the nerve to serratus anterior during this process. By medial retraction of the posterior paravertebral muscles the foramina can be opened by hemi-laminectomy. Closure after completion of the procedure is by re-uniting the muscles carefully over a drain or drains Dubuisson and colleagues (1993) describe the three potential complications: winging of the scapula, instability of the cervical spine if more than two facet joints are removed, and damage to various related structures. We certainly accept as firm indications: (1) In the thoracic outlet syndrome: prior operation by the transaxillary or supraclavicular routes for removal of the first thoracic or seventh cervical rib, when the posterior third of the rib remains; (2) In tumours of the plexus: tumours with intraforaminal and extra foraminal
7.6.3 The Postero-Lateral Route This, the posterior subscapular route, provides very good access to the most proximal parts of the nerves and in particular to the nerves in the intervertebral foramina (Kline and colleagues 1978). The patient is almost supine, with the limb on the affected side resting on a separate table. The incision is convex medially, truly parascapular. The scapula is freed by the division of the trapezius and rhomboid muscles and, if necessary, the levator scapulae and it is retracted laterally so
260
Surgical Disorders of the Peripheral Nerves
Fig. 7.37 The transclavicular exposure. Top left: the position of the patient and line of incision. Top right: elevation of the skin flaps with platysma and definition of sternocleidomastoid muscle (1). Middle left: The clavicle exposed and the phrenic nerve (2) and subclavian artery (3) traced. Middle right: the bone flap is elevated, taking care for the
subclavian vein (4). The trunks of the brachial plexus (5) are seen. Malleable retractors (6) are inserted deep to the manubrium before section of the first costochondral junction and division of the manubrium. Bottom: the tumour (7) is seen deep to the subclavian artery, where it enveloped the vertebral artery.
lateral components; (3) In radiation neuropathy when there is extensive change in the skin and deep tissues of the neck and chest wall: (4) In traumatic lesions when the evidence is that a reparable lesion is in or near the foramen. Our posterolateral approach owes more to Adson and Brown (1929) than to Kline and his colleagues (Fig. 7.38a–c). It affords a less extensive view to the more distal (lateral) part
of the supraclavicular plexus than does Kline’s. The patient is put in the lateral position, the affected side uppermost, and the limb is included in the field. The incision, convex laterally, is centred over the seventh cervical vertebral spine. The flaps are raised. The trapezius is divided close to the midline and the muscle is retracted laterally. The next layer then comes into view: the upper part of the rhomboids, and the lower part of
Operating on Peripheral Nerves
261
a
b Splenius capitis Rhomboids Levator scapulae Supraspinatus Acromion C7
Trapezius
C7 T1 T2
c
T3 T4 T5
Scalenus medius
T6
Fig. 7.38 Posterior approach to the left brachial plexus. (a) The incision. (b) The division of the trapezius. (c) Removal of posterior elements and detachment of scalenus medius to expose the proximal part
of the plexus. Note that the nerve to serratus anterior is shown. In many cases of proximal injury to the plexus, it will have been damaged with the main nerves.
the splenius capitis. The levator scapulae, running from the scapula to the upper cervical transverse processes, and the splenius cervicis, running from the third to the sixth thoracic vertebral spines to the upper cervical transverse processes, are rather lateral. The upper part of the rhomboids and the lower part of the splenius capitis are divided near the mid line and the erector spinae group is exposed. Lateral to this, the
transverse processes of the first thoracic and lowest four cervical vertebrae can be felt. The transverse processes and lateral masses are exposed by blunt dissection, with medial retraction of the erector spinae group. The back of the first rib is exposed at the lower end of the field, and the scalenus medius is cleared from its upper surface. That muscle is detached from its origin on the posterior tubercles of the lowest three of four cervical
262
transverse processes. The nerves are then shown distal to the transverse processes, the proximal branches of the fifth, sixth and seventh vertebrae going into the scalenus medius to form the nerve to the serratus anterior. Now the posterior tubercles of the fifth and sixth vertebrae and part of the transverse process of the seventh are nibbled away to show the most proximal parts of the nerves. If necessary, the first thoracic transverse process and the first rib can be removed to increase the exposure of the lowest part of the plexus. Later, if necessary, one or two interfacetal joints can be removed to show the nerves in their dural sheaths in the foramina. It is necessary to proceed carefully and methodically in this exposure, securing good haemostasis at each stage. It is not an easy exposure, but it is the preferred route when there is a very proximal lesion and access by the anterior route is barred by the sclerosis of a previous intervention. Two muscle layers and one subcutaneous layer are closed over a suction drain. This route opens possibilities for exposure of the central segment of the spinal nerve and the opportunities for repair of the roots of the spinal nerves will now be considered.
7.7 Repair of the Roots of the Spinal Nerves in an Avulsion Lesion In these cases the cord has already been damaged by the avulsion of one to five spinal nerves. The risks of interfering, however marginally, on this damaged cord are clear. The anterior spinal artery is maintained by radicular vessels which enter the spinal canal in company with the spinal nerves. These vessels must be ruptured in an avulsion lesion. It is likely that the incomplete Brown Séquard syndrome, seen in so many cases of avulsion lesions, is ischaemic in origin. This fact must be at the forefront of thinking about reconnection between the spinal cord and the avulsed spinal nerve. Some contraindications to direct repair or reimplantation of avulsed spinal nerves include severe associated injuries to the head, the chest or the viscera; rupture or occlusion of the subclavian artery or the vertebral artery; fracture of a cervical, or upper thoracic vertebra and any hint of cord lesion. These are hazardous operations and it is most unwise to add to those hazards by delay when fibrosis and the sealing off of the rent in the dura will add to difficulties. The posterior subscapular exposure has been described above, three other approaches are available. The endoscopic transforaminal approach to the anterior aspect of the spinal cord was opened by the experience of one of us working with Thomas Carlstedt in 1995. A 10 year old boy with a complete lesion of the brachial plexus was operated on 2 weeks from injury. C7, C8 and T1 were avulsed. A small arthroscope introduced through the foramen of C8, gave a surprisingly good view of the cord and of
Surgical Disorders of the Peripheral Nerves
the stump of the ventral root of C8. The presence of useful stumps of C5 and C6, and the state of the avulsed nerves persuaded us against reimplantation. Transforaminal endoscopy has been used on a number of occasions. A 2 mm diameter small joint arthroscope is used. The lower foramina are best suited for this manoeuvre because of their size and because of their relation to the vertebral vessels. Carlstedt (2007a) says that: “more information regarding remaining intradural root stumps or full root avulsions can be gained with this technique than through ancillary investigations”. Direct repair was possible in one case; in the others, short interposed grafts, using dorsal rootlets from the avulsed spinal nerves, were used. This method is usually possible only within a few days of injury. It is safe, it is relatively straight forward and it merits further attention. The possibility of repairing the ventral root by endoscopic methods through an enlarged foramen, with proper care for the vertebral artery, is attractive. The anterolateral transvertebral approach. Fournier and his colleagues have developed an antero-lateral approach to the ventral root entry zone (Fournier et al. 2001, 2005). The exposure is described as a “lateral interscalenic multi-level oblique corpectomy”. Pre operative vertebral angiography is essential. The patient lies supine, with the head slightly extended, and lightly rotated to the opposite side. The incision follows the posterior margin of the sternocleidomastoid muscle. The great vessels and the viscera are not formally displayed, as dissection passes deep to them. The cervical column is exposed between the scalenus anterior and the scalenus medius, by following the cervical spinal nerves. The phrenic nerve is isolated and scalenus anterior resected. The longus colli and the sympathetic chain are retracted medially. The vertebral artery is exposed and mobilised with great care using high magnification and it may be taken forwards or back. Now, part of the vertebral body and part of the adjacent intervertebral disc are removed to expose the dura mater. This is opened and nerve grafts which are passed behind the vertebral artery and the phrenic nerve are inserted through a small incision in the pia at the level of the ventral root entry zone, to a depth of about 2 mm. Plasma glue is used to support the position of the grafts. The dura mater is repaired. Stabilisation of the cervical spine is not necessary because the articulations have not been disturbed. Fournier and his colleagues recognise some limitations to this approach: it is not feasible to expose all five of the ventral roots; reimplantation of the dorsal roots is not possible. The lateral approach was used by Bonney and Jamieson in the first case of reimplantation in 1977 (see Chapter 9) and is described by Carlstedt and Birch (2000) and by Carlstedt (2007b) and it is possible to extend the exposure from the anterior spinal artery to the dorsal root entry zone as well as the supra and infraclavicular brachial plexus. The patient is placed laterally, (Kratimenos and Crockard 1993)
Operating on Peripheral Nerves
263
Spinal accessory n. Levator scapulae
Scalenus posterior Sternocleidomastoid
C5
a
Trapezius
b Levator scapulae
Brachial plexus
Scalenus posterior
Scalenus posterior
Erector spinae
N. XI
Longissimus (divided) Trapezius (divided)
Transverse proc. (post tub.)
c
Longissimus
d
Trapezius
Splenius
Levator scapulae
Fig. 7.39 The lateral approach to the brachial plexus. (a) The skin incision. (b) The trapezius muscle is split protecting the accessory nerve. (c, d) The longissimus is approached between the levator scapulae and
the scalenus muscles and it is split longitudinally to expose the posterior tubercles of the transverse processes (Drawn from Carlstedt and Birch 2000).
with the head supported in a Mayfield Clamp. The neck is slightly flexed away from the side of operation whilst the head of the table is elevated to about 30°. It is possible to tilt the table to assist anterior and posterior exposure. The skin incision is transverse supraclavicular extended to the spinous process of C5 (Fig. 7.39). It is possible for two surgeons to work simultaneously, the first exposing the brachial plexus in the posterior triangle of the neck, the second working towards the posterior aspect of the cervical spine. The posterior tubercles of the transverse processes of C4 to
C7 are exposed between the levator scapulae and posterior and medial scalene muscles, splitting the longissimus muscle longitudinally. The paravertebral muscles are detached from the hemilaminae and retracted dorso-medially. The dura is exposed through a hemilaminectomy. The medial part of the processes of the facet joints may be removed. The dura mater is incised longitudinally and held with stay sutures; the denticulate ligaments are detached laterally and held with stay sutures which permits very gentle rotation of the cord. Grafts are fed through the intervertebral canals
264
using fine vascular slings or a silicon catheter. These are implanted through slits in the pia mater and secured with plasma glue (Fig. 7.40). Carlstedt (2007b) provides a detailed account of the method of implantation and a compact disc (CD) showing the operation is supplied with his publication. The nerve graft can be split into separate bundles and these are introduced to a depth of between 1 and 2 mm: Carlstedt says “a small Roton instrument in the shape of a hockey stick, is useful in this manoeuvre; its “blade” being 1–2 mm in length, can be used as a depth gauge when implanting the root or the nerve graft. Implanted grafts are maintained in position by glue (Tisseel) or they can be stitched to the pia. The pia is slightly elastic and closes around the introduced tips of the nerve graft. In cases of root ruptures, the nerve graft is directly apposed to the trimmed end of the ventral root stump… if the foramen is blocked, the grafts are passed through the incision in the dura and outside the vertebral canal”. The dura is closed by a vein patch or a patch of artificial dura. An early case was complicated by a CSF leak. Since then a lumbar drain has been retained for up to 7 days.
Fig. 7.40 Grafts are drawn through the canals of C6 and C7 and implanted through short slits in the pia mater (Courtesy of Thomas Carlstedt).
Surgical Disorders of the Peripheral Nerves
7.7.1 The Spinal Accessory Nerve The incision is often dictated by the site through which the nerve was originally damaged. Often enough, the original incision can be extended either as a “Z” or in the direction of the lines of skin tension. The usual site of injury is in the posterior triangle between the sternocleidomastoid muscle and trapezius muscles, though occasionally the nerve is wounded in its course through the former muscle or proximal to it. The greater auricular nerve is the key to the exposure of the proximal stump which emerges from behind the sternocleido muscle about 5 mm cephalad. The nerve emerges from deep to the muscle as one trunk, and at this point a fine branch is seen which innervates the uppermost part of the upper one third of the trapezius. This must be respected. The course and relations of the nerve are remarkably constant (Fig. 7.41). The incision extends to the anterior face of the SCM, displaying the greater auricular and transverse cervical nerves. Exposure of the accessory nerve should begin in unscarred tissue, and the lesion should preferably be exposed after proximal and distal trunks have been defined. The proximal stump may have retracted deep to the sternomastoid. It can be identified anterior to the muscle or branches to the sternocleido mastoid are identified with a nerve stimulator. The distal part is found beneath the anterior part of the upper fibres of the trapezius where it must not be confused with branches of the supraclavicular plexus which pass obliquely or horizontally in front. Atrophy of the distal trunk is a common finding because of the long delay before diagnosis. There is no point in attempting end-to-end suture after
Fig. 7.41 The spinal accessory nerve emerges, as one trunk (1), from deep to sternocleidomastoid about 5 mm cephalad to the greater auricular (2) and transverse cervical (3) nerves. In this case the nerve had been transected 15 months earlier and there is atrophy of the distal stump (4).
Operating on Peripheral Nerves
265
resection: the gap is usually too great, and in any case, the mobility of the forequarter on the axial skeleton forbids the use of such a method because of the hazards of disruption. On a few occasions we have been forced to reinnervate the distal trunk by transfer of part of the lateral pectoral nerve.
7.7.2 The Suprascapular Nerve We prefer the lateral position and a transverse incision. The flaps are raised, and the supraspinous part of the trapezius muscle is detached from its origin and raised to show the supraspinatus muscle, or the fibres of trapezius can be split to show the underlying supraspinatus, which is lifted from its scapular origin to reveal the suprascapular nerve traversing the notch. The accompanying artery is usually superficial to the transverse ligament or bony bridge. Ochiai and his colleagues (1997) showed that the suprascapular nerve might be seriously damaged in several places and describe an exposure which exposes the nerve along its entire course as far as the infraspinatus and which can be extended to display the circumflex nerve anteriorly.
7.7.3 The Infraclavicular Part of the Brachial Plexus With the full opening of the delto-pectoral interval the pectoralis minor muscle is exposed crossing the field to its attachment on the coracoid process. Below and above it is the neurovascular bundle, covered by a layer of fascia and, in the lower part of the field, by the flat reflected tendon of the sternal part of the pectoralis major. The easiest way to define the interval between the pectoralis minor and the coraco brachialis is to pass the index finger behind the upper margin of pectoralis minor and feel for the interval. The tendon of pectoralis minor is divided. The muscle is drawn medially, care being taken of the medial pectoral nerve piercing it and going on to the pectoralis major. Now the fascia over the plexus is divided above and below the former site of the pectoralis minor. If it is absolutely necessary, the flat reflected tendon of the pectoralis major too can be divided. Thus the whole neurovascular bundle is displayed. The lateral cord is the most prominent component, the axillary artery is behind it and the axillary vein medial to it. In the lower part of the field the median nerve is formed from the contributions from the medial and lateral cords. Just at the formation of the lateral cord the lateral pectoral nerve arises to pierce the clavipectoral fascia and enter the pectoralis major. The posterior cord lies deep to the lateral cord and axillary artery; the medial
Fig. 7.42 The infraclavicular brachial plexus splayed over a massive lipoma. The axillary artery(1) is marked by the blue sling, the median nerve (2) by a red sling and the ulnar nerve (3) by a white sling. The medial cutaneous nerve of forearm (4) is seen and the radial nerve (5) lay deep to the tumour. The appearances before (above) and after (below) removal of the tumour (Courtesy of Marco Sinisi).
cord is deep to the axillary vein. Some mobilisation of both great vessels is necessary for the full display of the cords (Fig. 7.42). The musculocutaneous nerve arises from the lateral cord above the level of the coracoid process and runs laterally into the coraco-brachialis muscle. It may be a single branch but can consist of several branches arising at different levels from the cord. The posterior cord and its lateral and terminal branches are exposed between the lateral cord and the axillary artery. The three subscapular nerves are seen, and in the lower part of the field the separation of the trunk into radial and circumflex nerves is visible. Reflection of coracobrachialis from the tip of the coracoid process improves exposure of the anterior “door” of the quadrilateral space and permits exposure of the distal stump of a ruptured circumflex nerve. This display may be extremely difficult in the late case, especially after arterial injury.
266
7.7.4 The Circumflex Nerve In cases of rupture the anterior (proximal) stump is almost always found just below the coracoid process and repair sometimes requires exposure of the posterior (distal) stump through a separate posterior incision. Repair has to be effected by graft, and it is certainly easier to secure good placement and attachment through two incisions than through one giving limited access. The posterior circumflex artery is often ruptured by the head of humerus, in anterior dislocation. This can lead to serious bleeding, or to a false aneurysm. The distal stump of the nerve may be strangled in fibrosis and in these cases direct muscular neurotisation is useful as it is when the nerve has been avulsed from the muscle (Fig. 7.43).
7.7.5 Median and Ulnar Nerves in the Arm and the Axilla The incision crosses the anterior axillary fold and the axilla, and descends the medial side of the arm. The flaps are raised and the axillary fat displaced downwards. The neurovascular bundle is found in its sheath in the lower part of the axilla: most medial is the axillary/brachial vein; the nerves are grouped around the lower part of the axillary and the brachial arteries. In the upper arm, the medial cutaneous nerve of forearm (MCNF) lies anterior to the brachial vein and is the most superficial of the nerves within the neurovascular bundle. The nerve perforates the deep fascia in the middle part of the arm. The slender medial cutaneous nerve of the arm (MCA) runs outside and posterior to the neurovascular sheath. Communicating branches between the MCNF and MCA are common. The axillary artery is embraced by the
Fig. 7.43 Rupture of the circumflex nerve at the entrance to the quadrilateral tunnel. The proximal stump (1) and, somewhat unusually the distal stump (2) are visible.
Surgical Disorders of the Peripheral Nerves
two roots of the median nerve, which starts on its lateral side and crosses it in the arm. The musculocutaneous nerve most commonly arises as a single branch from the lateral cord at or below the level of the coracoid process, but it may consist of several branches arising at intervals along the line of the lateral root of the median nerve. It passes laterally into the coracobrachialis muscle and the flexors of the elbow. The ulnar nerve and the medial cutaneous nerve of the forearm are on the medial side of the artery. The former passes posteriorly about half way down the arm to pierce the medial intermuscular septum and to lie between it and the medial part of the triceps muscle, to which it usually gives a branch. Deepest of all is the radial nerve, which crosses anterior to the tendon of latissimus dorsi and then passes posteriorly between the long and medial heads of triceps. The nerve runs behind the humerus deep to the lateral head of the triceps and gains the lateral aspect of the lower part of the arm by piercing the lateral intermuscular septum. The neurovascular bundle can be traced up into the axilla to expose the cords and the origin of the circumflex and radial nerves.
7.7.6 The Radial Nerve The proximal part of the radial nerve is quite easily displayed in the upper part of the axillo-brachial incision; the distal part is easily found through an anterolateral incision in the lower part of the arm and by entering the interval between the biceps and brachialis medially and the brachioradialis and extensor carpi radialis longus laterally (Fig. 7.44). Finding the middle part - the part most likely to be in trouble - is rather more difficult. It is to be found through a posterior incision, and by separation of the superficial part of the triceps (the long and lateral heads) from its deep part (the medial head). The flaps are raised and the lower part of the deltoid muscle and the superficial part of the triceps muscle are exposed. The V-shaped interval between the upper parts of the long and lateral heads is now defined, by locating the upper part of the long head and following its lateral border distally. The “seam of the half sleeve” (Henry 1975) – that is, the junction of the long and lateral heads of the triceps – is now opened from the top towards the olecranon (Fig. 7.45a, b). Further exposure of the radial nerve is anterior to the lateral head of the triceps muscle, by entering the interval between the biceps/ brachialis and the brachioradialis/extensor carpi radialis longus. If there is difficulty in bridging a gap after resection, some length can be obtained by re-routing the distal stump towards the upper medial aspect of the arm deep to the biceps and brachialis muscles. For exposure of the more distal parts of the nerve the incision may be extended round the side of the arm to the medial side of the brachioradialis. Thence it runs across the anterolateral
Operating on Peripheral Nerves Fig. 7.44 Transverse section through the arm above the level of the insertion of the deltoid muscle, just below the level of the posterior movement of the radial nerve.
267
Biceps (short head) and coracobrachialis Musculocutaneous n.
Median n.
Biceps (long head) Deltoid Humerus Triceps (medial head)
Ulnar n. Radial n. Tendinous lamina
Triceps (lateral head) Triceps (long head)
aspect of the elbow into the forearm. With this anterior approach the radial nerve and its terminal branches are found in the interval between the biceps/brachialis and the brachioradialis/ extensor carpi radialis longus. There are important branches from this segment of the nerve. The nerve to brachialis, passing antero-medially is often seen about five finger breadths above the lateral epicondyle. The nerve to brachioradialis passes postero-laterally about 2 cm more distal. More distally still, is the branch to extensor carpi radialis longus and then, after that, branches, usually two, to extensor carpi radialis brevis (Fig. 7.46).
7.7.7 The Posterior Interosseous Nerve In the Henry (1975) approach, the incision is made on the posterolateral aspect of the upper forearm between the mobile mass of the brachioradialis and the radial extensors of the wrist, and the extensor communis digitorum. The interval between the extensor carpi radialis brevis and the extensor communis digitorum is then opened and the supinator muscle exposed. The posterior interosseous nerve passes between the superficial and deep parts of this muscle, to emerge at the lower margin of the superficial part and to run for about 4–5 cm before breaking up into its terminal (motor) branches. The point of emergence can quite easily be found, and the nerve then followed proximally by division of the superficial part of the muscle along its line (Fig. 7.47a, b).
7.7.8 The Lower Part of the Median Nerve The median nerve is easily displayed at and just below the elbow and in the lower part of the forearm through a sinuous incision winding down from the elbow. At elbow level it is found medial to the brachial artery. It then descends between the superficial and deep heads of the pronator teres The median nerve can be traced down into this tunnel, and can to some extent be mobilised by division of the deep (ulnar) head of the pronator teres. Having negotiated the pronator teres, the nerve runs down the forearm between the deep and superficial flexor muscles, loosely attached by areolar tissue to the deep surface of the flexor digitorum superficialis. It can be exposed in this part of its course by separating the flexor superficialis from its radial origin and retracting the muscle. The anterior interosseous nerve is now displayed. The nerve enters the hand just deep to and between the tendons of the palmaris longus and flexor carpi radialis. The “bottle neck” (Lanz 1993) of the carpal tunnel is formed posteriorly and laterally by the concavity of the carpus; in particular, the scaphoid, lunate, hamate, and pisiform bones. It is roofed by the flexor retinaculum and anchored medially to the pisiform and hamate bones and laterally, to the scaphoid and the trapezium. The width of the retinaculum is about 2.5 cm; its length is about the same. Proximally it blends with the ante-brachial fascia, and distally with the palmar aponeurosis. The tendon of the palmaris longus is attached to it anteriorly. A deep lamina on the radial side is attached to the medial lip of a groove on the trapezium. Between this layer
268
Surgical Disorders of the Peripheral Nerves
b
a
Deltoid
Teres major
Ulnar n.
Profunda a.
Deltoid
Triceps (lateral head)
Brachial a. Median n.
Triceps (long head)
Radial n. and branches to lateral and medial heads
Lateral head of triceps Tendinous laminae of long head (with cut fibres of lat. head)
Olecranon
Fig. 7.45 The exposure of the radial nerve in the arm. (a) The “seam of triceps” (AK Henry). The finger is introduced between the long and lateral heads and separated. (b) The opening of the “seam” reveals the
and the more superficial part lies the tendon of the flexor carpi radialis muscle in its synovial sheath. On the ulnar side a localised thickening of the ante-brachial fascia extends laterally from the pisiform bone as a superficial part of the retinaculum, passing superficial to the ulnar nerve and vessels to blend with the retinaculum lateral to them. The positions of the motor and the palmar cutaneous branches of the median nerve are very important for the surgeon. Lanz (1993) illustrates 12 variants of the main nerve and its motor branch. In most cases the motor branch arises from the radial side of the nerve in the carpal tunnel and runs laterally, soon branching to supply the short abductor of the thumb, and often part of the opponens and flexor brevis. A particular hazard for the surgeon is produced by an origin from the ulnar side of the nerve and a course across it, either deep or superficial, to the retinaculum. Wadsworth (1984) shows an excellent illustration of the variations in origin and course of the motor branch The palmar cutaneous branch arises from the median nerve about 7 cm above the wrist crease and runs distally, medial to the tendon of the flexor carpi radialis to innervate
radial nerve from the start of its course on the back of the humerus to the piercing of the lateral intermuscular septum.
Fig. 7.46 The radial nerve at the elbow exposed during repair of posterior interosseous nerve cut during operation for fracture of radial head. The nerve is seen in the valley between brachio-radialis and biceps lying on fibres of brachialis. The nerves to brachioradialis (1) to extensor carpi radialis longus and brevis (2) the proximal part of the posterior interosseous nerve (3) a further branch to extensor carpi radialis brevis (4) and the superficial radial nerve (5) are shown. A segment of the lateral cutaneous nerve of forearm (6) has been prepared as a graft.
Operating on Peripheral Nerves
a
269 Extensor carpi radialis brevis
b
Head of radius
Supinator
Extensor digitorum communis
Fig. 7.47 Exposure of the posterior interosseous nerve (After Henry 1975). (a) The incision (left) and the separation of the extensor carpi radialis brevis from the extensor communis (right). (b) The posterior interosseous nerve and supinator exposed.
the skin over the thenar eminence and the radial side of the proximal part of the palm. Damage to this can cause a profound, persistent painful state, with very troublesome hyperaesthesia and hyperalgesia, and even hyperpathia and allodynia (Semple and Cargill 1969; Carrol and Green 1972; Taleisnik 1973) (Fig. 1.28).
7.7.9 The Lower Part of the Ulnar Nerve As the ulnar nerve approaches the elbow, it runs behind the medial intermuscular septum and then passes subcutaneously behind the medial epicondyle of the humerus in the retro-condylar groove. Schroeder and Scheker (2003) define the “arcade of Struthers” thus: “the arcade was better described as a fibrous canal with an average length of 5.7 cm, its openings, at either end, were 3.9 and 9.6 cm proximal to the medial epicondyle”. The components of the arcade include not only the tough medial intermuscular septum of the arm but also the sheath of the medial head of triceps, which can be seen enveloping the nerve, as well as the internal brachial ligament. The ulnar nerve then passes superficial to the medial capsule of the elbow joint and deep to the arcade, joining the two origins of the flexor carpi ulnaris muscle (the arcuate ligament). This is the “cubital tunnel”. The nerve then runs between the flexor carpi ulnaris and
flexor digitorum profundus muscles, joining the ulnar artery in its course down the forearm. The muscular branches arise just below the elbow. The nerve is vulnerable to external pressure. It may be compressed between the arcuate ligament and the medial capsule and with relaxation of this ligament it may become hypermobile, dislocating forward when the elbow is flexed. Because of the excursion required of it during movement of the elbow, it may be stretched during flexion if its movement is restricted by changes due to damage by external pressure. Wadsworth (1977) dealt with the “external compression” syndrome of the ulnar nerve, drawing on Osborne’s (1970) exposition of the role of the arcuate ligament in different positions in the elbow. The ulnar nerve and artery pass down the forearm within a well defined sheath which is found deep to the anterior margin of flexor carpi ulnaris. The nerve and artery may become tightly compressed by swollen anoxic muscle and, later, by post ischaemic fibrosis. Arterial bleeding from a wound on the ulnar aspect of the forearm indicates damage to the ulnar nerve. The ulnar nerve divides into its superficial (sensory) and deep (motor) components at about wrist level. Both components run into Guyon’s space (the piso-retinacular space of Denman (1978)) (Fig. 7.48). The superficial branch passes superficially to supply the palmaris brevis and the skin of the medial two digits; the deep branch runs between the abductor and the flexor of the little finger to pierce the opponens and
270 Palmaris longus
Surgical Disorders of the Peripheral Nerves Hook of hamate bone
Pisohamate ligament
Ligamentous band Ulnar n. (deep branch)
Ulnar n. (superficial branch) Pisiform bone Flexor retinaculum Volar carpal ligament
Flexor carpi ulnaris Ulnar a. Ulnar n.
Fig. 7.48 Guyon’s space. Note the possibilities for entrapment at, and distal to, the volar carpal ligament.
then runs across the deep palmar space with the deep palmar arch, ending by supplying the adductor and flexor brevis of the thumb and the first palmar interosseous muscle. Deep as it is, the deep branch is vulnerable to wounds from glass or knife. Horsley (1899) describes a typical case: “It is the case of a young lady who succeeded in performing, by accident, an exceedingly difficult operation upon her own hand. She was engaged in wood carving, and was working with a very long and narrow chisel. The chisel slipped and stuck into the palm of the hand. She succeeded in just missing the cutaneous branch of the nerve and cut the deep branch”. Horsley sutured the nerve and the patient had complete recovery. The nerve is best followed by an incision beginning above the wrist, entering the ulnar side of the palm and curving across the distal palm in a crease. It is isolated in the upper part of the incision lateral to the tendon of flexor carpi ulnaris and medial to the ulnar artery and followed to its bifurcation. Then the deep branch is followed between the hypothenar muscles. Its deep course has to be revealed by mobilisation of the deep and superficial flexor tendons. The multitude of the motor branches of this nerve, to most of the intrinsic muscles of the hand, has to be borne in mind during this exposure, though it is impossible to avoid some of these in a scarred field.
7.7.10 Nerves in the Abdomen and Pelvis The lumbar plexus is accessible through the lower quadrant of the abdomen, by the same transverse muscle cutting incision
and extra-peritoneal approach that is used for lumbar sympathectomy (Kline et al. 2001a). The exposure may be helped by placing the patient in the lateral or semi-lateral position. The incision is made between the rib cage and the iliac crest. The external oblique is split and the deeper muscle is cut, with opening of the lateral part of the rectus sheath. The underlying fascia is delicately opened and the peritoneum is swept away from the abdominal wall and vertebral column. The lateral cutaneous nerve of thigh, the ilio-inguinal nerve and the iliohypogastric nerve are encountered approaching the lateral and posterior margin of the psoas muscle. The femoral nerve lies in a gutter between the iliacus and the psoas major. The obturator nerve runs medial to the psoas muscle. The lumbo-sacral trunk is closer to the mid line and deep to the great vessels. The spinal nerves passing to the femoral nerve run deep to the psoas muscle in much the same way as the cervical spinal nerves run deep to scalenus anterior and the muscle must be cut or reflected to display them. This approach gives good access to the lumbar plexus in the psoas muscle and to the femoral and obturator nerves on each side of the lower part of the psoas. Access to the lumbosacral plexus is rather restricted through this extra-peritoneal approach. A trans-peritoneal approach makes access easier, but the viscera have to be mobilised with the consequent risks of ileus after operation and of late complications from adhesions. Injuries to the lumbosacral plexus are often associated with forbidding scarring. The femoral nerve can be traced down to the level of the inguinal ligament from above through an abdominal incision and exposed again in the thigh through a separate anterior crural incision. Kline et al. (2001b) suggest that the crural incision can be extended laterally above the inguinal ligament and the lower abdominal muscles split to give an extraperitoneal approach. The sacral plexus We have no experience of the exposure described by Sedel (1975). Sedel attributes this “voie transiliaque” to Judet and comments that it is relatively destructive to the skeleton but that the exposure of the lumbo-sacral plexus and of the hypogastric venous plexus is excellent. The patient lies in a semi prone position. The incision is vertical, and runs from the inferior margin of the rib cage to the level of the greater trochanter. The iliac crest is exposed and then the gluteus maximus is detached from its insertion and reflected laterally and inferiorly so that the blade of the bone is exposed as high as the sciatic notch. The sciatic nerve and its accompanying vessels are carefully exposed and retracted out of harms way. The internal face of the ilium is exposed by subperiosteal dissection. The bone is now divided between the iliac crest and the sciatic notch. Sedel uses a Gigli saw with malleable retractors protecting deeper structures. It is now necessary to open the sacro-iliac joint by detaching the sacro-iliac ligaments and incising the capsule in the deep face the joint so that the bone flap can be reflected posteriorly with muscles still attached to the crest. The lateral flank of the fifth lumbar vertebra, and “la fossette de Cuneo et Marcille” which
Operating on Peripheral Nerves
a
271
b
Femoral n.
S1
Lumbosacral trunk
S2
Fig. 7.49 The trans iliac exposure of the sacral plexus (After Sedel 1975). (a) The position of the patient and line of incision left and the line of division of the ilium right. (b) Reflection of the fragments of the ilium provides access to the sacral plexus.
contains the lumbo sacral trunk, the obturator nerve, and the ilio-lumbar artery and vein is now accessible. The upper sacral roots appear below, running on the piriformis muscle (Fig. 7.49a, b). Closure of the wound commences with internal fixation of the Ilium followed by careful reattachment of gluteus maximus. The dorsal trans-sacral approach This was described by Linarte and Gilbert (1986). It offers exposure of the cauda equina, the sacral spinal nerves, and also the sciatic nerve which is seen at the sciatic notch after reflection of the origin of gluteus maximus. The patient lies prone. An incision is made from the spinous process of L3 curving somewhat as far as the ischial tuberosity. The para-spinal muscles are elevated from L4 and L5 and from the sacrum. The spinous processes are removed to expose the laminae and the dorsal sacrum as far lateral as the dorsal lateral foramina. The spinal canal can now be opened by hemilaminectomy (Fig. 7.50). This approach enables diagnosis of the level of lesion of the lumbar and sacral spinal nerves and offers the opportunity to place grafts onto the superior gluteal and the sciatic nerves, bypassing severe scar. Carlstedt (2007c) observed that the spinal nerves were either compressed or transected in fractures involving the sacral foramina, rarely avulsed.
Fig. 7.50 Display of cauda-equina after dislocation of hemi pelvis. Severe pain and extensive paralysis were relieved after liberating L4 and L5 from adhesions, 3 months after injury in a 28 year old woman (Courtesy of Marco Sinisi).
7.7.11 The Sciatic Nerve Two exposures are regularly used. In the first, the gluteus maximus muscle is split as in the posterior approach to the hip. This provides an adequate exposure of the sciatic nerve where it traverses the neck of the femur. Up to 15 cm of the trunk can be seen. A second, more extensive, approach is described by Henry (1975.) The prone position is used for both.
272
Surgical Disorders of the Peripheral Nerves
Fig. 7.52 The extended exposure of the sciatic nerve damaged by fracture/dislocation of the hip (After Henry 1975).
Fig. 7.51 Exposure of sciatic nerve by transgluteal approach. Above: The patient lies prone and the incision is made between the tip of the greater trochanter and the ischial tuberosity. Below: the nerve is exposed as it crosses the neck of the femur.
Splitting the gluteus maximus The incision runs obliquely about two finger breadths above the line drawn from the ischeal tuberosity to the tip of the greater tuberosity. The gluteus maximus is exposed and gently split along the line of its fibres noting and preserving branches of the inferior gluteal nerve and securing careful haemostasis. The muscle flaps are held apart with the large Deavers retractor. The nerve is seen crossing the posterior aspect of the neck of the femur, and it can be traced up to the sciatic notch if necessary by division of the piriformis (Fig. 7.51). The wound is closed by repair of the sheath of gluteus maximus, the fat and the skin. This approach is very useful in exploration of lesions of the sciatic nerve incurred during arthroplasty of the hip and the patient is encouraged to mobilise. Full exposure of the sciatic nerve and of the superior and inferior gluteal nerves and vessels is achieved through the more extensive operation described by Henry (1975). Henry compares the “gluteal lid” with a parallelogram whose shorter sides are almost longitudinal and whose longer (upper and lower) sides are oblique. The caudal edge is virtually unattached.
The incision can follow the “question-mark” shape advised by Henry, or can run from the iliac crest at the junction of the gluteus maximus and the iliotibial tract, obliquely down the back and lateral side of the buttock to reach the great trochanter and then turn medially to descend in the mid line of the thigh. It is necessary to look out for the posterior cutaneous nerve of the thigh, piercing the deep fascia in the upper part of the thigh and descending in the midline. It should be identified under the fascia below the lower border of gluteus maximus. Now, the gluteus maximus is detached from its insertion to the femur; the cephalic side of the muscle is separated from the ilio-tibial tract, and both tendinous and muscular insertions to the femur are divided. Then, with continued care for the posterior cutaneous nerve of the thigh, the gluteal lid is hinged back on its pelvic origin. Excessive traction upon the superior gluteal vessels must be avoided. The structures displayed are the gluteus medius and the short lateral rotators of the hip, the superior and inferior gluteal vessels and nerves, the pudendal nerve and vessels and the sciatic nerve. The sciatic nerve can be followed into and through the great sciatic notch by division of the tendon of the piriformis and retraction of the muscle. Downward extension of the vertical limb of this incision permits display of the whole of the sciatic nerve (Fig. 7.52). Careful closure of the wound is essential. In particular, the gluteus maximus must be reattached laterally and cephalad. One or two suction drains should be used. The subcutaneous tissues should be closed.
7.7.12 The Tibial and Common Peroneal Nerves in the Popliteal Fossa and Below The incision starts above the back of the knee and skirts the crease to return to the midline below the knee, to descend in
Operating on Peripheral Nerves
the midline for about 10 cm. The flaps are raised. Care must be taken of the sural nerve arising from the tibial nerve and descending in the midline, at first just under the deep fascia and piercing that in the proximal part of the leg. The tibial nerve and popliteal artery and vein are found above in the mid-line; the common peroneal nerve has at this level deviated laterally to lie close to the tendon of the biceps femoris. The gastrocnemius is split in the midline to expose the underlying soleus muscle, which also is split to show the nerve and vessels below the knee. The tibial nerve in the leg and behind the ankle is exposed through a straight or sinuous incision over the medial head of the gastrocnemius and medial to the tendo calcaneus. The medial head of the gastrocnemius is exposed, freed and retracted laterally to expose the medial part of the soleus. Then, the medial part of the soleus is mobilised by division of the medial “pier” of its tendinous arch and of its medial origin. The soleus is then retracted laterally, to expose the deep compartment of the leg with the tibial nerve and vessels (Fig. 7.53a, b). Exposure of the common peroneal nerve necessarily takes the popliteal incision laterally, to descend on the lateral side of the leg and permit exposure of the deep peroneal and superficial peroneal divisions in which the nerve terminates just below the neck of the fibula. The operator must remember that the common peroneal nerve is very close to the surface behind the head of the fibula and lateral to its neck. We have seen it partly divided by the initial skin incision. The nerve is embraced by fascia enveloping the biceps femoris and is displaced with that muscle in dislocation of the knee.
7.7.13 The Lower Tibial Nerve and the Plantar Nerves The lower part of the tibial nerve is easily found on the medial side of the ankle, under the flexor retinaculum, between the flexor tendons of the hallucis and the flexor digitorum. The terminal plantar branches are traced into the foot by division of the retinaculum and then by bringing back the abductor hallucis muscle after defining its superior edge and detaching it from its origin on the distal part of the retinaculum. The plantar nerves are found between the deep and superficial layers of the plantar muscles in the plane between, on the one hand, the two abductors and the flexor brevis digitorum and on the other, the flexor accessorius and the tendons of the long flexors (Fig. 7.54a, b).
7.8 Entrapment Neuropathy The concept of the entrapment syndrome is attractive; the belief that symptoms from such syndromes are easily relieved
273
by simple operation is especially espoused by operative masons. In practice, the matter is not so simple, because when things go wrong symptoms are likely to be severe, persistent, and resistant to treatment. No letter fills us with more gloom and apprehension as one requesting an opinion on symptoms persistent after operation for entrapment neuropathy. The proffered case is indeed likely to prove a poisoned chalice. The term “entrapment neuropathy” should probably be reserved for instances in which a small addition to the crowding produced by existing anatomical arrangements is liable to cause compression, constriction or distortion of a peripheral nerve or, by interfering with the ability of the nerve to glide, to cause traction injury. The situation of the median nerve deep to the flexor retinaculum provides an obvious example of such an anatomical arrangement, and the causes of lesions there provide a microcosm of the wider field. Comtet (2005) points out that “pure nerve compression is rare” and that it is usually compounded by traction, shear, or vibration as well as by the effects of generalised disease. In many cases, it appears , at first sight, impossible to distinguish a precipitating cause other than the anatomical situation of the affected nerve. However, the more rigorously investigation is pressed, the fewer true entrapments will be found. Investigation should be thorough because the recurrence of an “entrapment syndrome” may be the first sign of serious underlying disease, and both surgeon and patient may have cause, later, to regret a temporarily successful operation (Spinner and Spencer 1974; Gelberman et al. 1993.) Rosenbaum and Ochoa (1993) show, that in a combined series of 2,705 patients with “carpal tunnel syndrome” there were 321 with rheumatoid arthritis, 166 diabetics, and 94 with hypothyroidism, a total of 581 patients. In another 30, the condition was related to pregnancy or to an oral contraceptive. Most of our experience is drawn from failures of primary operations. Since 1979, more than 500 patients have been treated because of failure of, or complications of, first operation for entrapment neuropathy. Four causes head the list. 1. Error in diagnosis. The symptoms of entrapment neuropathy may, in fact, point to a proximal lesion of the nerve or to a generalised disease of nerve or muscle or to a systemic endocrine, neoplastic or circulatory disorder. Many errors can be prevented by the appropriate electrodiagnostic investigation (Chapter 6). 2. Inadequate decompression. 3. Damage to the nerve, to branches of the nerve, to adjacent nerves or vessels. 4. Provocation of the fibrosis which entraps the nerve with even more vigour than before operation. The gliding mechanism is destroyed and it is this often unavoidable complication of operation that underlines the failure of so many interventions. The rate of recurrence of symptoms increases with the duration of follow up. Scar is inevitable after operation. It is worse after clumsy surgery and worse still following a haematoma.
274
Surgical Disorders of the Peripheral Nerves
Fig. 7.53 Exposure of the tibial nerve in the leg (After Henry 1975). (a) The soleus exposed and the line for its detachment shown. (b) The tibial nerve and vessels revealed by the retraction of the soleus.
a
Popliteal a.
Gastrocnemius (medial head)
Popliteus Line of incision dividing medial pier of soleus arch Divided edge of soleus
Plantaris
Tibia
Tibialis posterior 2nd layer of deep fascia Flexor digitorum longus
Tibial n. Posterior tibial a. 1st layer of deep fascia
b Popliteus
Gastrocnemius (medial head)
Tibial n. Tibioperoneal trunk
Nerve to flexor digitorum longus
Posterior tibial a. and v. Tibia
Medial pier of soleus arch divided and turned back
Tibial attachment soleus turned back Peroneal a. and v.
Plantaris tendon
Tibialis posterior
1st layer of deep fascia
Flexor digitorum longus
Posterior tibial a. and tibial n.
Operating on Peripheral Nerves Fig. 7.54 Exposure of the plantar nerves through a medial incision (After Henry 1975). (a) Retraction of abductor hallucis reveals the medial plantar nerve. (b) Release of the tendon of the flexor longus hallucis gives access to the whole of the deep compartment of the sole.
275
a
Navicular tubercle
Master Knot of Henry
Long saphenous v. Flexor hallucis longus Flexor hallucis brevis Flexor digitorum longus
Medial plantar bundle
Flexor digitorum brevis
Abductor hallucis Lateral plantar bundle
Navicular tubercle
b
Divided Master Knot of Henry
Tibialis posterior
Short plantar ligament Tibialis anterior tendon Flexor hallucis brevis Flexor hallucis longus
Lumbricals Abductor hallucis
Peroneus longus tendon
7.8.1 The Thoracic Outlet Syndromes Wilbourn (1993) described thoracic outlet syndrome as one of the most confusing and certainly the most controversial of all peripheral disorders: “The confusion exists mainly because many physicians are unaware that several different entities are grouped under this one heading, entities that have little in common beyond their known or presumed lesions site”. Narakas (1993) described more than 18 different types of compression syndromes arising between the cervical spine and the inferior margin of pectoralis major and he proposed the general term “syndrome de la traversée cervico-thoraco-brachiale”. This
Lateral plantar bundle
Flexor Flexor Long Quadratus Medial digitorum plantae plantar digitorum plantar ligament brevis bundle longus
concept has been developed by Poitevin (2005a, b) who introduced the term “upper limb inlet” to include complete and incomplete tunnel like spaces spanning from the axial skeleton of the thorax and cervical spine to the lower border of the pectoralis major muscles. Poitevin (2005a) writes: “The scalene complex is a common mass running from the cervical vertebrae and reaching the first and second ribs. During development, this mass is fragmented in a variable fashion. As a result of this fragmentation, muscular bridges remain joining anterior and middle scalene muscles, passing in between the neurovascular bundles. These muscular bridges sub divide and reduce the interscalene space which is altered or diminished by
276
variations in the shape, the number and the extension of the insertions of scalene muscles and also by the presence of a cervical rib”. The lower part of the brachial plexus and the lower trunk in particular, joined at the first rib by the subclavian vessels, run a regular obstacle course on their way to the lower borders of the pectoralis minor muscle. Some of the more important obstacles are now described
Surgical Disorders of the Peripheral Nerves
Rhomboid lig.
Scalenus medius
Scalenus anterior
V
7.8.2 The Suprapleural Membrane (Sibson’s Fascia)
A Subclavius
In its pure form, the suprapleural membrane is a thin, fanshaped structure attached apically to the tip of the seventh cervical transverse process and peripherally to the medial edge of the rib. The extent of its peripheral attachment varies. The deep part of the membrane may be quite tough and aponeurotic. The scalenus minimus is the muscle that sometimes replaces the deep or posterior part of the membrane. The muscle may be up to 10 mm in diameter in its widest part, but is usually much smaller than that. The posterior border of the membrane forms, with the edge of the first rib, the foramen through which the first thoracic nerve escapes. The membrane covers the apical pleura and intervenes between it and the subclavian artery.
7.8.3 The First Rib The broad first rib, whose surface is more horizontal than vertical, extends from the head at the vertebral body, the neck and the costotransverse articulation, over the apical pleura to curve round to the costo-chondral junction and then to the manubrium (Fig. 7.55). The apical pleura is, in the healthy state, easily separable from the head and neck and most of the underside of the first rib. Anteriorly, it becomes rather adherent to the underside of the rib, in relation to the costochondral junction. The scalenus medius is attached to the upper surface of the rib posteriorly. The upper part of nerve to the serratus anterior is usually formed in the muscle by branches from the fourth, fifth and sixth cervical anterior rami; it emerges from the lateral surface of the scalenus where it is usually joined by the contribution from the seventh cervical nerve. In front of the scalenus medius the surface of the rib is, in the adult, grooved by the passage of the first thoracic nerve and lower trunk. There is a distinct tubercle at the attachment of the scalenus anterior, behind which is the slight depression for the subclavian artery. In front of the scalenus tubercle the subclavian vein runs over the upper surface of the rib. The intercostal muscles are attached to the
Serratus anterior
Scalenus posterior
Fig. 7.55 The upper surface of the left first rib, showing the sites of the arterial and venous crossings and the muscular attachment (After Frazer 1920).
edge of the rib. The lower part of the cervicothoracic (stellate) ganglion lies on the head of the first rib, deep to the proximal part of the vertebral artery arising from the subclavian. The major part of the first thoracic nerve is at first below the rib, then medial to its edge. The white pre-ganglionic ramus of the first thoracic nerve connects it to the stellate ganglion; the grey ramus (post-ganglionic) is less noticeable. The size and disposition of the scalenus anterior varies. It may bifurcate to include the artery, the muscle itself may be bulky and the tendon may curve posteriorly round the artery to form a kind of snare. Deep to the scalenus, a very thin fascia covers the artery. The phrenic nerve, chiefly derived from the fourth cervical anterior ramus, curves round the muscle to run down in front of it into the thorax, lying at first behind the internal jugular vein and crossing behind the subclavian vein. On the left the thoracic duct and on the right the right lymphatic duct enter the brachiocephalic (innominate) vein at the subclavian-jugular (tri-radiate) junction.
7.8.4 The Seventh Cervical Rib Parry and Eastcott (1992) and Narakas (1993) credit the second-century physician and anatomist Galen with the first description of the seventh cervical rib. A little later, Gruber (1869) published his classification of such ribs according to their size and connections. Narakas (1993) indicates an incidence of seventh cervical ribs of 0.004 to 1% and he says that cervical rib “is asymptomatic in 90% to 95% of persons”. The extent of the rib, which is rarely bilaterally symmetrical, ranges from a prolongation and pointing of the seventh cervical transverse process to a complete rib in all respects like a
Operating on Peripheral Nerves
277
TYPE I
C7
T1
Omohyoid TYPE II
TYPE III
C7
C7
T1
T1
TYPE IV
TYPE V
C7
C7
T1
T1
Fig. 7.56 Variations of the seventh cervical rib, ranging from elongation of the seventh cervical transverse process to a virtually complete additional rib (After Sargent 1921).
first thoracic rib. The classification proposed by Sargent (1921) is illustrated in Fig. 7.56. Arterial Thoracic Outlet Syndrome This was the first of the syndromes to be recognised and the first to be treated surgically. Cooper wrote (Cooper and Travers 1817): “I have
seen an exostosis arise from the 6th or 7th cervical vertebrae or perhaps from both”. He went on to describe the case of a woman who came into Guy’s Hospital with no pulse at the wrist or elbow and with “venous redness” of the upper limb, which was cold, “enumbed” and painful. There was a projection of the lower cervical vertebrae towards the clavicle, with consequent pressure on the subclavian artery. Coote (1861) reported the removal of a left seventh cervical rib causing aneurysm of the subclavian artery. Two physicians, Lewis and Pickering (1934) pointed to the mechanism of production of the arterial lesion – namely, intimal breakage with local thrombosis and distal embolisation. Arterial affection occurs as the subclavian artery crosses a “normal” first rib or a seventh cervical rib, commonly one with bony union to, or synchondrosis or synostosis with, a boss on the first rib at about its middle part. The vessel is distorted by the bone and constricted by a leash formed by the tendon of insertion of the scalenus anterior. There is commonly dilation of the artery distal to the point of constriction: that dilation may proceed to the formation of a true aneurysm (Wickham and Martin 1962) Complete obstruction of the main vessel with distal embolisation may lead to critical ischaemia. The most dangerous complication is that demonstrated by Symonds in 1927, that of contralateral hemiplegia, doubtless from embolisation from thrombus extending proximally to the carotid vessel. The arterial form of this outlet syndrome is a condition predominantly affecting young women; the occurrence of such symptoms in older persons must suggest a primary diagnosis of atherosclerosis or other systemic disorder . The symptoms are initially those of a Raynaud phenomenon with episodic blanching and temperature change; episodic pain with muscular activity; distal ulceration and necrosis. The extremity may be cooler than the unaffected limb; there is likely to be a pulse difference, even the absence of a pulse. Even if the hands are the same colour at rest, slightly dependant, there is likely to be blanching of the affected hand on elevation. The artery is prominent, displaced upward and anteriorly by the underlying rib. There is a loud bruit, there may be a thrill. Doppler ultrasound examination is likely to show abnormality: Parry and Eastcott (1992) named principally the finding of a clamped, monophasic velocity signal anywhere over the arterial tree of the upper limb. Radiographs will, of course, show the seventh cervical rib if one is present. Magnetic resonance angiography and duplex ultrasonography, have generally superseded arteriography in this and in other fields. These confirm the site and extent of the obstruction, the extent of stenosis and aneurysm formation, and the degree of distal occlusion (Fig. 7.57). Operation is required urgently in cases of critical ischaemia, and soon in cases in which ischaemia is threatened. It may be indicated when a cervical rib is present, causing no symptoms, but producing deformity of the subclavian artery
278
Surgical Disorders of the Peripheral Nerves
Fig. 7.58 Appearance at operation in two cases. Above: showing thrombosis, stenosis and post stenotic dilatation of the subclavian artery, with peri-arteritis (and distal embolisation) in a girl with a cervical rib. Below: another case of thrombosis in a young woman. The right subclavian artery has been opened to show the detached intima.
ing four stages of arterial affection in the presence of a seventh cervical rib. Their recommendations are as follows
Fig. 7.57 Angiographic appearances. Above: complete occlusion over cervical rib. Middle: complete occlusion at the outlet in a 40 year old woman, a heavy smoker. Below: obstruction of the artery at the outlet with elevation of the limb.
sufficient to cause a steady bruit over the vessel. Such a bruit must surely indicate distortion of the vessel which is likely, in time, to cause intimal breakage and thrombosis (Fig. 7.58). In cases with clear signs of arterial affection it is rarely sufficient simply to remove the seventh cervical or first rib and to rely on the decompression so afforded. Haug and Sanders (1991) follow Scher and colleagues (1984) in defin-
Stage 0. Rib(s) with no symptoms and no aneurysm should be left alone, though the presence of a bruit is an indication for angiography. Stage 1. Rib(s) with minimal stenosis and mild post stenotic dilation are treated with arterial decompression by resection of the abnormal rib. Stage 2. Rib(s) with aneurysm, intimal damage and mural thrombosis should be treated by resection of the rib and excision and repair or ligation and bypass of the aneurysm. Stage 3. Rib(s) with distal embolisation require distal embolectomy and cervico-thoracic sympathectomy in addition to rib resection and arterial recon struction. Parry and Eastcott (1992) recommend endarterectomy or resection and anastomosis for short stenoses, strip resection
Operating on Peripheral Nerves
for a large aneurysm, and saphenous vein grafting for more extensive lesions. A long by-pass graft may be necessary. We have no personal experience of such cases, but have seen critical ischaemia relieved by this intervention. The arterial thoracic outlet syndrome is uncommon rather than rare and it is potentially dangerous. The bruit from a prominent subclavian artery associated with underlying seventh cervical rib, cannot be ignored. Investigations need to be pursued to define the extent of distortion of the artery. The Neurogenic Thoracic Outlet Syndrome This is rare. As Gilliat (1984) pointed out a number of cases were seen at the National Hospital for Nervous Diseases, Queen Square, in the early years of the last century, with wasting of the thenar muscles which were attributed to a cervical rib. In fact, these patients probably had a carpal tunnel syndrome. The confusion was not resolved for more than 40 years until the application of nerve conduction studies defined the much more common disorder. Gilliatt put the annual incidence, in the general population, of the cervical rib syndrome with muscle wasting, as low as one in one million. The patients are usually young to middle aged women. The abnormalities are confined to one upper limb in the great majority. The symptoms are insidious, and often mild, so that presentation is delayed for some years, usually by the sudden realisation that the hand is wasted. There is pain in the medial aspect of the forearm. There may be some blunting of sensibility in that area and in the little finger which is frequently shorter than it is in the normal hand, suggesting that the affliction of the fibres in the lower trunk commenced before skeletal maturity. There is wasting of the superficial thenar muscles which sometimes extends to all of the small muscles of the hands (Fig. 7.59). There may be weakness of the flexor muscles of the forearm, most evident in those innervated by the ulnar nerve. We detected a bruit over the subclavian artery in 14 of our 38 cases of true neurogenic thoracic outlet syndrome. Radiographs of the neck show rudimentary cervical ribs or elongated transverse processes of the seventh cervical vertebrae. The diagnosis is confirmed by the neurophysiological evidence described in Chapter 6. Operation is indicated in these patients in the expectation of alleviation of pain, improvement in sensation and some improvement in strength but recovery of the wasted small muscles of the hand is rare. Neural Outlet Syndrome without plain motor signs. This is much more difficult. A large variety of symptoms has been attributed to compression at the outlet in addition to pain, paraesthesiae and numbness affecting the ulnar side of the forearm and hand, and circulatory disorder of the hand. We think that the clinician, having excluded causes outside the thoracic outlet, should first attempt to identify conditions which are or could be precursors of the “true” neural syndrome. Here is a typical example. A woman in her twenties or thirties presents with paraesthesiae affecting the ulnar side
279
Fig. 7.59 The classical neurological outlet syndrome: shortening of the little finger is often evident. Above: Severe wasting of the intrinsic muscles of the right hand in a 35 year old woman. Below: the right lower trunk is seen angulated over a “scalenus sickle”. There is extreme attenuation of the trunk, which, in the photograph, appears to be rather lateral to the sickle. When the nerve was first exposed at operation, the zone of attenuation was, in fact, at the site of angulation. There was slow improvement after decompression.
of her forearm and hand and mild circulatory changes in the hand. There is tenderness over the lower part of the plexus above the clavicle on the affected side but none on the unaffected side. There is an impression of fullness in this site of tenderness; tapping produces radiated paraesthesiae. There is a bruit over the subclavian artery at rest on the affected side but not on the unaffected side. The radial pulse on the affected side is labile, disappearing with bracing of the shoulder or elevation of the limb. Radiographs show elongation of the seventh cervical transverse process or some degree of seventh cervical rib. Magnetic resonance imaging shows distortion of the lower part of the brachial plexus (Panegyres et al. 1993). Electrophysiological examination shows F-wave abnormalities in the ulnar nerve. Lucky is the clinician who is presented with so full a picture! But it does happen and when it does, a period of “conservative” treatment is given,
280
Surgical Disorders of the Peripheral Nerves
recommend a first operation on the thoracic outlet; even the boldest may recoil from the prospect of a second operation in a scarred field. It may however be inevitable when (1) the primary operation limited to scalenectomy has plainly for a time relieved symptoms; (2) there are true signs of recurrent interference with structures at the outlet; (3) damage to nerves has been inflicted during the primary operation.
7.8.5 Considerations About Operation
Fig. 7.60 Entrapment of the lower trunk through abnormality of the suprapleural membrane and scalenus medius. Above: the right lower trunk is trapped between the aponeurotic edge of the scalenus medius posteriorly and the suprapleural membrane anteriorly. Below: a “scalenus sickle” and an anomalous scalenus anterior. The left lower trunk passes over the edge of the sickle and behind the part of the scalenus anterior that passes behind the subclavian artery.
but is no more than a formality, and operation is recommended when that treatment fails (Fig. 7.60). Far more difficult is the case in which the symptoms are more diffuse and the signs less clear. The finding of a bruit over the subclavian vessel on the symptomless side as well as on the affected side is at least discouraging, as is lability of the radial pulse on the unaffected side. Failure to gain support from electrophysiological evidence, from arteriography or from magnetic resonance imaging should encourage clinicians to give “conservative” methods a trial. Most difficult of all is the case in which operation, usually simple section of the scalenus anterior muscle, has been undertaken on a largely speculative basis and has at first succeeded, but then symptoms have recurred. All who operate regularly at the thoracic outlet know well the extraordinary amount of scarring which can follow even so simple an operation as anterior scalenotomy. No one should lightly
Three routes are available for access to the thoracic outlet and neurovascular bundle: (1) Anterior supraclavicular; (2) Transaxillary; (3) Posterior. The anterior route offers the best exposure of the plexus, the artery and the sympathetic chain, and of the proximal structures likely to distort or compress them. Its disadvantages are the difficulty of access for removal of the first rib and the restricted access to the subclavian vein. The incision is placed just above the clavicle. It is important that division of the scalenus anterior should be complete and should include the fascia at the back of the muscle, so as to permit display and later mobilisation of the subclavian artery. The suprapleural membrane is now visible. It is defined and divided so that the apical parietal pleura can be pushed away from the sides of the vertebrae and the inside of the chest wall. Once the vessel is mobilised, it is possible to display the plexus from the fifth nerve just lateral to the scalenus anterior to the first thoracic nerve winding round the medial side of the posterior part of the first rib to join the eight cervical nerve and form the lower trunk. At this stage it should be possible to see or feel any distortion or compressing agent such as a scalenus minimus, a scalenus medius band or a seventh cervical rib. A scalenus minimus or an aponeurotic thickening of the posterior edge of the suprapleural membrane should be divided. A scalenus medius band should be removed as follows. First, it is carefully cut at its attachment to the first rib below the first thoracic nerve then it is traced up behind the lower trunk, and between the eighth and seventh cervical nerves, to its attachment to the seventh cervical transverse process. Its deep surface is separated from the muscular part of the scalenus medius, its attachment to the transverse process is divided and it is removed. If the underlying muscle appears to be distorting the lower part of the plexus, its prominent part should also be removed. The question will now arise as to whether the decompression effected is adequate. Before the wound is closed the surgeon should certainly put a finger between the first rib and the clavicle and then manipulate the upper limb to make sure the finger is not trapped in any position of the shoulder girdle. If there is clear evidence of costo-clavicular compression, the first rib should be removed since the
Operating on Peripheral Nerves
opportunity to do this with relative ease will not occur again. If there is no such compression the rib should be left alone. Removal of the seventh cervical and first thoracic ribs. It is safer to remove the rib(s) piecemeal, dividing them into sections. Useful instruments include a 5 mm spinal punch, and bone nibblers angled to 60°. The seventh cervical ribs should be removed from the level of the tip of the transverse process to the attachment to the first rib wherever the latter may be and in whatever manner it may be formed. The posterior end is best reached between the seventh and eighth nerves; its middle between the eighth and first thoracic nerves, and the lower end below the first thoracic nerve, either behind or in front of the artery. Any articulating bony boss on the first rib should be removed with the attached tendon of scalenus anterior. Particular care has to be taken with the removal of the anterior part of the first thoracic rib since the subclavian vein can be nipped, and the pleura is often adherent here to the under surface of the rib. The rib is divided near the costo-chondral junction or separated from the costal cartilage. The importance of the extra-periosteal removal of ribs is stressed. The wound is closed over a suction drain. A few stitches reunite the fatty areolar tissue, the omohyoid muscle is repaired, and the platysma and skin are closed. A subcuticular suture is neat, but a well-placed and well-sutured wound heals well whatever type of suture is used. Dangers of operation: The spinal nerves and the trunks of the plexus are displaced forward and laterally by the abnormal rib(s) and are at risk during the exposure. Sometimes the first thoracic nerve is so attenuated over a scalenus medius band or seventh cervical rib that the surgeon may fail to see it and may accidentally divide it. It is difficult to avoid damaging the phrenic nerve by traction, so producing a temporary paralysis of the hemi-diaphragm. The subclavian artery is sometimes fragile, even when symptoms are predominantly neural. The most serious immediate danger is that of wounding the subclavian vein behind the clavicle. It may be difficult to effect repair in this restricted space because of the difficulty of achieving control. In these circumstances it may be necessary, while bleeding is controlled by pressure, to extend the exposure by elevation of the manubrio-clavicular flap. It is in the region of the vein and tri-radiate junction on the left side that the thoracic duct may be injured especially in a scarred field. This damage is usually recognised because of the leak of milky fluid. It is best to identify and repair the leak, because if the pleura has been opened there is a real danger of formation of a chylothorax. If the hole cannot be repaired, it can be plugged with a muscle graft. Damage to the pleura is a tiresome but sometimes unavoidable complication usually necessitating the introduction of an intercostal catheter connected to a water seal. Sometimes the hole can effectively be repaired, the air aspirated and the lung expanded, and the necessity for pleural drainage avoided.
281
The principal delayed complication of operation is the development of a painful neuropathy, with extensive scarring around the nerves In a proportion of cases, even when the utmost care is taken to avoid rough handling of the plexus, a painful and persistent neuropathy develops, resistant to treatment with analgesics and in particular to further non specific exploration for decompression. It is this complication which should be present in the mind of every surgeon about to propose operation for thoracic outlet syndrome, and in particular for an outlet syndrome not associated with a clearly defined lesion. The transaxillary approach is admirable for access to the subclavian/axillary vein and the middle and anterior parts of the first rib. Its use is of course mandatory on the rare occasions when there is distal obstruction in the axilla. We have never found it easy; we have always found access to the posterior part of the first rib is limited and we are not alone in discerning the risk to the plexus imposed by the position of the limb. Roos’ (1981) detailed description of the procedure is required reading. It is an admirable approach for a second operation following recurrence after a supraclavicular procedure. It is the obvious approach for relief of costo-clavicular compression causing predominantly “venous” symptoms. The positioning of the patient is very important. Best is a lateral position, with the table slightly rolled towards the surgeon. The assistant holds the affected limb at an angle of 90° to the axial line avoiding lateral rotation, and should be prepared to relax the position for a few minutes every 15 min. The incision is in the skin crease at the level of the third rib, just below the hair line. The flaps are raised and the chest wall and neurovascular bundle are displayed. As with all transaxillary procedures, it is as well to define early with the stimulator the position of the nerve to serratus anterior in the posterior part of the wound. The intercosto-brachial nerve running down the axilla is seen and preserved. Later, the superior thoracic artery and vein are tied and divided to allow access to the upper part of the axillary tunnel. With swab dissection the operator follows the chest wall, under the neurovascular bundle, into the apex of the axilla until the lateral edge of the first rib is felt. It is as well at this stage to verify identification of the rib on the screen: the second rib has been mistaken for the first. Roos has said that with abduction of the shoulder the axilla opens up like a oyster. He may perhaps, in writing this, have forgotten that for the unpractised hand opening an oyster is very difficult. Access does indeed depend critically on abduction of the shoulder, but the surgeon must at all times bear in mind the limits of tolerance of that position by the nerves of the plexus. We have found Deaver retractors useful, but Roos (1981) prefers the Heaney vaginal retractor as being less liable to damage the vein and muscles. The first rib, whose lateral margin is at first the only part easily visible, is cleared by swab dissection to reveal the
282
upper surface crossed by the first thoracic nerve and subclavian artery and receiving the attachment of the scalenus anterior. In front of this is the crossing of the subclavian vein. Anterior to the vein is the tendon of the subclavius muscle. Laterally, the edge of the first rib receives attachment of the intercostal muscles. Once these have been cleared from the rib, the apical pleura can be separated from its lower surface. The tendons of the subclavius and the scalenus anterior muscles are divided and the nerve, artery and vein are very carefully mobilised from the upper surface of the rib. The scalenus medius muscle is cleared from the upper surface of the posterior part of the rib, then the suprapleural membrane can be freed from the medial surface of the rib. In these proceedings the Howarth dissector and the small Doyen raspatory are useful. The aim is to remove the rib extra-periosteally from, anteriorly just in front of the subclavian vein to, posteriorly, behind the crossing of the first thoracic nerve. The anterior division is the easier: with the vein held up and the pleura depressed, the rib can be divided with fine bone cutters or simply separated from the costal cartilage. The posterior division is more difficult, because of the presence of the first thoracic nerve. This can be held and pushed out of the way during the division of the bone - usually, just distal to the costo-transverse joint. The rib is strong here: it can rarely be divided with a single bite of the bone cutters. Bleeding is checked before closure. The wound comes together very well with the lowering of the limb. The subcutaneous fascia and skin are closed over a suction drain; if the pleura has been opened, an intercostal catheter should be added. Dangers and complications: The chief immediate danger is that of damage to the subclavian vein. Bleeding is very severe. It must be checked with local pressure until the vein is sufficiently accessible proximally and distally to permit isolation of the segment and repair of the tear. The lesion most often suffered by the plexus during the operation is that inflicted by prolonged abduction and lateral rotation, neither of which should be permissible. It must be remembered that the first thoracic nerve is very vulnerable during the exposure and division of the posterior part of the rib. The posterior approach gives the best access to the posterior ends of the first rib and a seventh cervical rib, and allows the operator to separate the first thoracic nerve from the neck of the first rib. On the other hand, access to the vessels, the anterior part of the first rib and the distal part of the plexus is restricted. We have found it too difficult and too traumatic and bloody to be suitable as a primary route for the removal of a seventh cervical or first thoracic rib. Pain is common after this operation, doubtless because of the amount of muscle cutting and retraction necessary. Neural, vascular and pleural complications are rare. We prefer to reserve this approach for the removal of a remnant of the first rib or a seventh cervical rib left after a previous operation by the anterior route.
Surgical Disorders of the Peripheral Nerves
7.8.6 Carpal Tunnel Syndrome It was the paper by Brain, Wright and Wilkinson (1947) that opened British eyes to the existence of the condition and to the benefits of treatment by operation. Brain and his colleagues did not deal only with clinical features and results of operations; they also made measurements of variations of pressure in the carpal tunnel. It is interesting to note that those measurements showed that extension of the wrist produced a rise of pressure in the carpal tunnel much greater than that produced by flexion. Since 1947 the carpal tunnel industry has expanded enormously: so too has the understanding of the underlying causes. However, practice has tended to run ahead of the systematic search for an underlying cause. Decompression is, after all, superficially a simple, and generally successful operation, but the clinician must bear in mind the possibility of a systematic cause for neuropathy. Kretschmer and his colleagues (2009) provide an important and salutary account of the risks of endoscopic carpal tunnel release. Our experience with the method is confined to the treatment of its complications in 38 patients.
7.8.7 Technique of Operation (Open Method) General anaesthesia is necessary after failure of a primary operation. If a bloodless field is used the cuff should always be released and bleeding checked before closure. We rely on the effect of elevating the hand on pillows and prefer local anaesthetic. The incision should be planned to avoid damage to the palmar cutaneous branch of the median nerve. It starts proximal to the wrist crease, just to the ulnar side of the mid line. The median nerve is identified before it passes deep to the flexor retinaculum. This enables detection of aberrant branches, and avoids damage to the nerve if it has become adherent to the under surface of the retinaculum. The retinaculum is divided from proximal to distal under vision and the nerve, tendons and synovial membranes are exposed. The state of the nerve, tendons and synovial membrane is recorded. If there is any doubt about the condition of the synovial membrane, a piece is removed for histological examination. If a cuff is used, it is released at this stage and particular note is taken of the reperfusion of the nerve. Bleeding is checked and the skin is closed. We have not thought it necessary to resort to ligamentoplasty. There is no case for internal neurolysis and external neurolysis is only done if the nerve is adherent to adjacent structures. After the closure a padded dressing is applied. This is supplemented with a plaster slab holding the wrist in 20 degrees of extension. This is removed at about 48 h after operation (Fig. 7.61). The ulnar nerve at the elbow The diagnosis of focal ulnar neuropathy at the elbow may be easy, as in the case of a
Operating on Peripheral Nerves
Fig. 7.61 Release of the carpal tunnel. Our preferred incision is shown above. The operation is done using infiltration with local anaesthetic usually without tourniquet. The incision is placed to protect the palmar cutaneous nerve and it extends proximal to the wrist crease where the median nerve is identified. Middle: A tourniquet was used in this case and the nerve “flushes up” after release of the cuff. Note the zone of
283
constriction formerly deep to the retinaculum. Below: The wound is dressed with paraffin gauze and a plaster of Paris splint is applied holding the wrist in 20 degrees of extension. This reduces discomfort and also the risk of prolapse of the nerve. It is removed at 48 h after operation.
284
patient in his thirties with serious deformity of the elbow following previous injury, with motor and sensory signs in the distribution of the ulnar nerve, with tenderness of an ulnar nerve thickened at the elbow, and with marked slowing of conduction along the nerve at the elbow. It may be difficult, as in the case of a man in his sixties with weakness and wasting of the small muscles of both hands and with marked degenerative changes at the cervical spine. An easily made, though rare error, is to confuse “snapping” of an enlarged or abnormal medial head of triceps over the medial epicondyle with abnormal mobility of the ulnar nerve (Spinner and Goldner 1998). The more important and much more common difficulty is to determine the level or even the levels at which the nerve cell is being compressed, distorted or otherwise damaged. Is it in the anterior horn cell, in the spinal canal, in the intervertebral foramen, at the thoracic outlet, at the elbow or in the wrist or hand?
7.8.8 Operations for Ulnar Neuropathy at the Elbow Simple Decompression. (Osborne 1970), is adequate in most cases. Nathan et al. (1995) showed good or excellent relief after decompression of 164 nerves in 131 patients. At an average of 4.3 years good relief was maintained in 79%. Their predictors of good outcome were: absence of subluxation after operation, greater body weight and normal two point discrimination before operation. The ulnar nerve is displayed through a curved postero medial incision running distally from above the level of the medial epicondyle. A posterior branch of the medial cutaneous nerve of forearm (MCNF) seems always to be in the way and this must be seen and preserved. The arcuate ligament is divided and the fibres of the flexor carpi ulnaris are separated. The nerve is inspected before and after release of the cuff if this has been used, and the area is examined to exclude the possibility of a compressing agent, such as a ganglion. The cuff is released and haemostasis secured. Osborne (1970) sutured the arcuate ligament deep to the ulnar nerve to provide a bed for it, but we have not followed this practice. The skin is closed. There is no need for immobilisation; indeed, immobilisation can lead to tethering and to later trouble. Transposition. Transposition is necessary when there is severe deformity, extreme osteophytosis, recurrent subluxation over the medial epicondyle or recurrence of signs of nerve compression after simple decompression. Both the surgeon and the patient must recognise the limitations and hazards of this procedure. A bloodless field and general anaesthesia are necessary. The long postero-medial incisions starts 10 cm above the medial epicondyle and ends 10 cm below it. The anterior skin flap is raised in its full thickness. Special care must be taken to identify and preserve the posterior terminal branch of the medial cutaneous nerve of
Surgical Disorders of the Peripheral Nerves
forearm, which runs obliquely in an awkward position, distal to the medial epicondyle. Now the flexor origin and the undisplaced ulnar nerve are displayed. In the next step the ulnar nerve is mobilised, with its extrinsic axial vessels, from about 8 cm above to 6 cm below the epicondyle. Above the epicondyle by division of the overlying fascia; below it, the arcuate ligament has to be divided and the two parts of the flexor carpi ulnaris are separated. In the lower part the uppermost muscular branches are displayed and mobilised, either with the main nerve or away from it. Now, the medial intermuscular septum is excised. Subcutaneous transfer is recommended in most cases. The nerve is lifted from its bed and re-routed in front of the medial epicondyle and the flexor origin, care being taken to avoid kinking or obstruction at either end of the altered course. The cuff is deflated and bleeding is controlled. The two heads of the flexor carpi ulnaris can be brought together above the point of re-entry of the ulnar nerve, but there must be no constriction at that point. It is quite a good plan to put two or three sutures between the subcutaneous fat and the underlying aponeurosis of the flexor origin in order to retain the nerve in its new course. The skin is closed over a suction drain. In order to lessen adhesions around the transferred nerve, gentle movement of the elbow is encouraged on the day of operation. We identify three chief causes of failure of primary operation in 85 cases coming to reexploration as (1) inadequate resection of the medial intermuscular septum through an inadequate incision, (2) fibrosis induced by haematoma leading to tethering and constriction of the nerve and (3) damage to the medial cutaneous nerve of forearm (Fig. 7.62). We are sure that there is no place for intramuscular transposition and, increasingly, whether there is a case for deep muscular transposition. We have seen constriction, adhesion and distortion of the nerve in the sub muscular bed, just as we have seen them in the intramuscular bed. However, Lowe and MacKinnon (2004) provide a great deal of evidence to support their preference for submuscular transposition and clinicians considering this method should read their chapter before embarking on operations. The oblique course of the terminal part of the posterior branch of the MCNF is particularly well illustrated. Dellon and Coert (2004) provide a well illustrated account of the operation.
7.8.9 The Less Common Entrapment Syndromes – Upper Limb It is with some of the less common “entrapment neuropathies” that we see real difficulties in diagnosis. Nagano (1987) casts a shadow over the concept of entrapment of the anterior interosseous nerve by revealing, at electromyography, denervation in muscles beyond the territory of that nerve. The shadow is deepened by Seki and his colleagues
Operating on Peripheral Nerves
285
Fig. 7.63 Pelvic chondrosarcoma in a 17 year old girl who was initially treated for “neuralgia paraesthetica”.
7.8.11 Meralgia Paraesthetica
Fig. 7.62 Complications of anterior transposition of the ulnar nerve at the elbow. Above: A 43 year old man experienced worsening of pain after sub muscular transposition of the ulnar nerve. The nerve was found rigidly immobilised and constricted by dense fibrous tissue (white sling). A branch of MCF had been cut (red sling). After external neurolysis the nerve was placed subcutaneously in an unscarred bed. Below: A 28 year old man experienced worsening of pain and of symptoms after partial excision of medial epicondyle. The nerve was transposed into an intra muscular position at a second operation. His pain worsened. At third operation, the nerve was found tethered, immobilised, and angulated in scar tissue. The epineurial vessels were obliterated.
(2006) who describe fifteen cases of neuralgic amyotrophy presenting with affection of the anterior interosseous nerve. Nagano and colleagues (1996) describe another cause for anterior interosseous nerve palsy by demonstrating “hour glass like” constriction of the median nerve. Similar lesions have been seen affecting the posterior interosseous nerve. We have operated in only a few cases of “entrapment neuropathy” of either of these nerves and doubt that the intervention altered the natural course in any case.
7.8.10 Some Entrapment Neuropathies in the Lower Limb The difficulties facing the clinician in establishing the diagnosis are even greater in the lower than they are in the upper limb.
Entrapment of the lateral cutaneous nerve of the thigh at the level of the inguinal ligament has a long and distinguished history. It is in its character as a symptom of underlying disease that this “entrapment” is particularly important. Krikler and Grimer (1993) record the case of a pelvic chondrosarcoma presenting with “meralgia paraesthetica”, and indeed that was the mode of presentation to us in 1960 of a young girl with that affliction (Fig. 7.63). Dawson and colleagues (1990) rightly remark that “differential diagnosis consists of an anatomic exercise to establish whether the lateral femoral cutaneous nerve is the only structure affected”. The possibilities that must be excluded are (1) a general neuropathy; (2) a local manifestation of some other generalised condition; (3) a more proximal lesion of the second and third lumbar nerves; (4) a local condition affecting the lumbar plexus or the lateral cutaneous nerve within the abdomen. Perhaps in no other entrapment syndrome are there so many possibilities of missing an important and potentially serious underlying condition. We see this syndrome in its true form increasingly seldom. Our preference is for simple decompression, having sought unavailingly to relieve intractable pain in eighteen patients sent to us after section of the nerve.
7.8.12 Entrapment of the Pudendal Nerve Amarenco and colleagues (1988) first described the syndrome of entrapment of the pudendal nerve in “Alcock’s canal” or “perineal paralysis of the cyclist”. They described five cases, the first, bilateral, in a man of 38 who had
286
travelled “several hundred kilometres” on a cycle with a new saddle. Robert and colleagues (1989) made clinical, neurophysiological and anatomical studies and reported on the results of nine operations aided by the microscope. The pudendal nerve, arising from the second, third and fourth sacral nerves, leaves the pelvis through the great sciatic foramen and soon re-enters it through the lesser. In its course on the medial side of the ischium it may be trapped between the ischial spine or the sacro-spinous ligament and the sacro-tuberous ligament. Later the nerve or its branches may be trapped by the falciform edge of the ligament. Its vulnerability to external pressure during cycling is obvious; less so is a mechanism determining entrapment in the absence of such external pressure. Eight of Robert and colleagues’ (1989) patients were women; only one was a cyclist. Carlstedt (2008) has provided us with details of 20 cases of entrapment of the pudendal nerve successfully treated by operation. These patients were investigated by electromyography in conjunction with Dr Andrew Baranowksi (National Hospital for Nervous Diseases, Queen Square). Pudendal blocks were used which were followed by lasting relief in a few cases. Recurrence of symptoms after a successful block is a pre-requisite before advising operation. In most patients symptoms were caused by blunt injury, a fracture of the pelvis, or pelvic surgery. The patient is placed prone and the nerve is approached by splitting the gluteus maximus and identifying the inferior gluteal pedicle where it emerges below the piriformis muscle. The sacro-tuberous ligament is divided and the pudendal nerve is found by swab dissection through areolar tissue where it traverses the sacrospinous ligament. The ligament is now divided. Carlstedt has observed that the nerve seems to be stretched over the ligament and that it may be tethered by scar although no case of direct damage to the nerve has been encountered. The results have been gratifying. One patient who had endured severe pain for more than 12 years after operation for prolapse of the uterus, reported complete and lasting relief of her pain which had disappeared by 24 h after operation.
7.8.13 The Piriformis Syndrome The sciatic nerve, with the inferior gluteal and pudendal nerves and the inferior gluteal vessels, leaves the pelvis through the lower part of the greater sciatic foramen deep to the piriformis muscle running from the inner face of the bony pelvis to the great trochanter. The nerve passes over the tendon of the obturator internus muscle with the gemelli. There is an obvious liability to entrapment with hypertrophy or excess activity of the piriformis muscle. Stewart (2000) defines criteria for the diagnosis of this rather nebulous syndrome: the symptoms and signs arise from a focal lesion of
Surgical Disorders of the Peripheral Nerves
sciatic nerve; neurophysiological investigations confirm a focal lesion; a mass lesion is excluded by magnetic resonance imaging or ultrasound scanning; appropriate findings at operation, and relief of symptoms by that operation. Benson and Schutzer (1999) described 14 patients, in whom symptoms were provoked by direct blunt blow to the buttock. The piriformis muscle was found thickened or fibrosed in all cases. The symptoms were abolished in 11 patients who were all able to return to work. From Bendszus and colleagues (2003) comes a well illustrated report of three patients with refractory sciatica in whom high resolution MRI revealed compression of the sciatic nerve in the buttock by varicose gluteal veins. Severe pain in the leg and buttock was provoked by sitting or lying on the affected leg and rapidly relieved by standing and walking. The pain was abolished by decompression in two patients, in the third, pain subsided with carbamazepime. In our single case, a young man came with pain, paraesthesiae and motor weakness. There was supporting electrophysiological evidence of a focal lesion of the sciatic nerve. At operation, there was entrapment at the level of the crossing of the sciatic nerve by the piriformis muscle, division of which was followed by complete and lasting relief. We have seen eleven patients in whom pain was not relieved, or was even worsened by operation.
7.8.14 The Tarsal Tunnel Syndrome It was probably Lam (1967) who introduced the concept of entrapment of the tibial nerve beneath the medial retinaculum at the ankle. We should state that we have no real experience of dealing primarily with the syndrome; only that of dealing with the ill effects of operation upon the basis of that diagnosis. Katirji and Wilbourn (2005) say: “nonetheless, this is probably the most over diagnosed of the various nerve lesions of the lower limb. Thus, most patients referred to our EDX laboratories with this diagnosis are shown to have problems other than distal tibial neuropathies……. On a yearly basis, we evaluate more patients for failed TTS surgery than we do for failed carpal tunnel syndrome surgery, although the latter operation is far more common”. On the other hand, Dellon (2004) provides extensive evidence showing that the operation, when carefully performed, for the correct indications, is valuable in the treatment of diabetic neuropathy by relieving pain, restoring sensation and preventing ulceration and amputation. MacKinnon and Dellon (1988) described the indications for operation and offer a detailed illustrated account of the procedure. We reserve our position in this matter. Plantar pain, paraesthesiae and numbness may indeed be caused by compression of the tibial nerve under the retinaculum, but the possibility of
Operating on Peripheral Nerves
a more proximal cause must be excluded, and the diagnosis should be supported by the results of electrophysiological examination. When operation is done, the possibility that there may be a tumour or ganglion near the nerve or a tumour in it must be borne in mind. In one of our cases the cause of “tarsal tunnel syndrome” was a synovioma occupying the medial plantar nerve at the level of the malleolus. In that case the pain and the numbness of the great toe had led to a primary diagnosis of multiple sclerosis. A schwannoma, neurofibroma, ganglion, aneurysm of tibial artery has been revealed at exploration. Our experience of dealing with the failures of operation for tarsal tunnel syndrome is very discouraging. Five of the most severe pain states we have ever seen followed such procedures. Errors at the first operation included four lacerations of tibial nerve, five wounds of the calcaneal nerves and three arterial injuries.
7.8.15 “Morton’s Metatarsalgia” Perhaps some clinicians do not always remember that to section a nerve must be followed unavoidably and inevitably by formation of a neuroma. We have seen 14 patients who were much worse off after excision of Morton’s “neuroma” in the sole of the foot. Claustre and Simon (1978) indicate a course less drastic than neurectomy: they follow Gauthier and Dutertre (1975) in practising simple division of the metatarsal ligament through a dorsal incision
7.9 Pitfalls in Operating on Tumours of Peripheral Nerves “The trouble about these tumours is that they are just common enough to come within the experience of almost all surgeons and rare enough to cause embarrassment in diagnosis and treatment.” (Seddon 1975c) Some common difficulties in the treatment of two benign tumours, the schwannoma and the neurofibroma, and one tumour like condition, the ganglion, and also the malignant tumour of nerve are now described.
7.9.1 The Solitary Benign Schwannoma (Neurolemmoma, Neurinoma) Schwannomas arise from Schwann cells anywhere in the peripheral nervous system including the third to the twelfth cranial nerves close to their origin from the brain stem. Bilateral vestibular (eighth nerve) schwannomas are pathognomonic of neurofibromatosis Type II, schwannomas are
287
common in neurofibromatosis Type I. Most, be they sporadic or an expression of neurofibromatosis Type II, show deletion of the neurofibromatosis Type II gene on chromosome 22 (Giannini 2005; Meltzer 2001; Fletcher 2001). The schwannoma is enveloped by a true capsule which consists of the perineurium of the nerve bundle of origin, surrounded by a condensation of the deepest layers of the epineurium. This allows excision of the tumour without damage to the parent nerve although the rare plexiform variant of the schwannoma may infiltrate between nerve bundles. Knight and her colleagues (2007) studied 234 schwannomas and they found that distribution between the sexes was equal. Eighty per cent of patients were aged between 30 and 69 years whilst the mean age of the patients was 45.2 years. The site of origin is set out in Table 7.1. Four patients presented with tetraparesis due to an intraspinal extension. Deep seated tumours in the posterior mediastinum and the lumbo sacral spinal nerves and plexus can grow to an impressive size. The average delay before diagnosis was 5.2 years in the 13 large schwannomas reported by Abernathy and his colleagues (1986). Nawabi and Sinisi (2007) described 25 patients with schwannoma of the tibial nerve treated in the Peripheral Nerve Injury Unit. Only three were diagnosed within a year of presentation. The median interval to diagnosis was 48 months. The delay before diagnosis was, in one case, 30 years! Eleven patients had been referred to a Pain Table 7.1 Solitary benign schwannomas: 340 cases 1975–2007. Upper limb Cranial nerves and sympathetic chain
5
Supraclavicular brachial plexus
96
Infraclavicular brachial plexus
21
Circumflex nerve
1
Musculocutaneous nerve
2
Median nerve
32
Ulnar nerve
45
Radial nerve
20
Cutaneous nerve
13
Muscle or bone
3
Total
238
Pelvis and lower limb Lumbar spinal nerves
3
Lumbosacral plexus
7
Femoral nerve
8
Sciatic nerve
18
Tibial nerve
36
Common peroneal nerve
18
Cutaneous nerves
6
Muscle or bone
6
Total
102
288
Surgical Disorders of the Peripheral Nerves
Clinic. Four had operations to decompress the tarsal tunnel, and two more went through lumbar discectomy. All patients complained of pain, which was usually felt in the sole of the foot and a Tinel like sign was detected in all patients. The clinical diagnosis was confirmed by magnetic resonance or ultrasound scanning. The most frequent presentation of schwannoma is a swelling, which is painful to pressure. Progressive neurological deficit occurs only when the tumour arises in a confined space, as in intradural, extramedullary extension, or when it is deep to the clavicle. Diagnosis in the limbs is straightforward. The lump is mobile from side to side but not in the vertical axis of the limb. Percussion induces painful paraesthesiae in the territory of the nerve of origin, similar to the Tinel sign. This finding is the single most useful sign in the diagnosis of schwannoma. At times the tumour causes severe pain, particularly when it arises in the sole of the foot or in the buttock and is exposed to pressure on walking or sitting (Table 7.2). The tumour, when exposed, forms an eccentric oval swelling, usually less than 5 cm in diameter with nerve bundles stretched over it. It may be multinodular or plexiform, infiltrating the nerve of origin, so that enucleation is impossible. Larger tumours within the body cavities, the buttock or the axilla, develop cysts and show degenerative changes of haemorrhage, calcification and hyalinisation. The appearance of the epineurial vessels coursing over the capsule of the tumour is quite characteristic; they are engorged and tortuous. The cut surface is yellowish, and in larger tumours, is cystic and may be blood stained. Rarely the tumour is semi fluid, of the consistency of toothpaste, and appears to infiltrate between nerve bundles. The diagnostic histological feature is the organisation of Schwann cells either into areas of compact bundles (Antoni-A tissue) or a less orderly arrangement of spindle or oval cells in a loose matrix (Antoni-B tissue). There may be an abrupt change from one to the other. Pallisading or compact parallel rows of cells forming Verocay bodies are features of Antoni-A tissue (Weiss and Goldblum 2001a). In larger tumours nuclei may appear hypochromatic, large, and multi lobed (Ancient schwannoma) and the diagnosis of malignancy may be made on the basis of these findings, as it may also be for those schwannomas formed exclusively from Antoni-A tissue (the Table 7.2 Benign solitary schwannoma: symptoms and signs in 198 primary referrals 1984–2004. Symptoms and signs Number of patients (%) Painful swelling
194 (98)
Tinel-like sign
155 (81)
Spontaneous pain
60 (31)
Alteration in cutaneous sensibility
53 (27.7)
Weakness
19 (10)
Rapid increase in size Drawn from Knight et al. 2007.
25 (13)
“cellular” variety). Schwannomas, particular cellular areas, strongly express the S-100 protein, and this is particularly useful in distinguishing the large retroperitoneal tumours from such soft tissue sarcomas as leiomyosarcoma (Folpe and Gowan 2001). Electronmicroscopy shows Schwann cells forming stacks and layers of long cytoplasmic extensions surrounded by basement membranes. Nerve fibres are sometimes seen in looser Antoni-B tissue (Fig. 7.64).
7.9.1.1 Treatment The incision must be adequate so that the nerves and vessels are properly displayed above and below the tumour. With the help of the loupe or microscope, the surgeon should make a longitudinal incision in the epineurium over the most prominent part of the tumour. In nearly all cases it will possible to develop a plane of cleavage between the tumour and the nerve bundles and remove the tumour whole, without disturbing conducting tissue. Sometimes a small bundle may have to be divided to permit removal. Bleeding is checked with bi-polar diathermy in all parts of the wound. Earlier biopsy increases difficulties by causing scar formation in the plane between the tumour and the nerve bundles (Fig. 7.65). Access is a particular problem for those tumours in the upper most part of the spinal canal, those lying ventral to the denticulate ligament and for those with large intra spinal extensions. In these, it may be safer to remove the intraspinal extension at the first operation before returning to excise the extraspinal component at a second operation (Fig. 7.66). Crockard and Bradford (1985) developed the transoral, transclival route to remove a schwannoma lying anterior to the cranio-vertebral junction. Their patient, who presented with a severe tetraparesis, made a full recovery. The transclavicular approach is particularly valuable in the treatment of tumours involving the lower roots of the brachial plexus with extension adjacent to it or within the canal (Fig. 7.67). Abernathy et al. (1986) approached giant sacral schwannomas either by an anterior abdominal exposure or by a posterior retrorectal transacral route. They attempted to preserve the first three sacral nerves and the pudendal nerves, and found that unilateral section of the third sacral nerve could be performed without incurring severe significant neurological deficit. The risk of inflicting damage on the nerve during a well executed operation, is low but it is wise to warn patients about these risks and about the steps that can be taken to diminish those risks and about what can be done in the event. A measurable defect in function was found in five of our cases after operation. In two of these, haematoma from epineurial vessels compressed the nerve. There was partial recovery after urgent decompression. In three more cases of tumours arising at the wrist or in the hand, removal of the bundle entering the tumour led to some loss of sensation or
Operating on Peripheral Nerves
289
Fig. 7.64 Histopathological features in benign schwannoma. Above left: Antoni-A and B tissue, palisading and Verocay bodies. Haematoxylin and eosin × 250. Above right: ancient changes. There is hyalinisation of the walls of the small blood vessels, haemorrhage, haemosiderin pigment and cystic change. Haematoxylin and eosin × 250. Below left:
S-100 positive staining of Schwann cells within a plexiform schwannoma. The myelinated nerve fibres were demonstrated by solochrome cyanin × 160. Below right: elctronmicrograph showing long interleaving Schwann cell processes × 2250 (Courtesy of Dr Jean Pringle).
of power. Sawada and his colleagues (2006) found that the risk of increasing neurological deficit by operation was greater in cases where there was already sensorimotor disturbance. There is a measurable risk of recurrence: Knight et al. (2007) found four cases out of a total of 234. In two of these the tumour was poorly encapsulated and “infiltrative”; excision was incomplete at first operation in two cases of intraextraspinal tumour. Malignant transformation is rare. Woodruff and his colleagues (1994) describe two cases, and reviewed seven “acceptable” reported examples.
exploration, but the risks to conducting tissue attaching to removal must be exposed to the patient. Spinner and his colleagues (2003) have clarified this rather mysterious disorder in a very helpful manner. A study of peroneal intraneural ganglia demonstrated the importance of the articular branch of the nerve showing that this is the pathway for synovial fluid passing from the proximal tibio-fibular joint into the trunk of the nerve. The “pedicle” is, in fact, the articular nerve. A characteristic cystic transverse limb of the articular branch has been defined as a pathognomonic sign for peroneal intraneural ganglia (Spinner et al. 2006a). Spinner and his colleagues extend this concept to ganglia at other sites including the shoulder (Spinner et al. 2006b) and the tibial nerve at the ankle (Spinner et al. 2007) (Fig. 3.37). Exploration will soon reveal the extent of the attachment of the ganglion to the nerve. If it is early clear that (1) the swelling is a ganglion and (2) the ganglion is intimately associated with the nerve, it is important to identify the pedicle (Parkes 1961) which is so dilated that the sheath is translucent, and follow this to the proximal tibio-fibular joint. The
7.9.2 The Intraneural Ganglion This is often the source of needless maiming. The diagnosis of ganglion of the common peroneal nerve must be borne in mind when a patient comes with a pain in the leg, a swelling in the region of the nerve and possibly weakness of the anterior tibial muscles. There is no alternative to the recommendation of
290
Surgical Disorders of the Peripheral Nerves
Fig. 7.65 Schwannoma of the upper trunk of the left brachial plexus. Top left: The swelling was mobile, tender, and percussion evoked painful “pins and needles” sensation into the forearm and thumb. Right: The
tumour exposed at operation. Note the central position, with uniform expansion of the nerve. Bottom left: The tumour during enucleation. There was no neurological deficit after operation.
pedicle is divided between transfixion sutures as close as possible to the joint which invariably leaves a defect in the joint capsule. The pedicle is excised and the main nerve decompressed by incision and expression of contents. Failure to identify and remove the pedicle will lead to recurrence. In such cases we have interposed flaps of fascia and fat between the joint and the main nerve.
and fibrous. The dilated and tortuous epineurial vessels so characteristic of the schwannoma are absent. Neurofibromas do not possess a true capsule and staining for S-100 protein is less intense and less uniform than in the Schwannoma. Light microscopy shows interlacing bundles of elongated cells with wavy dark staining nuclei interspersed with strands of collagen (Weiss and Goldblum 2001a) (Fig. 7.68). There are three chief reasons why neurofibromas are clinically significant. First, they may grow to a very large size causing pressure on adjacent structures; they may also extend along a spinal nerve into the spinal canal. The largest benign solitary neurofibroma that we have removed measured 22 × 8 × 7 cm. Next, the neurofibroma of a trunk nerve is not separable from nerve bundles, which become intimately involved within it. Enucleation is not possible. Finally there is a slight
7.9.3 The Solitary Neurofibroma The solitary neurofibroma arising in a major nerve trunk is easily recognisable. The tumours appear uniformly greywhite and translucent, and in consistency they are rubbery
Operating on Peripheral Nerves
291
Fig. 7.66 Dumb-bell tumour (schwannoma) in the cervical region. Upper left: Plain radiograph shows an erosion of pedicles and enlargement of the foramina. Upper right: Digital subtraction angiography shows distortion of the vertebral artery. Below: Magnetic resonance imaging shows the extent of the tumour at levels C2/3 and C3/4. The tumour was approached first by laminectomy of C4 toC6, allowing the intradural extramedullary component to be excised (Mr R Bradford, Royal Free Hospital). Six weeks later the remainder was excised by antero-lateral approach, and the spine stabilised with bone graft and plate.
but definite risk of malignant transformation which has been estimated at around 1% (Weiss and Goldblum 2001a). We have found that the large peripheral neurofibromas arose from slender nerves adjacent to the brachial plexus or major nerve trunks and their removal was relatively straightforward causing little loss of function. In four of the dumbbell tumours conduction in the spinal nerve had been lost before operation, so that removal did not add to neural deficit. In other tumours
arising within the cervical spinal nerves, the greater part of the tumour was removed, preserving displaced nerve bundles. Painful recurrence indicates malignant transformation. Case report: A 34 year old man presented with a palpable lump in the posterior triangle which had been present for at least 12 months. It had increased in size and was causing pain. There was no clinical evidence of neurofibromatosis nor any suggestive family history. The tumour appeared to
292
Fig. 7.67 Massive schwannoma in the mediastinum compressed the brachio-cephalic veins. It was successfully removed by the transclavicular exposure.
Fig. 7.68 Solitary benign neurofibroma. Above left: the matrix is fibrillary and there are comma shaped nuclei. Haematoxylin and eosin × 250. Above right: S-100 positive staining. Below left: a “cellular” neurofi-
Surgical Disorders of the Peripheral Nerves
arise from the first rib with no extension into the spinal canal. At operation the tumour (10 × 8 × 8 cms in dimension) arose from the first rib, distorting and stretching the whole of the plexus and accessory nerve and the subclavian artery, indenting the pleura but not involving it, and merging into the scalene muscle. The tumour and rib were removed, with affected soft tissues. The pathological diagnosis was benign neurofibroma with foci of myxoid degeneration. There was no evidence of malignancy, no hypercellularity and very scanty mitoses, but there was however extension into the medullary cavity of the first rib. Eight years later, a routine radiograph of chest showed two nodules in the upper and lower lobe of the left lung. Upper lobectomy was done; histological appearances were identical with those of the first tumour. Four years later, the patient represented with a further mass in the left posterior triangle of the neck with increasing pain and weakness in the left upper limbs. MRI showed massive recurrence. The tumour was incompletely removed and in this specimen there was pleomorphism, small areas of multi-nucleated tumour cells but the mitotic rate remained low at less than 1 per 10 high powered fields. A course of radiotherapy was completed. A further partial excision of tumour from the left chest wall was performed 2 years later: histological
broma. There is some pleomorphism. Haematoxylin and eosin × 250. Below right: these cells stained positively for S-100 (Courtesy of Dr Jean Pringle).
Operating on Peripheral Nerves
293
Fig. 7.69 Large solitary neurofibroma. Malignant change. The final recurrence 14 years later. Left: MR scan. Right: fibroblast like cells predominated. × 5600 (Electronmicrographs prepared by Mr Stephen Gshmeissner).
appearances now were frankly malignant, with loosely arranged spindle and round or oval cells, nucleur pleomorphism and a mitotic index of 1 per 10 hpf. Sixteen years after the original presentation, extensive metastases in abdomen and chest were evident. Chemotherapy was given (Fig. 7.69). The clinical behaviour of this tumour, invading the first rib and scalenus anterior, suggests that it was malignant from the outset. The slow but inexorable progression in this case is reminiscent of the behaviour of some other soft tissue sarcomas and reminds us that neurofibromas are always potentially malignant. We have operated on seven similar cases. These were massive tumours extending the length of between three and six vertebrae. The transclavicular approach, sometimes supplemented by removal of the first rib, provided adequate access in all. The neurological deficit after excision was trivial in five cases; it was significant in two. One patient suffered paralysis of gleno-humeral muscles and the second had paralysis of long flexor muscles and small muscles of the hand.
7.9.4 Malignant Peripheral Nerve Sheath Tumours (MPNST) MPNST are amongst the most terrible of soft tissue tumours. Increased understanding of their origin and improved methods of diagnosis have not altered their prognosis. These patients have but one chance and that chance is lost by delay in diagnosis, by inappropriate or inadequate biopsy or by inadequate excision. Attempts to preserve function by restricting the resection or by repairing the excised nerve are futile and dangerous. Most of these tumours arise from cells within the endoneurium, and in particular, from Schwann cells. The Schwann cells are formed from a neural crest, not from the mesoderm, and it may be unwise to follow,
uncritically, the principles of treatment developed for soft tissue sarcomas. The perineurium is the true compartment of most MPNST’s and when the tumour has burst out of the epineurium, extending into muscle or bone, or axial blood vessels, then attempts at “limb sparing” operations are usually futile. In one of our cases tumour was detected within the perineurium 20 cm proximal to the main mass in frozen section biopsies. There are three groups. The most dangerous are those arising in patients with neurofibromatosis, and these account for about one third of the total. The risk of a malignant or a central nervous system tumour developing in patients with NF1 has been measured in extensive population studies by Huson and her colleagues (1989a, b) and by Sorensen et al. (1986). It lies between 3% and 5%. Dall (2007) investigated 123 cases of MPNST treated, between the years 1979 and 2002, at the Royal Orthopaedic Hospital, Birmingham, and the Peripheral Nerve Injury Unit at the Royal National Orthopaedic Hospital. Thirty three patients (27%) had NF1. The median age of the patients with NF1 was 26 years compared with 53 years for the sporadic MPNST. The 5 year survival rate for the NF1 patients was 32% compared with 60% in the cases of sporadic MPNST. Tumours in NF1 tend to be more centrally located, and are usually larger. It is often very difficult to detect a malignant tumour arising within a plexiform neurofibroma at an early stage. Pain, sudden enlargement and interference with conduction are common in plexiform neurofibromas. These difficulties have been eased by the work of Rosalie Ferner and her colleagues working at Guy’s, King’s and St Thomas’ Hospitals. (Ferner et al. 2000) Fluorodeoxyglucose positron emission tomography (FDG PET) was used to detect malignant tumours arising within 23 plexiform neurofibromas in 18 patients. Seven malignant peripheral nerve sheath tumours were diagnosed in five patients. The next identifiable group are those tumours arising years after irradiation. (Sordillo and colleagues 1981,
294
Ducatman et al. 1986). We have indentified five such cases. In the case of a 63 year old man who presented with massive swelling deep to and below the clavicle full mantle radiotherapy had been used for Hodgkin’s disease 15 year previously. When we saw him the history and physical signs admitted only one possible diagnosis, that of MPNST. Two biopsies had already been performed, one by needle and one at open operation, both of these were incorrectly reported as benign. The tumour arose from the medial cord. Frozen sections revealed tumour in the eighth cervical nerve root, 15 cm proximal to the main mass. Forequarter amputation had been considered but had to be abandoned because the chance of cure had already passed because of dissemination of the tumour throughout the field. The third group, the solitary MPNST, are somewhat less dangerous. The peak incidence is in the fourth, fifth and sixth decades and the tumour tends to arise more peripherally. There is no significant differences between the sexes. The head, neck and upper limb are the most common sites, accounting for over 40% of tumours; the lower limb is the site of approximately one third, the remainder are found within the trunk, usually arising from spinal nerves and in close relation of the spinal column (Weiss and Goldblum 2001b).
Surgical Disorders of the Peripheral Nerves
ing the longitudinal spread of tumour within the perineurium and the nerve trunk.
7.9.4.2 Pathology Detailed and beautifully illustrated accounts of the histological characteristics can be found in Weiss and Goldblum (2001b, c). The most common histological appearance is reminiscent of fibrosarcoma with masses of spindle shaped cells organised into sweeping bundles or lamellae. Irregularity of the cells and a characteristic wavy or buckled appearance of the nuclei is often found. Metaplastic differentiation occurs in about 20% of tumours. Bone, cartilage and muscle are the most common tissues seen; epithelioid differentiation into glandular or squamous tissue is less common. Woodruff and colleagues (1973) described three new cases of rhabdomyosarcomatous change, the so called triton tumour, and reviewed the seven previously published cases. They agreed with Masson and Martin (1938) in proposing that muscle is formed from malignant Schwann cells. Nine of their patients were dead within a year of diagnosis (Fig. 7.70).
7.9.4.3 Principles of Operation 7.9.4.1 Clinical Features The cardinal clinical feature of MPNST is pain. This is usually attended by clear evidence of impairment of nerve conduction. A tender and enlarging mass is found in all but the most deeply located tumours. Errors in diagnosis are not uncommon; there are numerous references to unnecessary laminectomy or excision of a cervical rib before a correct diagnosis was made. Nerve conduction studies are not useful. CT and MRI scan are always valuable in showing the extent of tumour and in distinguishing between benign and malignant. Myelography has a place in defining intraspinal extension and is extremely good for this when combined with CT scan. Moser and Parrish (2001) issue an important caveat. In most cases radiological features cannot distinguish reliably between benign and malignant soft tissue tumours. It is imperative that a final interpretation of CT scan or MRI findings is not given until the cross sectional imaging has been correlated with plain radiographs. Failure to adhere to this principle results in serious errors in management, The clinical features and findings at operation are more important in diagnosis of MPNST than are histological characteristics, most of all in the so called low grade malignant neurofibromata. Detection of metastases, notably in the lungs, has obvious implications for the planned treatment. Conventional methods of staging these tumours must be used with discretion because of the risk of underestimat-
“Because the proper treatment for a neurosarcoma should not be less than a radical excision the biopsy should be of the whole specimen and of a considerable length of the nerve trunk on either side of it – provided it is possible to make the diagnosis of malignancy on the clinical evidence and on the appearances displayed at operation”. (Seddon 1975c) Biopsy must never be a matter of unthinking routine and it should be seen as a preliminary step before urgent and definitive excision of the tumour (Fig. 7.71). Wide excision of the trunk nerve is the correct treatment for a tumour which is still confined within the epineurium. Examination of frozen section biopsies, taken during the operation, is essential to establish adequate clearance. It proved necessary to section the spinal nerve at the foramina to secure clearance in seven of our cases and resection of the intraspinal segment of the nerve was required in three more. We think that it is futile to attempt repair of the resected nerves. The object of the operation must be wide resection of the parent nerve with adequate margins of clearance (Bolton et al. 1989). The prospect of restoring useful function in the foot or hand by grafting gaps in excess of 20 cm, in a field which will, in all likelihood, be irradiated, is negligible. We have seen three cases where inadequate resection followed by grafting of the affected nerve led to a fatal outcome. One of these was the case of a 32 year old man, in good general health, who was reviewed 12 months after an operation
Operating on Peripheral Nerves
Fig. 7.70 Histopathological findings in four malignant peripheral nerve sheath tumours. Above left: a proportion of the tumour cells are expressing myoglobin, in particular, larger cells. Triton tumour. × 400 (Courtesy of Dr Kim Suvarna). Above right: another triton tumour.
during which an 8 m segment of the medial cord was resected and grafted. This resection was wholly inadequate. The tumour soon recurred, and enveloped the neurovascular bundle. Forequarter amputation was performed. Tumour extended 14 cm proximal to the upper pole of the tumour and it had infiltrated the nerve graft. Amputation is probably indicated when the epineurium has been breached, and in rapidly growing recurrence. Lusk, Kline and Garcia (1987) recommended forequarter amputation for MPNST arising within the brachial plexus. We agree that this is necessary with recurrence after apparently adequate excision or when there is such involvement of the plexus that no function can be preserved. It may be necessary for palliation even if it cannot cure; the pain caused by malignant infiltration of the brachial plexus is terrible. Fungation of tumour must be prevented (Fig. 7.72). The outcome of treatment in our 56 cases of MPNST is bleak. Of the 15 tumours in known NF1, nine died within 2 years of treatment. One patient died from a second intrapelvic MPNST after successful forequarter amputation. Two more patients have active disease at less than 12 months from operation. Of the 41 patients with sporadic tumours eight
295
There is S-100 positive staining. Mitoses are evident. × 400. Below right: S-100 positive staining showing glandular differentiation × 400. Below left: only a few cells show S-100 positive staining in this tumour × 400 (Courtesy of Dr Jean Pringle).
Fig. 7.71 Open biopsy was performed on a lump in the thigh in an 87 year old woman. She experienced intense pain and rapid progression of neural deficit. The lesion was re-explored 1 month later, the tumour had burst out of the epineurium and infiltrated the muscles of the thigh and extended to the skin. (Courtesy of Marco Sinisi).
died within 2 years and 6 patients are known to have active disease. Twenty one appear free of disease at a minimum of 4 years from operation. Of those “cured” twelve were treated
296
Fig. 7.72 MPNST in a 53 year old man with NF1. Forequarter amputation was performed to relieve his pain. His death was caused by erosion into the carotid vessels. Below: the tumour is densely cellular with moderate pleomorphism and high mitotic index. Haematoxylin and eosin × 400 (Courtesy of Dr Jean Pringle).
by excision of the nerve, six by amputation of the limb and three by forequarter amputation. Surgical ablation was impossible in six cases because of intra-spinal extension. Death was caused by invasion of great vessels, of the spinal canal or viscera in seven cases, and in six more the threat of
Surgical Disorders of the Peripheral Nerves
this remains. Death was caused by pulmonary and visceral metastases in eight cases. Primitive Neurectodermal Tumours: Our experience with these is too small to be worthy of discussion. En bloc excision is impossible in most large primitive neurectodermal tumours (PNET) extra-osseous Ewing’s tumours and in neuroblastomas (Figs. 7.73 and 7.74). Removal of as much of the tumour as is possible may increase the chance of a positive response to chemo and radiotherapy. Edward Kiely has described his approach to the abdominal and pelvic neuroblastomas in an inspiring Hunterian lecture (Kiely 2007). Two hundred and thirty four children were operated, almost 60% of these presented with distant metastatic disease (stage 4). Kiely points out that these tumours do not extend deep to the adventitia of major blood vessels and his technique rests on developing the plane, by meticulous scalpel dissection, between the tumour and the media. Dissection starts below the lower limit of the tumour: “this usually means the common external iliac vein on the right side or the corresponding artery on the left side”. The dissection advances proximally precisely in the mid line of the vessel. The major visceral arteries are treated in a similar fashion, and the tumour and adventitia are incised along the midline of these vessels and along each of their branches as far as necessary. The lumbo sacral trunk and the obturator nerve are identified and are separated from the tumour without great difficulty. At least one hypogastric nerve is preserved wherever possible. Clearance of the coeliac and superior mesenteric arteries inevitably removes the sympathetic nerve supply to the gut and post operative diarrhoea is common. There is also sympathetic paralysis in the ipsi lateral lower limb. There were two post operative deaths. Removal of at least 95% of tumour was achieved in 90% of children. Twenty six of the 29 children presenting with tumours in the midline or crossing the midline or with contra lateral involvement of lymph nodes (Stage 3) were alive 5 years after operation. The 5 year survival for children presenting with Stage 4 disease was 30% (Fig. 7.75).
Operating on Peripheral Nerves
Fig. 7.73 Neuroepithelioma. Left: the tumour arose from the sciatic nerve in a 14 year old boy with NF1. The significance of his intense pain and the swelling in the upper thigh was not grasped for several months. Above right: another example of tumour arising from the sci-
297
atic nerve in a 19 year old man. Middle: there are areas of rosette formation. Haematoxylin and eosin × 400 (Courtesy of Dr Jean Pringle) Below right: intense mitotic activity. Electronmicrograph × 8075 (Prepared by Mr Stephen Gshmeissner).
298
Surgical Disorders of the Peripheral Nerves
Fig. 7.74 Extra osseous Ewing’s tumour arising from the sympathetic chain. Left: the extent of the tumour is demonstrated by MRI. Right: the tumour consists of undifferentiated cells, many of these containing glycogen × 7000 (Prepared by Mr Stephen Gshmeissner).
Fig. 7.75 A massive neuroblastoma extending from C5 to T4 enveloped the subclavian vessels and the brachial plexus in a 14 year old girl,. About 90% of the tumour was excised through an extended transclavicular approach. The operation was done with Mr R Spicer
(Children’s Hospital, Bristol). Above: the approach to the tumour after elevation of the flap. The subclavian artery was freed from one of the lobes of the tumour. Below: Although the first thoracic nerve was excised little defect was evident 20 months later.
References
Amarenco G, Lanoë Y, Ghnassia RT, Goudal H, Perrigot M (1988) Syndrome du canal d’Alcock et névralgie périnéale. Rev Neurol 144:523–526 Barros D’sa AAB (1992) Arterial injuries. In: Eastcott HHG (ed) Arterial surgery, 3rd edn. Churchill Livingstone, Edinburgh, pp 355–413 Barros D’sa AAB (1982) A decade of missile-induced vascular trauma. Ann R Coll Surg Engl 64:37–44. Hunterian Lecture. Delivered at the Royal Army Medical College on the 30th April 1980 Bendszus M, Rieckmann P, Perez J, Koltzenburg M, Reiners K, Solymosi L (2003) Painful vascular compression syndrome of the sciatic nerve caused by gluteal varicosities. Neurology 61 October(1 of 2):985–987
Abernathy CD, Onofrio B, Scheithauer B, Pairolero PC, Shives TC (1986) Surgical management of giant sacral schwannomas. J Neurosurg 65:286–295 Addas BMJ, Midha R (2009) Nerve transfers for severe nerve injury. In: Spinner RJ, Winfree CJ (ed) Neurosurgery clinics 20: Peripheral nerves; injuries, pp 27–38 Adson AW, Brown GE (1929) The treatment of Raynaud’s disease by resection of the upper thoracic and lumbar sympathetic ganglia and trunks. Surg Gynae Obst 48:577–603
Operating on Peripheral Nerves Benson ER, Schutzer SF (1999) Post traumatic piriformis syndrome: diagnosis and results of operative treatment. J Bone Jt Surg 81A:941–949 Birch R, St. Clair Strange FG (1990) A new type of peripheral nerve lesion. J Bone Jt Surg 72B:312–313 Birch R (1987) The place of microsurgery in orthopaedics. In: Catterall A (ed) Recent advances in orthopaedics 5:165–186 Birch R, Bonney G, Marshall RW (1990) A surgical approach to the cervico-dorsal spine. J Bone Jt Surg 72B:904–907 Birch R, Raji ARM (1991) Repair of median and ulnar nerves. J Bone Jt Surg 73B:154–157 Bircher M, Lewis A, Halder S (2006) Delays in definitive reconstruction of complex pelvic and acetabular fractures. J Bone Jt Surg 88B:1137–1140 Bland-Sutton J (1930) Injuries of the peripheral nerves. In: The story of a surgeon. Methuen & Co Ltd., London, pp 34–37 Bolton JS, Vauthey JN, Far GH, Sauter EI, Bowen JC, Kline DG (1989) Is limb-sparing surgery applicable to neurogenic sarcomas of the extremities? Arch Surg 12:118–121 Bonney G, Jamieson AM (1979) Reimplantation of C7 and C8. Communication – Au symposium sur le plexus brachial. Int Microsurg 1:103–106 Brain WR, Wright AD, Wilkinson M (1947) Spontaneous compression of both median nerves in the carpal tunnel. Lancet 1:277–282 Brushart TM (1999) Nerve repair and grafting: Histochemical identification. In: Green EP (ed) Operative hand surgery, 4th edn. Churchill Livingstone, New York, pp 1388–1389 Camp SJ, Milano R, Sinisi M (2008) Intractable neurostenalgia of the ulnar nerve abolished by neurolysis 18 years after injury. J Hand Surg 33E:45–46 Carlstedt T (2007a) Endoscopy. Central nerve plexus injury. Imperial College Press, London, pp 114–116 Carlstedt T (2007b) Intraspinal exposure. Ibid. Imperial College Press, London, pp 118–124 Carlstedt T (2007c) The lumbosacral plexus – expouse. Ibid. Imperial College Press, London, pp 124–131 Carlstedt T (2008) Pudendal nerve entrapment. Experience with 20 cases, personal communication Carlstedt T, Birch R (2000) Surgical management of nerve root injuries. In: Schmidek HH (ed) Operative neurosurgical techniques, 4th edn. Saunders, Philadelphia, PA, pp 2369–2376 Carroll RE, Green DP (1972) The significance of the palmar cutaneous nerve at the wrist. Clin Orthop Rel Res 83:24–28 Claustre J, Simon L (1978) La maladie de Thomas Morton. Syndrome canalaire. Rhumatologie 91:283–287 Comtet JJ (2005) Etiology and physiopathology of nerve compression. In: Tubiana R, Gilbert A (eds) Tendon, muscle and nerve. Taylor & Francis, Abingdon. UK. ISBN 1-85317-494-7 Chapter 19, pp 303–312 Concannon MJ, Kester CG, Welsh CF, Puckett CL (1992) Patterns of free radical production after tourniquet ischaemia: implications for the hand surgeon. Plast Reconstr Surg 89:846–857 Cooper A, Travers B (eds) (1817) On exostosis. In: Cooper A, Travers B (eds) Surgical essays, 2nd edn. Cox & Son, Longman Hurst Rees Orme, London/A Constable & Co, Brown Edinburgh. Glasgow: Smith & Son. Dublin: Hodges & McArthur, p 171 Coote H (1861) Exostosis of the left transverse process of the seventh cervical vertebra, surrounded by blood vessels and nerves, successful removal. Lancet 1:360–361 Crockard HA, Bradford R (1985) Transoral transclival removal of a schwannoma anterior to the cranio-vertebral junction. J Neurosurg 62:293–295 Crumplin M (2007) Amputation. In: Men of steel. Surgery in the Napoleonic wars. Quiller Press, Shrewsbury, ISBN 978 1 904057 949, Chapter 10, pp 287–314 Crumplin MKH, Harrison A (2005) A fatal shot: the death of Admiral Horatio Viscount Lord Nelson. Ann R Coll Surg Engl Suppl 87: 302–308
299 Dall GF, Porter DE, Prasad DV, Foster L, Birch R, Grimer RJ (2007) Survival in malignant nerve sheath tumours: A comparison between 123 cases of sporadic and neurofibromatosis Type 1 associated tumours. Personal Communication Dawson DM, Hallett M, Millender LH (1990) Lateral femoral cutaneous nerve. In: Entrapment neuropathies, 2nd edn. Little Brown, Chicago, p 301 Dellon AL (2004) Diabetic neuropathy: review of a surgical approach to restore sensation, relieve pain, and prevent ulceration and amputation. Foot Ankle Int 25:749–755 Dellon AL, Coert JH (2004) Results of musculofascial lengthening technique for submuscular transposition of the ulnar nerve at the elbow. J Bone Jt Surg 86A(Supple I/Part 2):169–179 Delorme E (1915) The treatment of gunshot wounds of nerves. Brit Med J 1:853–855 Denman EE (1978) The anatomy of the space of Guyon. Hand 10:69–76 Dubuisson AS, Kline DG, Weinshel SS (1993) Posterior subscapular approach to the brachial plexus. J Neurosurg 79:319–330 Ducatman BS, Scheithauer BW, Piepgras DG (1986) Malignant peripheral nerve sheath tumour: A clinic-pathological study of 120 cases. Cancer 57:2006–2021 Duval P (1921) Exposure of the great vessels in the neck. Attributed by Fiolle JJ, Delmas 1921. In: Cumston CG (trans; ed) The surgical exposure of the deep seated blood vessels, Heinemann, London, p 63 Dyck PJ, Thomas PK (eds) (2005) Preface to the 4th edition of peripheral neuropathy, Elsevier Saunders, Philadelphia, PA, pp xvii–xviii Dyck PJ, Dyck PJB, Engelstad J (2005) Pathological alterations of nerves. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn, Elsevier Saunders, Philadelphia, PA, Chapter 32, pp 733–829 Esmarch F (1873) Über Künstliche Blutleere Bei Operationen: (on artificial bloodlessness during operations). Volkmann’s Sammlung Klinischer Vorträge, Chir. Band 1: No: 58 373–384. Translated with commentary by: Gohritz A, Vogt, PM.2006. J. Hand Surg 31B: 390–396 Ferner RE, Lucas JB, O’doherty MJ, Hughes RAC, Smith MA, Cronin BF, Bingham J (2000) Evaluations of 18 fluoro deoxyglucose positron emission tomography (18 FDG PET) in the detection of malignant peripheral nerve sheath tumour in Neurofibromatosis 1. J Neurol Neurosurg Psych 68:353–357 Fiolle J, Delmas J (1921) In: Cumston CG (trans; ed) The Surgical exposure of the deep seated blood vessels. Heinemann, London, pp. 61–67 Fletcher JA (2001) Cytogenetic analysis of soft tissue tumours. In: Weiss SW, Goldblum JR (eds) Enzinger and Weiss’s soft tissue tumours, 4th edn. Mosby. St Louis: Chapter 5, pp 125–146 Folpe AL, Gowan (2001) Immunohistochemistry for analysis of soft tissue tumours. In: Weiss SW, Goldblum JR (eds) Enzinger and Weiss’ soft tissue tumours, 4th edn. Mosby, St Louis, Chapter 8, pp. 119–246 Fournier H-D, Mercier P, Menei P (2001) Lateral interscalene multilevel oblique corpectomies to repair ventral root avulsions after brachial plexus injury in humans: anatomical study and first clinical experience. J Neurosurg Spine 95:202–207 Fournier H-D, Mercier P, Menei P (2005) Repair of avulsed ventral nerve roots by direct intraspinal implantation after brachial plexus injury. Hand Clin 21:109–118 Frazer JE (1920) The anatomy of the human skeleton, 2 edn. J&A Churchill, London Frykman GK, Adams J, Bowen WW (1981) Neurolysis. Orthop Clin North Am 12:325–342 Gauthier G, Dutertre P (1975) La Maladie de Morton: syndrome canalaire: 74 cas opérés sans résection du nevrome. Lyon Méd 223:917–921 Geisinger KR, Abdul-Karim FW (2001) Fine needle aspiration biopsies of soft tissue tumours. In: Weiss SW, Goldblum JR (eds) Enzinger and Weiss’s soft tissue tumours, 4th edn. Mosby, St Louis, pp 147–188 Gelberman RH, Eaton R, Urbaniak JR (1993) Peripheral nerve compression. J Bone Jt Surg 75A:1854–1878
300 Giannini C (2005) Tumours and tumour like conditions of peripheral nerves. In: Dyck PJ, Thomas PK (eds.) Peripheral neuropathy, 4th edn. Elsevier, Philadelphia, PA, Chapter 115, pp. 2585–2606 Giddins GEB, Drew SJ, Birch R (1998) The use of facsimiles in acute patient transfers: experience with brachial plexus patients. Injury 29:257–259 Gilliatt RW (1984) Thoracic outlet syndrome. In: Dyck PJ, Thomas PK, Lambert EH, Bunge R (eds) Peripheral neuropathy, 2nd edn. WB Saunders, Philadelphia, PA, pp 1409–1424 Gruber W (1869) Über die Halsrippen des menschen, mit vergleichendanatomischen Bemerkungen. Mémoires de l’Académie Impériale de St Petersbourg, VIIe Serie 13:1–52 Gschmeissner SE, Pereira JH, Cowley SA (1991) The rapid assessment of nerve stumps. J Bone Jt Surg 73B:688–689 Haug CE, Sanders RJ (1991) Arterial TOS. In: Sanders RJ, Haug CE (eds) Thoracic outlet syndrome, JB Lippincott Company, pp 211–231 Hennen J (1820) Principles of military Surgery, 2nd edn, Edinburgh, pp 255–256 Henry AK (1975) Extensile exposure, 4th edn. Churchill Livingstone, Edinburgh Holmes S (1881) A study in scarlet. In: Baring-Gould WS (ed) The Annotated Sherlock Holmes, John Murray, London, p 168 Horsley V (1899) On injuries to peripheral nerves. Practitioner 63: 131–144 Huson SM, Compston DAS, Harper PC (1989a) A genetic study of von Recklinghausen Neurofibromatosis in South East Wales. II. Guidelines for genetic counselling. J Med Genet 26:712–721 Huson SM, Compston DAS, Clark P et al (1989b) A genetic study of von Recklinghausen’s neurofibromatosis in SE Wales. 1. Prevalence, fitness, mutation rate and effect of parental transmission on severity. J Med Genet 26:704–711 Katirji B, Wilbourn AJ (2005) Mononeuropathies of the lower limb. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn, Elsevier Saunders, Philadelphia, PA, Chapter 61, pp 1487–1510 Kato H, Minami A, Kobayashi M et al (1998) Functional results of low median and ulnar nerve repair with intraneural fascicular dissection and electrical fascicular orientation. J Hand Surg 23A:471–482 Kiely E (2007) A technique for excision of abdominal and pelvic neuroblastomas. Hunterian Lecture. Ann R Coll Surg Engl 89:342–348 Kirkup J (2007) A history of limb amputation. Springer-Verlag, London Klenerman L (2003) The Tourniquet manual – principles and practice. Springer Verlag, London Kline DG, Kott J, Barnes G, Bryant L (1978) Exploration of selected brachial plexus lesions by the posterior subscapular approach. J Neurosurg 49:872–879 Kline DG (2008a) Nerve action potential recordings In: Kim DH, Midha R, Murovic JA, Spinner RJ (eds) Kline and Hudson’s nerve injuries, 2nd edn, Saunders, Elsevier, Philadelphia, PA, Chapter 5, pp. 75–86 Kline DG (2008b) The dorsal scapular nerve, In: Kim DH, Midha R, Murovic JA, Spinner RJ (eds) Kline and Hudson’s nerve injuries, 2nd edn, Saunders, Elsevier, Philadelphia, PA, Chapter 5, pp 285–287 Kline DG, Hudson AR, Kim DH (2001a) Ilioinguinal iliohypogastric and genitofemoral nerves. In: Atlas of peripheral nerve surgery. WB Saunders, Philadelphia, PA, (Revised reprint) pp 157–162 Kline DG, Hudson AR, Kim DH (2001b) Femoral nerve. In: Atlas of peripheral nerve surgery. WB Saunders, Philadelphia, PA, (Revised reprint) pp 151–156 Knight DMA, Birch R, Pringle J (2007) Benign solitary schwannoma. A review of 234 cases. J Bone Jt Surg 89B:382–7 Kratimenos GP, Crockard HA (1993) The far lateral approach for ventrally placed foramen magnum and upper cervical spine tumour. Brit J Neurosurg 7:129–140 Kretschmer T, Antoniadis G, Richter H-P, König RW (2009) Avoiding iatrogenic injuries in endoscopic carpal tunnel release. In: Spinner
Surgical Disorders of the Peripheral Nerves RJ, Winfree CJ (Guest editors), Neurosurg Clin N Am (Peripheral nerves: Injuries) 20:65–72 Krikler SJ, Grimer RJ (1993) Pelvic chondrosarcoma presenting with meralgia paraesthetica. Hosp Update August 1993(460):464 Lam SJS (1967) Tarsal-tunnel syndrome. J Bone Jt Surg 49-B:87–92 Lanz U (1993) The carpal tunnel syndrome. In: Tubiana R (ed) The Hand Volume IV, WB Saunders, Philadelphia, PA, pp 463–486 Lee JY, Schwartz DM, Anderson DG, Hilibrand AS (2006) Epidural haematoma causing dense paralysis after anterior cruciate corpectomy. J Bone Jt Surg 88A:198–201 Lewis T, Pickering GW (1934) Observations upon maladies in which the blood supply to the digits ceases intermittently or permanently and upon gangrene of the digits: observations relevant to so-called “Raynaud’s disease”. Clin Sci 1:327–366 Linarte R, Gilbert A (1986) Trans-sacral approach to the sacral plexus. Peripher Nerv Repair Regen 4:17–20 Lister J (1879) Effects of the position of a part on the circulation through it. Br Med J 1:923 Lowe JB, MacKinnon SE (2004) Ulnar nerve entrapment at the elbow. In: Winn HR. Youman’s neurological surgery, 5th edn, Saunders, Philadelphia, PA, Chapter 256, pp 3897–3920 Lusk MD, Kline DG, Garcia CA (1987) Tumour of the brachial plexus. Neurosurgery 21:439–453 MacCarty CS (1951) Two-stage autograft for repair of extensive damage to sciatic nerve. J Neurosurg 8:319 MacKinnon SE, Dellon AL (1988) The tarsal tunnel syndrome. In: MacKinnon SE, Dellon AL, Surgery of the peripheral nerve, Thieme Stuttgart, Chapter 12, pp 305–318 Masson P, Martin JF (1938) Rhabdomyomes des Nerfs. Bulletin de L’Association Francaise pour l’Etude de Cancer 27:751–767 Meltzer PS (2001) Molecular genetics of soft tissue tumours. In: Weiss SW, Goldblum Enzinger JR (eds) Weiss’s soft tissue tumours, 4th edn, Mosby. St Louis, Chapter 4, pp 103–124 Millesi H (1981) Interfascicular nerve grafting. Orthop Clin North Am 12:287–301 Montgomery AS, Birch R, Malone A (2005) Sciatic neurostenalgia: caused by total hip arthroplasty, cured by later neurolysis. J Bone Jt Surg (Br) 87-B:410–411 Moser RP, Parrish WM (2001) Radiologic evaluation and soft tissue tumours. In: Weiss SW, Goldblum JR (eds) Enzinger and Weiss’s soft tissue tumours, 4th edn. Mosby, St Louis, pp 45–102 Nagano A (1987) Anterior interosseous nerve palsy. J Jp Soc Surg Hand 3:894–897 Nagano A, Shibata K, Tokimura H, Yamamoto S, Tajiri Y (1996) Spontaneous anterior interosseous nerve palsy with hourglass-like fascicular constriction within the main trunk of the median nerve. Hand Surg 21A:266–270 Narakas AO (1988) The use of fibrin clot glue in repair of peripheral nerves. Orthop Clin North Am 19:187–199 Narakas AO (1987) Thoughts on neurotisation or nerve transfers in irreparable nerve lesions. In: Tertzis JK (ed) Micro reconstruction of Nerve Injuries. WB Saunders, Philadelphia, PA, pp 447–454 Narakas AO (1993) Cervicobrachial compression. In: Tubiana R (ed) The hand volume IV. WB Saunders, pp 352–389 Nathan PA, Kenistin RC, Meadows KD (1995) Outcome study of ulnar nerve compression at the elbow treated with simple decompression and an early programme of physical therapy. J Hand Surg 20B:628–637 Nawabi DH, Sinisi M (2007) Schwannomas of the posterior tibial nerve. J Bone Jt Surg 89B:814–816 Nixon M, Jackson B, Varghese P, Jenkins D, Taylor G (2006) Methicillin resistant staphylococcus aureus on orthopaedic wards: Incidence, spread, mortality, cost and control. J Bone Jt Surg 88B:812–817 Nylen CO (1972) The otomicroscope and microsurgery 1921–1971. Acta Otolaryngol 73:453 Oberlin C, Beal D, Leechavengvongs S, Salon A, Dauge MC, Sarly JJ (1994) Nerve transfers to biceps muscle using part of ulnar nerve
Operating on Peripheral Nerves for C5–C6 avulsion of the brachial plexus: anatomical study and report of four cases. J Hand Surg 19A:232–237 Ochiai N, Nagano A, Mikam V, Yamamoto S (1997) Full exposure of the axillary and suprascapular nerves. J Bone Jt Surg 79B: 532–533 Ömeroğlu H, Uçaner A, Tabak AY, Guneg O, Cimŏglua B, Gűnel U (1998) The effect of using a tourniquet on the intensity of postoperative pain in forearm fractures. Internat Orthop 22:369–373 Orgell MG (1987) Epineurial versus perineurial repair of peripheral nerves. In: Tertzis JK (ed) Microreconstruction of nerve injuries. WB Saunders, London, pp 97–100 Osborne GV (1970) Compression neuritis of the ulnar nerve at the elbow. Hand 2:10–13 Owen E. (1975) Microsutures. Presented and used during the Royal Coll. Surg (Eng) Microsurgical Workshop organised byEarl Owen, John Cobbett and Richard Ackland Panegyres PK, Moore N, Gibson R, Rushworth G, Donaghy M (1993) Thoracic outlet syndromes and magnetic resonance imaging. Brain 116:823–841 Parkes AR (1961) Intraneural ganglion of the lateral popliteal nerve. J Bone Jt Surg 43B:784–790 Parry EW, Eastcott HHG (1992) Cervical rib and thoracic outlet syndrome. In: Eastcott HHG (ed) Arterial surgery, 3rd edn. Churchill Livingstone, Edinburgh, pp 333–353 Petrie A (2006) Statistics in orthopaedic papers. J Bone Jt Surg 88B:1121–1136 Poitevin LA (2005a) Thoracic outlet syndrome, scalene complex and Interscalene passages. In: Tubiana R, Alain Gilbert, Taylor & Francis, Abingdon, United Kingdom, ISBN 1-85317-494-7, Chapter 21, pp 321–332 Poitevin LA (2005b) Anatomical bases for brachial plexus and subclavian artery compression. In: Tubiana R, Gilbert A, Taylor & Francis, Abingdon. United Kingdom, ISBN 1-85317-494-7, Chapter 20, pp 313–320 Reissis N, Stirrat A, Manek S, Dunkerton M (1992) The terminal branch of the posterior interosseous nerve: a useful donor for digital nerve J. Hand Surg 17B:638–640 Robert R, Labat JJ, Lehar PA, Le Borgne J, Le Barbin JY (1989) Réflexions cliniques, neurophysiologiques et thérapeutiqies à partir des donnés anatomiques sur le nerf pudendal (honteux interne) lors de certaines algies périnéales. Chirurgie 115:515–520 Rodger NAM (2004) The Command of the Ocean. A naval history of Britain. 1649–1815. First published by Allen Lane 2004. Published in Trade Paperback by Penguin Books 2005 and in this format by Penguin Books in 2006, pp 308–309 Roos DB (1981) Essentials and safeguards of surgery for thoracic outlet syndrome. Angiology 32:187–192 Rosenbaum RB, Ochoa JL (1993) Carpal tunnel syndrome and other disorders of the median nerve. Butterworth Heinemann, Oxford Santos-Pallazzi C (1991) The quality of the stumps in post ganglionic ruptures of the spinal nerves. International Symposium on Brachial Plexus Injuries, Lausanne Sargent P (1921) Lesions of the brachial plexus associated with rudimentary ribs. Brain 44:95–124 Sawada T, Sano M, Ogihara H, Omura T, Mivra K, Nagano A (2006) The relationship between pre-and post operative symptoms, operative findings and post operative complications. J Hand Surg 31B:629–634 Scher LA, Veith FJ, Haimovici H, Samson RH, Ascer E, Gupta SK, Sprayregen S (1984) Staging of arterial complications of cervical rib: guidelines for surgical management. Surgery 95:644–649 Schroeder HP, Scheker LR (2003) Redefining the “Arcade of Struthers”. J Hand Surg 28A:1018–1021 Schwartz DM, Auerbach JD, Dormans JP et al (2007) Neurophysiological detection of impending spinal cord injury during scoliosis surgery. J Bone Jt Surg 89A:2440–2449
301 Seddon HJ (1975a) Common causes of nerve injury. In: Surgical disorders of peripheral nerves, 2nd edn. Churchill Livingstone, Edinburgh, pp 67–88 Seddon HJ (1975b) Operative treatment. In: Surgical disorders of peripheral nerves, 2nd edn. Churchill Livingstone, Edinburgh, pp 250–286 Seddon HJ (1975c) Nerve tumours. In: Surgical disorders of peripheral nerves, 2nd edn. Churchill Livingstone, Edinburgh, pp 154–171 Sedel L (1975) Voies d’abord des nerfes du member inférieur. Med Chir Technol Chirurg Orthop (Paris) 445301:1–8 Seki M, Nakamura H, Kono H (2006) Neurolysis is not required for young patients with a spontaneous palsy of the anterior interosseous nerve. J Bone Jt Surg 88B:1606–1609 Semple JC, Cargill AO (1969) Carpal tunnel syndrome. Results of surgical decompression. Lancet i:918–919 Smith PF (2006) Louis Stromeyer (1804–76) German orthopaedic and military surgeon and his links with Britain. J Med Bibliography 14:65–74 Sood SK, Balasubramani AN, Harrison BJ (2007) Percutaneous biopsy of adrenal and extra-adrenal retroperitoneal lesions: beware of catecholamine secreting tumours. Surgeon 5:279–281 Sorensen SA, Mulvihill JJ, Neilson A (1986) Nation wide follow up of Recklinghausen neurofibromatosis survival and malignant neoplasms. New Eng J Med 314:1010–1015 Sordillo PP, Helson L, Hajdu SL et al (1981) Malignant schwannoma – clinical characteristics, survival and response to therapy. Cancer 47:2503–2509 Spinner M, Spencer PS (1974) Nerve compression lesions of the upper extremity. Clin Orthop Relat Res 104:46–67 Spinner RJ (2008) Operative care and technique. In: Kim DH, Midha R, Murovic JA, Spinner RJ (eds) Kline and Hudson’s nerve injuries, 2nd edn. Saunders, Elsevier, Philadelphia, PA, Chapter 6, pp 87–106 Spinner RJ, Amrami KK, Kliot M, Johnston SP, Casanas J (2006a) Suprascapular intraneural ganglia and glenohumeral joint attachments. J Neurosurg 104:551–557 Spinner RJ, Atkinson JLD, Teil RL, Birch R et al (2003) Peroneal intraneural ganglia: the importance of the articular branch. A unifying theory. J Neurosurg 99:330–343 Spinner RJ, Dellon AL, Rosson GD, Anderson SR, Amrami KK (2007) Tibial intraneural ganglia in the tarsal tunnel: is there a joint connection? J. Foot Ankle Surg 46:27–31 Spinner RJ, Desy NM, Amrami K (2006b) Cystic transverse limb of the articular branch: a pathognomonic sign for peroneal intraneural ganglia at the superior tibio fibular joint. Neurosurgery 59:157–166 Spinner RJ, Goldner RD (1998) Snapping of the medial head of triceps and recurrent dislocation of the ulnar nerve J. Bone Jt Surg 80A:239–247 Stewart JD (2000) Focal peripheral neuropathies, 3rd edn. Lippincott Williams & Wilkins, Philadelphia, PA. ISBN 0-7817-1717-5 Strange FG, St C (1947) An operation for pedicle nerve grafting. Brit J Surg 34:423–425 Symonds CP (1927) Cervical rib: thrombosis of subclavian artery, contralateral hemiplegia of sudden onset, probably embolic. Proc R Soc Med 20:1244–1245 Taleisnik J (1973) The palmar cutaneous branch of the median nerve and the approach to the carpal tunnel. J Bone Jt Surg 55A:1212–1217 Tarlov IM (1944) Autologous plasma clot suture of nerves. JAMA 126:741–748 Vandeput J, Tanner JC, Huypens L (1969) Electrophysiological orientation of the cut ends in primary peripheral nerve repair. Plast Reconstr Surg 44:378–383 Wadsworth TG (1977) The external compression syndrome of the ulnar nerve at the cubital tunnel. Clin Orth Rel Res 124:189–204 Wadsworth TG (1984) Variations of the motor recurrent branch of the median nerve. In: Birch R (ed) DM Brooks operative surgery. The hand. Butterworth, London, p 481
302 Weiss SW, Goldblum JR (2001a) Benign tumours of peripheral nerves. In: Weiss SW, Goldblum JR (ed) Enzinger and Weiss’s soft tissue tumours, 4th edn. Mosby. St Louis, Chapter 30, pp 1111–1208 Weiss SW, Goldblum JR (2001b) Malignant tumours of the peripheral nerves. In: Weiss SW, Goldblum JR (eds) Enzinger and Weiss’s soft tissue tumours, 4th edn. Mosby, St Louis, Chapter 3, pp 1209–1264 Weiss SW, Goldblum JR (2001) Primitive neurectodermal tumours and related lesions. In: Weiss SW, Goldblum JR (eds) Enzinger and Weiss’s soft tissue tumours, 4th edn. Mosby, St Louis, Chapter 32, pp 1265–1322 Wickham JEA, Martin P (1962) Aneurysm of the subclavian artery in association with cervical abnormality. Br J Surg 50:205–209
Surgical Disorders of the Peripheral Nerves Wilbourn A (1993) Thoracic outlet syndromes. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 3rd edn. Elsevier, Philadelphia, PA, pp 936–942 Woodruff JM, Chernik NL, Smith MC, Millett JB, Foote WF (1973) Peripheral nerve tumours with rhabdomyosarcomatous differentiation. (Malignant “triton” tumours). Cancer 32:426–439 Woodruff JM, Selig AM, Crowlely K, Allen PW (1994) Schwannoma with malignant transformation: a rare distinctive peripheral nerve tumour. Am J Surg Pathol 18:882–895 Young JZ, Medawar PB (1940) Fibrin suture of peripheral nerves. Lancet 2:126
8
Compound Nerve Injury
We define compound nerve injuries as those in which damage to a nerve or nerves is associated with major damage to other tissues or organs such as skin, muscle, the skeleton, viscera and major vessels at the same site. There are some obvious categories. 1. Penetrating missile injuries 2. Severe open wounds, partial or complete amputation 3. Nerves injured by fracture or dislocation, open or closed, notably at the shoulder, elbow pelvis, hip and the knee. 4. Burns The timing of nerve repair in compound injuries remains controversial. Horn (1959) practised primary suture of median and ulnar nerves at the Accident Hospital in Birmingham, reserving grafting for more severe proximal injuries. From 1956 onwards primary suture of severed nerves was practised at St Mary’s Hospital whenever it was possible. There, in 1965, Michael Laurence achieved an outstanding result from primary repair of all flexor tendons, radial and ulnar arteries and median and ulnar nerves sectioned at the wrist. O’Brien (1975), one of the founders of modern microsurgical work, proved that primary repair of nerves was better even in the most severe injuries, in reattached limbs amputated by crush or traction. We think that the general contra-indications to primary repair include:
stumps need to be removed before repair in the tidy wound. The difficulties in preparing the stumps in traction rupture have been exaggerated. It is a simple matter, on the day of injury, to remove the tips of the stumps until a recognisable architecture of nerve bundles is revealed. This usually coincides with the level at which conduction can be demonstrated unless ischaemic conduction block has occurred following wounding of the axial artery or a compartment syndrome in the distal part of the limb. In the distal segment of the nerve it is a simple matter to move a nerve stimulator centrifugally away from the stump until the appropriate muscular response is seen. The recognition of the level of physiological continuity in the central portion of the nerve involves moving the stimulator centripetally away from the tip of the rupture whilst recording responses through electrodes secured to the skin of the scalp or over the spine. Delay before repair of a divided nerve is damaging particularly when there is associated injury to the adjacent artery. The best result that we have seen following repair of sciatic nerve was achieved by Mr Plaha, orthopaedic surgeon, London, who sutured the transected sciatic nerve in the left thigh, and also the right common peroneal nerve at the knee during an emergency operation (Fig. 8.1) for severe bleeding from multiple stab wounds in both lower limbs.
1. The poor condition of the patient. 2. Inexperience of the surgeon and lack of suitable facilities. 3. Most but not all of penetrating missile injuries. Ouellette (2000) recommends primary repair of nerves ruptured by civilian gunshot wounds in children. Delayed primary repair is the current policy is followed by the surgeons of the Royal Army Medical Corps (RAMC) in war wounds. 4. Most burns some aspects of thermal and cold injury are related in Chapter 3. 5. Cases of severely contaminated wounds from machinery, road accidents or agricultural injuries in which the prevention of sepsis must be the priority. The risk of sepsis is far less in the tidy wound and less still in the closed traction rupture. Little, if any, of the divided nerve
Fig. 8.1 Function 18 months after emergency suture of sciatic nerve in the left thigh and of the common peroneal nerve at the right knee by Mr Plaha (London).
R. Birch, Surgical Disorders of the Peripheral Nerves, DOI: 10.1007/978-1-84882-108-8_8, © Springer-Verlag London Limited 2011
303
304
Surgical Disorders of the Peripheral Nerves
One of the best results that we have seen following repair of a closed rupture of the common peroneal nerve was that achieved by Mr Roussow, orthopaedic surgeon (London) who, on the day of injury, grafted the rupture at the same time as he repaired the lateral ligament complex of the knee. Function was excellent by 22 months by which time the patient considered his function normal. This is a remarkable result, the defect in the nerve was over 15 cm. Our material is drawn from around 3,500 patients passing through the fracture service of St Mary’s Hospital, Paddington between 1965 and 1991 and the Peripheral Nerve Injury Unit at the Royal National Orthopaedic Hospital between 1965 and 2008. Most patients were referred from other hospitals. Our observations do not permit inferences about the incidence of nerves injured by fractures or dislocations nor about the course to spontaneous recovery in non operated cases. Some elements of the findings from the cases seen before 1998 are set out in Tables 8.1 and 8.2. Arterial injury or ‘compartment syndrome’ complicated about one fifth of these. Spontaneous recovery occurred in about one half of the cases of circumflex palsy, in 40% of the radial palsies and in only one third of the lesions to the common peroneal nerve. One hundred and eighty six patients sustained nerve injuries during operations for skeletal injury from a total of 2,330 cases (8.1%). A more detailed record (Chapter 5) was used to collect data from 1,271 patients seen after the year 2000. The most disturbing finding in the later group is the increase
Table 8.1 Operations in compound nerve injuries. St Mary’s Hospital – Royal National Orthopaedic Hospital 1965–1998. Number of patients Arterial injury or compartment syndrome 1,816
322
Open fracture/ dislocation
359
110
Penetrating missile wound
155
51
Closed fracture/ dislocation
Total 2,330 483 186 nerves were injured during operation for skeletal injury.
Table 8.2 Circumflex, radial and common peroneal nerves injured by fracture/dislocation. Operated and non-operated cases (1965–1998). Nerves showing Nerve Number Arterial injury/ good spontaneof nerves “compartment ous recovery syndrome” Circumflex
380
41
198
Radial
738
92
301
Common peroneal
318
98
104
1,436
231
603
Total
in the number of iatrogenous injuries. Three hundred and twenty three patients (25.4%) sustained such injuries. There are three general factors which determine the final outcome of nerve injury and repair and which strongly influence the timing of that repair. They are: the nature of wound and the risk of sepsis; vascular injury, and loss of skin.
8.1 The Wound One of the achievements of Robert Jones was to secure agreement amongst the surgeons on the Allied side on the Western Front in 1917 of a common policy of treatment of missile wounds. The elements of this included debridement, in its true sense of unbridling or decompression, wound excision and delayed closure. Hull and Rob (1997) write: ‘such were the infection rates following amputation and primary closure of the stump in the early months of the First World War that the chief surgeon of the British Expeditionary Force issued an order that forbad primary suture’. Wright (1915) and Fleming (1915) at St Mary’s Hospital, described the evolution of colonisation in war wounds which came from sporebearing clostridia, non spore-bearing gut organisms and pyogenic cocci. Clostridia was detected in 90% of wounds at the first dressing station; pyogenic cocci in 15%. Mellor et al. (1997) remind us that contamination and delay is a characteristic of battle wounds so that the natural history in the untreated cases was putrefaction followed by gas gangrene, or by cellulitis and septicaemia which was later recognised as ‘hospital’ streptococcal infection. Clarke (1959) emphasised that much of the primary contamination of wounds comes from the clothes and the skin of the patient: ‘this became important during World War Two when surgeons transferred their attentions from the Libyan Desert to the cultivated terrain of Italy. In the desert the techniques of wound exploration based on World War One experience appeared no longer necessary and this was unfortunately attributed to the use of sulphonamides. Experience in Italy showed this to be an illusion; full wound exploration and decompression became essential. The difference depended not only on the germ free desert but also on the fact that the wounded men were often wearing not more than a thin pair of shorts. A dangerous feature in Italy was often the germladen clothing in the depths of the wound.’ Clarke emphasised that debridement is to properly extend the wound and he argued against a needless excision: ‘the scalpel cannot and need not excise all micro-organisms’. These important lessons were soon forgotten and it seems that the principles of wound surgery must be relearned during conflict: at times it seems as if some civilian surgeons never grasped them. Cooper and Ryan (1990), in commenting on the trend in civilian practice to minimal surgical
Compound Nerve Injury
intervention and primary closure of wounds with increasing reliance on antibiotics, say ‘it is timely to remind our civilian colleagues that this approach, transferred to the battle field will result in catastrophe for many of the wounded’. No surgeon will overlook the excellent advice from Gustilo and Anderson (1976): ‘if there is the slightest doubt in the surgeon’s mind as to whether there has been adequate debridement of the wound after an open fracture, the wound should not be closed regardless of the type of open fracture. For the surgeon who manages only an occasional open fracture, the safe rule is not to close the wound.’ We have had to treat seven cases of gas gangrene in which fundamental principles had not been followed. For these patients only urgent amputation or radical excision of infected tissues availed. Case Report: An 18 year old man suffered a closed traction lesion of the brachial plexus with rupture of the axillary artery, and closed fractures of humerus and both forearm bones (Fig. 8.2). Circulation was restored through a prosthesis. At his arrival to St Mary’s Hospital, 18 h after this operation, the diagnosis was immediately obvious. He had a heliotrope complexion, a extraordinary sang froid and a racing pulse. Gas was palpable under the skin of the arm. Emergency amputation was performed through the upper humerus; the skin flaps were closed 72 h later. The errors which led to this outcome include: failure to stabilise the skeleton; failure to decompress the limb; and to a lesser extent use of a prosthesis rather than a vein graft. Case Report: A 28 year old man suffered fracture of femur, complicated by a sciatic nerve palsy in a road traffic accident. The fracture was open, and the skin wound which was 3 cm in diameter was sutured before the patient was transferred to us. He arrived with gas in the muscles of the thigh. Excision of all of the flexor muscles of the thigh was necessary. One of us is reminded of the demonstration by Peter London, at the Accident Hospital in Birmingham: on being presented with a thigh after the fracture had been reduced and the skin closed, London immediately manipulated the limb so that the ends of the fracture were forced through the wound. Road tar and a cigarette stub were impaled onto a spike of the femur.
Fig. 8.2 Gas gangrene.
305
8.1.1 War Wounds: Current Practice Eardley et al. (2009) restate the principles of the treatment of war wounds based on their experience in current conflicts. ‘the aims of war wound management are to prevent infection and enable safe and timely closure. All war wounds are considered to be contaminated. The culture medium is eliminated by wound excision, decompression and antibiotics.’ They set out four principles. I. The effects of ballistic injury extend beyond the area of the wound and thorough primary and secondary surveys with careful records are essential. II. The administration of antibiotics, of prophylaxis against tetanus and of agents to stabilise blood coagulation are essential steps during the initial phase of resuscitation. III. Wound surgery is staged. Whilst the objects of the first operation include the removal of contaminated and devitalised tissue this must be undertaken with a view to subsequent procedures particularly in relation to the provision of soft tissue cover. IV. The outlook for recovery within the injured limb is largely determined by the success of the first operation. Eardley and his colleagues identify eight stages which are essential for success. 1. Preparation: a pneumatic tourniquet may be applied before non sterile cleansing. Towels are placed so that there is no hindrance to the exposure of axial vessels or of potential vein graft. This is done rapidly. 2. Lavage is performed using 6–9 L of warmed normal saline at low pressure, not by pulsatile delivery. 3. Excision of skin is confined to that which is clearly dead at the edges of the wound. 4. Extension of the wound by longitudinal incisions, suitably modified across joints reveals the true extent of ballistic injury. 5. Inspection which permits removal of all non viable tissue, requires exposure of all structures in the wound cavity. Damaged muscle is excised back to healthy tissue, recognised by its contractility, consistency, colour and capillary bleeding. Unattached bone fragments of foreign material are removed and the search for small metallic fragments and intact bullets should not be prolonged. On the other hand mud, fragments of clothing and organic material must be removed. Main vessels are repaired. Frayed tendon ends are trimmed smooth. Disrupted nerves are carefully tagged by a fine suture to each other or to adjacent soft tissues. No nerves are repaired at the first operation. 6. Decompression is extensive. In the upper limb this should include opening the carpal tunnel and other compartments within the hand. The risks from swelling within the muscle compartments are increased by ventilation during evacuation.
306
Surgical Disorders of the Peripheral Nerves
7. Stabilisation of fractures is performed early. All methods have their advantages and disadvantages; the poor application of any one of them usually leads to a poor result. 8. Dressing, ALL WOUNDS ARE LEFT OPEN. They are lightly packed with fluffed gauze dressing. The injured limbs are elevated. 9. The ‘second look’ operation is performed at 3–5 days later. Dressings are left undisturbed so long as the patient remains apyrexial and does not experience pain or undue swelling of the limb proximal to the wound. Now it may be possible to proceed to delayed primary closure (DPC) depending on the condition of the wound, the patient and the requirement for more elaborate methods of fracture stabilisation and skin cover (Figs. 8.3 and 8.4). Eardley and Stewart (2010) describe the application of these principles in a seminal paper describing the early management of ballistic hand trauma. They emphasise that excision must be precise and limited to dead tissue and also that any viable tissue, such as fragments of bone which have retained their blood supply or flaps of skin which are still innervated must be kept as potential ‘building blocks’ in later reconstruction. Early rehabilitation is central to the concepts developed by Eardley and Stewart and the hand is splinted in the position of function with the wrist extended, the metacarpophalangeal joints lightly flexed and the proximal interphalangeal joints extended before and after the first operation. An important observation is made about the organisation of the operating team: ‘when an estimated 2 h or more of damage control surgery of the abdominal cavity or a lower limb surgery is anticipated, we use a team of three surgeons, one of whom has expertise in the hand and operates solely on the injured hand’. The policy described by Eardley and Stewart is driven by the fact that so many of the wounded have suffered severe injuries or even amputation of the lower limbs so that it is important to restore the highest level of function possible in the upper limbs and hands.
8.2 The Vascular Lesion Advances in arterial and venous surgery, with the introduction of antibiotics, blood transfusions and improvements in anaesthesia are the basis of modern surgical practice particularly in the treatment of severe wounds. It is a source of dismay that trends in training make it less likely that orthopaedic surgeons of the future will be as capable of dealing with the injured artery, or for that matter coping with the problem of skin loss as were those who contributed to these advances. Bonney (1963) and Kirkup (1963) successfully dealt with femoral arterial injury caused by fractures of the femur. We think that the first successful replant of the hand was performed by Mr David Harris, an orthopaedic surgeon, and that
Fig. 8.3 Penetrating missile wounds of leg treated by debridement (above) external fixation (middle) and rotation of a full thickness skin flap to cover the bone (below) (Courtesy of Major K Brown RAMC).
one of us was perhaps the first to successfully replant an amputated thumb in the United Kingdom. We think that Bonney (1963) ‘Surgeons treating fractures of long bones must be prepared to treat associated vascular injury’, London (1967a) ‘there is nothing so difficult about sewing up an artery as to make it the strict preserve of the vascular specialist’, and Birnstingl (1982) ‘the ability to repair blood vessels should now be part of the repertory of every accident surgeon’, are right.
Compound Nerve Injury
307
Fig. 8.4 Blast wound of hand, treated by emergency repair of the ulnar artery, simple stabilisation of the skeleton and delayed closure of the skin. Fig. 8.6 This soldier suffered multiple fragment wounds. He held his face together with one hand and he used the other to compress the left femoral artery which had been transected.
Fig. 8.5 A glass wound to the axilla transected the axillary artery. Spasm of the stumps prevented catastrophic bleeding.
Hunter (1793) noted the effect of transection of major arteries in the limbs of the boar and the dog. The vessels constricted ‘before the animal weakened’ (Figs. 8.5 and 8.6). Hallowell (1761) wrote to John Hunter telling him of repair of a brachial artery by means of a Farriers stitch. Murphy
(1897) was probably the first successfully to repair a severed femoral artery. The foundations of modern surgical technique were laid by Carrel (1902), who was awarded the Nobel prize, and by Guthrie (1912), whose contributions were no less significant. Pringle (1913) used vein grafts to repair two cases of aneurysm. Sherrill (1913) described successful suture of a brachial artery ‘by the method of Carrell employed by Crile’ and there are occasional reports in these years from German surgeons such as Lexer (1913). The First World War saw important work. Makins (1919) advised suture of arteries in selected cases. The Serbian surgeon Subbotitch (1913), described repairs of 135 arterial injuries and of 32 arteriovenous fistulae during the Balkan Wars. The Russian surgeon Weglowski (1925) described 51 vein grafts for arterial repair in the wounded of the Eastern Front. The Second World War inevitably saw advances. The work of the Vascular Injuries Committee of the Medical Research Council, chaired by Sir Thomas Lewis, was summarised in the MRC War Memoranda No. 13 (1944). We
308
think it should be re-issued, slightly modified, to all accident and orthopaedic surgeons: ‘The memorandum is intended for guidance of those who have had only limited previous experience of the early treatment of arterial wounds.’ It described pathological changes and principles of diagnosis, and strongly recommended axial anastomosis for stab wounds of vessels and the use of artificial tubes where tension was too great for end to end suture: ‘In the war of 1914– 1918 Tuffiers cannula usually failed in its purpose, because thrombosis occurred within a few hours; the use of heparin, however should prevent this.’ This technique was designed to maintain blood flow for 3–4 days before removal of the tube and ligation of the artery, giving time for increased collateral circulation. The memorandum referred to early work with free vein grafts, commenting that heparin was necessary, and adding ‘they are considered unsafe for use in heavily contaminated wounds and are likely to be applicable only in special vascular injury centres’. However, ligation was widely practised and the results of this are shown by DeBakey and Simeone (1946, 1955) in a series of nearly 2,500 arterial injuries. Three hundred and forty six from 502 limbs were amputated after ligation of the popliteal artery. Overall, nearly 50% of limbs came to amputation after ligation of an axial vessel. The results following arterial repair in 120 cases were a good deal better even in those difficult times when the average interval between wounding and operation was 10 h. Arterial repair was regularly practised during the Korean War, a conflict in which the incidence of arterial injuries in those surviving wounding increased and the interval from injury to operation dropped to 6 h. An amputation rate of 13% was recorded by Hughes (1958) and by Jancke (1958). One significant publication describing experiences in civilian practice came from Porter (1967) who described 100 cases of arterial injury treated at the Accident Hospital in Birmingham between 1956 and 1965. Eighteen of 19 major vessels were successfully repaired. In 48 cases the condition of the limb was such that amputation was necessary. Morris et al. (1960) reported 220 cases. Arterial repair was successful in 86%. Fourteen major vessels were ligated early in this series, which led to five amputations, hemiplegia in two of three ligated carotid arterial lesions and two deaths. The conflicts in Vietnam and Ulster saw an evolution of emergency arterial repair so that failure should be the exception. In 1966 the Vietnam Vascular Registry was established at the Walter Reed Hospital, carrying on a tradition of centralisation of information so permitting analysis and improvement, a tradition established in this country and now, sadly, abandoned with the same alacrity as the bandwagon of audit, outcome studies and clinical governance rolls inexorably forwards. Rich et al. (1970) reviewed 1,000 arterial injuries: in 42% there were associated peripheral nerve injuries; there were fractures in 28.5%. Hewitt et al. (1973) and Rich et al. (1975) discussed 558 false aneurysms and arterio-venous fistulae. The popliteal artery remained a problem; an
Surgical Disorders of the Peripheral Nerves
amputation rate of 30% was recorded by Rich et al. (1974). The principles of urgent arterial repair are set out in Barros d’Sa’s (1992) outstanding chapter in Eastcott’s work describing intra-luminal shunts for the early restoration of blood flow to the leg after knee-capping injury in which a close range shot gun blasts the contents of the popliteal fossa and fragments the femur and tibia. He comments that shunts have revolutionised the treatment of vascular injuries complicating fractures of the lower limbs. Barros d’Sa set out the principles of treatment in 188 major vascular injuries in the lower limbs of 118 patients in his Hunterian Lecture of 1982 (Barros d’Sa 1982). These include: revascularisation within 4 h; fasciotomy of all four compartments; vein grafting for the artery, and repair of at least one major vein. This led to an amputation rate of 5% for cases of popliteal artery arterial injury, the lowest of any civilian or military series yet published. Most patients were admitted within 30 min, indeed one half were admitted within 15 min of injury. Shunts permit adequate fixation of a fracture before definitive repair of the vessels. We are in no doubt that Barros d’Sa’s Hunterian Lecture stands as one of the most important contributions to the surgery of injury ever published. We make no apology for repeating the comments of Hughes and Bower (1961) who write of the incidence of gangrene after ligation of a major artery: ‘It is realised that these percentages, derived from battlefield wounded may be slightly higher than may be encountered in many injuries in civilian life; these should serve as a grim reminder of the dangers of ligation of major arteries and may deter some individuals so inclined. We have been amazed at the lack of interest and knowledge of many surgeons who have no idea as to which vessels may be safely ligated.’ Harsh words! Sadly we have found them applicable today. Perhaps now is as good a time as any to comment on arterial spasm. We have never seen a case of spasm alone producing obstruction in an adult; we have seen it in children. As to sympathetic block, or sympathectomy, Seddon (1964) when writing of acute ischaemia said ‘let us hope that the completely futile sympathetic block will not have been done’. Birnstingl (1982), vascular surgeon to the Royal National Orthopaedic Hospital for some years, said ‘sympathetic block is useless’; ‘sympathectomy has no place in the treatment of acute ischaemia; spasm should never be diagnosed unless the artery has been exposed at operation and either distended with saline or opened to exclude an intimal tear’. There is, too, the view of Eastcott and colleagues in his monumental work Arterial Surgery (1992): ‘sympathetic block or denervation are practically useless in the treatment of acute limb ischaemia’. The concept that arterial spasm is induced by the sympathetic nerves was demolished by Kinmonth and his colleagues (1956) who were unable to induce spasm in large arteries by stimulation of the sympathetic system or dispel it by sympathectomy, and finally laid to rest by Holden (1979) and Eadie (1979) in two articles of exceptional importance.
Compound Nerve Injury
309
Table 8.3 Arterial injuries in association with nerve lesions: including the supraclavicular plexus and penetrating missile wounds (1975–2006). Closed Open Total Repaired Total Repaired – 203 – 122 – 374 – 291 4
0
4
1
Subclavian
82
40
13
9
Axillary
63
54
35
27
Brachial
17
11
35
31
Radial and ulnar
5
2
224
198
Common femoral
0
0
4
3
Superficial femoral
2
2
4
4
Profunda femoris
2
1
2
1
Vertebral
Inferior gluteal
2
0
1
0
14
9
32
10
Posterior tibial
3
1
7
4
Anterior tibial
9
2
13
3
Popliteal
Table 8.4 Traumatic false aneurysms and arterio-venous fistulae with associated nerve lesions 1875–2006. Aneurysms – 75 Arterio-venous fistulae – 13 Region – artery Posterior triangle of neck
5
2
Axillary
44
3
Brachial
11
3
Iliac
1
2
Femoral
3
1
Profundus femoris
2
0
Inferior gluteal
1
0
Popliteal
5
2
Posterior tibial
2
0
Anterior tibial
1
0
Our own experience in the treatment of arterial injuries complicated by nerve lesions of one sort or another is set out in Tables 8.3 and 8.4. Methods of exposure and repair are described in Chapter 7. Arterial lesions are sufficiently common a complication of injuries to three joints, the shoulder, the knee, and the elbow as to merit separate consideration.
8.2.1 The Shoulder ‘Arterial injury should be suspected when there is a large and sometimes pulsatile axillary swelling even though a radial pulse be still perceptible’ (London 1967b). As Barros d’Sa(1992) pointed out it is sometimes difficult to recognise critical ischaemia after injuries to the axillary artery because
of the extensive anastomotic network of collateral vessels around the shoulder: ‘the distal limb may look perfectly viable to the extent that even the pulse may be maintained … failure to recognise and surgically correct an axillary artery injury may result in an expanding false aneurysm and marked compression of the cords and nerves of the brachial plexus within a limited space’. The persistence of peripheral pulses with normal capillary refill may lead the unwary clinician to neglect the underlying arterial injury. The consequences of neglect of a late bleed from a false aneurysm were described in a report (Syed and Williams 2002) of an 80 year old woman who suffered a fracture of the humerus. Bleeding recommenced after 4 months leading to erosion of the humerus. The diagnosis was not made for 2 weeks, by which time the upper limb was ischaemic and disarticulation of the shoulder was necessary (Fig. 8.7). The dangers of late closed reduction were shown by Flaubert (1827) who described attempted late reduction of a dislocation by traction applied by six (!) medical students which led to rupture of the axillary artery, avulsion of the brachial plexus, injury to the spinal cord and death (perhaps this experience stimulated his son to describe the fate of the wretched Hippolyte in Madame Bovary); and by Calvet et al. (1942) who collected 90 cases of arterial injury from dislocation, in 68 of whom late reduction had been attempted. There was a 50% mortality. Watson-Jones (1936) reported a fatal outcome from rupture of the axillary artery in a man with recurrent anterior dislocation of the shoulder. Bigliani et al. (1991) said that ruptures of the axillary artery, secondary to shoulder fracture or dislocations, account for 7% of all arterial injuries. Puri et al. (1985) made a significant observation: ‘the intima had fractured cleanly and circumferentially’. Mullett et al. (1998) are not alone in urging swift action and avoiding delay for the sake of angiography. Amongst others, Gallen et al. (1984), Nash et al. (1984) Laverick et al. 1990 and Helm and Watson (2002) urge alertness and swift action (Fig. 8.8). Emadian and Lee (1996) say ‘Regardless of the mechanism, vascular injury must be suspected in all cases of anterior shoulder dislocation where there is evidence of plexopathy or isolated neuropathy … delayed diagnosis and treatment invariably leads to permanent neurological deficit and morbidity’. In one of our cases, of fracture/ dislocation of the shoulder in a 59 year old man, the patient’s complaints of pain and the signs of deepening of lesion were not appreciated for several days. Angiography was done which showed bleeding from the subscapular artery. Two unsuccessful attempts were made to occlude this by embolisation. We first saw him at 6 weeks. He was in right heart failure. His haemoglobin was 4.9 gm per 100 mL. He was in great pain and there was a total and deep plexopathy. An MR scan which accompanied him showed an enormous haematoma occupying the axilla and the arm and suggested continuing bleeding. Treatment was simple: the axillary artery was exposed and controlled above pectoralis minor and the brachial artery exposed and controlled in the arm. More than 4 L
310
Fig. 8.7 Leaking aneurysm 20 weeks after fracture of the neck of the humerus in an 83 year old woman. She made a good recovery after repair of the defect. Above left. Radiograph on day of injury. Top right.
of altered blood were removed from the sac. The defect in the axial vessel was, at most, 2 mm in diameter and it was closed by direct suture. His pain was relieved. There was gradual recovery of the nerves but the small muscles of the hand never recovered. A much happier outcome was achieved in another case because of the alertness of Andrew Unwin (London), acting surgeon to the Ascot race course. A champion jockey fell at a hurdle and a horse following stamped on his left shoulder. There was immediate severe pain. We saw him within 12 h. Pain was intense: there was complete paralysis of deltoid and weakness and sensory loss involving the median and radial nerves. There was no bruit. The radial pulse was present. There was ecchymosis, bruising, and swelling in the lower part of the axilla. A clinical diagnosis of bleeding from one of the offsets of the axillary artery was made and this was
Surgical Disorders of the Peripheral Nerves
Radiograph at 18 weeks. Below left. MR scan at 18 weeks. Below right. Angiogram on the day of operation.
Fig. 8.8 Rupture of axillary artery from fracture of proximal humerus in a 68 year old woman. The robust collateral circulation contributed to delay in diagnosis. A large haematoma in the axilla caused pain and deepening of the nerve lesion.
Compound Nerve Injury
311
Fig. 8.9 A 78 year old woman sustained a double circumferential rupture of the intima of the axillary artery of the intervening thrombosis: in effect a floating segment of the vessel. Repair was successful.
confirmed by exposing and controlling the axillary artery using the inferior portion of the exposure of Fiolle and Delmas. The posterior circumflex humeral artery, which was 2.5 mm in diameter, had ruptured and 1.5 L of fresh blood were removed from the axilla and the quadrilateral tunnel. The artery was ligated. The ruptured circumflex nerve was repaired. The median and ulnar nerves recovered within days and there was early relief of pain. The first signs of reinnervation of the posterior deltoid were detected at 3 months after operation and by 9 months the general recovery was good, he was back riding and winning. By 15 months the range of movement at the shoulder was full, and the power of forward flexion, abduction and extension at the shoulder was measured at 90% of the uninjured shoulder (Patel et al. 2001). An
invited reviewer criticised the conduct of this case because angiography had not been performed. We thought this very odd: that bleeding was occurring in the axilla was beyond doubt and rapid exposure and control of the first part of the axillary artery was the obvious, unavoidable, and correct course. Embolisation would have done nothing to stem the bleeding from the distal artery which measured, in diameter, 2.5 mm, nor would it have decompressed the affected nerves. Repair of the circumflex nerve would have been delayed. The first injury in closed traction is fracture of the intima, which in more violent injury progresses to rupture of all coats of the vessel (Fig. 8.9). The intimal injury can be seen easily as a pale crescentic line, the damaged segment of the artery filled with thrombus. Attempts to restore flow using embolectomy catheters are invariably futile in such cases and cause such extensive longitudinal damage to the intima that later repair is impossible. Torrential bleeding after rupture of the whole vessel is unusual in the closed traction lesion. The tips of the artery generally contract down but in one case the proximal stump of the vessel was plugged by an adjacent ruptured nerve trunk. There was great haemorrhage when this plug was removed. The length of damage ranged from 3–28 cm. The brittle atheromatous vessels of older patients are vulnerable even in low energy injuries. Arteriography, whilst always valuable, should not unduly delay operation. A reasonable circulation to the skin is not the issue: what is important is whether muscle is perfused. Failure to restore flow through damaged axillary or common brachial artery within 6 h of injury is almost always followed by, at least, severe ischaemic fibrosis. In most of our cases of closed infraclavicular lesion we did not proceed to arteriography because the clinical evidence was clear. In these cases the posterior triangle was neither swollen nor deeply bruised; the subclavian pulse was palpable whereas in the infraclavicular fossa there was swelling with bruising of the skin and the brachial pulse was absent. The axillary artery was ruptured deep to pectoralis minor at the level of the coracoid process. It is important to remember that these severe complications may follow low energy injuries especially in an older patient. Stenning and his colleagues (2005) described 16 cases where arterial injury occurred after dislocation of the shoulder or fracture of the proximal humerus (Table 8.5). The diagnosis was made by the delayed onset of nerve palsy or the deepening of the nerve lesion whilst under observation.
Table 8.5 87 nerve lesions in 20 patients with arterial injury from low energy injuries to the shoulder. Median Ulnar Radial Musculocutaneous Circumflex
Total
Conduction block (Neurapraxia)
9
17
18
16
9
69
Favourable degenerative (Axonotmesis)
10
2
1
1
2
16 (recovery was incomplete in 4 cases)
Unfavourable degenerative (Neurotmesis)
1
0
0
0
1
2
20
19
19
17
12
87
Drawn from Stenning et al. (2005).
312
All patients had severe pain. Nine ruptures of the axillary artery were repaired by reversed vein grafts; in one, this was undertaken at the primary treating hospital. Of the seven ruptures of branches of the axillary artery, the defect was oversewn in five and repaired by vein patch in two. Anticoagulants were never used. The onset of nerve palsy was evident within 6 h in six patients and within 24 h in five more. It ranged from 1 to 7 days in two limbs and from 8 days to 16 weeks in seven more. In nearly every case the nerve palsy was initially incomplete and there was steady worsening in the ensuing 12–24 h. This deepening in cases of late onset of palsy is diagnostic of secondary haemorrhage. Most of the affected nerves recovered rapidly (conduction block); the remainder recovered more slowly(axonotmesis).
8.2.2 The Elbow Morris et al. (1960) described 53 wounds to the brachial artery. Forty nine of these were repaired; three came to amputation. Five arteries were ligated. From these there were two amputations and one death. We have seen 52 cases and have repaired 42 arteries. Seventeen arteries were ruptured in closed lesions. We agree with Barros d’Sa (1992) that the technique must be meticulous, using interrupted sutures under magnification. Some ischaemic fibrosis was detectable in the 22 patients in whom the brachial artery was not repaired or where delay before restoration of flow exceeded 6 h. The collateral circulation at the elbow is more robust than that at the knee. The alternative pathways include the profunda brachii artery which accompanies the radial nerve through the lateral intermuscular septum where it becomes the radial collateral artery. The superior ulnar collateral artery accompanies the ulnar nerve; in the adult it may be as wide as 1.5 mm at its origin hence its regular use for the free vascularised ulnar nerve graft. The inferior ulnar collateral artery is regularly seen during medial approaches to the elbow as a vessel which pierces the intermuscular septum to curl around the humerus between the triceps and the bone. On occasion, there is a substantial artery running with the median nerve. These vessels anastomose with recurrent branches from the ulnar and radial arteries. This network may maintain perfusion of the hand and the extensor muscles in cases where the brachial artery has been divided in the lower one third of the arm when there is no associated fracture or dislocation, but the situation is very much worse if the ulnar or radial nerve is entrapped within the fracture or compressed by haematoma, because collateral circulation will diminish or altogether cease. A similar situation obtains if a surgeon is so unwise as to apply a suprasystolic cuff in a pulseless limb. We are particularly grateful to Mr Philip Coleridge Smith (London) for his advice about flow through peripheral arteries and also for informing us about the work of Wajcberg and his colleagues (2006) who used high resolution
Surgical Disorders of the Peripheral Nerves
ultrasonography to measure the rate of flow through the brachial artery in adults and in children. Flow was calculated by multiplying the velocity-time interval of the Doppler flow signal by the heart rate and the cross-sectional area of the vessel according to Laplace equation: BF(blood flow) = [p × (D/2)]2 × FV(flow velocity). The mean diameter of the brachial artery in children aged between 4 and 5 years is 2.7 mm, which provides a resting flow of about 200 mL per min. The reader will at once note the significance of the diameter of the vessel from the equation of Laplace and this factor is emphasised by Poiseuille’s law which states that flow through a vessel is affected by three variables; the radius of the cylindrical vessel, the total tension in the wall and the pressure gradient. Poiseuille’s law is the physical law describing the volume of flow (F) of an incompressible uniform viscous liquid, where R is the internal radius of the tube, P the pressure difference between the two ends, h the dynamic fluid viscosity and L the total length of the tube: F=
π R 4 [DP ] 8η L
The diameter of the superior ulnar collateral artery at the elbow in children is no more than 1 mm. This calibre provides flow of about 20 mL per min assuming that the pressure gradient is the same as that in the brachial artery itself. These facts must be borne in mind by any clinician inclined to the view that cessation of flow through the brachial artery is a matter of little consequence. Case Report: A 20 year old woman fell downstairs suffering fractures of the distal humerus (Fig. 8.10). On arrival at hospital the limb was noted to be pulseless. The fracture was fixed through a posterior approach. The artery was not explored and the forearm was not de-compressed. Twelve hours later an operation was performed attempting to restore circulation, which failed. The patient was advised to undergo cervical sympathectomy, advice which she declined. She was then referred to St Mary’s Hospital where prompt decompression of the extensor muscles saved them. The flexor muscles of the forearm were necrotic; indeed, they were liquefied and were totally excised.
8.2.3 The Knee The arrangement of the popliteal artery, of its branches and the paucity of collateral vessels makes it particularly vulnerable to injury, adds greatly to the difficulties of repair and leads to the most severe consequences in the event of failure. We think these are amongst the most technically difficult of peripheral arterial injuries. Eastcott’s phrase ‘the lifeline of the leg’ is clearly shown by the amputation rate in such civilian series as Orcutt and colleagues (1983), who found survival of the leg after repair of the popliteal artery in 95% of the penetrating
Compound Nerve Injury
wounds, but in only 70% of those from blunt injury; and by a recorded amputation rate following popliteal artery injury from dislocation of the knee of from 30% to 80% (Lefrac 1976; Green and Allen 1977; Alberty et al. 1981) (Fig. 8.11).
313
Four fractures are notably associated with arterial injury. Arterial rupture occurs in 3% of supracondylar fractures of the femur, in 30% of dislocations of the knee, in 10% of fractures of the proximal tibia, and fractures of the proximal part of the fibula are associated with damage to the anterior tibial artery. The technical difficulties of repair increase with injuries close to the trifurcation of the popliteal artery. In such cases there is more likely to be gross injury to the skin, muscle and nerve. Case Report: A 58 year old man was knocked down by a car and suffered open fractures of femur and tibia, with rupture of the popliteal artery (Fig. 8.12). The limb was critically ischaemic and had been so for 2h by the time the patient reached theatre. The skeleton was rapidly stabilised by a Küntscher nail passed from the hip to the ankle. Then with the patient prone the femoro-popliteal trunk was exposed. The vessels were atheromatous, and the gap between the prepared stumps was 15 cm. A reversed vein graft from the distal femoral artery to the posterior tibial and to the anterior tibial (by means of a branch) successfully restored circulation. The fractures healed; persisting discharge from the tibial fracture ceased at 9 months after removal of the nail. The patient had no pain, normal function of the ankle and foot and a range of flexion of the knee between 0° and 80° (Figs. 8.13 and 8.14).
8.2.4 The False Aneurysm and Arteriovenous Fistulae Fig. 8.10 Damage to the brachial artery associated with severe supracondylar fracture of the humerus in a 20 year old woman. There is extensive occlusion of the brachial artery caused by intimal damage, thrombosis and spasm. There was infarction of all the flexor muscles of the forearm.
Fig. 8.11 Dislocation of the knee in a 68 year old woman was treated by manipulation and plaster of Paris. The appearance of the foot on removal of the splint some days later is shown.
Complete section of the artery does not inevitably lead to exsanguination as the vessel may constrict in spasm: an observation first made by John Hunter. The partial wound is
314
more likely to cause catastrophic bleeding but may also lead to the formation of a false aneurysm from a breach of the arterial wall. The sac contains thrombus, its wall is formed from perivascular soft tissues which may include nerve
Fig. 8.12 Open fractures of the femur and tibia, with rupture of the popliteal artery. Internal fixation and reversed vein graft to the anterior and posterior tibial arteries.
Fig. 8.13 The popliteal artery was ruptured by a fracture/ dislocation of the knee in a 31 year old man. The main artery was successfully repaired. The anterior tibial artery was ligated. Most of the anterior compartment of the leg had to be excised.
Surgical Disorders of the Peripheral Nerves
trunks. Aneurysms may also occur in a closed traction lesion of the infraclavicular plexus because of avulsion of such an offset as the suprascapular, circumflex, humeral or suprascapular arteries. We have encountered 33 such cases. Aneurysms and fistulae lead to a general decline of nerve function which is all the more rapid with expansion of the vascular lesion. Nerves embedded within the sac of a false aneurysm or fistula are badly damaged; recovery is always imperfect, and may not occur at all. Those more remotely affected will recover after prompt treatment. Pain indicates that nerves are closely related to the false sac; in these, complete loss of autonomic function is the rule even when deep pressure sense and other modalities are still partially preserved (Dunkerton and Boome 1988). In some of our cases diagnosis was delayed and in nine cases of false aneurysm the preoperative diagnosis was neurotmesis of an adjacent main nerve. Case Report: A 37 year old man suffered a perforating injury in the delto-pectoral groove from fine steel wire. There was immediate loss of sensation in the hand. There was no local swelling. The patient was seen by us 3 months after the injury, with loss of elbow flexion, complete median palsy and a tender swelling in the axilla with a positive Tinel sign radiating into the territory of the nerve. There was a good radial pulse. At operation a false aneurysm of the axillary artery was found. The musculo cutaneous nerve had been severed, the median nerve was wrapped around the wall of the sac. The sac was excised, the artery repaired by reversed vein graft, the musculo-cutaneous nerve was grafted, a median neurolysis was done. Good circulation was maintained. At 9 months the biceps muscle regained power to MRC 5. There was a Tinel for the median nerve in the middle part of the forearm, with recovery in the superficial forearm flexors.
Compound Nerve Injury
315
Fig. 8.14 The popliteal artery and vein were ruptured by dislocation of the knee in a 28 year old man. The vessels were successfully repaired and a four compartment fasciotomy was completed within 6 h of the injury.
Larger arterio-venous fistulae produce the local and systemic responses described by Sumner et al. (1992): ‘the drop in total peripheral resistance is the essential patho-physiological aberration’. This fall leads to a drop in diastolic pressure, an increase in pulse rate, increase in cardiac output, and ultimately to heart failure. There is hyperventilation. The local effects are a reflection of deprivation of blood which may lead to venous insufficiency. The part is swollen, dry and discoloured. The diagnosis is made by listening to the murmur which is continuous but louder in systole. The level of arterial injury is not necessarily related to the entry wound. One false aneurysm was caused by a gunshot wound in the opposite arm, 4 years previously: a fragment in this arm migrated around the chest wall and embedded itself in the wall of the axillary artery of the opposite limb. In another case of high velocity missile injury the entry point was at the anterior axillary line whilst the exit wound lay at the spine of the scapula. Arterio-venous fistula between subclavian artery and vein developed where they traversed the first rib. Case Report: A 28 year old soldier was wounded in the posterior triangle of the neck by a fragment from an exploding tank. He presented with severe pain, and complete loss of function of C5 and C6. A rumbling bruit in the neck with palpable thrill diagnosed an arterio-venous fistula. The lesion was successfully embolised by Dr Al Kutoubi (St Mary’s Hosppital). There was rapid reduction of pain, with considerable spontaneous recovery of the deficit in C5 and C6 (Fig. 8.15). Case Report: The following description is typical of a true causalgia. One of our patients, a man of 27, was shot in the
arm by a bullet from a hand gun during a street robbery. He was an innocent bystander. He experienced immediate, intense, burning pain in the arm, forearm, and hand, which worsened over the following days. Nothing relieved it. We first saw him at four weeks after injury. He was quiet, withdrawn, answering questions in monosyllables. He had lost weight, and was exhausted from lack of sleep. He protected the affected arm against his body; he could not tolerate examination, nor pressure or contact from his sleeve. Light draught provoked intolerable pain. The hand was swollen, mottled
Fig. 8.15 A bullet caused a fistula between the internal iliac artery and vein which was successfully embolised by Dr Al Kutoubi.
316
with alternating red and blue discolouration. There was profuse sweating. The nails were coarse, the hair on the back of the hand had overgrown. He had learned that some protection came by keeping the hand and forearm wrapped in a cool, moist cloth. There were deep but incomplete lesions of the median and ulnar nerves. There was a pulsatile swelling in the upper arm, and this, with a continuous rumbling bruit, led to the diagnosis of a large arterio venous fistula (Fig. 8.16). He was in right heart failure and the veins of the neck were so distended that sympathetic blocks or sympathectomy could not be considered. The fistula was successfully corrected by excision of 12cm of the axillary artery and replacing it with a reversed vein graft and by suturing the defects in the axillary vein. The median nerve was embedded in the wall of the fistula, the epineurium was disrupted and individual bundles were splayed out over sac. The ulnar nerve was more remote from the fistula. Pain relief was early, dramatic and enduring.
Surgical Disorders of the Peripheral Nerves
At three years the patient noted only imperfect sensation in the thumb and index but little else abnormal. Case Report: Lateral meniscectomy through the arthroscope was performed in a 31 year old man. He experienced intense pain on awakening which was not controlled by morphine. The leg became swollen, blue and cold. He was given anticoagulants on the suspicion of deep vein thrombosis. The true diagnosis of false expanding aneurysm of the lateral genicular artery was not made until it was too late to prevent massive ischaemic fibrosis of all muscles in the leg. We saw him 5 years after the event. One of us was sharply condemned by the lawyers and doctors involved in this case for our temerity in suggesting that 5 years without any form of active treatment was a touch unreasonable! Case Report: A 55 year old woman developed intense pain after removal of a lipoma in the axilla (Fig. 8.17). We saw her at 2 weeks because of suspected damage to the median nerve. The hand was red and dry, but not insensible; there was weakness of the flexor muscles of the forearm and the intrinsic muscles of the hand. The radial pulse was diminished. The arteriogram showed a block in the distal part of the axillary artery. At operation, a false aneurysm was displayed; its wall was formed from the brachial sheath and the median nerve which was splayed out: the ulnar nerve was displaced and compressed. Sumner, Eastcott and Rich (1992) emphasise that arteriography can be misleading in the peripheral aneurysm and that ultrasonography may be the more valuable investigation: of course first one must think of the diagnosis.
8.2.5 Ischaemia and the Nerve
Fig. 8.16 Large arterio-venous fistula in the axilla.
It is difficult, in the clinical situation, to separate the effects on the nerve of pure ischaemia from those of ischaemia associated with compression, traction and distortion. Certainly we see: (1) ischaemia of the nerve trunk itself. Seddon (1975a) described several cases. In one case of severe Volkmann’s ischaemic contracture, amputation was performed: the branches of the median nerve within the palm of the hand, where they were not compressed, were completely infarcted. Lamerton’s (1983) case is interesting as an example of painful ischaemia of the sciatic nerve relieved by the reconstruction of the internal iliac artery. (2) Ischaemia associated with compression or stretching of the nerve by haematoma or aneurysm, which is worse when the nerve is embedded in the sac. (3) Ischaemia of nerves compressed distally by swollen muscle. Parkes (1945) recognised the effect of compression by swollen muscles and noted venous obstruction rather like strangulation of a hernia leading to irreversible changes in the nerve at between 12 and 24 h. An intact intrinsic system may nourish a considerable length of the nerve trunk. In one case in which the entire flexor
Compound Nerve Injury
317
Fig. 8.17 Iatrogenous false aneurysm in the brachial artery.
compartment of the forearm was excised for ischaemia the median nerve survived because of the longitudinal supply over a length of 24 cm. The nerve recovered. In another case, in which the posterior muscles of the thigh were excised for gas gangrene the sciatic nerve was maintained by its longitudinal supply over a length of 38 cm. There was useful recovery. In one of our early hand replants we noticed that the median artery was so large that its repair provided sufficient circulation to the hand while more elaborate reconstruction went on. One of the best descriptions of the effect of ischaemia upon nerves, and upon nerves alone, and of the rapid deepening of the defect is provided by Wilbourn (2005). He, with Tsao described a compartment syndrome following axillary arteriography in 2003. Wilbourn suggests that bleeding into the medial brachial fascial compartment, which extends from the axilla to the elbow and which is bounded by the tough medial intramuscular septum and the axillary sheath, is responsible for the majority of infraclavicular plexopathies following axillary regional block and possibly for most of the neuro-vascular injuries which result from closed or penetrating missile injuries into this region. Sensory symptoms precede the motor. Reversible ischaemic conduction block is responsible for the initial symptoms but the lesion soon deepens into a degenerative one and this is marked by paralysis. The diagnosis rests upon the history and upon detecting
haematoma or ecchymosis in the axilla or arm. Wilbourn says: ‘distal pulses are normal, as they are with most compartment syndromes because the elevated pressure, although sufficient to collapse the vaso nervora, is far below mean arterial pressure. Ultrasound, MR, and CT may reveal the vascular lesion, but, considering the very brief time available for surgical decompression before irreversible nerve damage occurs, obtaining these is rarely justified.’ Patients explored within 4 h from onset of symptoms were eight times more likely to recover completely than those who were observed or treated with delayed operations. In these, there was permanent residual defect often accompanied by severe pain (Fig. 8.18). The lumbosacral plexus is vulnerable to haematoma. Allieu and colleagues (1986) studied the effects of haematoma on the femoral and sciatic trunks. Their advice is salutary: ‘conservative treatment of a haematoma is to be condemned’. Ultrasound is useful in diagnosis: decompression is a matter of urgency. The reader is referred to the excellent review of lesions of the lumbo sacral plexus from Donaghy (2005) who recognises two distinct syndromes of compression of the lumbo sacral plexus by retro peritoneal haemorrhage. In the first bleeding within the psoas muscle damages the obturator and femoral nerves. In the second, haematoma within the iliacus muscle compresses the lumbo sacral plexus. Because the
318
Fig. 8.18 The brachial artery was occluded by embolism in a 58 year old man. Embolectomy and fasciotomy were carried out 8 h later. There is fibrosis of the muscles of the forearm and continuing pain.
iliacus fascia is less capacious than that overlying the psoas muscle the effects are more severe. Donaghy describes how to distinguish between these two syndromes by clinical examination: ‘psoas muscle haematomas do not cause profound pain on forced hip extension, the hip is not held flexed, no haematoma is palpable or visible in the iliac fossa or groin, and there is weakness of the thigh abductors indicating involvement of the obturator nerve’. He emphasises that the important differential diagnosis in these syndromes is abscess secondary to appendicitis or following laparotomy. Urgent retrieval of a nerve from a joint or fracture, urgent correction of expanding haematoma, or urgent decompression in compartment syndrome is usually followed by rapid and complete recovery. Delayed action or inaction may lead to such damage of the nerve that repair is needed, or to irretrievable damage to the nerve, the muscle and other target organs. Effects on the tissue – compartment syndrome: When an artery is occluded there is death of tissue. It is wrong to apply the term compartment syndrome to the events which follow ligation of the popliteal or anterior tibial arteries, to the gangrene of peripheral vascular disease or the gangrene of arterial embolus. However, muscles confined within semi-rigid spaces bounded by bone and inelastic fascia are vulnerable to perfusion block, which may occur as a result of arterial
Surgical Disorders of the Peripheral Nerves
injury and in the event of delayed restoration of circulation, but it may also occur because of swelling within the compartment, as in bleeding from a fracture or haemophilia or infusion of fluids. The swelling following unaccustomed exercise or after the prolonged compression sustained in coma is more akin to the effect of delayed restoration of circulation. All these effects may be exacerbated by tight external bandages or splints. The term compartment syndrome was introduced by Matsen (1975) This is useful for it draws attention to the ‘final common pathway’ of ischaemia (Pellegrini and McCollister Evarts (1991). Of course, abnormal events precede ischaemia of muscle. There is collapse and block of flow through low pressure systems, first the lymphatic and then the venous. There is now tissue hypoxia and derangement of fluid and electrolyte exchange across cell membranes so that fluid passes from vessels to the extra vascular compartment. Barros d’Sa (1992) emphasises that the increased permeability of vessels from hypoxia leads to increase of pressure within compartments by exudation. The final event is the closure or obstruction of perfusing arterioles when tissue or extra vascular pressure exceeds the critical closing pressure of those vessels (Burton 1951). The vicious cycle is complete (Eaton and Green 1972). Fluid entering the compartment cannot get out; it continues to leak into the compartment until the pressure within that compartment is so high that inflow is blocked. The tissues will now die. Ellis (1958) found 4.3% of ischaemic contractures in his series of tibial shaft fractures and Owen and Tsimboukis (1967) found a clawing of toes from contracture of the deep flexor compartment in no less than 10% of such fractures. In the upper limb the compartments most at risk are the flexor compartment of the forearm, then the extensors of the forearm, the small muscles of the hand, especially the interosseous muscles and adductor pollicus, and the pronator quadratus. We have seen cases involving the deltoid and pectoralis major. In the lower limb the deep flexor compartment of the leg leads the way, followed closely by the anterior compartment. Then follows the other compartments of the leg, and the small muscles of the foot; the syndrome is not rare in the gluteal muscles and it also , on occasion, involves the muscles of the thigh. Holden’s review (1979) is particularly good. One cannot add to his criteria for clinical diagnosis and action which are, first and foremost, pain persisting after reduction of fracture and refractory to moderate analgesia: and tense swelling of the afflicted compartment. We agree with Holden that clinical suspicion is the most important indication for intervention and this must be acted upon before the onset of neural deficit. Ragland et al. (2005) described 24 cases of compartment syndrome in the new born. Urgent decompression is recommended – most of these infants required multiple operations but even so they were left with permanent defects in growth, nerve function and skeletal development.
Compound Nerve Injury
319
8.2.6 Iatrogenous Ischaemic Injury We have seen an alarming increase in the number of these cases over the last 10 years. Table 8.6 outlines 164 cases of post ischaemic contracture treated by palliative operations since 1980. More than one half of these were seen after 2000. A main artery was damaged during operation in 78 patients (Fig. 8.19). Most of the wounds to the popliteal and anterior tibial arteries occurred during elective arthroplasty, ligament reconstruction or arthroscopy of the knee. The posterior tibial and peroneal arteries were most often injured during operations for fracture. Correction of deformity or limb lengthening, using frames, accounted for some serious injuries throughout the leg. The 84 cases of compartment syndrome were related to operation; those directly caused by fracture, crush, coma or sepsis are excluded. Delayed intramedullary nailing of closed fractures of the tibia was responsible for 18 cases. It seems that some clinicians persist in the belief that oxygenation of muscle and nerve enclosed within rigid osseofascial compartments can be assessed by noting return of circulation to the nail bed. Others persist in the false notion that compartment syndrome is a distinct entity from the effects of anoxia caused by interruption of flow through an axial artery. The difference is one of degree. The final common pathway is the same: anoxia causes breakdown of homeostasis across the membrane of the cells of the capillaries, muscles, Schwann cells and axons. The annotation on compartment syndrome by Klenerman (2007) is compulsive reading: ‘In 1896, Starling had pointed out that under normal conditions, a state of near equilibrium exists at the capillary membrane, whereby the amount of fluid filtering outward through arterial capillaries equals the quantity of fluid which is returned
Table 8.6 164 cases of iatrogenous post ischaemic contracture of leg and foot treated by operation (1980–2007). A: Cause Lesion of popliteal artery
20
Lesion of anterior tibial artery
26
Lesion of posterior tibial artery
32
Lesion of peroneal artery “Compartment syndrome” Total
2 84 164
B: Compartments affected All four compartments
24
Anterior and lateral
46
Deep flexor
55
Deep and superficial flexor
32
Small muscles of foot
7 164
More than one half of these cases were seen after 2000.
Fig. 8.19 Massive false aneurysm from the lateral geniculate artery caused by the arthroscope during lateral meniscectomy. This was not diagnosed for three days.
to the circulation by re-absorption at the venous ends of the capillaries.’ To Klenerman’s abnegation of the ‘emergence of evidence based orthopaedics’ might be added that some recent meta analyses and ‘Cochrane reviews’ are flawed by ignoring earlier work, work published in other languages or the contributions made by other disciplines. Valuable monographs and reference texts are excluded. ‘Evidence-based’ medicine which relies on the partisan and selective review of recent literature is an ugly distortion of the work of Cochrane. The earliest symptom of peripheral ischaemia is pain, often so intense that it does not respond to morphine and alteration in sensibility in the extremity soon follows. The earliest fibres to suffer are the largest, those conveying proprioceptive and vibration sense. Deepening of sensory loss and the first signs of weakness of muscle are proof positive of critical ischaemia. One of the effects of anoxia is to increase the permeability so that to the true ischaemia caused by cessation of arterial supply is added that caused by rising pressure within the fascial compartment. The pathophysiology of ischaemia and the risks from re-perfusion are reviewed by Barros d’Sa and Harkin (2005) who outline the grading system for ischaemia developed by the Society for Vascular Surgery and the International Society of CardioVascular Surgery. Class 2A describes a limb which is
320
threatened, in which symptoms are limited to a mild sensory loss. Limbs with pronounced sensory loss, mild to moderate motor loss, even with audible Doppler signals are immediately at risk: for these, delay is unacceptable. The recommended treatment for Class 3 limbs, where Doppler flow is absent and where there is paralysis with total sensory loss, is delayed amputation which is performed after resuscitation as attempts to restore blood flow would lead to hyperkalaemia and myoglobinuria. The classification sets out treatment of the lower limb in which circulation has been imperilled by thrombo-embolism. Perhaps it is time to replace the much abused term ‘critical ischaemia’ by this classification. McQueen and Court-Brown (1996) have clarified the role of intracompartmental monitoring. Many of the cases outlined in Table 8.6 could have been prevented by their method. Chant (2005), in commenting on the recent recommendations from the Vascular Surgical Society (2001) says: ‘it notes the introduction of so-called “clinical governance” and goes on to say “general surgeons who are not vascular specialists undertake no active emergency arterial surgery, and it is difficult for them to justify treating vascular emergencies under the scrutiny of clinical governance”. In other words, if a general surgeon in the UK now performed an operation for a vascular emergency, and were that patient to die or perhaps lose a leg unnecessarily, then that surgeon may well be open to litigation.’ Chant adds a comment which must bring a chill to the heart of doctors and patients alike: ‘if you choose to live or work in a remote area, then it is unlikely that you will get the very best of medical treatment. That is a reality of life.’ Our experience suggests patients do not need to live in the Outer Isles to receive sub standard care. One wonders what might have been said about the actions of an experienced surgeon who encountered profuse bleeding whilst operating on a fracture of the proximal tibia. A cuff had been placed about the thigh but it was not inflated. He immediately packed the wound, inflated the cuff and turned the patient prone. The incision was extended, the popliteal artery and its branches were exposed and then, having secured control of the vessels, the tourniquet was released and he sutured the wound of the vessel. The operation of internal fixation of the fracture was then completed. The patient made an uneventful recovery. We think that the surgeon’s conduct in this difficult and indeed dangerous case was wholly admirable. Case Report: A 26 year old man sustained a tidy wound in the mid arm of his dominant upper limb. The brachial artery and median and ulnar nerves were transected. Spasm from the divided artery prevented copious bleeding and an opinion was taken from a vascular surgeon who considered that no action was necessary because the circulation to the skin of the hand was adequate. We were asked to see this patient at 8 months after his injury for consideration of repair of the nerves. They were exposed: the state of the proximal and of the distal stumps was such that no repair was worthwhile. This is a
Surgical Disorders of the Peripheral Nerves
tragedy because primary repair by any competent surgeon would have virtually guaranteed a high level of function. Matters are not helped by the imposition of changes, in reduction or withdrawal of medical staffing. Continuity of care has been abolished. A 33 year old man, who had previously served in the Armed Forces, was operated for reconstruction of a ligamentous injury at the knee. The operation concluded at 7 pm, and this is relevant for no member of the operating team was on call that night. The patient became aware of severe pain and increasing numbness of the foot 2 h after returning to the ward. At 4 h pain was intolerable, and resistant to opiates. The duty doctor was not a member of the orthopaedic department and took no action. The operating surgeon saw him early the next morning, (Saturday) and immediately took the patient to theatre where the leg was decompressed by open fasciotomy of all four compartments. The anterior compartment was infarcted and it had to be excised. Absence of a pulse in the presence of a fracture is an indication for exploration: pain, tense and tender swelling is an indication for decompression of a compartment even in the presence of a distal pulse. Sensible monitoring of intracompartmental pressure in the leg is useful. The correction of fixed ischaemic deformity is outlined in Chapter 13.
8.3 Skin The safe and regular closure of major defects in the skin is the third and most recent technical advance in the surgery of wounds. It rests upon the recognition that large areas of skin can be elevated on a pedicle of one artery and its vena comitans (as in the groin flap); that blood supply may pass to skin through perforating vessels from underlying muscle (the major myocutaneous flaps); or through deep fascia and intermuscular septa, (the fascio-cutaneous flaps) (Hallock 1992). McGregor and Jackson described the groin flap in 1972. This represented a real revolution for those brought up with the difficulties of split skin grafts or the cumbersome random or tubed flaps. This, the first axial pattern flap, remains exceptionally valuable in the emergency treatment of the hand and the distal forearm. During these years microsurgical techniques advanced so that there are now a large number of flaps which are in regular use for free tissue transfer. In 1986 Godina described the outcome in no less than 572 patients following ‘microsurgical’ reconstruction of the injured extremities, who were divided into three groups. Free flap transfer was performed within 72 h of the injury in Group 1, between 72 h and 3 months after the injury in Group 2 and later than that in Group 3. The failure rate of the flaps was 0.75% in Group 1, 12% in Group 2, and 9.5% in Group 3. Post operative infection occurred in 1.5% of Group 1, 17.5% of Group 2, and 6% of Group 3 patients. The average time to
Compound Nerve Injury
bone union was 6.8 months in Group 1, 12.3 months in Group 2, and 29 months in Group 3. The average of the length of hospital stay and the number of operations was far lower in Group 1. The evidence is decisive: delay only makes things worse. Godina’s contribution is a very great one, and his work is gracefully acknowledged in the review by Lister (1988) of emergency flaps. Lister and Scheker (1988) describe 31 emergency free flaps which were used for severe injuries to the upper limb. Twenty nine flaps survived, 27 of the 31 patients returned to work, 18 of them to their original employment. Gopal (2000) and his colleagues treated 84 cases of severe open fractures of the leg by debridement and excision of the wound extending to healthy tissue, stabilisation of the fracture and early soft tissue cover by a vascularised flap of muscle. The results are impressive and these workers observe that immediate internal fixation with healthy soft tissue cover is safe and that that most of the problems encountered were caused by delaying soft tissue cover beyond 72 h (Fig. 8.20). It is a considerable mistake for fracture or orthopaedic surgeons to decline knowledge of these advances. With vascular repairs, they must remain part of the repertoire of any accident surgeon. Surgical details, indications and complications are reviewed by Gilbert et al. (1990) who described flaps for the upper limb, and by Oberlin et al. (1994) for both upper and lower limbs. These last authors used a charming grading of between one and four for technical difficulty, akin to mountaineering terminology (or for the more sedentary reader, the Michelin star system). Grades one and two encompass those
Fig. 8.20 A 28 year old woman suffered a gunshot wound to the elbow destroying the brachial artery and the median nerve. These were repaired and skin cover provided by a free latissimus dorsi myocutaneous flap. Operation done with Professor Roy Sanders.
321
flaps which have served us well and which do not need elaborate equipment nor extensive experience in micro-vascular work, neither do the pedicle flaps, admirably described by Mathes and Nahai (1979). Grades 3 and 4, on the other hand, are for the expert! One most important lesson is that nerves which have been repaired or which are embedded in scar require healthy, full thickness and unscarred skin cover. Nothing less will do (Fig. 8.21 a, b). With these lessons in mind we turn to the first group of ‘compound nerve injuries’.
Fig. 8.21 Secondary revision of repair of median nerve and brachial artery with full thickness skin cover. A 23 year old man sustained a penetrating missile wound to the left antecubital fossa which destroyed the brachial artery and median nerve. The artery was repaired in the field hospital, the median nerve grafted five days later and the wound covered by split skin graft. The vein graft thrombosed at about 48 h after repair and the wound became infected. The nerve did not recover and the patient experienced severe pain. 10 months later the split skin graft was excised (above), a large neuroma had formed at the proximal stump of the median nerve (seen below) and occlusion of the vein graft was confirmed. A free full thickness skin flap was inserted and the vascular pedicle was used to repair the brachial artery so restoring through flow. The median nerve was, again, grafted. Pain was relieved and Tinel’s sign progressed distally at the rate of 2 mm a day. By 22 months he had recovered accurate and rapid localisation to the tips of the thumb, index and middle fingers, vaso- and sudomotor function and recovery of the thenar muscles. His power grip, 60%, was due in part to a high median tendon transfer (Courtesy of Mr Roderick Dunn, Salisbury).
322
8.4 Penetrating Missile Injuries Delorme (1915) then Inspector General of the Medical Services of the Armies of France outlined a method of treatment for shell and bullet wounds of nerves based on three principles: resection of scar until a healthy bed is secured; excision of damaged nerve until healthy stumps are reached; and tension free suture by adequate mobilisation and flexion of adjacent joints or grafting. His paper was heavily criticised. The verdict of history goes to Delorme and against his detractors. Tinel (1917) supported the proposals of Delorme adding ‘when the distance between the segments of the nerve trunk is too great to permit direct suture the only legitimate operation is nerve grafting as recommended by J and A Déjerine and Mouzon’. He further added ‘there is nothing but nerve tissue can serve as a conductor for regenerating axis cylinders’ and roundly castigated ‘mischievous’ operations such as lateral implantation, transplantation of a motor nerve into a sensory one and isolation of nerve by foreign body. Only the interposition of a ‘muscular-or better still, a fatty-layer’ between a nerve and callus or bony projection is permitted. It seems difficult to add to this. Owen-Smith (1981) and Omer (1991, 1998) point out that the effects of these injuries were well known in the nineteenth century. In 1894 Horsley (Adams 1982) showed that the wound was caused by transmission of energy from bullet to tissues and that the cause of tissue destruction was related to the velocity and spin of the missile. Woodruff (1898) introduced the term ‘cavitation’ for the damage caused at a distance from the wound track because of the creation of a cavity in the target into which air was sucked. Omer (1991) showed that neither energy transfer nor wound severity could be judged by the velocity of the missile alone. Indeed, he (Omer 1974, 1982) found, from a prospective study of 595 gunshot wounds, that spontaneous recovery occurred in 69% of both low velocity and high velocity wounds and that there was little difference in the time for recovery between the two groups. The Red Cross Wound Classification (Chapter 5) focuses the mind on the wound, not the weapon; Lindsey (1980) pithily condemns ‘the idolatry of velocity, or lies, damn lies and ballistics’. A number of writers have described the dangerous nature of the close range shotgun wound. Raju (1979) reported 39 shotgun wounds and 72 gunshot vascular injuries over 13 years, finding the shotgun injuries much the most severe. Shepard (1980) emphasised the significance of the mass of shotgun pellets in his series of 42 cases; so did Luce and Griffen (1978) who reported 77 cases of penetrating missile injuries of the upper limb collected over 7 years. Forty five of these patients had nerve lesions; the shotgun wounds fared particularly badly. Of the nine patients with combined injury to the brachial plexus and major vessels only one had a useful result and three came to amputation. We should not forget the lesson given us by McCready et al. (1986) who found good
Surgical Disorders of the Peripheral Nerves
recovery of the nerves after prompt decompression of haematomas compromising the brachial plexus in their series of 34 cases of open wounds of the subclavian and axillary vessels. The devastating effect of close range shot gun wounds was shown by Stewart and Kinnimonth (1993) from the Royal Infirmary in Glasgow, who treated 23 patients with 28 shotgun wounds to limbs in the course of 4 years. Their valuable paper uses the Red Cross Wound Classification. No limbs were amputated for vascular injury alone but three amputations in the lower limb were considered necessary because of the associated injury to sciatic or to tibial nerve. The functional outcome was disappointing in three more cases of shotgun injury to the forearm in which median and ulnar nerves had been torn. These were treated by early grafting, skin cover was achieved by free full thickness flap transfer. The authors comment ‘nerve repair was unsuccessful and left the patients with significant pain and disability two years after injury’. Stewart (Stewart and Birch 2001) recommends the term ‘massive energy transfer’ (MET) to describe those wounds, which with injuries caused by blast, do not fit conveniently into any method of classifying wounds (Fig. 8.22). Seddon (1975b) described the varying injury to a nerve following ‘through and through’ missile injuries. ‘the shock wave has caused more or less disturbance of function and structure in the nerve, ranging from fleeting paralysis (many combatants remark that the injured limb “went completely dead” for a few minutes) to neurapraxia and axonotmesis, and to internal disruption sufficient to cause moderate or severe intraneural fibrosis; all this short of actual severance of the nerve. The longitudinal extent of the fibrosis is a reflection of the distortion that the nerve has suffered and in this there is probably an element akin to traction’ (Figs. 8.23 and 8.24). Seddon thought that secondary repair is obligatory because only after a few weeks could the extent of intraneural fibrosis be recognised. In his series of 379 median and ulnar nerves damaged by missiles no less than 70% of them had been divided or partially divided. In commenting on nerves injured by missiles he said ‘patients who present with
Fig. 8.22 Shot gun blast to the neck. Bleeding from the first part of the subclavian artery was controlled through the transclavicular exposure.
Compound Nerve Injury
Fig. 8.23 Military rifle bullet wound. The entry wound, on the medial side of the knee was quite small, the exit wound was much larger. The nerve had not been transected, there was no compound nerve action potential detectable across the lesion but the lesion was not resected because of distal progression of Tinel’s sign. The nerve recovered.
323
group of scientists in Oxford who closely collaborated with Seddon, then in charge of the Wingfield Morris special hospital. They included Young, Medawar, Holmes, Sanders, Gutmann and Guttmann. Young summarised some of this work in 1993. The first evidence of axonal transport was revealed. The rate of nerve regeneration, the effects of disuse, the regeneration of proprioceptors, the maturation of regenerated fibres under retrograde influence and nerve grafting were amongst the fields of enquiry. Much of this clinical and the laboratory work was assembled and analysed by the Nerve Injuries Committee under the chairmanship of George Riddoch and it appeared in 1954 in Report No. 282 from the Medical Research Council, ‘Peripheral nerve injuries’. The quality and the scope of this document remain unsurpassed. ‘Aids to the examination of Peripheral Nerve Injuries’ appeared in 1943 based on the work of Riddoch, Ritchie Russell, Learmonth and McArdle at Gogarburn Peripheral Nerve Injury Centre near Edinburgh. This slim volume is now in its 4th edition under the direction of Michael O’Brian. All clinicians should possess it; even better, they should use it.
8.4.1 Suture and Grafting During World War Two: The MRCR Evidence
Fig. 8.24 Transection of the median and ulnar nerves by a bomb fragment were treated by delayed primary suture (Mrs Webb, Birmingham). Neurolysis was carried out 10 weeks later. Recovery for both nerves was good.
complete paralysis require exploration anyway, because there is no other means of knowing whether the nerve has been completely severed. Discovering it to be in continuity is fortuitous. The debate then centres on finding reason why resection and suture should not be performed.’ The lessons learned from the First World War were implemented soon after the beginning of the second. Hugh Cairns, Professor of Surgery in Oxford, was put in charge of injuries to the nervous system. Five nerve injuries centres were established as part of the emergency medical service. The three in England were directed by Rowley Bristow, Platt and by Seddon; the two in Scotland by Learmonth and by Illingworth. In spite of the hazards of war and dictates of geography concentration of cases, continuity of treatment and supervision was very good. Much fundamental work was undertaken by a
Zachary (1954) set out the results of 1,441 sutures performed in the five Special Hospitals. The follow up was remarkably good, with 1,108 patients presenting for review at a minimum of 3 years after the operation. Zachary points out: ‘the lapse rate for patients suffering from radial nerve lesions was nearly twice as great as that for any other group (82 out of 195, 42%). Radial paralysis is very disabling: no victim of it who had passed through a nerve injury centre would be ignorant of the possibility of tendon transplantation if the result of nerve suture was disappointing. Therefore, there is a strong probability that in this group at any rate, some patients defaulted because they were well satisfied rather than because they were disappointed.’ The results were graded by strict, even stringent, criteria and close reading of the Special Report reveals some results which would be pleasing to the most exacting modern clinician. In the case of the common peroneal nerve Zachary considers that a patient who regains active extension and eversion of the ankle against resistance as only M1(3) because of persisting paralysis of the extensor muscles of the toes. Most recent publications consider that a patient able to walk without an orthosis in a normal shoe has a pretty good result. Zachary described some findings which are of general interest. (1) Many patients continue to improve for up to 5 years after operation. All must be followed for at least 3 years before assessment of recovery can be made. (2) Delay before repair is most harmful. (3) High velocity injuries are
324
worse because of greater destruction of nerve tissue and also because more of these wounds are complicated by sepsis. (4) The more proximal the lesion the poorer the prospect of recovery. (5) About 50% of all cases of suture made a useful recovery by the exacting criteria used throughout the study. Three hundred and eighty four sutures of the ulnar nerve were reviewed at 5 years. Strong interosseous function was seen in 24% and useful recovery in 46% more. Recovery of interosseous function was at least useful in 78.5% of 141 low sutures of the nerve. There were 350 repairs of the median nerve. Ninety per cent of the high sutures (above elbow) regained useful grasp and a strong abductor pollicus brevis was found in one third. Useful sensation (S2+) was restored in 53%. Thirty eight per cent of patients regained some two point discrimination and lost any overreaction. The results of the 114 radial nerve sutures followed for a minimum of 5 years are particularly interesting: nearly 90% regained extension of the wrist and 61% regained some extension of the digits. Independent action for the thumb and the index finger was restored in 36%. The sciatic nerve was considered in terms of its two components, the tibial (132 cases) and the common peroneal (264 cases). Zachary seems to have been agreeably surprised by the results for the tibial nerve, with 56% of patients regaining powerful heel flexion and 80% showing some response to pin prick in the sole of the foot. Thirty six per cent of the common peroneal nerves recovered so that patients were able to walk without a brace or modified shoe, but 59% achieved this result when the level of suture was low, the delay short and the nerve defect small. There were 12 amputations following suture of the sciatic nerve because of deformity of the foot and ulceration on the sole: ‘amputation was performed for poor function of the limb and not primarily for poor neurological recovery’. Zachary also graded the recovery for 107 sutures of the sciatic nerve by the level of function. A good result implied no ulceration, no more than light pain, the ability to walk more than 1 mile and the ability to carry out work necessitating a fair amount of walking or standing. An excellent result meant that the patient could walk for 5 miles and could run, could carry out work requiring walking or standing and took no time off for at least 6 months, because of trouble with the leg. There was some discrepancy between neurological recovery and function: ‘eight patients with no recovery whatever had “good” function of the limb’.
8.4.2 Grafts Seddon (1954) described the results of 67 cases operated between January 1941 and October 1947. Nearly all of these were done at the Wingfield Morris Hospital, in Oxford. The outcome is analysed with the same precision that is a feature of
Surgical Disorders of the Peripheral Nerves
the memorandum as a whole. Twenty four main nerves were grafted using cutaneous nerves, by the method which Seddon described as cable grafting, perhaps a rather unfortunate term which has led to persisting misinterpretation since. As we have seen (Chapter 7) Seddon did not bundle the grafts together. The lesion of the main nerve was partial in six cases. These were unfavourable cases with large defects in nerves and frequent arterial injuries. The operations were successful in five of the six partial lesions and in eight of the eighteen complete lesions. There were four complete failures. Good recovery was recorded in one of five cases of repair by graft of the upper trunk of the brachial plexus; there was some recovery in a second case. A portion of a main nerve, usually from a proximal stump of an otherwise irreparable lesion was used to repair 15 median nerves. Useful function was restored in nine of these cases; there was no recovery at all in only two. Some results were remarkably good. In one case a 7 cm gap within the median nerve was bridged by using one half of the proximal stump of the nerve. The long flexor muscles had been destroyed. By 39 months the abductor pollicus brevis was graded at M4, there was recovery of sensibility without overreaction, fair localisation and two point discrimination averaging 7 mm. In another case (M.66) the proximal stump of a hopelessly damaged ulnar nerve was used to bridge a 7 cm gap in the median nerve. By 30 months the thenar muscles were graded at M3, there was recovery of sensibility and localisation without overreaction and two point discrimination lay between 15 and 20 mm. On the other hand the results using this method for repair of the trunk nerves of the lower limb were generally unsuccessful. Three out of the four nerve pedicle grafts succeeded. The successful case of grafting of the brachial plexus (patient Q.4) is, we think one of the first reported. Seddon (1975c) acquired enormous experience in the repair of war injuries to nerves, for in addition to the 379 cases of median and ulnar nerve repair he also described outcome in over 240 cases of missile injury to the sciatic nerve or its major divisions. Clearly, Seddon felt exploration was indicated in those cases with no clear evidence of clinical recovery as soon as the patient’s condition permitted. But Seddon used grafts for nerve repair at a time when this was not fashionable. The extensive experience of the military surgeons of the USA is brought together in the monumental work of Woodhall and Beebe (1956). On the basis of 3,418 nerve sutures in war wounds they said that: ‘delay in suture involves a loss of, on average, about 1% of maximal performance for every six days of delay’. Those clinicians, for there are some around, who countenance delay before dealing with injury to a main nerve might care to refresh their memories from the voluminous earlier published work. Omer has written extensively about peripheral nerves injured by penetrating missile wounds (1974, 1991, 1998); his works are required reading. Much of his work relates to
Compound Nerve Injury
the war in Vietnam. In 1974 he reported that only four from 34 grafts, followed for a minimum of 12 months, demonstrated return of functional activity but perhaps the period of follow up was rather short. He, with Eversmann (Omer and Eversmann 1994) came to the view that pedicle nerve grafting gave results superior to free nerve grafting. Omer (1974) sutured 83 nerves finding useful recovery in 40%. The worst group was the above elbow high velocity injury. Brown (1970) sutured 135 nerves in the upper limb; 44% of these regained useful function. The ulnar nerve fared worse but the continued long term study of the patients operated by Brown (Omer and Eversmann 1994) showed continuing improvement in function. Several cases of high repair of the ulnar nerve regained function in the small muscles of the hand at intervals exceeding two years after the operation. Omer (1998) says that: ‘the longest time elapsed before functional return was eight years after nerve suture. These well-documented late returns of neurological function to the intrinsic muscles of the hand are contrary to all results previously reported in the medical literature.’ The harmfulness of delay was emphasised by both Brown and Omer. Armed conflict within or between States continues and these have led to a number of significant contributions. Secer and his colleagues (2008) analysed 2,106 patients treated over the course of 40 years. Nerve injuries were equally divided between those caused by fragments and those caused by missiles. The median nerve accounted for 605 lesions, more than one quarter of the whole. 641 (30.4%) occurred in the lower limb. There were ruptures with separation of stumps or ruptures in continuity in 747 cases (35.5%), and partial ruptures in 191 nerves (9.1%). Lesions were recovering in 413 (19.6%). The most common cause of lesion was compression or entrapment within scar. There were 859 of these, 40.8% of the whole. Three main causes of nerve injury are identified: a shock wave leading to cavitation; affection from arterial injury, and direct impact upon the nerve. The characteristics of the lesion in continuity are described, and colour, consistency, and the state of the epineurial vessels are all pointers to the likelihood of recovery. Fibrosis was particularly severe in cases complicated by sepsis. Recovery was best for the tibial nerve, followed by the musculocutaneous and radial nerves. It was worst for injuries to the brachial plexus and the common peroneal nerve. The conclusion of this extensive work is important: earlier repair is more effective and there is a ‘strong correlation between functional recovery and relief of pain’. Gousheh (1995) relates his experience in the treatment of 369 injuries from the war between Iran and Iraq. Major arterial injury, false aneurysm or arterio-venous fistula complicated 74 of these. Using his own assessment Gousheh considered that results were good in 45 of 54 repaired spinal nerve elements, and in 86.5% of 107 cord elements. He found that repair of the ulnar nerve was worthwhile and concluded that results from repair in these injuries are better than those
325
caused by closed traction lesion. Gousheh et al. (2008) report 684 cases of injuries to the divisions of the sciatic nerve, incurred during the same war. 116 tibial nerves were repaired by graft, and 80 by direct suture. On occasion a segment of the common peroneal nerve or the ulnar nerve from an amputated limb was used. 308 lesions in continuity were extricated from scar or other entrapment and the general results for the tibial division were good. Powerful heel flexion returned in 86.3% of them and only a few cases were complicated by overreaction. 74 common peroneal nerves were grafted, 144 were sutured, and external neurolysis proved necessary in 272. Recovery for these nerves was remarkably similar to that found by Zachary. Useful dorsiflexion of the ankle returned in 46% of the grafts and in 41.7% of the sutures. The level of lesion was an important factor in outcome for both divisions. Gousheh and his colleagues observe that one reason for the vulnerability of the common peroneal nerve is that its vessels lie peripherally whereas those to the tibial nerve ‘lie in the crevices between the bundles’. The particular vulnerability of the common peroneal nerve is explored further by Hamdan et al. (2008) who studied no less than 1,463 civilians injured in Iraq over the course of 3 years. Ninety three per cent of these injuries were caused by bullets and 97% of the lesions occurred within the thigh. The tibial nerve was involved far less frequently; 64 cases were complicated by fracture of femur and/or arterial injury. Hamdan et al. refer to the tethering of the nerve above by the piriformis and found the nerve passing through the muscle in as many as one third of cases. The nerve below is tethered by the biceps and peroneus longus muscles. The nerve is smaller, it is endowed with less connective tissue and the blood supply is less rich than it is for the tibial nerve which is cushioned by the fat pad in the popliteal fossa and in the lower segment of the thigh. Roganovic and his colleagues from the University hospital in Belgrade offer extensive and important evidence drawn from experience in the recent Balkan wars. A paper describing 586 lesions of the ulnar nerve seen over three years, appeared in 2004 (Roganovic 2004). Operations were not performed in 181 of these. Bullets were responsible for just under 40% of the cases, fragment wounds for 60.9%. About one quarter of the nerves were ruptured in a different plane from the passage of the missile. Ruptures in continuity were characterised more by unusual colour, consistency, and poor circulation than they were by the usual neuroma formation: ‘such nerves were thinned, soft, “boil like”, or conversely, diffusely fibrotic especially if long-term local infection existed within the vicinity of the nerve’. There were 43 sutures and 85 grafts. Fifty six of 128 repairs restored material improvement in function, indeed the result was graded excellent in 10 cases. Deleterious factors include: the defect between the stumps; delay; and associated injury to the median nerve. A defect in the nerve exceeding 4.5 cm and a delay exceeding 5.5 months proved particularly deleterious. Roganovic points out that the
326
detection of a compound nerve action potential (CNAP) across the lesion is not necessarily associated with a good outcome. He draws similar conclusions from his study of 81 repairs of the median nerve (Roganovic 2005a). A high level injury, a gap exceeding 5.5 cm and a delay in excess of 4.5 months were materially deleterious to recovery. His study of the common peroneal nerve appeared in 2005 (Roganovic 2005b). No less than 702 cases were seen in the three years 1991 and 1994. The policy of emergency treatment in most of these cases included stabilisation of the skeleton by external fixation, early provision of good skin cover and approximation of the nerve stumps. Operations upon the nerve were performed in 472 cases; there were 157 repairs. Roganovic repeats his earlier observation, and it is an important one, that the nerve stumps were often found in different planes from the track of the missile and that they were: ‘sometimes grabbed by adjacent callus or abundant scar tissue’. Only 7 nerves could be sutured; grafts were necessary in 120. There was a high incidence of associated injuries: fractures in 45.9%, a major artery in 24.4%, and severe soft tissue injury in 24.8%. Results were useful in 57.4% of repairs where the gap was less than 4 cm. There were no useful results where the gap exceeded 8 cm. Useful function was regained in 47.9% of repairs performed within 4 months of injury but in only 3 of the 61 later repairs. Roganovic emphasises the importance of proper primary treatment of the wound and he recommends identification and formal tethering of the nerve stumps. In all, about one third of these repairs restored function, a proportion similar to that recorded by Zachary and by Seddon. Roganovic et al. (2007) discuss the association between false aneurysm and peripheral nerve lesions. 2,239 injuries to peripheral nerves were observed in the course of three years; 300 of these (13.4%) were complicated by arterial injury. There were 90 cases of false aneurysm which were directly associated with an adjacent nerve lesion in 22. Four of these were located in the axillary artery, seven in the brachial, three in the femoral and eight in the popliteal artery. The severe pain presenting in ten patients responded to correction of the arterial injury and the lesion of the nerve. Roganovic and his colleagues make this very significant observation: the lesion of the nerve progressively deepens with time and if a delay of more than 3–5 days is permitted a conduction block of the nerve is transformed into a much deeper injury. David Kline and his colleagues in New Orleans have acquired a great experience in the treatment of civilian gunshot injuries. Kline and Hudson (1995) describe over 150 cases of wounds to the lumbo sacral plexus, the femoral nerve, the sciatic nerve and its divisions. In 1989 David Kline (Kline 1989) reported 140 wounds of the brachial plexus (90 operated cases), and in nearly one third of these there was major vascular injury. 125 nerve elements were repaired and recovery was useful for C5, C6 and for C7 lesions, and for lesions to the upper and middle trunks and the posterior and
Surgical Disorders of the Peripheral Nerves
lateral cords but results of repairs of C8 and T1, of the lower trunk or the medial cord were generally rather poor except in children. Kline emphasises the role of detecting conduction across the lesion. Compound nerve action potentials (CNAP) were detected in 76 such lesions, and these were left alone. There was useful spontaneous recovery in nearly all of these. The later paper from New Orleans, (Kim et al. 2004) provides more information about 118 cases of gunshot wounds to the brachial plexus. 156 elements were repaired. Recovery was good in 14 of the 21 sutures and in 73 of the 135 grafts. Kim and his colleagues offer clear indications for operation after such injuries: the presence of causalgia or other severe pain; suspected or proven arterial injury or false aneurysm or fistula, and the failure of progression towards recovery for lesions of C5, C6 and C7 or their derivatives. It is possible to offer some general observations from these publications which together provide information about thirteen thousand cases of penetrating missile wounds. The first operation to the wound itself is most important; arterial injuries must be treated and sepsis must be prevented. General factors mitigating against the likelihood of recovery include: delay before repair; extensive damage to the nerve, to the adjacent soft tissues and to the skeleton; the presence of arterial injury, and the level of the lesion. It is clear too that some nerves do particularly badly; amongst these the common peroneal nerve heads the field. We continue to use the admirable operations of nerve pedicle transfer developed by Strange (1947) for otherwise irreparable injuries of median and ulnar nerve and the analogous operation developed by MacCarty (1951) for otherwise irreparable injuries to both divisions of the sciatic nerve. These methods are described in Chapter 7.
8.4.3 Experience from St Mary’s and the Royal National Orthopaedic Hospitals Between 1975 and 2000 we operated on 117 patients and repaired 116 ‘elements’. The indications for operation were: 1. In emergency – for severe wound and /or bleeding 2. Aneurysm or arterio-venous fistula with nerve lesion 3. For severe pain 4. For deep defect in a nerve or nerves The injuries were complicated by wounds to major arteries in 41 patients.
8.4.4 The Brachial Plexus Stewart (Stewart and Birch 2001) has studied 58 injuries of the brachial plexus, 51 operated cases, and his analysis of the
Compound Nerve Injury
327
different types of missile injuries is particularly helpful. He recognised three groups: 1. Bullet wound, usually from the military rifle or machine gun (32) 2. Fragment, from explosion shell or grenade (11) 3. Blast from close range shot gun wound (15) There was a high incidence of associated injuries of adjacent structures which had been treated by operation before referral. Major vascular lesions (24%) and chest injuries (38%) predominate. Emergency neurovascular repair had been performed in three cases. Injury to the spinal cord occurred in three patients and two others, with complete plexus lesions, had intradural injury and a Brown Séquard syndrome. The Red Cross Wound Classification was used to score, retrospectively, the features of each wound. The wounds were graded according to the amount of tissue damage: grade 1, low energy transfer (LET); grade 2, high energy transfer (HET); and grade 3, massive energy transfer (MET). The wounds were then typed according to the structures involved, soft tissue, vital structures or fractures. The nominal category of the wound was derived from a combination of the grade and type to give an indication of the complexity and severity. Shot gun blasts caused complex HET or MET wounds with severe and diffuse neurological injury. Bullets generally cause HET wounds but there were fewer disruptions of the plexus, most were single nerve injuries. Fragments usually caused small LET wounds. The indications for further operation were, in order of priority, known or suspected vascular injury (16), severe and intractable pain (33), and persistent deep loss of function in a major nerve trunk or trunks (51). The seven patients showing recovery were not operated. One of these suffered from causalgia but he experienced spontaneous remission some months after injury, a phenomenon described by Barnes (1954). The neurovascular axis was usually approached by the exposure of Fiolle and Delmas, but the transclavicular approach (Chapter 7) was used in three patients with
arterial injury. False aneurysms or arterio-venous fistulae were repaired in 13 patients. The first arterial repair was revised in two more. Normal flow was restored in 13 cases. One hundred and three damaged nerve elements were encountered. Some nerves were wholly transected, others partially so. Others showed the effects of displacement and traction, lying in a tortuous path resembling a ‘barley sugar’ (or Jacobean table leg) appearance of a traction lesion. The diagnosis for transected nerves did not present a problem but the lesion in morphological continuity certainly did. The distinction between degenerative lesions of favourable from those of unfavourable prognosis rested on the demonstration of intact bundles traversing the lesion, the detection of distal muscular contraction from stimulating the nerve proximal to the lesion and the presence of compound nerve action potentials (CNAP) traversing it. Examples of prolonged conduction block were seen in cases in which nerve trunks were displaced or compressed by an expanding haematoma or false aneurysm. In these, the diagnosis was established by demonstrating persisting conduction in the distal trunk. The ruptured nerves were grafted; external ‘neurolysis’ was used for nerves which were constricted or distorted in scar tissue. A fine catheter was placed adjacent to the plexus in patients presenting with severe pain, allowing infiltration of local anaesthetic for up to five days after operation. One or more elements of the plexus were repaired in 36 patients. Repair of the spinal nerves or of the upper, middle and lower trunks produced good or useful results in 10 from 14 patients. Repair of the divisions, cords or terminal branches of the plexus produced good or useful results in 16 of 22 patients. Three of the six repairs of the medial cord or ulnar nerve were rated as useful or better. The result was poor in all cases where there was an injury to the spinal cord. ‘Neurolysis’ or, as it should be termed, decompression, was generally useful and significant improvement in neurological function was seen in the 13 patients after repair of a subclavian or axillary aneurysm or fistula (Table 8.7).
Table 8.7 Penetrating missile wounds of the brachial plexus. Outcome Results of repair of 56 nerve elements in 36 patients Spinal nerve trunks Cords, nerves No. of No. of No. of No. of nerves patients nerves patients
Results of decompression of 47 nerves in 23 patients Spinal nerve trunks Cords, nerves No. of No. of No. of No. of nerves patients nerves patients
Good
2
2
1
1
13
7
5
4
Useful
12
8
20
15
11
3
15
7
Poor
15
4
4
4
0
0
0
0
Unknown
0
0
2
2
0
0
3
2
29 14 27 22 24 10 23 13 Results of repair were poor in five of the cases of spinal cord lesion. Pellets or fragments in the spinal canal were evident in these poor results Drawn from Stewart and Birch (2001).
328
8.4.5 Pain Stewart recognised four syndromes in 35 patients presenting with severe pain (see Chapter 12). Ten patients presented with causalgia, which was caused by aneurysm or fistula in five. All of these patients experienced rapid improvement or abolition of their pain after operation which included repair of ruptured nerves in eight cases. Sympathectomy was performed twice, in earlier years. This method was abandoned after we realised that correction of the vascular lesion, extrication of the nerves and repair of divided nerves with prolonged infusion of local anaesthetic after the operation achieved the desired result. All 19 patients with neurostenalgia experienced very great improvement after operation. The rapid relief of neurostenalgia after liberating nerve trunks from entrapment in scar tissue or callus, or after removal of a missile fragment, is characteristic. In all of these cases the nerve trunk was intact and the lesion was neurapraxia, or at worst, axonotmesis. In neurostenalgia the nerve is, in some way, irritated, tethered, compressed or ischaemic and treatment of the cause relieves pain. On the other hand, post traumatic neuralgia caused by rupture or partial transection of a nerve in four patients proved more difficult. Two patients experienced improvement in their pain, two did not. Case Report: A 21 year old man sustained multiple LET gunshot wounds. Emergency surgery included a right hemicolectomy, right thoracotomy and enucleation of an eye. A right supraclavicular wound was not explored. He discharged himself from hospital, but presented 6 weeks later with severe pain and profound loss of function in his right arm and hand. There was no major neurological deficit. The bullet was lodged in the transverse process of the sixth cervical vertebra; the pain was improved for about 12 h by blockade of the stellate ganglion. Digital subtraction angiography showed no evidence of vascular injury (Fig. 8.25). At operation, a bullet was found to be embedded in the scalenus anterior muscle, adjacent to the seventh cervical nerve. There was no evidence of direct injury to the spinal nerve, or to the trunks of the plexus, but considerable fibrosis surrounded the middle and lower trunks of the subclavian artery. The bullet
Surgical Disorders of the Peripheral Nerves
adjacent to the seventh cervical nerve was removed and a cervical sympathectomy was carried out. There was dramatic relief of pain after operation and at follow up at two years there was good function. The bullet adjacent to C7 caused diffuse, severe pain in the arm and functional disability. At the time, the pain was considered to be causalgia, and a sympathectomy was undertaken because of the good response to the sympathetic block. In retrospect, we considered the pain was probably a neurostenalgia; its dramatic relief with recovery of function should probably be attributed to removal of the bullet adjacent to the spinal nerve. Case Report: A 24 year old woman sustained an MET supraclavicular wound from a shot gun blast to her left brachial plexus with rupture of all five spinal nerves and a Brown-Séquard syndrome (Fig. 8.26). Within weeks of her injury she gave a clear description of post traumatic neuralgia. At operation, 2 months after the wounding, all five spinal nerves were grafted. Her pain remained intense for 6 months but it settled after 14 months. Two years after injury there was some recovery of finger flexion, the sternal head of pectoralis major, of deltoid, latissimus dorsi and triceps. Tinel signs were observed for the median and ulnar nerves in the middle of the forearm, and for the radial nerve, at the elbow. The injury to the spinal cord had improved considerably, but there was still some clinical evidence of Brown Séquard syndrome at four years after injury. Further recovery was largely confined to C8 and T1, with useful function in the sternal head of pectoralis major, triceps and flexor digitorum superficialis. Transfer of the third and fourth intercostal nerves to innervate serratus anterior achieved power in that muscle of MRC grade 5, and there was no longer a requirement for flail arm splints. Protective sensation, comprising warm perception, delayed pressure sense and some pin prick sensation, was present below the elbow. This was an unexpected and encouraging result after a MET wound. Since there were no pellets within the spinal canal, the Brown Séquard syndrome was thought to reflect an injury to the brachial plexus. Interruption of vessels accompanying the lower roots of the plexus may have provoked a partial and perhaps transient episode, affecting the ipsi lateral part of the cord. We consider that repair of the plexus was at least partially responsible for the relief of her pain.
8.4.6 The Peripheral Nerves
Fig. 8.25 Causalgia caused by a bullet which was embedded in the transverse process of C7.
Our earlier experience is summarised in Table 8.8. Severe pain was the presenting symptom in more than one half of patients. Causalgia was recognised in three cases of upper limb injury and in four more cases within the lower limb. Neurostenalgia, from embedded pellets or retained foreign bodies within the nerve trunk or displacement by the sac of a
Compound Nerve Injury
329
Fig. 8.26 Close range shot gun blast caused massive destruction of the paravertebral muscles. There were no pellets within the spinal canal. Table 8.8 Penetrating missile wounds of the peripheral nerves in the upper and lower limbs 1975–2000. Wound type No. of Nerves Vessels Fractures patients repaired repaired
Skin loss (flaps)
Fragment
15
13
8
5
3
Bullet
26
24
10
17
4
Blast
25
23
10
16
10
Total
66
60
28
38
20
Nerves transected: level of lesion Upper limb Median
Lower limb 18
Arm
Forearm
Hand
11
2
5
Sciatic
17
Hip
Thigh
Knee
Leg
7
9
1
0
Ulnar
18
13
2
3
Femoral
3
2
1
0
0
Radial
9
6
2
1
Tibial
7
0
4
3
0
Musculocutaneous
3
3
0
0
Common peroneal
10
1
3
5
1
haematoma or aneurysm was seen in the remainder. Pain was relieved by correction of the cause (Fig. 8.27). Bullet wounds are more likely to be associated with a fracture, and the treatment of these is often difficult. In seven cases where external fixation had been adopted we were faced with the problems of non union and sepsis. There must be no short cuts in the emergency treatment of the wound; debridement must be thorough and adequate irrespective of the chosen method of fixation to reduce the risk of sepsis.
Case Report: A 17 year old boy was shot in the arm with a Kalashnikov bullet. External fixation was used to stabilise the humerus. We first saw him at seven days, with deep sepsis in the arm which extended to the pin track site. Salmonella Type 5 was cultured from the fracture site, which was excised and packed with antibiotic impregnated cement beads. Two weeks later repair of the brachial artery and of the median and radial nerves was possible. At 3 months the non union was treated by internal fixation and interposed bone graft. The facture united; there was, ultimately, useful function in the hand.
330
Surgical Disorders of the Peripheral Nerves
Fig. 8.27 Sacular aneurysm from a bullet wound to the subclavian artery which was repaired using a Dacron graft (George Bonney 1964).
8.4.7 Recent Experience Nearly all service men and women injured in Iraq or Afghanistan pass through the Defense Medical Rehabilitation Centre at Headley Court and those with nerve injuries are seen in a dedicated clinic and investigated using the methods described in Chapter 5. Information is entered on the compound nerve injury form (see Fig. 5.80) suitably modified to the study of war wounds. The methods of grading results have been described in Chapter 4 and later in this chapter. Incidence, cause and distribution: The incidence of nerve injuries is about 12% which is higher than that seen in earlier conflicts. One explanation for this is the lower mortality amongst the wounded in current conflicts. This stands at 88.2%, compared to 76.4% for those wounded in the Vietnam war and 69.7% for those injured in World War II (Eardley and Stewart 2010). The incidence of nerve injuries is higher
in the lower limb than it was in the MRC Report because of the preponderance of wounds caused by blast or explosion. Most of the lower limb injuries occurred in massive energy transfer wounds (Table 8.9). The lesion of the nerve: the lesions of the nerves have been grouped according to the predominant quality of that lesion. Apart from those cases where the nerve was torn apart there were few pure lesions. Two patterns of conduction block have been recognised. The first is caused by distortion, displacement and traction upon the nerve by the passage of a nearby bullet or fragment. The second is caused by blast and recovery is prolonged and more complete for larger nerve fibres (see Chapter 3) (Table 8.10). Pain: No case of causalgia required active treatment although several patients related a history consistent with the diagnosis and its spontaneous resolution. Post traumatic neuralgia persisted in four patients with lesions to the superficial
Compound Nerve Injury
331
Table 8.9 Ballistic peripheral nerve injuries by nerve (a) and level (b) 232 nerves in 91 patients. (a) Upper limb (133 nerves) Above clavicle
Below clavicle
Buttock/hip Thigh and below
Ulnar
29
High: Above FCU
10
Cervical sympathetic chain
1
Inter: Above wrist
17
Spinal accessory
3
Low: Wrist and Hand
2
C5
7
Radial
22
C6
8
High : above triceps
5
C7
7
Inter: above PIN
14
C8
6
T1
4
Dorsal scapular
1
Suprascapular
6
Nerve to serratus anterior
1
Circumflex
6
Musculocutaneous
4
Radial
22
Median
25
Ulnar
29
Lower limb (99 nerves) Pelvis
Table 8.9 (continued)
S1
1
S2
1
S3
1
Inferior gluteal
5
Inferior gluteal
5
Femoral
3a
Tibial
38
Common peroneal
39
Sural
4b
Saphenous
2b
(b) Level by main nerves Common peroneal nerve
39
Buttock
7
Thigh
7
Knee
16
Leg
8
Ankle and foot
1
Tibial
38
Buttock
7
Thigh
6
Knee
14
Leg
8
Ankle and foot
3
Median
25
High: above superficial flexor branch
13
Inter: Between superficial flexor branch and wrist crease
10
Low: Wrist and Hand
2
Low: PIN 3 a The three femoral nerves were injured below the branches to ilio psoas muscle. b Sural and saphenous nerves were injured in the leg. Table 8.10 The cause, and significant associated injuries at the level of nerve lesion: 216 nerves (85 patients). Cause (216 nerves) Associated injury (85 patients) Explosion or blast
150 (69%
Main vessel
27 (31.5%)
Penetrating missile wound from bullet or fragment
66 (31%)
Fracture
48 (56.5%)
radial, the sural or saphenous nerves. Post traumatic neuralgia complicated two cases of below knee amputation in which the neuroma had become adherent to the scar. Pain was abolished by revision of the stump. Neurostenalgia, caused by the strangulation of a nerve by fibrosis deep to split skin grafts was more common (11 patients) and this was the single most important indication for second operation which involved the excision of split skin graft and replacement by a full thickness skin flap. Recovery: Post ischaemic fibrosis was identified in only one patient. In this case emergency repair of the brachial artery maintained flow for about 2 days before thrombosing. The moderate fibrosis of the flexor muscles of forearm was corrected by serial splinting and stretching. In two patients direct damage to the flexor muscles of the leg required subsequent step elongation of the affected tendons. The nerves: Recovery for the nerves was graded by the methods set out in Chapter 4 and later in this chapter (Table 8.11). Recovery was good in all cases of conduction block and in most of the examples of axonotmesis. Recovery in six of the nine primary sutures was good (three sciatic, one median, one ulnar, one tibial) and was poor in only one case of suture of a radial nerve. Recovery was good in 7 of the 26 grafts, it was poor in 10. Results in the three cases of direct muscle neurotisation and in the one case of frozen muscle graft were fair. The quality and speed of recovery in the successful primary sutures was remarkable and these included three repairs which were performed in the field hospital. Decompression and neurolysis proved necessary in 23 nerves and in 10 patients this involved replacement of scar by a full
332
Surgical Disorders of the Peripheral Nerves
Table 8.11 The nature of the lesion in 211 nerves and the outcome of that lesion in the 199 nerves adequately followed. The lesion Outcome Conduction block (neurapraxia)
75 (36%)
Good: including 13 repairs and 21 neurolyses
133 (67%)
Favourable degenerative (axonotmesis)
81 (38%)
Fair: including 15 repairs and 2 neurolyses
34 (17%)
Unfavourable degenerative (neurotmesis)
55 (26%)
Poor: including 11 repairs
32 (16%)
thickness skin flap. There were 21 good results. Early relief of pain and rapid recovery of function was evident in all patients following revision of the scar. Ramasamy and his colleagues (2009) conclude that: 1. Loss of sensation in the sole of the foot is not an indication for amputation because of the high incidence of prolonged conduction block or axonotmesis of the sciatic or tibial nerves. 2. Proper treatment of the wound at the first operation with restoration of circulation and perfusion, with stabilisation of the skeleton together with a rigorous approach to the control of pain enables early repair of divided nerves. 3. Neurophysiological evidence is valuable in confirming the level and the depth of lesion. Recovery is best monitored by clinical examination and, in particular, by the behaviour of the Tinel sign. Persisting neuropathic pain is a strong indication for exploration of the nerve. 4. Nerves and vessels must be protected by full thickness healthy skin. The condition of the patient and of the limb often precludes elaborate and lengthy operations for the purpose of skin cover. In these cases its seems better to defer the nerve repair until that has been done. (Seddon 1954) (Figs. 8.28 and 8.29).
Fig. 8.28 Bilateral lesions of the median nerve from penetrating missile wounds. The right median nerve and brachial artery were repaired by Mr Power (Birmingham). Recovery was good. The injury to the left median nerve was accompanied by destruction of the flexor muscles and fractures. Recovery was poor.
8.5 Neurovascular Injuries: Amputation Revascularisation In these injuries the prognosis of the nerve is determined above all else by successful restoration of arterial perfusion. The most extreme example is, of course, the amputated part. Malt (1964) and Chen, Ch’en and Pao (1963) successfully reattached an amputated hand. Komatsu and Tamai (1968) reported the first successful reattachment of amputated thumb and Cobbett (1969) the first successful free transfer of a great toe to replace a lost thumb. Buncke, Buncke and Valauri (1991) reviewed the technical developments which permitted these achievements and indeed were spurred on by them. Chief amongst these was
Fig. 8.29 Penetrating missile wound to right forearm destroyed the flexor muscles and damaged both median and ulnar nerves. The median nerve was grafted and a lateral thigh flap (Mrs Webb Birmingham) was performed at 5 days. Both nerves recovered. There was never, at any time, pain. There was a partial lesion of the median nerve in the left forearm.
the regularly successful repair of vessels of 2 mm in diameter or less, by interrupted sutures of 9×0, 10 × 0 or 11 × 0 filaments on needles of 2–4 mm in arc and 50–100 microns in
Compound Nerve Injury
calibre. Kleinert and colleagues were amongst the first to describe these techniques in limb salvage (1963). Chen, Meyer, Kleinert, Beasley (1981) reported the results of a co-operative study between Shanghai, Louisville and Zurich in 181 cases. The best results followed replantation of limbs amputated through the distal forearm and wrist with useful results in 79% to 83% of cases; the worst results were from replantation of the limb amputated or avulsed at shoulder level. Avulsion injuries fared very much worse than guillotine or crush injuries. The extent of vascular damage is, of itself, no barrier to successful replantation. In one of our cases a 28 year old woman threw herself under a train in an attempt at suicide. The dominant hand was amputated at the junction of the distal one third and proximal two thirds of the forearm. The gap between healthy vascular stumps ranged from 5 to 12 cm and five vein grafts were used for the radial and ulnar arteries and for three veins. Primary suture of nerves was possible because of bone loss of about 3 cm. Circulation was restored and by two years recovery for the median nerve was fair and for the ulnar nerve good (Fig. 8.30). The function of the hand was limited by the poor excursion of long tendons from extensive disruption of the muscle bellies but this result, together with those from two other successful replantations of the amputated hand performed between 1979 and 1982 proved to our satisfaction that primary suture even in these most severe cases gave results far superior to those following delayed repair
Fig. 8.30 Replantation of a hand amputated through the forearm by a railway wheel. Recovery of median and ulnar nerves was good. A subsequent free skin flap was put in by Mr David Evans. The function is shown before reconstruction of long flexor and extensor tendons.
333
Failure to restore flow through a main axial vessel is almost always followed by muscle fibrosis to some degree; in the lower limb the effects are catastrophic. But there is a generally depressing effect on nerve recovery when the lesion is associated with rupture of the adjacent artery. Shergill (Shergill et al. 2000) provided evidence about the deleterious effect of arterial injury in his study of 260 repairs of the radial nerve. Thirty five out of the 55 repairs in cases of associated vascular injury failed, that is very nearly two thirds. Osborne (Osborne et al. 2000) came to a similar conclusion in his study of 85 repairs of the musculocutaneous nerve. We shall return to these papers later in this chapter. Incised wounds of nerves and vessels in the limbs are best treated by primary repair. Failing that urgent repair of the artery with an early return to the nerves is nearly as good.
8.5.1 The Brachial Plexus The tidy wounds from knife, glass and scalpel are amongst the most rewarding of all nerve injuries to repair. When the opportunity for repair within hours or days of injury is grasped, the results for C5,C6 and C7 wounds are excellent; better by far than equivalent results even for early repair of more distal nerve trunks ruptured by traction injuries. They are worthwhile for C8 and T1 too. We emphasise that in virtually no other nerve laceration is the harmfulness of procrastination so clearly shown as in the supraclavicular stab wound. On the other hand the nature of the wound is such that other and more pressing problems frequently arise. Setting aside the cases of damage by scalpel there are three distinct patterns. When the blade or bottle is thrust from above down (as it was in Hector’s case) the lower trunk, subclavian vessels and lung are damaged. Such wounds may be as rapidly fatal as they were for the Prince of Ilium. When the blade is thrust towards the face or neck the fifth, sixth and seventh cervical with phrenic nerves are damaged: in addition to the jugular and carotid vessels, the trachea and the oesophagus are at risk. However in the cases coming to us the vertebral body shields the viscera, the knife point glides laterally, severing the spinal nerves close to their exit from the intervertebral foramina. Lastly, the lateral thrust divides the upper and middle trunks, the nerve to serratus anterior and the accessory nerve. The temptation to suture the severed nerves directly should be resisted. Grafts are better, even if only half a centimetre in length. Some examples follow: Case Report: A 26 year old man was stabbed in the neck. The internal jugular vein was ligated by the receiving surgeon who gave us the opportunity to repair the neural lesion within 36 h. The fifth, sixth and seventh cervical nerves had been severed close to their foramina: the phrenic nerve had also been severed. Twenty four grafts were used to repair the
334
Surgical Disorders of the Peripheral Nerves
Fig. 8.31 The outcome 3 years after repair, 36 hours after injury, of stab wound of right C5, C6 and C7 and the phrenic nerve.
defect of 2 cm. By three years the only defect at the shoulder was lateral rotation which was restricted to 40° by capsular contraction. Sensation in thumb, index and middle was good without hypersensitivity. The power of deltoid, biceps, triceps and wrist extension was 80% of the uninjured side: the phrenic nerve recovered, there was accurate localisation of touch with two point discrimination less than 8 mm. in the thumb, index and middle fingers (Fig. 8.31). Case Report: A 21 year old nurse was assaulted and stabbed on her way home from duty in her Accident and Emergency Unit. Mr Maelor Thomas (Kings College Hospital) transferred her to us so that repair was performed within 24 h from injury. The knife had severed the upper and middle trunks and the nerve to serratus anterior. Nineteen grafts were used to bridge the defect of 1 cm. At four years this patient considered function in her upper limb to be normal. Measured power of serratus anterior, of the muscles of the shoulder, arm and forearm were MRC Grade 5: power grip in the afflicted (dominant) hand was 90% of the uninjured hand. Two point discrimination sense was 6–8 mm in the median territory of the hand. Early recovery of sympathetic function was a striking feature. Case Report: A 22 year old man was stabbed in the neck with a broken bottle and clearly remembered the downward thrust of the weapon which entered the neck just above the medial one third of clavicle. At emergency exploration
copious bleeding from the offsets of subclavian vessels was controlled. The seventh and eighth cervical nerves were severed with about 50% of the sixth cervical nerve. The glass had scored the first rib, lacerating the pleura. Sixteen grafts were used in the repair. At two years the one significant defect of function was thumb opposition which was mitigated by muscle transfer. Power grip for the whole hand was 85% of normal, pinch grip between thumb and index 75% of the uninjured (dominant) side. There was no pain and no serious sensory deficit (Fig. 8.32). A striking feature of these and of other cases or urgent repair of stab wounds is the speed of recovery, the absence of pain and a level of recovery superior to repair of more distal wounds of trunk nerves. Stab wounds involving the infraclavicular plexus often cause torrential bleeding. This can be reduced by firm pressure to the wound and by compression of the subclavian artery where it crosses the first rib (Guthrie’s manoeuvre) (Fig. 8.33).
8.5.2 The Closed Infraclavicular Lesion It is perhaps in the case of the traction lesion of the infraclavicular brachial plexus that the deleterious effect of
Compound Nerve Injury
Fig. 8.32 The outcome at 2 years after repair, on the day of injury, of a stab wound of left C6, C7 and C8.
Fig 8.33 The outcome 2 years after repair, on the day of injury, of a stab wound of the circumflex, musculocutaneous and radial nerves, and of the axillary artery. A flexor to extensor transfer was performed at 9 months.
arterial injury and the profoundly harmful effect of delay before nerve repair is most clearly expressed. This is among the worst of all true peripheral nerve injuries. It is
335
characterised by the violence of the cause of the lesion, by the high incidence of rupture of the subclavian or axillary arteries (about 30% of our cases) and by the complexity of the nerve lesion. It is common to find nerve trunks ruptured at different levels up and down the limb, and it is not rare to find them avulsed from muscle bellies. The distinction according to level is to some extent artificial. There is overlap between the supraclavicular, the retroclavicular and the infraclavicular injury in closed traction lesion, with double lesions of supraclavicular lesions combined with more distal ruptures in about 10% of cases. Two patterns can be discerned. The first, which is more common, is caused by violent hyperextension at the shoulder. There is almost always a fracture of shaft of the humerus or injury to the gleno humeral joint; the incidence of vascular injury is high; the level of proximal rupture is deep to pectoralis minor which acts as a guillotine on the neurovascular bundle. The second pattern is even more severe and even more dangerous. In these injuries the forequarter is virtually avulsed from the trunk. There is disruption of the anterior arch of the shoulder girdle. The subclavian artery suffers and it is usually torn in its first or second part. There is usually avulsion of the eighth cervical and first thoracic nerves which is combined with ruptures of the cords of the plexus more distally. The forequarter drops down and away from the trunk because of the savage disruption of musculo-skeletal supporting elements. There is often a lesion of the spinal accessory nerve and of the nerve to serratus anterior. There is, almost always, a phrenic palsy. There must be no delay in the treatment of these limb and life threatening injuries. Restoration of ventilation and control of bleeding are critically important and robust stabilisation of the injury to the skeleton is necessary to enable repair of the main artery and , if circumstances permit, the main nerves. Guy Broome (Carlisle) demonstrated one such case at the first meeting of the British Plexus Club in Warrington in 2006. The patient, a pedestrian, was knocked down by a car and sustained a ‘floating shoulder’ with fracture/dislocation of the acromio-clavicular joint and complex fracture of the glenoid and the scapula body. These were rapidly stabilised before successful repair by vein graft of the rupture of the third part of the subclavian artery. A rupture of the lateral cord of the brachial plexus was then primarily grafted. Perfusion was restored, there was no ischaemic fibrosis of the forearm muscles, the repair of the lateral cord was successful and this, with good spontaneous recovery through the traction lesion of the medial cord, permitted later successful flexor to extensor transfer. The gleno-humeral joint was arthrodesed 15 months after the injury. The patient, now at work, is free of pain and has a useful arm (Fig. 8.34). Of course the overwhelming priority is to repair the vessel and it may not be feasible to deal with the nerve injury at the same operation (Fig. 8.35). The injuries to the nerves in these cases are extremely complex. It is common to find nerve trunks ruptured at
336
Surgical Disorders of the Peripheral Nerves
Fig. 8.34 Incomplete avulsion injury of the right forequarter. The shoulder girdle was smashed, and the first part of the axillary artery ruptured (above left). The lateral cord was ruptured (above right). The
function at 2 years is shown below. The gleno humeral joint was arthrodesed (Courtesy of Guy Broome, Carlisle).
different levels up and down the limb; it is not rare to find them avulsed from muscle (Figs. 8.36 and 8.37). The advantages of urgent repair of the vessel and of the nerves are overwhelming. They include ease of identification of ruptured structures; the use of the nerve stimulator to detect more distal nerve injuries and the ability to diminish the gap between nerve stumps, which retract widely after rupture (Fig. 8.38). The exposure of Fiolle and Delmas is ideal (see Chapter 7). Cavanagh et al. (1987) reviewed the results of the repair of nerves in 89 such cases operated over a ten year period. They reckoned that infraclavicular lesions accounted for 20% of all brachial plexus lesions referred. In 46% there was damage to the axillary artery; in 52% there was bony injury. They took as indications for operation signs of arterial injury and evidence of severe local trauma with a deep neural lesion. A good case for primary repair was made out. Our later work has confirmed the last proposition. 91 nerve ruptures were repaired. In some
later cases, nerve repair was abandoned because of technical difficult or because the state of the tissues in the arm was so bad. As might be expected, the results were a good deal better for the suprascapular, the circumflex, the radial and the musculocutaneous nerves than they were for the median and the ulnar nerves. Results were very much better when the nerve was repaired within 14 days of injury than in those repaired later (average time in this group was 12 weeks), and this was particularly true for the median and ulnar nerves. These are striking results but it must come as no surprise that the effects of delay are worse in more severe injuries. The methods used for grading results are described in Table 8.12. The traction lesion is the worst of all. Urgent repair offers the only realistic prospect for useful recovery (Tables 8.13 and 8.14). Case Report: A right handed metalworker, aged 21 years, lost control of his motor cycle on a wet road at a speed of 30mph. His right shoulder struck a lamp post. Mr Davies,
Compound Nerve Injury
337
Fig. 8.35 Ruptures of the third part of the left subclavian artery and of the right superficial femoral artery were successfully repaired by Mr Butler Manuel and Mr Sandison (Hastings), and this enabled early repair of the left brachial plexus.
Fig. 8.36 Severe closed infraclavicular traction lesion. The axillary artery was ruptured at the level of the coracoid. All the trunk nerves were ruptured at different levels in the upper limb.
Fig. 8.37 Severe closed infraclavicular lesion. The axillary artery is ruptured at the level of the coracoid. Musculocutaneous, and ulnar, nerves were ruptured and the distal stumps retracted into the arm.
vascular surgeon at Charing Cross Hospital, recognised rupture of the third part of the subclavian artery, a fracture of clavicle and an extensive nerve lesion. He considered that circulation to the limb was adequate for survival and provided us with the opportunity to proceed to repair of all damaged structures on the day of injury. The exposure of Fiolle and Delmas enabled rapid control of the great vessels and identification of the damaged nerves. The suprascapular and the circumflex nerves were ruptured, there was a lesion in continuity of the median and musculocutaneous
nerves. At exposure there was still distal conduction, but this disappeared whilst we were repairing the subclavian artery. Conduction returned with minutes of restoration of flow through the vessel. The subclavian vein had been lacerated by a bone fragment and was easily closed by suture. The suprascapular and circumflex nerves were repaired. The forearm was decompressed by open fasciotomy. The muscles were heading for critical ischaemia but repair of the artery and decompression swiftly restored them to a healthy condition.
338
Surgical Disorders of the Peripheral Nerves Table 8.13 Results of repair of 55 median and ulnar nerves in closed infraclavicular traction lesion. Early: within 14 days Late: after 15 days; mean (33 nerves) 22 weeks (22 nerves) Good
2
0
Fair
25
9
Poor 6 13 These nerves were ruptured in 42 of the 89 patients. The axillary artery was ruptured in 31 cases (drawn from Cavanagh et al 1987).
Table 8.14 216 high repairs of median and ulnar nerves in adults and children 1975–2000 (%). Outcome Tidy Untidy Closed Total wounds wounds traction Fig. 8.38 Rupture of the axillary artery, and the radial nerve, associated with closed fracture of the upper humerus. Both were repaired (George Bonney 1962).
Median 117 cases
His intense pain disappeared by the following morning. By 3 years he considered function throughout his upper limb normal (Fig. 8.39). These cases are demanding, and they require the closest possible collaboration between the surgeons involved. The ‘nerve’ surgeon must be ready to travel, so as to assist the first surgeon and then to complete the nerve repair. The happy outcome related above must be contrasted with others. Table 8.12 Grading of results in median and ulnar nerves repaired in the infraclavicular region, the axilla and the arm (high lesions) Median Good
Long flexor muscles MRC 4 or better Localisation to digit, without hypersensitivity Return of sweating
Fair
Long flexor muscles MRC 3 or 3+ Protective sensation. Moderate or no hypersensitivity Sweating diminished or absent
Poor
Long flexor muscles MRC 2 or less Protective sensation but severe hypersensitivity or no sensation
Ulnar Good
FCU and FDP little and ring MRC 4 or better Intrinsic muscles MRC 2 or better Localisation to little and ring fingers. No hypersensitivity Return of sweating
Fair
FCU and FDP little and ring fingers MRC 3 or 3+ No intrinsic muscle function Protective sensation little and ring fingers No, or moderate hypersensitivity Little or no sweating
Poor
FCU and FDP little and ring fingers MRC 2 No intrinsic muscle function Protective sensation with severe hypersensitivity or no sensation No sweating
Good
8 (36.4)
6 (16.2)
3 (5.2)
17 (14.5)
Fair
10 (45.4)
16 (43.3)
22 (37.9)
48 (41)
Poor
4 (18.2)
15 (40.5)
33 (56.9)
52 (44.5)
Total
22
37
58
117
Ulnar 99 nerves Good
5 (33.3)
5 (14.3)
0 (0)
10 (10.1)
Fair
7 (46.7)
16 (45.7)
24 (49)
47 (47.5)
Poor
3 (20)
14 (40)
25 (51)
42 (42.4)
Total 15 35 49 99 1. There were 75 arterial injuries (58% of patients). 2. 20 of the 27 good results followed repair within 48 h of injury. Seven of these results were in children. 3. Results of repair at 3 months or more after the untidy wound or closed traction lesion were uniformly bad. 4. Operation was not considered, or attempted repair was abandoned, in 31 patients presenting with a delay of 3 weeks or more because of ischaemic fibrosis or sepsis.
A 31 year old slater sustained severe injuries to his left upper limb in a motor cycle accident which included rupture of the axillary artery and rupture of the radial and musculocutaneous nerves. There were fractures of the humerus, of the radius and the ulna. The artery was not repaired. The nerves and vessels were explored 18 weeks later. The operation was tedious and difficult. The stumps of the ruptured nerves were widely separated, the gap in the case of the radial nerve exceeded 12 cm. The condition of the median and ulnar nerves was poor, they were constricted but also stretched. Both nerve repairs failed. Anterior transfer of the triceps for elbow flexion had to be reversed because the patient found that he was worse off. Two attempts at flexor to extensor transfer failed because of the condition of the flexor muscles of the forearm. The muscles were affected to some extent by post ischaemic fibrosis; the nerves were damaged by ischaemia so that the deep afferent pathways never recovered. Case Report: A 30 year old woman suffered a closed fracture dislocation of the shoulder of her dominant right upper limb.
Compound Nerve Injury
339
confirmed this in their study of 150 cases. The repaired artery splints by drawing the stumps together, overcoming elastic recoil and reducing tension upon the repair. Birch and Raji 1991 described 108 repairs of median and ulnar nerves in tidy wounds, in adults, in the region from the elbow to the wrist. There were 48 primary sutures: in these a policy was followed of repair of the ulnar and the radial arteries, of muscles and of tendons but also of restoration of the gliding synovial plane around each tendon and between the tendon and the nerve (Table 8.15). The injuries were more severe in the primary repair group. Results for the whole nerve were described by the method of Seddon with the introduction of an excellent grade for those patients who considered function normal (Tables 8.16 and 8.17). The evidence provided supports the statement that ‘primary repair is best’. Even though there were far more arterial injuries in the primary repair group it was only amongst these that excellent results were obtained. The twelve poor results were found in the delayed group where there was little to choose between graft and secondary suture. Later work confirmed these findings (Table 8.18).
8.6 Nerves and Bone and Joint Injuries
Fig. 8.39 The right subclavian vessels with the suprascapular and circumflex nerves were ruptured by an open fracture of the clavicle. They were repaired on the day of injury. Function at 30 months.
There were fractures of humerus, of radius and of ulna. An angiogram revealed a tear of the axillary and also of the brachial artery. No repair was done. We saw her 48 h later by which time the condition of the limb was such that emergency amputation was indicated. Mr Norbert Kang (Wexham Park), who was visiting us at that time, fashioned a large skin flap based on residual flow through the profunda brachii artery so avoiding the necessity of disarticulation through the gleno humeral joint. The median and ulnar nerves provide an example of the significance of arterial flow throughout the upper limb. Leclerq and his colleagues (1985) showed that, in a study of 68 repairs, repair of the ulnar artery at the same time as repair of the nerve, was much better. Merle and his colleagues (1986)
Nerves are injured by damage to the adjacent skeleton by: traction from displacement which commonly ends in rupture, laceration by a fragment of bone, entrapment within the dislocated joint or in a fracture and late entrapment and compression from callus. On the whole dislocations are more damaging. It seems to be widely assumed that the prognosis for nerves injured in this way is good but this is not generally the case. Our own indications for exploring the nerve are governed by the violence of the injury which indicates the extent of displacement of bones and their fragments, by the depth of lesion and pain. A surgeon who considers that the fracture requires open reduction and internal fixation will rarely regret exposing the afflicted nerve at the same time and we think that a nerve palsy is added indication for open reduction. Seddon (1975b) had this comment to make in speaking about nerves injured in the arm and at the elbow. He thought that recovery could be awaited if two conditions were met. ‘The first is reasonable apposition of the bony fragments and the other complete certainty that there is no threat of ischaemia of the forearm muscles.’ In referring to sciatic palsy in fracture of the femoral shaft he said ‘it is wise always to explore the nerve’. Notable studies of nerve lesions and skeletal injury appeared in the inter-war years. Platt (1928) from the fracture service at Ancoats Hospital, Manchester recognised such unfavourable injuries as that to the median nerve in supracondylar fracture of the elbow and that to the common peroneal nerve in avulsion fracture of the fibular styloid.
340
Surgical Disorders of the Peripheral Nerves
Table 8.15 Details of 108 nerve injuries (NI) in 95 adult patients (P): tidy wounds 1977–1985. Primary: 35P 48NI Secondary: 25P 25NI
Graft: 35P 35NI
Mean age in years
28.75
29
29.5
Nerve
Median 27: Ulnar 21
Median 15: Ulnar 10
Median 14: Ulnar 21
Distance from pisiform in cm. (mean)
6.1
9.5
7.2
Mean delay to repair in weeks
–
34.5
35.5
Gap after resection of neuroma (cm)
–
2.5
5.5
Interval to review in weeks (mean)
174
150
200
Arterial injury
28P, 43 arteries
8P, 8 arteries
20P, 20 arteries
Tendon or muscle injury
32P
14P
26P
Repair at time of nerve repair 32P 14P Gap measured with elbow extended and wrist at 20° dorsiflexion. P - patients. NI- nerve injuries. Drawn from Birch and Raji (1991).
26P
Table 8.16 Methods of grading of results of median and ulnar nerves injured below the elbow (low). Grade Motor Sensory Excellent
Power MRC 5 No wasting or deformity No trophic changes
Good
Power MRC 4–5 Abolition of paralytic deformity Minimal pulp wasting
Fair
MRC 3 or more. Some sweating. Pulp wasted
Poor
MRC 3 or less. No sweating, Trophic changes *2PD = two point discrimination.
Function indistinguishable from normal hand. Good stereognosis, no hypersensitivity. 2PD* equivalent to uninjured digits
Good M5, S4
Accurate speedy localisation. Can recognise texture or objects. Minor cold sensitivity and hypersensitivity. 2PD < 8mm at tips of fingers
Good M5, S3+
Accurate localisation to digit. No stereognosis. 2PD >8 mm. Significant cold sensitivity and hypersensitivity
Fair. M3, S3
No sensation or severe cold sensitivity and hypersensitivity
Bad M0,1, S0, 1 or 2
Table 8.17 Results of 108 repairs in 95 adult patients: tidy wounds (1977–1985). Grade Nerve Primary repair Delayed repair Graft 35 48 nerves 25 nerves nerves Median
3
0
0
Ulnar
5
0
0
Median
16
5
4
Ulnar
15
2
7
Fair: 39 nerves
Median
7
8
8
Ulnar
2
5
9
Poor: 10 nerves
Median
0
2
2
3
5
Excellent: 8 nerves Good: 49 nerves
Ulnar 0 Drawn from Birch and Raji (1991).
Equivalent on Seddon’s grading
Watson-Jones (1930) analysed more than 100 nerve lesions from amongst 5,000 consecutive patients seen in the Liverpool Fracture Service over the course of 2 years. The nature of injury to the ulnar nerve from fracture of the medial epicondyle, to the common peroneal nerve from fracture of the fibula and the vulnerability of the median nerve distal to
Table 8.18 Repair of 264 median and ulnar nerves in tidy wounds injured between the distal wrist crease and elbow in adults aged between 16 and 65 years (1979–2004). Repair type (number of cases). Outcome Primary Delayed Graft Total repair repair Excellent
13
2
2
17
Good
52
15
34
101
Fair
25
30
52
107
Poor
2
16
21
39
92 63 109 This includes the cases in Tables 8.15 and 8.17.
264
pronator quadratus is clearly described. Platt and Watson Jones were strongly in favour of exploration of nerves injured by fractures. Siegel and Gelberman (1991) review the subject thoroughly, finding 85% of nerve palsies recovering spontaneously from closed fractures and 65–70% doing so after open fractures. However ninety per cent of those nerves which went on to recovery had done so by 4 months. These cannot have been wholly degenerative lesions. Siegel and
Compound Nerve Injury
Gelberman set out their indications for intervention which include (1) the fracture needs internal fixation; (2) there is associated vascular injury (3) wound exploration of an open fracture is necessary and (4) a fracture or dislocation is irreducible. We might add to this two more: (5) the lesion deepens while it is under observation, and (6) the lesion occurred during operation for internal fixation. The association between displaced fragments of bone and serious nerve lesions is relevant too. Goldie and Powell (1991) described a case where the median nerve was transfixed by a fragment of the distal radius. They intervened urgently, on the basis of a near complete nerve lesion and radiological evidence of a skeletal cause. Recovery was good. In Omer’s prospective study (Omer 1974) 83% of nerve palsies from closed upper limb fractures recovered; no less than 90% of these had done so by 3 months. If there was no recovery by 7 months then there would be none. Seddon’s own figures (Seddon 1975b) are interesting: 146 from 212 of cases of nerves injured in fractures or dislocations of the upper limb spontaneously recovered to near normal levels but less than one half of his 57 cases of nerve palsies after skeletal injury in the lower limb did so. One third of his own series of radial nerves so injured did not recover but of course these were referred cases. Seddon (1975b) related the information provided by Böhler from the fracture services of Salzburg and Vienna. There were 57 cases of radial palsy in 765 closed fractures of the humeral shaft, an incidence of 7.4%. Spontaneous recovery occurred in 47, so the incidence of unfavourable radial lesions in this series is less than 2%.
8.6.1 The Nerve and the Pattern of Fracture Lambert’s (2005) further opinion on the subject of radial nerve palsy associated with fractures of the shaft of the humerus offers clinicians insights which extend well beyond his theme. Unfortunately this article is available only on the electronic version of the journal. Lambert identifies three important features. First is ‘anatomical fixity’: ‘nerves are injured more readily at or near zones of relative fixity of the nerve within the surrounding tissue. Thus, peri-articular fractures and dislocations are associated with the higher risk of nerve and vascular injuries. Nerves are relatively fixed by the muscular branches entering muscle, the shorter the branch to the muscle, the greater fixity the nerve has relative to the mobile tissues around it.’ The length of the radial nerve trunk between the relatively fixed point of the nerves to the lateral head of triceps and the next fixed point where the nerve pierces the lateral intermuscular septum lies between 10 and 15 cm: ‘injury of some degree is certain if the nerve is stretched between 1.5 and 2.0 cm in the longitudinal axis of the humerus’. Next comes the fracture pattern: ‘short
341
oblique, transverse, segmental, and open fractures, in which the intermuscular septum is torn (indicative of the energy of the injury), and the fracture inherently unstable, are associated with a higher incidence of radial palsy than long spiral fractures in which the intermuscular septum may not be torn’. The third element discussed is the blood supply to the nerve. The radial nerve may become relatively ischaemic because of compression by the intermuscular septum: ‘a similar picture occurs in lesions close to the intermuscular septum where direct injury is a result of compression of the nerve trunk and indirect injury through hypoperfusion both contribute to the syndrome of neurostenalgia, the burning painful paraesthesiae associated with sensory motor nerve palsy of variable depth: decompression of the nerve may dramatically relieve the pain’. Lambert concludes that the outlook for a radial nerve injury is far better if exploration and repair are performed within 4 weeks of transection or rupture: ‘perineural fibrosis, intraneural ischaemic fibrosis, and hypoperfusion in the surrounding ischaemic penumbra of the zone of the injury all contribute to the difficulty of exploration and successful nerve repair in the late presentation’. These observations are general to the field of nerves injured by fractures or dislocations. We agree with his final comment: ‘meta-analyses can be useful, but should not supplant clinical observation and decisions based on the logical analysis of fractures type and nerve function in the individual patient, aided by electrophysiological investigation where necessary. Knowledge of, and therefore comfort with, the surgical exposure of a radial nerve is clearly important: lack of either might actually be the most important factor in the decision to treat the nerve expectantly.’ The view that injured nerves can be neglected for 3 months seems to be growing, presumably to ‘see what happens’. Such a policy is baleful, indeed it is deplorable. Each patient should be considered as an individual and not subjected to the rote of an algorithm. In most cases clinical examination will answer the following questions: 1. Is this lesion to the nerve complete or incomplete? 2. Is the lesion one of conduction block or is it one of Wallerian degeneration? Repeated examination during the ensuing 4–6 weeks will usually distinguish between those nerve palsies which are showing signs of recovery from those which are not. We have seen, over the last 10 years, several hundreds of patients in whom the injury was of such a nature that rupture of a nerve could and should have been anticipated. We have seen far too many patients who have lost their means of earning a living or of maintaining their independence because of neglect of the first principles of treatment of the consequences of radial or common peroneal palsies. Matters are not helped by the rise of subspecialisation. The case now described is but one of many.
342
Case Report: A 71 year old woman fell in the street sustaining a displaced fracture at the neck of the humerus. The distal fragment was displaced medially and clearly impinged upon the neurovascular axis. She developed pain, which worsened and over the next eight days she experienced a deepening palsy of the median, musculocutaneous, the radial and ulnar nerves. No action was taken because the ‘shoulder surgeon’ was on leave. Delayed hemi arthroplasty was followed by gradual improvement in her pain over the course of the next 6 months but this did not disappear and recovery for the ulnar nerve in particular was poor. We think that this patient’s age had something to do with her neglect, for had she not been a particularly articulate person it is possible that no active treatment would have been offered. Clinicians should not forget that neurophysiological investigations may confirm, or modify, a clinical diagnosis and it has to be said that the service provided by some electrodiagnostic departments to fracture services is poor. We have seen far too many patients who have waited for months before the investigation was done and we have seen patients who, on their own initiative, paid for the investigation to be done privately rather than wait for months. It is fortunate for them that they were not excommunicated from the National Health Service because of their ‘queue jumping’. Case Report: A 78 year old woman fell at home dislocating her elbow. This was promptly reduced in a nearby Accident and Emergency Department but she experienced intense pain and developed a complete median palsy. A letter was written to her family practitioner by the first surgeon five days later which said: ‘I expect she is going to have problems with pain control … she may well consult you regarding this aspect of her care’. Her family practitioner sought a second opinion which was provided 6 weeks after the injury. The second surgeon asked for urgent neurophysiological investigation. These were not performed for 4 weeks and did no more than confirm the clinical diagnosis that the median nerve, entrapped within the elbow joint, was by now destroyed. We follow the following principles in the approach to nerves injured by skeletal injury: 1. Clinical examination usually permits the distinction between lesions of conduction block from deeper, degenerative lesions. 2. A diagnosis of neurapraxia should not be made in the presence of significant neuropathic pain. This suggests that the noxious agent is still at work upon the nerve. 3. Deepening of the lesion whilst under observation indicates bleeding until proven otherwise. 4. The diagnosis of neurapraxia cannot be regarded as secure until the persistence of conduction in the distal segment of the nerve is demonstrated after about eight days from the injury.
Surgical Disorders of the Peripheral Nerves
5. An advancing Tinel’s sign distinguishes the degenerative lesion of favourable prognosis (axonotmesis) from the unfavourable lesion (neurotmesis) in the main nerves of the limb and in particular, the median, the ulnar, the radial, and the common peroneal and tibial nerves. This investigation should enable the clinician to come to a view about the likelihood of recovery at no later than 6 weeks from the injury. Unfortunately, the circumflex nerve cannot be examined by this method because is so deeply seated. 6. Lesions of the sciatic, the tibial and the common peroneal nerves are usually associated with high energy injuries. The likelihood of recovery in the untreated case is generally bad. These three nerves should always be exposed during operation upon the fracture or dislocation (Fig. 8.40). 7. If a surgeon elects to convert a closed fracture to an open one by whatever technique, then the lesion of the nerve should be exposed. The nerve, indeed the nerve with the adjacent artery, may be in the fracture or in the joint. Both will certainly be displaced from their normal position (Fig 8.41). 8. Whilst the presence of nerve palsy is not necessarily an indication for operation the presence of a complete lesion of the main nerve adds to the argument for open treatment of the fracture.
Fig. 8.40 ‘Closed’ intramedullary nailing of fractured shaft of femur. The sciatic nerve was exposed 5 months later. It was in the fracture.
Compound Nerve Injury
343
Understanding of their rôle, and of the consequences following lesions has been increased by such studies as those provided by Comtet, Hertzberg and Alnaasan (1993); by Narakas (1993), and by Coene (1985). Pain is usual after injuries to the accessory nerve and the nerve to serratus anterior. Now is as good a time as any to dispel a commonly held misconception that the two peripheral nerves without a cutaneous sensory component are purely ‘ motor’ nerves. This is quite wrong. Both contain large numbers of myelinated and non myelinated afferent fibres. Biopsies of the suprascapular nerve weeks after proven preganglionic injury to the fifth and sixth cervical nerves showed that over 30% of the larger myelinated fibres survived, those presumably responsible for proprioception with cells in the dorsal root ganglion. Lesions of the spinal accessory nerve, and of the nerve to serratus anterior are caused only rarely by skeletal injury, but the effect upon the shoulder girdle function is so severe that it seems appropriate to discuss them here. ‘Winging’ of the scapula is usual after injury to either nerve and should not be ascribed to paralysis of the serratus anterior. The distinction between the two is easy. In accessory palsy the active inferior scapulo-humeral angle is narrowed, usually to about 30° and the scapula drops downwards and away from the mid line (see Fig. 5.61). In paralysis of serratus anterior the scapula is drawn upwards and towards the mid line. The active ISHA is greater than the range of active abduction, a finding which is characteristic for the nerve to serratus anterior and unlike all other nerve palsies involving the shoulder girdle.
8.6.2.1 The Spinal Accessory Nerve
Fig. 8.41 ‘Closed’ intramedullary nailing of fracture of long bone. The tibial nerve and the posterior tibial artery were in the fracture. The muscles of the deep flexor compartment were fibrosed causing severe clawing of the toes.
8.6.2 The Shoulder Girdle and Gleno-Humeral Joints Bonnel (1989) estimated that one quarter of the nerves of the brachial plexus pass to this complex of joints which permits an extraordinarily wide range of movement for the upper limb. This places the nerves at risk as the spinal nerves passing to the brachial plexus are the weakest link in the suspensory chain. One cranial and three peripheral nerves are of particular significance: the spinal accessory (11th cranial), the nerve to serratus anterior, and the suprascapular and circumflex nerves.
This nerve is vulnerable to the attention of surgeons with inadequate knowledge of topographical anatomy and it is too often divided during lymph node biopsy. Williams et al. (1996) described the unmistakable syndrome of: pain, drooping of the shoulder, and restricted abduction of the shoulder, in 43 iatrogenous cases. The scapula drops downwards and away from the midline (Chapter 5). Camp (2010) has kindly provided her findings from 89 more iatrogenous cases. The surgeon responsible made the diagnosis only rarely. Most of the lesions were recognised by orthopaedic surgeons, some by neurologists and ten were made by solicitors seeking treatment for their clients. The delay before repair of the nerves was, on average, fully 18 months. Repair of the nerve was generally successful, often strikingly so. Most patients experienced early relief of pain and in most there was considerable improvement in function. Pain relief and improvement in function was regained by repair so late as four years after the injury (Fig. 8.42). The spinal accessory nerve appears to be less vulnerable to the harmful effects of delay before repair
344
Surgical Disorders of the Peripheral Nerves
supraclavicular brachial plexus, where the fifth, sixth and seventh cervical nerves are avulsed from the spinal cord. The position of the scapula is characteristic. It is drawn upwards, and towards the spine by the unopposed action of trapezius and the levator scapulae muscles. Abduction is restricted, usually to less than 90°, but the inferior scapulo-humeral angle widens as the weight of the upper limb thrusts the scapula towards the mid line. Repair of this nerve, either by suture, graft or by transfer of the deep divisions of intercostal nerves, is attended with a higher success rate than that seen after repair of any other peripheral nerve. This is fortunate because muscle transfers for the paralysed serratus anterior muscle are unreliable and provide only limited function.
8.6.3 The Clavicle
Fig. 8.42 The left spinal accessory nerve was repaired 4 years after injury. Function at 14 months. Table 8.19 Repair by graft of 98 iatrogenous lesions of the spinal accessory nerve. Grade Function Pain Excellent
Normal
None
8
Good
Abduction ³150°
Occasional – no analgesics
58
Fair
Improved, but abduction <150°
Regular analgesics
22
Poor
No improvement, Unchanged 10 persisting pain The result was graded poor in any patient with persisting severe pain irrespective of any improvement in function Drawn from Williams et al. (1996) and Camp (2010)
than the peripheral nerves as a whole. It seems that most orthopaedic surgeons are well aware of these facts for it is generally they who make the diagnosis after iatrogenous section of the nerve. The diagnosis was made by the operating surgeon in no more than 10 of more than 150 iatrogenous cases that we have seen. This is a pity because the results of repair of this nerve is generally good and certainly far better than those obtained by muscle transfer (Table 8.19).
8.6.2.2 The Nerve to Serratus Anterior This too is vulnerable to attentions of surgeons although most cases occur in very severe traction injuries to the
Fractures of the clavicle are very common in severe neurovascular injuries to the supra and infraclavicular brachial plexus although as Seddon (1975b) points out ‘they are produced by the violence that damages the plexus; the fractures are incidental though the break in the clavicle removes the last defense of the plexus against the distracting force’. However fracture of the clavicle itself is sometimes associated with wounding of the nerves and of the vessels deep to it although the subclavius muscle seems to act as an effective buffer in most fractures. A fragment of the bone may be displaced so that it perforates the muscle and this can be dangerous. Case Report: A 22 year old right handed mechanical engineer fell from his motor cycle and sustained an open fracture of the left clavicle. The nerve injury was extensive and rupture, at least, of the suprascapular, the circumflex and the musculocutaneous nerves was recognised by Martin Milling (Newport) who gave us the opportunity to operate 5 days later. The suprascapular nerve had, in fact, been avulsed from the supraspinatus muscle, and it was reimplanted into that muscle. The ruptured circumflex and musculocutaneous nerves were grafted. There was a degree of constriction of the divisions of the brachial plexus. This was relieved. He regained powerful flexion of his elbow, and a useful shoulder with abduction in excess of 120° from the successful reinnervation of supraspinatus and some innervation into the deltoid muscles (Fig. 8.43). The nature of injury to the nerves to the shoulder in particular was such that repair would have been impossible in a delayed exploration. This patient returned to full time work with slight modifications as a mechanical engineer. Kitsis et al. (2003) described the neurological complications of nonunion or malunion which led to operation in 17 patients who presented with pain and progressing neurological defects. These were caused by compression or irritation of the nerves deep to the clavicle. Removal of exuberant callus or treatment of the nonunion relieved symptoms and improved function in 12 cases (Figs. 8.44 and 8.45).
Compound Nerve Injury
Fig. 8.43 Open fracture of the left clavicle caused rupture of the circumflex and the musculocutaneous nerves and avulsion of the suprascapular nerve from the supraspinatus muscle. The circumflex and musculocutaneous nerves were grafted. The suprascapular nerve was implanted into the supraspinatus. Function at 22 months.
8.6.4 Dislocation of the Gleno-Humeral Joint The injury to the infraclavicular plexus associated with anterior dislocation of the shoulder is produced by the intrusion of the head of the humerus into the axilla, beneath the cords and the terminal branches of the plexus. The medial and posterior cords are predominantly affected; the lesion deepens and becomes more extensive while the humeral head is permitted to stretch the plexus. It seems that the large cords and the main nerves are less susceptible to rupture than are the smaller suprascapular and circumflex nerves. In a few cases the lesion or part of it is permitted to deepen from recoverable axonotmesis to irrecoverable neurotmesis through failure to recognise the dislocation and the intrusion of the humeral head. The lessons are clear: recognise the anterior dislocation and reduce it before there is neural affection; if closed reduction is impossible, be prepared to reduce by open operation; recognise neural affection and act instantly to remove its cause; intervene without delay in cases in which unrecognised dislocation has produced a deep and extensive lesion of the infraclavicular part of the plexus. The common course of such lesions, properly treated, is progressive recovery. Watson-Jones (1936) saw 34(15%) nerve lesions in 231 gleno-humeral dislocations during 5 years, treated in the
345
Liverpool Royal Infirmary. The fracture clinic at St Mary’s Hospital recorded 302 dislocations treated in the years 1941– 1953. Fracture of the greater tuberosity was noted in 58 of these, and a complete tear of the rotator cuff in seven more. There were seven circumflex palsies, one musculocutaneous palsy and two more diffuse nerve injuries. Coene (1985) reviewed 14 series of shoulder dislocations, and identified 292 nerve lesions from 2,594 cases. Pasila et al. (1978, 1980) prospectively studied 238 dislocations seen over a period of 3 years. There were 50 nerve injuries, 21 of the circumflex nerve. The cuff was torn in 28 patients. Thirty four of the nerve lesions and all of the cuff tears were found in patients aged more than 50 years. Whilst 7 of the nerves recovered within 1 month 15 did not proceed to full recovery. Rockwood et al. (1991) recorded an incidence of circumflex palsy in 30% of shoulder dislocations. De Laat et al. (1994) detected electromyographic abnormalities in 45 from 100 dislocations. The circumflex nerve headed the field, followed by the suprascapular and musculocutaneous nerves. EMG changes were more common in the elderly and in those with obvious haematomas. There was no spontaneous recovery of the nerve lesion in eight patients; not an insignificant number in these cases of low energy injury. Gumina and Postacchini (1998) found that the incidence of cuff tear was as high as 60% in 108 patients aged more than 60 years. There were 11 circumflex palsies, all of which recovered. Leffert and Seddon (1965) and Seddon (1975b) found that many closed injuries to the circumflex nerve went on to spontaneous recovery. Blom and Dahlback (1970) described 24 cases of isolated circumflex nerve lesions; all of these went on to full recovery even though electromyographic abnormalities were found in nearly all. Three hundred and twenty six nerve lesions were indentified in 100 consecutive cases of fracture/dislocation of the glenohumeral joint seen at the Royal National Orthopaedic Hospital in the years 2002–2005 (Table 8.20). These were selected cases and they were referred because of nerve and vascular injury. Twenty four patients were aged 60 years or more and 36 were older than 50 years. Forty three of the injuries were low energy. The tuberosities were fractured in 42 patients; in two of these there was also a tear of the rotator cuff. The cuff was torn in 15 more patients. Operations were performed in 53 patients because of suspected bleeding, severe pain, or because of suspected rupture of nerve or nerves.
8.6.4.1 Vascular Injuries Rupture of the artery, or one of its offsets, or an expanding haematoma was encountered in 14 patients. The following is a fairly typical example. A 20 year old, right handed, landscape gardener sustained a dislocation of his right shoulder whilst playing football. A diagnosis of continuing bleeding was made by William Bodey (Hillingdon) on the basis of severe pain and deepening of the nerve lesion and
346
Surgical Disorders of the Peripheral Nerves
Fig. 8.44 Severe pain and a C5, C6 palsy were caused by a fragment of bone in open fracture of the right clavicle (below right). Bone graft and internal fixation were performed at 3 months (above right). Function 2 years later (left).
he confirmed this clinical diagnosis by ultrasonographic examination. Operation was done 6 days later. We encountered a false aneurysm of the axillary artery where the posterior circumflex vessels had been avulsed. No nerve was ruptured, rather they were compressed by haematoma or splayed over or otherwise distorted by the sac. Correction of the arterial lesion was straight forward. The nerves were decompressed. His pain rapidly resolved and the suprascapular, circumflex, median, musculocutaneous and radial nerves recovered fully over the course of the next 9 months (axonotmesis). The ulnar nerve was never affected.
remaining nerve lesions were more or less equally divided between conduction block (137) and axonotmesis (157). The outcome was recorded for 263 nerves in patients followed for at least 18 months. Seventeen of 19 repairs performed within 14 days of injury made a good recovery. (Figure 8.46) seven of the 11 repairs performed later than that were fair or poor. In all there were 26 poor results. These included 17 lesions of axonotmesis which fared badly because of delay in decompression and the development of fixed deformities in the hand caused by pain and by neglect. Two nerves were found embedded in the torn subscapularis muscle and two more had become displaced into the joint. These had not been ruptured and were left alone. It is likely that they should have been repaired.
8.6.4.2 The Lesion of the Nerves On the whole, the prognosis was good but for many recovery did not commence until nerve trunks had been released from haematoma, scar or a thick axillary sheath. There were 32 ruptures, the circumflex nerve accounting for 19 of these. The
8.6.4.3 Pain The response to pain after decompression was very good. Twenty five patients experienced early and substantial relief.
Compound Nerve Injury
347
Fig. 8.45 Fracture of the right clavicle. Displaced fragments lacerated the subclavian artery and the brachial plexus.
The final functional outcome in the 71 cases followed for a minimum of 2 years is set out in Table 8.21. The main causes of poor function include a failure of recovery in the suprascapular nerve and inadequate treatment of the displaced greater tuberosity or the tear of the cuff. Severe fixed deformity which was by no means confined to older patients required many months of treatment in 18 cases. Combined injuries of the circumflex and suprascapular nerves is not uncommon. The diagnosis is never easy and it
is made even more difficult by fracture of the greater tuberosity or rupture of the rotator cuff. Many patients with complete paralysis of the deltoid are able to elevate the arm if the abductor apparatus is intact. Conversely, the poor elevation caused by damage to that mechanism is often attributed to circumflex palsy. Early diagnosis of rupture of the circumflex nerve remains one of the most difficult conundrums in examination of the peripheral nerves. The nerve is deeply seated and Tinel’s sign is rarely detectable. The patient is in pain, voluntary movement is restricted by that pain as well as by the restraint of any sling. The extent of loss of cutaneous sensation is variable; there may be no complete anaesthesia. Patients with an intact suprascapular nerve and an intact rotator cuff are able to initiate abduction; many can fully elevate the arm (Figs. 8.47 and 8.48). Wynn Parry (1981) found that the range of abduction was full or nearly so in 145 patients with deltoid paralysis. Seddon (1975b) was a little more cautious: ‘this perfect abductor action of the supraspinatus is rare; it is more usual to find abduction to about 155°, with the arm a little in front of the coronal plane of the body’. However we have seen more than 40 patients with deltoid paralysis able to maintain full elevation in medial rotation (see Chapter 5). The power of forward flexion and of abduction is reduced to about 40% of normal but extension, with the shoulder at 90° of abduction, is 10% or less. One reliable test is to examine the posterior one third of the deltoid with the fingers and ask the patient to thrust the arm back. It should be possible to detect activity in the muscle even when the arm is supported in a sling. Detection of injury to the suprascapular nerve with or without the complication of rotator cuff tear is much easier. The patient cannot initiate abduction. The head of the humerus migrates proximally. Prudnikov (1994) provides a helpful analysis of combined injuries to the nerves and the rotator cuff in 32 operated cases. The nerve lesion was confined to the circumflex nerve in 20 patients, it also involved the suprascapular nerve in 12 more. Two important physical signs are described. The head of the humerus drops inferiorly; the inferior scapulo-humeral angle is so reduced that the patient is
Table 8.20 Nerve lesions in 100 consecutive cases of shoulder dislocation. 2002–2005. Nerve Number Nature of lesion in 326 nerves Conduction block Axonotmesis Neurotmesis
Outcome in 263 nerves Good Fair Poor
Suprascapular
57
32
25
0
52
50
0
2
Circumflex
84
21
44
19
68
50
8
10
Radial
48
19
25
4
37
33
2
2
Musculocutaneous
36
18
16
2
31
29
0
2
Median
50
25
22
3
38
33
1
4
Ulnar
51
22
25
4
37
28
3
6
32
263
223
14
26
326 137 157 1. 63 nerve lesions remain under observation. 2. Rupture of an artery or expanding haematoma was treated in 14. 3. 27 patients presented with severe neuropathic pain.
348
Surgical Disorders of the Peripheral Nerves
Fig. 8.46 Rupture of the axillary artery and of the musculocutaneous and circumflex nerves during dislocation of the right shoulder. The function at 26 months after repair done at six days. Table 8.21 Function in 71 shoulders followed for a minimum of 24 months (2002–2007). Ranges of movement in degrees. Mallet Narakas Forward flexion Abduction Lateral rotation Medial rotation Inferior scapulo (normal 15) (normal 13) Active Passive Active Passive Active Passive Active Passive humeral angle Active Passive Median
13
11
145
170
125
160
40
55
90
90
Mean
12.4
9.4
121
147
114
141
42
55
85
86
Range
5–15
2–13
0–180
40–180
0–180
20–180
0–90
0–90
30–90
30–90
unable to move the arm away from the body. Prudnikov contrasts this with isolated cuff tears where the ISHA opens to 30° or more, and with isolated deltoid palsy, where elevation is maintained. Prudnikov says: ‘Néanmoins, il ne s’agissait pas d’une pseudoparalysie veritable mais d’une sorte d’épaule ballante car il y avait une perte de la capacité de maintenir un angle scapulo-huméral constant lors des tentative d’élévation du bras’: ‘nevertheless, this does not amount to a true pseudo paralysis but rather to a type of floating shoulder because there is inability to maintain the scapulo-humeral angle whilst attempting elevation of the arm’. The cuff was repaired in 22 cases of circumflex palsy. Elevation of the arm in excess of
114
159
120° was restored in 20 patients, even though the circumflex nerve did not recover at all in 4 of them. The ability to initiate abduction in a patient with a combined lesion of the circumflex and suprascapular nerves indicates recovery in the latter (Figs. 8.49 and 8.50). The suprascapular nerve passes away from the upper trunk about 3cms above the clavicle. It is not unusual to find it arising entirely from the fifth cervical nerve. It passes laterally and posteriorly deep to the omohyoid to the scapular notch, entering the supraspinous fossa deep to the superior transverse ligament, and it traverses the fossa deep to the supraspinatus muscle to wind around the lateral border of the spine of the
Compound Nerve Injury
Fig. 8.47 An example of shoulder function with no recovery into the deltoid. Rupture of the right axillary artery, and of suprascapular, circumflex and musculocutaneous nerves. Operation on the day of injury. The circumflex nerve could not be repaired. Function at 20 months. The power of forward flexion was 50%, that of abduction 40% and that of extension was 5% of the uninjured side.
scapula to enter the infraspinous fossa. The nerve contains between 3,000 and 4,000 myelinated nerve fibres. Its course renders it particularly vulnerable to traction lesions and we have seen it so injured where it branches from the upper trunk, in the posterior triangle, at the supraspinous notch and within the supra- and infraspinous fossae. The nerve is essential for abduction and lateral rotation at the gleno-humeral joint. The circumflex nerve is the terminal branch of the posterior cord and it contains between 6,000 and 7,000 myelinated nerve fibres which pass through the fifth and sixth cervical nerves. The nerve divides into two branches within the quadrilateral tunnel. The anterior division continues around the neck of the humerus to innervate the anterior deltoid, the larger posterior branch innervates teres minor and the posterior deltoid. One cutaneous branch, the upper lateral cutaneous nerve of the arm, is a useful landmark where it pierces the deep fascia over the posterior border of the deltoid. Coene
349
Fig. 8.48 Dislocation of shoulder caused rupture of the suprascapular nerve, the circumflex nerve was avulsed. Operation at 48 hours. The circumflex nerve was implanted into the deltoid. The range of movement at 18 months after operation is shown. There was some recovery into the anterior one third of the deltoid.
(1985) described the course and the relations of the nerve in his excellent monograph: ‘in the axillary region the nerve has a free course in loose fatty tissue, but when it turns around the subscapularis muscle, it is captured in a tunnel formed by the fascia of the subscapularis muscle cranially, the teres major muscle caudally and the coraco-brachialis muscle laterally. The fascii of these muscles join at the entrance to the quadrilateral tunnel, where they firmly surround the hitherto independent axillary nerve and posterior circumflex vessels which meet the nerve at this point. This neurovascular complex turns around the inferior border of the subscapular tendon, passes cranially over the superior border of the teres major tendon, and enters a horizontal “tube”.’ The fixity of
350
Fig. 8.49 Initiation of abduction indicates early recovery into the suprascapular nerve. Early recovery in the left suprascapular and circumflex nerves 4 months after fracture dislocation of the shoulder. The cuff is intact.
the nerve in the quadrilateral tunnel makes it particularly vulnerable to injury by forward displacement of the head of humerus. Bleeding from the posterior circumflex vessels strangles the nerve and leaves it embedded in scar tissue with the consistency of concrete (Fig. 8.51). Ochiai et al. (1997), in a significant paper describing treatment of combined injuries of the circumflex and suprascapular nerves, showed that the suprascapular nerve may be serially damaged at several places. They recommended that the nerve should be exposed along its entire course as far as the infraspinatus. Mikami et al. (1997) reported the results of nerve grafting in combined injuries of the circumflex and suprascapular nerves in 33 cases. The nerves were exposed through the extended approach described by the same authors which entails osteotomy of the coracoid and detachment of the posterior deltoid and trapezius. Fifty five nerves were ruptured, the remaining 11 were lesions in continuity. The patients were operated
Fig. 8.50 Dislocation of the left shoulder. The lesion of the suprascapular nerve recovered. The repair of the circumflex nerve failed. The power of forward flexion was 50%, that of abduction 40% and that of extension 15% of the uninjured side.
Surgical Disorders of the Peripheral Nerves
Fig. 8.51 Rupture of the circumflex nerve. The posterior circumflex humeral vessels (1) were ruptured. The proximal stump of the nerve (2) is displayed deep to pectoralis minor. The distal stump is embedded within a fibrosed quadrilateral tunnel.
within 4 months of injury and results were generally good. Bonnard et al. (1999) reported 120 cases operated in Lausanne by Narakas; 106 circumflex nerves and 21 suprascapular nerves were repaired. A good or useful result was obtained in 85% of the repairs but there was a dramatic falling off with delay and these authors recommend that repair is done within 3 months of injury. Twenty eight injured cuffs were repaired at the same time as the nerves. Twenty one arteries were repaired. Some circumflex nerves are irreparable because the nerve has been avulsed from the muscle or because the distal stump is too scarred. It is fortunate that two techniques are available which help in this situation: muscular ‘neurotisation’, and nerve transfer (Chapter 7). Spilsbury and Birch (1996) reported their findings in 129 nerve injuries in 98 of our patients. There were 62 ruptures of the circumflex and 22 of the suprascapular nerve. Lesions in continuity were seen in 26 circumflex and in 19 suprascapular nerves. In 31 patients both nerves were damaged and in at least eight more there was an associated rupture of the rotator cuff. Open wounds from stabs, missiles or surgeons accounted for 25 nerves in 16 patients. The results were described using the MRC grading of power, and the scoring system of Narakas. A myometer (Chapter 5) was used to measure strength and stamina in 28 patients. The suprascapular fared better than the circumflex nerve after grafting or decompression of lesions in continuity. Spilsbury thought that 25 of the 56 grafts of the circumflex nerve made a good recovery compared with 20 of the 24 grafts of the suprascapular nerve (Table 8.22). Only 6 of the 23 lesions in continuity of the circumflex nerve went on to good recovery whereas eight of the 16 lesions in continuity of the suprascapular nerve did so. It is clear that measuring recovery into the deltoid by the range of movement is inadequate. Restoration of abduction and lateral rotation provides a true indication of recovery of
Compound Nerve Injury
351
Table 8.22 Grading and results of operation on the circumflex and suprascapular nerves (From Spilsbury and Birch 1996). Outcome Circumflex (56 repairs) Suprascapular (24 repairs) Active range against resistance Good
Deltoid MRC 4 or better
Fair
Deltoid MRC 3+
Poor
Less than above
25
Abduction ³120°
23
Abduction 90–120°
Abduction ³ −120°
20
Lateral rotation ³20°
Abduction 90–120°
2
Lateral rotation 0–30° 8
Less than above
2
the suprascapular nerve. It is also evident that grading recovery of the muscles of the shoulder by the MRC system for muscle strength offers only limited information. Spilsbury measured strength and stamina, by repeated movements against resistance, and unveiled a more unfavourable picture. The stamina of the shoulder with paralysis of the deltoid was no more than about 30% of the normal side and it ranged between 40–45% in those cases where there had been some recovery through lesions in continuity. Where the circumflex nerve had been grafted the shoulder achieved no more than about 50% of normal stamina although the result was much better in the cases of urgent repair. Stamina of the shoulder after successful repair of the suprascapular nerve in the presence of an intact circumflex nerve reached at least 70% of normal. The most important cause of failure is untreated rupture of the rotator cuff or a displaced fracture of the greater tuberosity (Fig. 8.52). It appears that a more vigorous approach towards palsies of the suprascapular and circumflex nerves is required. If a patient with a clinical diagnosis of circumflex nerve lesion is unable to abduct the shoulder then either the suprascapular nerve is damaged or the rotator cuff is torn. Probably the best single investigation to disentangle these elements is an MR
scan which not only demonstrates damage to the rotator cuff but also indicates denervation of deltoid and supraspinatus muscles. The suprascapular and circumflex nerves can be seen respectively at the notch and in the quadrilateral space. Ultrasonography, when it is used by skilled clinicians as soon after injury as possible may bring about a radical improvement in the treatment of these complex and disabling lesions. We have found that the Mallet chart which was originally used for measuring shoulder function in birth lesions of the brachial plexus is also useful in recording shoulder function in the adult. A simple hand held myometer provides useful information about power and its use adds but a minute or two to the time required for the examination in a clinic.
8.6.5 The Radial Nerve and Fractures of the Humerus The radial nerve is the largest terminal branch of the brachial plexus. Between its origin and its entrance into the spinal groove the nerve receives fewer arteries than elsewhere along its course (Ramage 1927). The first 8–10 cm of the nerve may not have a nutrient artery; if so this segment is wholly supplied by descending intra neural channels and it will be relatively avascular if transected at its origin. The profunda brachii artery accompanies the nerve in the spiral groove. Bleeding from it compresses the nerve. The nerve lies closest to the bone where it pierces the lateral intermuscular septum as it passes through a short tunnel bounded by bone and unyielding fascia. Here, the nerve is tethered; here is a common level of rupture, entrapment or compression. Platt (1928) recognised that prolonged traction on a nerve tethered in this way will destroy it. Diagnosis of the level of the lesion is straightforward. Triceps is paralysed in high axillary lesions. The lateral head of triceps is paralysed when the nerve is damaged posterior to the humerus. If the lateral head of triceps is active but the brachioradialis and extensor carpi radialis longus muscles are paralysed, then the lesion lies between the lateral intermuscular septum and about 5 cm above the lateral epicondyle. The sensory loss in high lesions involves not only the dorsal skin of the web space of the thumb but also the skin of the lateral aspect of the elbow and proximal forearm (Chapter 5).
8.6.6 Incidence
Fig. 8.52 Dislocation of the right shoulder. Seven months later the greater tuberosity was reattached and the circumflex nerve implanted into the deltoid muscle. Function at 4 months.
The St Mary’s records reveal five cases of radial palsy from 186 consecutive cases of closed fracture of the shaft of the humerus seen between 1941 and 1954. Ekholm and his colleagues (2006) from the Karolinska Institute analysed 401
352
fractures from a population of 1.8 million, treated over 2 years. Ninety eight percent of the fractures were closed. The incidence of radial palsy was 8% and fractures in the middle and distal shaft were most likely to be responsible: ‘however, the risk of radial nerve palsy seems to be greatest in the less common fracture of the distal shaft in which the nerve is trapped in the lateral inter muscular septum and is therefore particularly vulnerable’. These figures are remarkably close to those provided to Seddon (1975b) by Böhler. Shao et al. (2005) reviewed 21 series and recorded a prevalence of radial nerve palsy in 11.8% of more than 4,500 fractures. Spontaneous recovery occurred in 411 from 581 patients ‘treated conservatively’. It is not clear how the fractures were treated in these cases. About 30% of the nerves did not recover, a similar proportion to that related by Seddon and as he pointed out cases were sent to him because the nerve was not recovering; they were selected. The figure from Shao et al seems rather high. All of the 17 palsies detected by Sonnenveld et al. (1987) from a total of 111 closed fractures of shaft of humerus recovered spontaneously. Pollock et al. (1981) and Larsen and Barfred (2000) studied, between them, 60 radial palsies caused by fractures of the shaft of the humerus, one of which was open. Two nerves were ruptured and two were entrapped. Fifty six recovered spontaneously. Holstein and Lewis (1963) suggested that the radial nerve was particularly vulnerable to injury in fractures of the distal one third of the humerus: ‘for only a short distance near the lateral supracondylar ridge is the nerve in direct contact with the humerus, and it is in this area that the nerve pierces the lateral intermuscular septum before passing onto the surface of the brachialis muscle. From our anatomical dissections, the nerve has least mobility at this point, and, in our opinion, it is this lack of mobility that is a prime factor contributing to the nerve injury.’ The observation is an important one. Livani et al. (2006) set out their method in six cases of fractures of the distal third of the humerus with associated radial palsy. The nerve was exposed through a short incision over the fracture, and two separate incisions were developed to permit placing of a long bridging plate. The nerve was found interposed or compressed or deviated to such an extent that spontaneous recovery was: ‘impossible, improbable or unpredictable’. Vichard et al. (1982) recommended exploring the nerve at the same time as fixing the fracture and they emphasise that one in four radial nerves injured by fractures of the humeral diaphysis will not recover. They found 6 ruptures and 4 lesions in continuity in 26 cases. Venouziou et al. (2007) agree: in 22 cases they found that 8 nerves were intact, 12 had been seriously damaged, and 2 were entrapped in the fracture.
8.6.6.1 The Open Fracture There is general agreement that the open humeral shaft fracture with radial nerve palsy requires exploration of the nerve
Surgical Disorders of the Peripheral Nerves
at the same time as debridement of the wound (Shah and Bhatti 1983; Venouziou et al. 2007; Vichard et al. 1982; Seddon 1975b; Siegel and Gelberman, 1991). Foster et al. (1993) found 9 nerves lacerated or interposed between the fracture fragments in 14 open fractures; Ring et al. (2004) explored the nerve in 11 cases of open fractures; 6 were transected. Pollock’s (1999) case provides some lessons of general interest. In a motor cycle accident a young man suffered an open fracture of the mid shaft of humerus which was treated by wound excision and external fixation. At 4 months non union was established. There was no sign of recovery for the radial nerve. Tinel’s sign was static at the fracture site. Whilst correcting the non union by compression plate and bone graft we found that the nerve had been transposed through the fracture. It was attenuated and encased in callus over a length of 8 cm lying in front of the humerus. There was no conduction across the lesion. The nerve was freed and returned to its anatomical position. It went on to recover as axonotmesis. Tinel’s sign advanced 300 mm in 120 days. This case is but one of many which confirm that lesions of intrinsically favourable prognosis, be they conduction block or axonotmesis, will recover only if the cause of that lesion has been removed.
8.6.6.2 Recovery in Favourable Lesions The alert clinician is usually able to detect favourable signs within 6 weeks of injury. Bowden and Sholl (1954) in a remarkably thorough investigation, found that the mean interval between injury and the detection of activity in the appropriate muscle was 64 days for lesions in continuity and 96 days after suture. Shah and Bhatti (1983) studied 56 cases of radial palsy in fractures of the humerus and recorded evidence of clinical recovery in 10 in the first week, in 12 between 1 week and 1 month, and in 13 more between the first and the third month. Operation was performed in 22 cases. Three nerves were ruptured, and 10 were entrapped. Shaw and Sakellarides (1967) observed the onset of recovery in as short a time as 2 weeks in partial lesions and in 5 weeks in complete lesions. Ring et al. (2004) saw signs of recovery at 7 weeks in their cases of lesions in continuity.
8.6.6.3 Experience at St Mary’s and the Royal National Orthopaedic Hospital Shergill et al. (2000) reported 291 cases of repair of the radial and posterior interosseous nerves. Most of these were operated at St Mary’s Hospital between 1975 and 1994. Thirty one patients were lost to follow up. Sixteen of the 18 repairs of the posterior interosseous nerve achieved a good result.
Compound Nerve Injury
353
There were 242 repairs of the main nerve. The injuries were classed into four groups: 1. Open tidy wounds from glass, knife or scissors (73). 2. Open untidy wounds, from penetrating missile injury, open fracture or fracture/dislocation and/or contaminated wounds (52). 3. Closed traction injuries, usually associated with fractures of the humerus (62). 4. Injuries which include an associated lesion of the axillary or brachial artery (55). The results were graded by a simplified version of the MRC system (Zachary 1954) and was based on the extent of recovery of the proximal and the distal muscles. A ‘good’ result of repair of a high lesion, above any branches to the triceps, required powerful elbow extension and extension of wrist against gravity whereas a ‘good’ result for repairs of the nerve below the nerves to triceps requires powerful extension of the wrist of the fingers and of the thumb. The method used for grading results, and those results are summarised in Tables 8.23 and 8.24. The nature of the wound, the severity of injury and, in particular, associated arterial injury are very important. The result was poor in 15 (21%) of the open tidy group, but in 35(64%) of the arterial injury group. The main vessel had been successfully repaired in 12 patients in the first hospital. We performed urgent arterial repair in 22 more. Repair of the artery was delayed in 16 more patients because the vessel had not been repaired primarily (6), because the repair had failed (8) or had been complicated by a false aneurysm (2). The artery was never
repaired in five patients. There was some degree of ischaemic damage to the muscles of the forearm in all cases in whom there was a delay of more than 8 h before restoration of axial flow. Of the 77 nerves repaired within 14 days of injury, 49% achieved a good result. Only 28% of the 169 late repairs achieved this grade; the mean delay was 190 days (15–440). There were no good results from repairs when the interval exceeded 12 months.
8.6.6.4 More Recent Findings Two hundred and twenty seven compound lesions of the radial nerve were studied between 2000 and 2007 (Table 8.25). ‘Tidy’ wounds were not included so the material is close to Shergill’s groups 2, 3 and 4, although the incidence of arterial injury or bleeding (30 cases, 15.3%) is rather lower in this than in the earlier series. The increasing numbers of iatrogenous injuries was a startling and dismaying finding; there were 54 (23.8%), all occurring in the course of operations for fracture or the complications of fracture.
8.6.6.5 The Tinel Sign A strong static Tinel sign indicated rupture or entrapment in 72 nerves. In 41 of these the static sign was converted into a progressing one after extricating the nerve from the fracture or other cause of distortion. The remaining 31 nerves were
Table 8.23 Grading of outcome in lesions of the radial nerve. For high lesions above nerves to triceps For intermediate lesions, between medial head of triceps and brachio radialis Good
Elbow extension ³ MRC grade 4
Wrist extension ³ MRC grade 4
Wrist extension ³ MRC grade 3
Finger and thumb extension ³ MRC grade 3
Fair
Elbow extension ³ MRC grade 3
Wrist extension ³ MRC grade 3
For low lesions, the posterior interosseous nerve Full independent extension of the digits ³ MRC grade 4 Extension of digits ³ MRC Grade 3
Finger and thumb extension ³ MRC grade 2
Wrist extension MRC grade 2 Poor Less than above Based on Shergill et al. (2010).
Less than above
Table 8.24 Results of repair of 242 radial nerves. Wound Number Good Fair of nerves No. (%) No. (%)
Less than above
Table 8.25 227 compound lesions of the radial nerve. 2000–2007. Level of lesion Depth of lesion Poor No. (%)
Open “tidy”
73
28
38
30
41
15
21
Open untidy
52
13
25
13
25
26
50
Closed traction
62
19
31
17
27
26
42
Associated arterial injury
55
12
22
8
14
35
64
Total 242 72 Drawn from Shergill et al. (2010).
30
68
28
102
42
High (above lateral head of triceps) 64
Conduction block 27
Intermediate
Axonotmesis
129
97
Low 34 Neurotmesis 103 1. Excluding nerves injured by shoulder dislocation. 2. Excluding nerves injured by fracture/dislocation at the elbow in children aged less than 15 years. 3. The total of iatrogenous injuries was 54 (23.8%). 4. The total of nerves repaired was 72 (31.7%). 5. The total of nerves extricated or decompressed (“neurolysis”) was 53 (23.2%).
354
Surgical Disorders of the Peripheral Nerves
Fig. 8.55 Fifteen centimetres of the radial nerve was avulsed by a drill during operation upon fracture of shaft of humerus. The nerve was grafted 2 weeks later (Marco Sinisi, RNOH). Function at 13 months.
Fig. 8.53 Ten weeks after primary suture of radial nerve the strong Tinel sign was static. A further repair by graft was performed which was successful.
The method of repair has not changed and it would seem that the two groups are broadly comparable. The one difference is the interval between injury and repair which was, on average, 190 days in the earlier series; it was 58 days in this.
8.6.6.7 Iatrogenous Lesions
Fig. 8.54 A progressing Tinel sign proved misleading in this case of radial palsy. Three quarters of the nerve was trapped within the fracture site.
repaired (Fig. 8.53). A strongly progressing Tinel sign was followed by spontaneous recovery in 103 cases, including those 41 nerves which progressed after neurolysis. There were five misleading progressing Tinel signs. In two of these the nerve was explored because the anticipated recovery into the muscles had not occurred; part of the nerve was caught within the fracture (Fig. 8.54). In the other three cases the extensor muscles of the forearm were damaged by ischaemia or by direct injury.
The nerve is vulnerable to the attentions of surgeons (Verga et al. 2007; Lin et al. 2003; Samardzic et al. 1990; Shaw and Sakellarides 1967; Shah and Bhatti 1983). In most of our cases the nerve was damaged because it was not exposed in operations for internal or external fixation. The most common level of lesion was at the junction of the upper two thirds and the lower one third of the arm. Damage was inflicted by drills passed blindly for locking screws or for Kirschner wires. In three cases more than 10 cm of the nerve was avulsed (Figs. 8.56 and 8.57). Prognosis is determined by the nature of the lesion and by the speed of correction. We have already seen (Chapter 3) examples of recovery in cases of conduction block or axonotmesis where an alert surgeon recognised and corrected the cause. In another case the radial nerve was damaged by cement which had been used in the treatment of a
8.6.6.6 Results of Operation Decompression or extrication (neurolysis) was performed in 53 nerves (23.2%) and recovery was good in all of these except in three cases of ischaemic damage to the forearm. Seventy two nerves were repaired (31.7%). Thirty six repairs (50%) achieved a good result (Fig. 8.55). This is a good deal better than the earlier series, in which 32 from 114 repairs (28.1%) of closed traction and the open untidy wound were graded good.
Fig. 8.56 Rupture of the radial nerve in a closed fracture of the humerus. The stumps are embedded within the fracture which had been nailed.
Compound Nerve Injury
355
Radial nerve palsy is, for many patients, far worse than median or ulnar lesion. These cases require close and careful supervision by the surgeon to mitigate that disability and to prevent fixed deformity. It will not do to pass the responsibility to therapists or ‘rehabilitation services’. Well made dynamic splints are valuable. In some patients they make the difference between holding on to their job or losing it, or maintaining their independence or losing that.
8.6.8 The Musculocutaneous Nerve Fig. 8.57 The radial nerve crushed by the plate inserted for fractured shaft of humerus.
pathological fracture in the proximal one third of the humerus. The first surgeon re-explored the nerve on the following day and found that the nerve was lying close, but not immediately adjacent, to extruded cement. The cement was trimmed away. The nerve recovered. In a case of intramedullary nailing of fractured shaft of humerus, a complete radial palsy was noted on the evening of operation. The nerve was exposed, it was found crushed within the fracture site and was extricated. The duration of compression was about 12 h. Recovery was complete (axonotmesis). Operating upon a closed fracture converts it into an open fracture no matter how small the incision used. Operating through small incisions or by means of the arthroscope compels the surgeon to work blind in cases where normal topographical anatomy has been distorted by the injury.
8.6.7 Conclusions Tinel’s sign is particularly useful in following the nerve palsy as it precedes the first signs of returning muscle function in cases of closed fractures of the shaft of the humerus treated by closed methods. The interval before we detected clinical evidence of recovery into the first distal muscle in the 52 non operated cases with adequate information was as follows: 15 (28%) within 35 days; 17 (32.6%) between 36 and 70 days, and 20 (38.5%) between 71 and 105 days. These findings were recorded when we first saw the patient and it is reasonable to assume that the clinical evidence was detectable some days or weeks before that meeting. More nerves are being injured during operations for fracture of the humerus. It is wise always to expose a nerve which is not working before proceeding to fixation of the fracture. The actions of the surgeons in the cases related earlier is commendable. They took the chance to retrieve the situation and prevent the deterioration of potentially favourable lesions into irretrievable ones.
It may be that our earlier views about this nerve were rather too sanguine. Many patients are able to flex their elbow against resistance without any biceps at all because of the contribution from brachioradialis and from that part of the brachialis muscle innervated by the radial nerve. The nerve is frequently variable in its origin and in its course. In one case we found it springing directly from the 4th cervical nerve; we have seen a number of cases where it arose directly from the median nerve in the arm as one or as several branches. Osborne and his colleagues (2000) examined 85 repairs in adults and in children which were performed between 1968 and 1997 (Table 8.26). Most of the operations were done at St Mary’s Hospital. The injuries were classed into three types: the open ‘tidy’, the open untidy, and closed traction. The results were also correlated with associated arterial injury. Osborne showed that the type of the injury was the most important factor determining the result; 12 of 13 open ‘tidy’ lesions gave good results compared with 30 of 48 closed traction lesions. The results were better when the nerves were repaired within 14 days of injury and when the Table 8.26 Results of repairs of 85 musculocutaneous nerves by type of injury (%). Outcome Injury Total Open tidy Open Closed untidy traction 13 nerves 24 nerves 48 nerves Good: Elbow flexion
12 (92.3)
15 (62.5)
30 (62.5)
57 (67)
0 (0)
7 (29.2)
10 (20.8)
17 (20)
8 (16.7)
11 (13)
³MRC grade 4 Full range of flexion Fair: Elbow flexion ³MRC grade 3+ Range of flexion £100 Poor. Less than 1 (7.7) 2 (8.3) the above Drawn from Osborne et al. (2000).
356
grafts were less than 10 cm in length. The effect of an associated fracture was bad, with a good outcome following in 23 out of 43 cases (53.5%) compared with 34 from 42 (81%) in those repairs where there was no fracture. Fractures of the mid shaft of the humerus may damage the biceps muscle. The presence of arterial injury proved to be relevant also: 25 out of the 43 (58.1%) repairs in this group achieved good result compared with 32 from 42 (76.2%) in cases where there was no arterial injury. Results were good in 21 of the 27 (77.8%) repairs done within 14 days of injury compared with 34 from 55(61.8%) carried out later than this. Osborne pointed out that ‘in those patients with apparently normal elbow flexion from a strong brachioradialis, the power of supination is reduced. For this reason the power of supination should be included as part of the assessment of the function of biceps’. He went out to point out the possibility of a lesion at two levels: ‘in a number of cases in which there has been a traction injury to the supraclavicular brachial plexus, bruising in the upper part of the arm along the tract of the musculocutaneous nerve caused us to suspect a peripheral lesion of that nerve. This suspicion was confirmed at exploration. These two level lesions were not included in this study’. Excepting these caveats it is the case that repair of this nerve is usually rewarding to the patient and the surgeon alike and the adoption of nerve transfer for otherwise irreparable lesions has improved the chances of restoring function (Chapter 7)
8.6.9 Elbow The elbow is a notorious area of difficulty in treating nerve injuries, especially in children. The close proximity of all trunks to bone makes them vulnerable. The ulnar nerve is particularly at risk in elbow dislocation, particularly so when the medial epicondyle is displaced into the joint (Fig. 8.58).
Fig. 8.58 A 43 year old man. The appearance of the ulnar nerve after removal from within the elbow joint. Recovery was poor and complicated by severe pain.
Surgical Disorders of the Peripheral Nerves
Fig. 8.59 Dislocation of the elbow in an 8 year old boy. The posterior interosseous nerve was trapped within the joint.
All three nerves are at risk of displacement into the dislocated joint, the posterior interosseous or radial nerves are at risk in the Monteggia fracture/dislocation (Fig. 8.59). All nerves at risk from entrapment within fractures. The median nerve is often entrapped with the brachial artery in the supracondylar fracture. Pain which persists, or worsens, after reduction of the fracture suggests impending critical ischaemia, or continuing irritation of the nerve, or both. Case Report: A 12 year old boy fell at school, sustaining a fracture of the medial epicondyle. He experienced immediate and intense pain. There was a dense ulnar palsy. An operation was done during which the fracture was stabilised with Kirschner wires but the nerve was not traced. He remained in very severe pain until another surgeon re-explored the elbow 6 weeks later. The ulnar nerve was embedded within the fracture and it was extricated. There was some improvement in pain, there was no recovery for the nerve. This was repaired by graft 14 weeks after the original injury. Recovery was good but fell short of normal. Had the nerve been exposed at the first operation the lad would have recovered full hand function. Our material is set out in Table 8.27. Some details of 118 palsies seen between 1975 and 1998 were presented in 2000 (Birch and Achan 2000). Twenty two nerves came to repair; seven of these had been transected during the first operation upon the fracture. Seventeen of the nerves damaged in supracondylar fractures were found entrapped within the fracture or impaled on a bone spike. The remainder were compressed by swelling or fibrosis, at or distal to the fracture, and recovery after decompression was uniformly good. Ramachandran and his colleagues (2006) provided more detailed information about 37 palsies in 32 children with supracondylar fracture who were seen between 1998 and 2002. Twenty three of these nerves were injured during the first operation. Of the ten nerves damaged during ‘closed’ reduction and percutaneous pinning, repair was necessary in
Compound Nerve Injury
357
Table 8.27 165 lesions of children’s nerves in fractures or dislocation at the elbow 1975–2002. Skeletal Nerves affected Median Ulnar Radial/posterior interosseous 21
Supracondylar fracture
53
54
Fracture of medial condyle
2
11
Dislocation
13
2
3
Monteggia fracture-dislocation
0
2
4
68 69 28 1. 56 other nerve lesions associated with ischaemic contracture are described in Table 8.28. 2. 75 nerves were injured during operation upon the fracture, either directly by wire, knife or scissors or indirectly by compression or entrapment within the fracture.
four cases and in four more nerves were extricated from the fracture or from dense fibrosis. Of the two cases incurred during open reduction one nerve recovered spontaneously the day before the planned operation whilst neurolysis was necessary in the second. The one palsy incurred during closed manipulation and reduction came to neurolysis. We have seen not one case of isolated anterior interosseous palsy amongst the 295 children’s nerves injured by fractures or dislocations involving the distal humerus seen since 1975. It may be that the fibres in the median nerve destined for the anterior interosseous nerve are selectively vulnerable to injury at the elbow; it is more likely that a partial lesion of the whole median nerve was not detected at initial presentation. The anterior interosseous nerve is particularly susceptible to ischaemic anoxia because of its location in the depths of flexor compartment and it is likely that some reported cases are examples of anoxic conduction block. Sixty eight more children with 74 injuries to the nerves about the elbow have been seen since 2003. There were 40 lesions of the ulnar nerve, 19 of the radial and 15 affecting the medial nerve. Sixteen nerves had been partially or completely transected, during operation. Twenty four other nerves were damaged during operation by failure to relieve compression or entrapment. One recent case is illuminating. A 28 month old girl sustained a supracondylar fracture of the humerus. The fracture was reduced and stabilised with Kirschner wires. There was no radial pulse before this operation. It did not return afterwards. There was a complete lesion of the ulnar nerve after operation and deepening defects of the radial and the median nerves. The elbow was explored (Marco Sinisi, RNOHT) 5 days later by which time there was complete palsy of all three nerves. The brachial artery and median nerve were sharply compressed by the bicipital aponeurosis. The radial nerve was severely compressed by swollen muscle. The epineurial circulation was obliterated in both median and radial nerves. The ulnar
nerve was found transected by the Kirschner wire. The brachial artery became pulsatile within minutes of decompression. There was rapid recovery for both median and radial nerves. By 9 months, there was no evidence of post ischaemic fibrosis of the forearm muscles; the ulnar nerve was recovering. Here indeed was a close brush with disaster! It is not easy to justify performing an open reduction and internal fixation in a pulseless limb without exposing the artery and even less so using a tourniquet. However, in this case, the first surgeon grasped the urgency of the situation and provided the opportunity to rectify matters, so avoiding the catastrophe of post ischaemic fibrosis.
8.6.9.1 Volkmann’s Contracture and the Supracondylar Fracture One might have hoped that this serious complication would now be no more than an historical curiosity at least in civilian practice in ‘advanced’ countries (Fig. 8.60). However it seems that there are some surgeons who remain quite sanguine about loss of distal pulses after reduction of a supracondylar fracture in a child and that some would go so far as to operate upon a child with a tourniquet in place, to proceed to internal fixation of a fracture without exposing the artery or the afflicted nerve. Seddon (1975a) described 38 cases in whom the notes that came with the child were very complete and he observed that: ‘paralysis was by far the most frequent manifestation of ischaemia, with depression of sensibility (a certain indication of nerve involvement) and a reduction or loss of voluntary power’. An absent pulse was recorded in only eight of these cases. Blakey et al. (2009) have studied 26 children who presented to the first hospital with a ‘pink pulseless hand’ following supracondylar fracture and who
Fig. 8.60 The neglected pulseless limb in supracondylar fracture in an 11 year old boy. Operation at 12 days. The median nerve (1) and ulnar nerve (2) were found constricted and ischaemic where they entered an infarcted muscle.
358
Surgical Disorders of the Peripheral Nerves
Table 8.28 Volkmann’s ischaemic contracture: the nerve lesion in 26 children. Nerve affected Cause Depth (56 nerves) Median: 24
Ulnar: 19
Radial: 13
Outcome motor
Outcome sensory
Trapped in fracture: 9
Conduction block: 6
Good: 7
Good: 20
Ischaemia and compression: 14
Axonotmesis: 17
Fair: 10
Fair: 4
Transection: 1
Neurotmesis: 1
Poor: 7
Poor: 0
Trapped in fracture: 3
Conduction block: 8
Good: 7
Good: 16
Ischaemia and compression: 15
Axonotmesis: 10
Fair: 10
Fair: 2
Transection: 1
Neurotmesis: 1
Poor: 2
Poor: 1
Trapped in fracture: 1
Conduction block: 10
Good: 10
Good: 11
Ischaemia and compression: 12
Axonotmesis: 3
Fair: 2
Fair: 2
Transection 0
Neurotmesis: 0
Poor: 1
Poor: 0
were later referred to us during the years 1983–2003 (Table 8.28). The brachial artery had been explored, as an emergency, in four of the children. Recovery was good in three of these, there was mild contracture in the fourth which responded to stretching. The artery was explored and repaired after an interval of 48 h in the fifth child in whom later step elongation of the flexor tendons proved necessary. Twenty three of the children were referred with ischaemic contracture of the muscles of the forearm and hand. Internal fixation of the fracture had been performed at the first operation in 21 children. Some of the findings from the study by Blakey and Biant are now described. The Pulse: The wrist pulses were recorded as absent in all 26 children when they first presented. These were restored in two cases by urgent decompression and in one by emergency repair of the artery. A weak radial pulse was noted in nine children when they were discharged from the first hospital. The pulse was absent or scarcely perceptible in 15 of the children when we saw them. The brachial artery and the affected nerves were explored in 21 children. The artery was constricted by dense scar tissue deep to the bicipital aponeurosis (12) or trapped within the fracture (9). In all cases pulsatile flow returned after decompression of the vessel. The artery was trapped in the fracture in four children, the vessel was narrowed to a cord. Pulsation returned at about 15 min after release of the vessels which were bathed in a solution of papaverine. The Muscle Injury: The flexor compartment was worst affected. Palliative operations were necessary in 22 children. These included flexor muscle slide (17); step elongation of the flexor tendons (4) and a free functioning gastrocnemius muscle transfer (1). The extensor compartment was least affected; the five cases were successfully treated by serial stretching and splinting. The intrinsic muscles were affected in nine hands, and operation became necessary in four of these by release of the contracted adductor pollicis(three cases), or release of the interosseous muscles (one child). The Lesion of the Nerve: Fifty-six main nerves were injured. Not one isolated anterior interosseous palsy was
encountered. Ischaemia of the nerve which was compounded by compression from swollen or infarcted muscle caused the lesion in 41 nerves. Thirteen nerves were trapped in the fracture. Two nerves had been divided by a Kirschner wire during internal fixation. The depth of the lesion varied between different groups of nerve fibres and some preservation or recovery of modalities of cutaneous sensibility and of vasoand sudomotor function was noted at the first examination in 14 of the median and ulnar nerves. Conduction block was the predominant lesion in 10 of the 13 radial nerves. The lesion was degenerative in 30 nerves, two of which required repair. Recovery of cutaneous sensibility and of small muscle function in the hand was generally good after decompression but recovery into the flexor and extensor muscles of the forearm was determined by the extent of post ischaemic fibrosis. In general the nerves recovered far better than the flexor muscles of the forearm after decompression. Pain: The children, or their parents, recollected intense pain in the first 12 h after the injury, which often proved resistant to standard analgesia. The pain was felt in the whole forearm and hand. Growth and Deformity: The extent of growth disturbance was evident only at long term review. Children with a seemingly good early result following corrective surgery had recurrence of deformity with significant defects at skeletal maturity. These injuries were caused by loss of flow through the brachial artery. The affect upon the small muscles of the hand and upon the extensor muscles of the forearm cannot be attributed to compartment syndrome. All of these catastrophes could have been prevented.
8.6.9.2 Prevention One of us (RB) follows the method demonstrated to him in 1975 by Mr Peter French, orthopaedic surgeon to St George’s Hospital. It was used in more than 30 cases of supracondylar fracture at St Mary’s Hospital between 1979 and 1991. If the
Compound Nerve Injury
radial pulse does not return after one attempt at closed manipulation then the brachial artery is exposed through an incision in the distal one third of the arm in the groove between the biceps and triceps muscles. The branches of the medial cutaneous nerve of forearm are encountered in close relation to the larger brachial vein. In front and in a slightly deeper plane lies the median nerve. The brachial artery lies a little deep to the nerve. Both structures are traced down to the elbow and exposed by dividing the bicipital aponeurosis. In most cases no more than this is required and the fracture can now be stabilised. On occasion the artery and the median nerve is caught up within the fracture. It is a simple matter to extricate these at urgent operation. Arterial spasm is a dangerous diagnosis but it does occur in the brachial artery in the child. In some cases it proved necessary to be patient for 10 min or so, keeping the vessel warm and moist, and bathing it with a solution of 1% papaverine. In cases where the pulse is unstable and disappears with flexion of the elbow, Charnley’s method is used. Light skin traction is applied to the skin of the forearm and the limb elevated from a stand holding the elbow at about 45 degrees of flexion. This method was also used in elbows so very swollen that the idea of operation was too daunting.
8.6.10 Iatrogenous Injuries in the Adult Severe pain was a common complication of the 84 nerve lesions in adults. The risks to the posterior interosseous nerve during postero-lateral approach and of ischaemic injury to the ulnar nerve during operations for displaced supracondylar fractures are well known. The introduction of new methods has, of course, brought with it new risks. We have treated five trunk nerves which had been transected during arthroscopic procedures at the elbow, two median, two radial and one ulnar. Six radial nerves, in the distal arm, were transected by the drill during operations of ‘closed’ reduction or during the application of an external frame. In one case the drill avulsed 18 cm of a radial nerve. It is likely that these nerves had been displaced by fracture haematoma, or entrapped within the fracture. It has to be said that there was unaccountable delay before recognition of the event in several of these cases. A different approach is exemplified in one recent case. An experienced surgeon, whilst operating with a displaced supracondylar fracture in a young man found that the ulnar nerve had been inadvertently transected during the operation. He completed the fixation of the fracture and then performed a meticulous primary suture using epineurial sutures. At his request we followed that patient. Recovery was good and the patient was never, at any time, troubled by pain. Tardy ulnar palsy was recognised by Panas (1878). Gay and Love (1947 ) described 100 cases with an average age of onset of 22 years. Holmes and Hall (1978) described five
359
cases in which the onset of the lesion lay between 2 months and 3 years. Wadsworth (1977, 1982) who has written definitively on this subject said ‘this is the most important complication of injuries to the capitular epiphysis ... even in the well treated case there is no absolute certainty that tardy ulnar palsy may not follow in adult life; however this probability is minimised by proper care’.
8.6.10.1 Conclusions 1. Circulation must be restored. Rupture of a brachial artery from a supracondylar fracture in the child is rare. The artery is usually compressed by the bicipital aponeurosis or it has become entangled within the fracture. The collateral circulation about the elbow is, at best, tenuous. Ischaemia is inevitable if the brachial artery is occluded, and if a main nerve or nerves, with attendant vessels are jammed in the fracture. 2. Neuropathic pain or severe pain persisting after reduction of a fracture or deepening of a nerve lesion indicates impending ischaemia until proven otherwise. 3. If the skeletal injury is complicated by a nerve palsy and if the surgeon thinks it necessary to expose the fracture then it is wisest to trace the nerve affected. Entrapment or other distortion of the nerve is more likely than rupture. 4. A deep lesion of a nerve evident after completion of operation strongly suggests that the nerve has been cut or entrapped. 5. The ulnar nerve in the adult requires careful handling. The collateral vessels are better taken with the nerve during anterior transposition.
8.6.11 The Forearm The main threat to the median nerve in fractures of the forearm is that from compression and ischaemia. The embracing flexor muscles and its mid line course generally protect the nerve from laceration by bone fragments. The ulnar nerve lies much closer to bone and it is tethered above in the cubital tunnel and below by the dorsal cutaneous branch. The posterior interosseous nerve is at risk during operation for fractures of the proximal radius. The extensor muscles themselves may be seriously damaged in open fractures or by penetrating missile wounds. The cutaneous nerves, including the superficial radial nerve, the lateral cutaneous nerve of forearm, the dorsal branch of the ulnar nerve and the palmar cutaneous branch of the median nerve are particularly at risk during exposure of the skeleton. Huang et al. (1998) described a case of transection of the median nerve which was probably caused by a bone fragment in an angulated greenstick fracture of the proximal ulna
360
and radius. The nerve was repaired at four and a half months and a good result was achieved. Huang et al. suggest that the nerve was lacerated by a bone fragment before it sprang back into place and they identified five more reports of similar cases. Stahl et al. (1997) described three cases of injury to the ulnar nerve caused by closed fractures of the mid forearm with anterior angulations. The nerves were explored, two were entrapped within the fracture, one was partially severed. Sound advice is given in this good paper: the nerve should be explored when the palsy is associated with anterior angulation of the fracture. Neiman et al. (1998) described two cases of ulnar palsies caused by closed forearm fractures. They went on to spontaneous recovery. We have operated in 14 children with deep lesions of the median and ulnar nerves. In most of these the fracture was so displaced that internal fixation was necessary. The onset of the nerve palsy was not always clear but it seems that it occurred during the operation in 11 of the children. In seven it was clear that the epineurium had been caught up by the bone fragments and drawn into the fracture. Recovery followed release of the nerve in five cases, repair was necessary in the remainder.
8.7 The Lower Limb The prognosis for nerves injured by fractures and dislocations in the lower limb is much worse than it is for the upper limb, probably because of the great violence which usually causes fracture or dislocation of the pelvis, or fracture of the long bones. Seddon (1975b) said, of complete sciatic palsies associated with fracture of the femoral shaft that: ‘it is wise always to explore the nerve’. This principle might be extended to complete lesions of the common peroneal and tibial nerves especially when operation is proposed. The marked rise in iatrogenous vascular injuries incurred during treatment for fractures in the lower limb has already been described. There has been a similar increase in the number of iatrogenous nerve injuries. It seems that the expansion of knowledge and the growth of specialisation, whose effects have deliberately been intensified by the operations of lawyers and the law have created new dangers for patients. Some clinicians may have forgotten that disease does not come neatly labelled as, for instance, ‘medical’, ‘surgical’, ‘orthopaedic’, or even ‘knee’ or ‘foot and ankle’ but as a patient with symptoms and signs which have to be construed. The case described by Blakey and Biant (2008) provides one example. A 28 year old woman with anterior cruciate ligament deficiency and an unstable knee underwent an arthroscopically assisted right anterior cruciate ligament reconstruction using autologous hamstring tendon. She developed a painless common peroneal palsy. We saw her at
Surgical Disorders of the Peripheral Nerves
3 months from operation. The palsy was complete, there was a strong Tinel sign in the mid thigh with weakness of the lateral hamstrings. She had developed a pressure sore from a rigid orthosis applied to the insensate skin of the leg. The nerve was found divided 20 cm proximal to the tip of the fibula. It had been divided by the blunt ended tendon stripper which was used to prepare the tendon graft. The gap between the prepared stumps was 7 cm in length, this was bridged by six sural nerve grafts. Recovery was slow but ultimately good so that she was able to walk comfortably without any orthosis or modification of her shoe (see Fig. 5.31).
8.7.1 The Femoral Nerve The femoral nerve is rather like the spinal accessory nerve in its susceptibility to the attentions of doctors. Kim and Murovic (2008) record 67 iatrogenous lesions in 100 operated cases. The nerve is at risk during operations in the lower abdomen and pelvis, during operations for hernia and it is especially at risk during operations for unrecognised benign nerve tumours. The nerve is vulnerable to ischaemia during operation for arterial reconstruction and also from haematoma following puncture of the femoral artery. It is too frequently injured during femoral nerve block. A high lesion of the femoral nerve is far more disabling than a sciatic palsy. A patient without hip flexion, who is unable to extend the knee, faces great difficulty in walking. The level of the lesion is signified by the presence or absence of hip flexion and by the extent of loss of cutaneous sensibility(Chapter 5). Severe pain after injury is common. Kim and Murovic observed useful recovery of function after eight from nine sutures of the nerve and after 22 of 36 grafts. Tsuchihara et al. (2008) obtained excellent results following repair in two cases where lengths of the nerve had been excised with a benign schwannoma. Repair was performed in one patient 1 month after the injury, a defect of 13 cm was grafted. In the second patient repair was performed at 3 months after the injury and the gap between the nerve stumps was 14 cm. This patient experienced early relief of his pain. Our own experience is set out in Table 8.29. The injury was complicated by an arterial lesion in seven cases and by
Table 8.29 Femoral nerve repairs (adults). Outcome
Number of cases
Good: flexion at hip and extension at knee ³MRC Grade 4
8
Fair: flexion at hip and extension at knee ³MRC Grade 3
6
Poor: no useful recovery
9
Total
23
Compound Nerve Injury
361
sepsis in three more and the results following repair were poor in seven of these cases. Case Report: A surgeon encountered a tumour within the femoral nerve above the inguinal ligament and removed this with 9cm of the nerve. The tumour was a benign schwannoma and the nerve was repaired by graft ten days later. Recovery was rapid and the power of knee extension was normal 14 months later. A similarly happy outcome was seen in the case illustrated in Fig. 7.6, when the damaged nerve was repaired two days after the injury. It seems that the femoral nerve is rather vulnerable to ischaemia after arterial bypass. In one such case aortofemoral bypass was carried out by a most experienced surgeon and it was at his request that the femoral nerve was explored because of the ensuing complete, but painless, femoral palsy 3 weeks later. The nerve was intact but there was only poor flow through the epineurial vessels. Recovery was poor.
8.7.2 The Lumbo Sacral Plexus The brachial plexus is vulnerable because of the mobility of the forequarter on the trunk so that the attachment of its constituent spinal nerves to the spinal cord is the weakest link in the chain. By contrast the lumbo-sacral plexus is protected by the stability of the pelvic girdle. The ilium is bonded to the lumbo-sacral spine by massive ligaments and there is no joint between ilium and pubis. These features are one reason for the relatively scanty writing about and only occasional report of repair of injuries to the lumbo-sacral plexus. Such injuries can only occur from massive injury of a life threatening nature. There is too the matter of technical difficulty. Some clear descriptions of the injury come close on 100 years after descriptions of the equivalent lesions in the upper limb. Finney and Wulfman (1960) wrote of the case of avulsion of lumbar spinal nerves confirmed by myelography. Barnett and Connolly (1975) described what they thought was the fourteenth published case: a patient sent to them after below knee amputation who continued in intractable pain. The stump had been explored, and sympathectomy performed. The diagnosis was made at six years after injury when myelography revealed the true cause. Pain was eased by transcutaneous nerve stimulation. Radiographs confirmed that the original injury had caused displacement of the hemipelvis (Fig. 8.61). The anatomical disposition of the lumbo-sacral plexus is set out by Horwitz (1939) who performed 228 dissections and described 13 variations from the standard pattern, noting that these variations were related to the formation of the lumbar spine so that the shorter (sacralised) spine was associated with a pre-fixed plexus: the longer spine with an extra vertebral body was associated with a postfixed plexus formation.
Fig. 8.61 A 43 year old man experienced severe shooting pains, ‘like lightning’ into his leg and foot for 7 years after fracture of pelvis. A clinical diagnosis of intradural injury of L5 and S1 was confirmed by magnetic resonance scan.
London (1967c) emphasised that the nervi erigentes were particularly vulnerable within severe pelvic fractures ‘judging from the fact that about one man in twenty is more or less impotent (but not sterile) after major fractures’. He also described cases of sciatic paralysis from narrowing of the sciatic notch and a case of ‘an unimportant looking fracture that was shown to have the sciatic nerve trapped in it’. The displacement of the fracture was far greater than the radiographs suggested. Räf (1966) recognised that the double vertical fracture is associated with a particularly high incidence of nerve injuries. Huittinen and Slätis (1972) described a series of 31 patients with nerve injuries, from a total of 68 patients with unstable fractures which included fractures of the pelvic ring and of the sacro-iliac joint. L5 and S1 were most commonly affected and six patients were incontinent. They estimated that the incidence of significant nerve injuries from all pelvic fractures was about 10–12% and the natural history for spontaneous recovery in their cases was poor. Huittinen (1972) went on to necropsy studies of 42 cases of pelvic fracture. There had been unstable vertical fractures in 38 and nerve injuries were revealed in 20 of these. Fifteen nerves were ruptured, 21 were stretched. The spinal nerves were damaged in 6 cases, the anterior rami in seven and the trunk nerves in 27. The lumbo-sacral trunk and the superior gluteal nerves were the most vulnerable (23 cases). Injuries
362
to these two nerves were associated with hemi-pelvic dislocation with outward rotation and upward displacement. Comminution of the lateral sacral mass caused compression of the sacral nerves and the intra-dural injuries were caused by comminution and displacement of the sacro-iliac joints. He described the lumbo-sacral trunk as ‘three to four centimetres long and oval in cross section about 10 mm wide and 4 mm thick. It contains about 40 funiculi and many vasa nervorum. It lies under the psoas major muscle from which it descends obliquely in front of the lateral sacral mass, traverses the front of the sacro-iliac joint behind the common iliac vessels and enters the sacral plexus at the upper border of the piriformis muscle. The lumbo-sacral trunk participates in the nerve supply of the 4th and 5th lumbar myotomes and dermatomes and when it has been damaged these are the segments where motor, sensory and vaso motor disturbances are to be expected.’ Moossy et al. (1987) made an important contribution in their description of six cases of avulsion of spinal nerves from the conus presenting with intractable pain. These patients had suffered violent injuries. There had been three amputations, two by hemi-pelvectomy. Two patients were paraplegic. The pain was characteristic of deafferentation, being constant and burning in nature, but there was also paroxysmal pain. Five of these six patients were cured or eased of their pain by coagulation of the dorsal root entry zone(DREZ lesion). Gibbons et al. (1990) described the patterns of nerve injury in 44 patients with fractures of the sacrum treated during two years. Thirty two fractures were through the wing of the sacrum or through the foramina. Nerves were injured in one quarter of these. The lesions were usually unilateral and involved lumbar or sacral spinal nerves. Twelve fractures extended into the central canal; in these, nerve injuries were more common(about 60%) and much more severe. They were usually bilateral and bowel and bladder incontinence occurred in six of the 12 patients. It is suggested that accurate reduction and internal fixation of the fractures may have contributed towards partial recovery, a view supported by Reilly et al. (1996) who studied 83 patients with unstable pelvic injuries who were treated over the course of 3 years. Nerves were injured in 21% of these. There was some recovery in all of the patients by 24 months after injury. About one half of patients with nerve injuries went on to complete recovery. These findings are rather encouraging and they can be attributed, at least in part, to a policy of early anatomical reduction and adequate stabilisation of unstable pelvic ring fractures. We are indebted to Mr Martin Bircher (St George’s Hospital, London) who has provided information about 489 cases with fractures or dislocations of the pelvis treated in his Institution during three years. Forty two (8.6%) of patients with injuries to the pelvic ring suffered injuries to the lumbo sacral plexus. The injury was classified using the Tile and the Burgess and Young
Surgical Disorders of the Peripheral Nerves
classifications. Thirty seven (88%) of patients with nerve injuries had suffered unstable, Tile type C, fractures. The mean delay from injury to reduction and internal fixation was 11 days. Complete recovery was found in 16(38%) patients in whom fractures were reduced to less than 6 mm of displacement. Some improvement was seen in 11 patients. In 15 patients the lumbo sacral palsy remained complete. Those patients who showed no recovery for the nerves had less accurate reduction of the fracture or the sacro iliac joint reduction, with a mean displacement of 8.8 mm and in these the interval between injury and operation was longer, mean 24 days. Rodgers et al. (2009) conclude that accurate fracture reduction and stabilisation, achieved without delay, creates a better environment for nerve healing and in those injuries with complete nerve lesions delay and incomplete reduction predict a poor prognosis. Meyer (1982) repaired the third and fourth lumbar nerves of a young woman, severed by a glass, at 3 months after injury. Five years after repair function was close to normal throughout the lower limb. Brunelli and Vigasio (1986) described 12 cases outlining indications for extra-and intra-peritoneal approach with repair. Our experience is limited to attempted repair of eleven intra-pelvic injuries, four from penetrating wounds and seven occurring during operation or angiography. The technical difficulties were severe. Probably the largest experience of repair of these injuries is that of our colleague Thomas Carlstedt. His work ‘Central Nerve Plexus Injury’ 2007 is required reading. Carlstedt (2007) advises that the clinical diagnosis of the level of injury in the lumbo sacral plexus is very much more difficult than it is in brachial plexus lesions. Examination of two nerves is helpful in distinguishing between intraspinal lesions and intrapelvic injuries. An intact gluteus medius (superior gluteal nerve, L5, S1) suggests an intrapelvic lesion. If the abductors of the hip are active and if sensation is preserved in the back of the thigh (posterior cutaneous nerve of thigh) then a high lesion of the sciatic nerve is likely. Carlstedt recommends magnetic resonance imaging in place of computerised tomographic myelography; the latter is not only extremely unpleasant but also potentially dangerous in the days after injury (Fig. 8.62 a, b). Some impressive results have been achieved in the complete lesion. Restoration of function in the flexor muscles of hip, the extensor muscles of knee and the abductor muscles of the hip may be sufficient to enable the patient to walk independently. This can be achieved by nerve transfers or by grafts from stumps of the spinal nerves to the femoral and the superior gluteal nerves. Lang Borges and Carlstedt (2004) described ten cases where serious injuries to the lumbo sacral plexus were treated by repair of intraspinal ruptures of ventral roots, by nerve transfers or by repair of intrapelvic lesions. One such case is described in Chapter 4.
Compound Nerve Injury
a
363
b
Fig. 8.62 (a, b) A 21 year old woman sustained a near complete lumbo sacral plexus lesion from fracture dislocation of the pelvis. She experienced intense pain. An intraspinal exploration was performed 6 months later (a) Below left: the fracture dislocation Above: postoperative radio-
graph. Below right: CT myelogram. (b) The lumbo and sacral roots were densely adherent to the dura. There was dramatic relief of pain following neurolysis and later some recovery of function. (By courtesy of Marco Sinisi RNOH).
Case Report: A 22 year old woman suffered an intraspinal lesion of L5 to S2 from a severe pelvic fracture complicated by disassociation of the sacro-iliac joint. The proximal stumps of S1 and S2 were found by laminectomy and reconnected by grafts to the gluteal nerve. By about 1 year after operation she was able to walk without support and with only a modest limp and reinnervation of the gluteal muscles was confirmed by electromyographic examination. Case Report: A 3 year old girl sustained a complete, intraspinal, lesion of the sacral plexus on the right side in a road traffic accident. The injuries to the pelvis and to the pelvic viscera were severe. Three months later the lesion was repaired by grafts placed between the ruptured spinal nerves within the spinal canal and the distal elements of the sacral plexus. She regained good power in the muscles of the hip and the posterior compartment of the thigh but there was little recovery of the muscles of the leg.
These are difficult injuries to diagnose and to treat, the technical problems are forbidding but results such as those obtained by Carlstedt, show that a more vigorous approach is justified in patients suffering from severe neural deficit or who are in severe pain (Fig. 8.63).
8.7.3 The Hip The lesions sustained by the sciatic nerve in posterior dislocation of the hip are analogous to that inflicted on the cords of the brachial plexus by anterior dislocation of the head of humerus. The nerve is stretched. It is ruptured only rarely. The traction force may extend proximally so that the superior gluteal nerve, the sacral or the lumbo-sacral plexus is affected in as many as one third of cases. The initial lesion is
364
Fig. 8.63 Functional recovery into the flexor muscles of the knee and ankle at 3 years after repair of intraspinal rupture (By courtesy of Thomas Carlstedt, RNOH).
predominantly conduction block or axonotmesis. Unless the head of femur is reduced rapidly the traction is compounded by ischaemia from persisting stretching of the vessels of the nerve which will be worsened by haematoma. The nerve lesion deepens from one of conduction block to one of degeneration, from one of favourable degenerative injury (axonotmesis) to an unfavourable one (neurotmesis). The extent of injury to the nerve is determined by the violence of the initial displacement and by the duration of application of the force.
Surgical Disorders of the Peripheral Nerves
acetabulum than in uncomplicated dislocation (ThompsonEpstein Type 1). Letournel and Judet (1981) recorded 57 nerve injuries in 469 of their patients treated by open reduction and internal fixation, a pre-operative incidence of 12% which rose to 17.4% in the posterior column group. Their recommendations about diminishing the risks to the nerve during operations for fracture dislocation repay study. Sciatic palsy occurred in 30 of their patients after operation and recovery was poor in 13 of these. They found that meticulous handling of the nerve, and combining femoral traction with flexion at the knee during operation reduced the incidence of intra-operative injury from 18% to 9%. Case Report: A 43 year old pathologist suffered posterior fracture dislocation of the hip complicated by complete sciatic palsy. The fracture was never reduced and when we came to expose the nerve at 10 months from injury we found it intact although stretched over the posterior aspect of the displaced acetabulum (Fig. 8.64). There was no recovery. Once again we are indebted to Martin Bircher who has released to us data as yet unpublished. Four hundred and fifty six cases of fracture/dislocation of the hip were treated at St George’s Hospital from the 1st January 2004 to the 31st December 2006. Nerve lesions, usually of the sciatic nerve, were identified in 29 (6.3%). All of these injuries were seen in fractures which involved the posterior wall, the posterior column or both . Twenty three (79%) were associated with the more complex Letournel fracture patterns. The nerves fully recovered in nine patients (31%). In these the mean accuracy of reduction was 1.6 mm. Incomplete recovery was found in 15 patients. There was no recovery in five patients with complete nerve palsy; in these, the mean fracture reduction was 2.5 mm and there was significantly longer delay to operation of 32 days against the overall mean delay of 16 days. Rodgers et al. (2009) conclude: ‘acetabular fractures involving the posterior wall or column have a high
8.7.3.1 Incidence Although the sciatic nerve is the most commonly injured, Gruson and Moed (2003) identified 4 cases of femoral palsy from 726 operated cases of acetabular fracture. Epstein (1980) in his personal series of 830 cases found 68 nerve injuries, an incidence of 8%. There was good recovery in 43% but the outcome was poor in 32%. Siegel and Gelberman (1991) found a range of between 6% and 20% in published series commenting that injuries to the sciatic nerve were a good deal more common in types 2, 3 and 4 of the Thompson-Epstein classification (1951) in which dislocation is associated with fractures of the rim or floor of the
Fig. 8.64 Neglected fracture dislocation of the hip complicated by lumbo sacral plexus lesion.
Compound Nerve Injury
incidence of neural injury. Accurate fracture reduction and stabilisation, achieved without prolonged delay, affords a good neural outcome for these patients. In similar injuries with complete nerve palsy, delay and sub optimal surgical reduction predicts a poor prognosis.’ The head of the femur must be reduced as soon as possible, and this must not await referral to an ‘expert’. The sooner the sciatic nerve is relieved from stretching and compression by the displaced head and fracture fragments the better. London(1967c) provides an excellent account of closed reduction in posterior dislocation: ‘The easiest way back for the head of the femur is the reverse of its outwards journey. The traditional method of pulling upwards on the flexed thigh ignores the fact that the hip dislocates most easily from a position of more or less flexion, adduction and medial rotation. It is recommended, therefore, that while attempts to reduce the dislocation may start with the hip flexed to a right angle, if they do not succeed in this position, the thigh should be gently flexed and adducted whilst continued pull on it. It is often helpful to rotate the hip gently inwards and outwards at the same time.’ Badger (1959) thought that ‘sciatic nerve damage is an indication for early manipulation followed immediately by open reduction of the fracture and repair of the sciatic nerve if practicable’. Siegel and Gelberman (1991) described ‘a specific fracture pattern that requires early exploration is one with displaced acetabular bone fragments, where there is an associated primary sciatic nerve injury. At exploration, these fragments have been found commonly compressed or impaling the nerve.’ London (1967c) emphasised the importance of neuropathic pain: ‘a striking feature of two cases in the writer’s experience has been pain so severe as to reduce the sufferers to plaintive wretchedness. Mounting doubts about their fortitude was swiftly dispelled when their sciatic nerves were released from the bony vice and the pain dramatically relieved.’ Another dramatic example of this was case 2 from Birch and Strange (1990). The patient suffered severe pain with peroneal palsy after successful operation upon a fracture of the pelvis. Nearly three years later the first operation the sciatic nerve was exposed at the notch. Its lateral part was narrowed to a translucent web by callus and the latter was removed. Pain was abolished; the nerve made a wholly unexpected recovery over the next 3 weeks. Issack et al. (2007) described how ten patients with painful sciatic neuropathy experienced relief of pain and some improvement in function by releasing the sciatic nerve from scar tissue and heterotopic bone. Jiang et al. (2002) recommend that ‘open reduction and internal fixation combined with nerve exploration and neurolysis should be used as early as possible for severe sciatic pain’. Our experience comprises 25 operations for exploration of the sciatic nerve damaged in fracture dislocations of the hip. Operations were performed in complete lesions; partial lesions with severe pain; lesions occurring after
365
Fig. 8.65 Rupture of the sciatic nerve from a fracture of the proximal femoral shaft.
open reduction and for lesions of increasing depth. This approach was justified in those cases where the nerve was entangled around a bone fragment or entrapped within the joint or compressed by callus. In four cases the nerve was found transected and was repaired but useful recovery followed in only one of these. As far as we can see exploration of the nerve did not improve the outcome in cases where the nerve had been violently stretched but had not ruptured. Injury to the lumbo-sacral plexus is likely in such cases. Rupture of the nerve is uncommon. Kline (1998) and his colleagues described 57 cases of sciatic palsy complicating fracture/dislocation of the hip. In 27 of these the lesion extended to the lumbo sacral plexus. The nerves were explored in 30 patients. The tibial nerve was ruptured in 6, the common peroneal nerve in 8, a total of 14 nerves from 60. These nerves were repaired by graft. Recovery was considered good in 21 of the 30 tibial nerves but in only 10 of the common peroneal nerves (Fig. 8.65).
8.7.4 The Common Peroneal and Tibial Nerves The two main nerves forming the sciatic trunk are functionally distinct and they often appear as quite separate structures as high as the greater sciatic notch. Lesions of the common peroneal nerve have acquired a bad reputation which may be, in part, explained by the high level of function which is necessary for a good result. A good result after repair of the nerve requires that the patient is able to walk normally and that he or she has no pain (Table 8.30). Such an outcome was seen in just over one third of the 205 cases reported by Zachary and by Woodhall and Beebe in an even larger series. The requirements for a worthwhile recovery in a lesion of the tibial nerve are set rather lower. Most repairs
366
Surgical Disorders of the Peripheral Nerves
of this nerve are followed by recovery of flexion of the heel and protective sensation in the plantar skin (Table 8.31). The tibial nerve is less exposed to traction or to laceration by a fragment of bone in the thigh and in the popliteal fossa but it is particularly at risk in the leg especially in the lower part of the deep flexor compartment where it runs close to the tibia and adjacent to the posterior tibial artery. The common peroneal nerve is tethered by the piriformis above and below at the neck of the fibula. The fascia around the biceps femoris muscle sweeps around to embrace the nerve. The proportion occupied by bundles is higher than in the tibial nerve: the proportion of connective tissue is less. Kadiyala et al. (2005) showed that the segment of the nerve just above and at the neck of the fibula contains only a few fine blood vessels, whereas the tibial nerve at the knee has an abundant supply from a series of arcades springing from the popliteal and tibial vessels. The deep peroneal nerve passes anteriorly and almost horizontally before continuing between tibialis anterior and extensor digitorum longus where it becomes closely applied to the interosseous membrane. There is added risk from the course of the anterior tibial artery, which
Table 8.30 Assessment of recovery of the common peroneal nerve. Motor recovery Sensory recovery Good
Dorsiflexion and eversion ³MRC 4
No spontaneous pain; no hypersensitivity
Fair
Dorsiflexion MRC3+ to 4 or Eversion MRC 3+ to 4 or Dorsiflexion and eversion MRC 2 to3
No spontaneous pain, no, or only mild, hypersensitivity
Poor
Less than the above
Spontaneous pain or significant sensitivity
is the end artery of the anterior compartment of the leg. The vessel passes between the heads of tibialis posterior, through the oval aperture in the proximal part of the interosseous membrane into the anterior compartment where it lies on the interosseous membrane in the proximal two thirds of its course immediately adjacent to the deep peroneal nerve. Niall et al. (2005) provide valuable information about the prognosis of peroneal palsy in dislocation of the knee. Fifty five patients with knee dislocation were treated in the years between 1994 and 2001. There were 14 palsies (41%). All of these were seen in the 34 cases with anterior or antero-medial dislocation in which both cruciate ligaments and the posterolateral structures of the knee were torn. Four nerves were ruptured. In three lesions in continuity the damaged segment was less than 7 cm, and these nerves recovered. Recovery was partial or altogether absent in the seven lesions with longer segments of damage. There were only two cases of arterial injury; one intimal tear of the popliteal artery was treated expectantly, one complete rupture of the popliteal artery was successfully repaired. Severe adduction injury at the knee inflicts traction upon the nerve which is pulled anteriorly by the biceps muscle in avulsion fractures of the fibular styloid (Fig. 8.66) (see also Fig. 5.2). This is exemplified by the case described by
Table 8.31 Grading of outcome for the tibial nerve: all levels. Good
Heel flexors MRC 4 or better. Tibialis posterior, flexor digitorum longus, and flexor hallucis longus MRC 3. Return of sweating. No fixed deformity, no trophic disturbance. Normal shoes
Protective sensation with no more than mild hypersensitivity; no spontaneous pain
Fair
Heel flexors MRC 3+ or better. No useful recovery in long flexor muscles. Incomplete return of sweating. No serious trophic disturbance, no fixed deformity. Normal shoes with or without foot drop splint for CPN palsy
Protective sensation, hypersensitivity not interfering with daily activities
Poor
No useful motor recovery; and /or significant fixed deformity; and/or trophic ulceration
No return of sensation; and/or significant hypersensitivity interfering with daily activities
Fig. 8.66 Rupture of the left common peroneal nerve in a girl of 17 years following severe adduction of the knee; appearance 5 days after injury (George Bonney 1959).
Compound Nerve Injury
367
Table 8.32 319 repairs of the common peroneal and tibial nerves 1979–2006 in adults and children (%). Common peroneal nerve Tibial nerve Tidy Untidy Closed Tidy Untidy wound wound traction wound wound
Closed traction
Good
29 (60.4)
18 (38.3)
40 (33.3)
87 (40.5)
17 (68)
17 (37)
6 (18.2)
40 (38.5)
Fair
12 (25)
15 (32)
28 (23.3)
55 (25.6)
6 (24)
20 (43.5)
16 (48.5)
42 (40.4)
Poor
7 (14.6)
14 (29.7)
52 (43.4)
73 (33.9)
2 (8)
9 (19.5)
11 (33.3)
22 (22.1)
Total
48
47
120
215
25
46
33
104
Montgomery et al. (2005) The cruciate ligaments were reconstructed three years after injury in a young woman. She awoke with a complete, but painless, foot drop. We saw her at 3 months when a Tinel sign was noted in the posterior fossa. The biceps tendon was displaced anteriorly. At operation the common peroneal nerve was pulled anteriorly by the displaced muscle. The tendon graft had encircled the nerve, compressing it to about one third of its normal diameter. The graft was divided, so releasing the nerve. The graft was then repaired. Recovery proceeded as axonotmesis and was complete by 5 months. Bottomley et al. (2005) studied the course of the common peroneal nerve in 54 patients coming to operation for repair of the postero-lateral structures. The nerve was pulled both proximally and anteriorly by the displaced muscle in 16 of the 18 patients who had sustained either with avulsion fracture of the fibular head or avulsion of the biceps tendon. Bottomley says: ‘whenever bone or soft tissue avulsion from the fibular head is suspected, the surgeon should expect an abnormal position of the common peroneal nerve and appreciate the increased risk of iatrogenic damage’. Here is a clear warning against endoscopic or ‘minimal access’ work, which is difficult enough when anatomical relations are normal, but even more so when they are not. The sciatic trunk is so large that repair by suture is often the only way forward. Fortunately this is usually possible when that trunk has been ruptured in the thigh or at the knee. The sooner that this is done the better. If the condition of the wound or of the patient is such that nerve repair cannot be contemplated at the first operation then approximation of the stumps by a 3 × 0 nylon filament is most helpful in preventing their retraction. From time to time suture is impossible even with the hip in the neutral position and the knee flexed to 90°. The gap between the stumps can then be lessened by passing sutures through the epineurium and bridging the diminished gap by grafts. Suture of either nerve in the leg is generally possible only in urgent repair of lacerations. Our results in 319 repairs are set out in Table 8.32. The outcome is determined by two factors which outweigh all others. First comes the cause, with the traction ruptures faring worse. Results are bad in the ischaemic leg. The delay before repair is the second most important factor in recovery. Sixty of the 153 traction ruptures were repaired within 14 days of the
injury and the result was graded good in 30 of these. This result was seen in only 18 of the 93 traction ruptures repaired at a time exceeding 14 days from injury. Kim et al. (2004) describe the experience acquired by David Kline and his colleagues in New Orleans in 318 operated patients. The lesions were caused by traction or fracture in 51% of the cases. 138 Nerves were grafted. The result was useful in 75% of repairs where the gap was less than 6 cm, in 38% where the gap lay between 6 and 12 cm and only in 16% when the gap exceeded 13 cm. There were 16 good results in 19 cases where suture was possible. Neurolysis was followed by recovery in 121 lesions in continuity exhibiting a compound nerve action potential (CNAP) traversing the lesion.
8.7.5 More Recent Experience at the Royal National Orthopaedic Hospital Some characteristics of the injury in 200 consecutive cases of compound lesions of the common peroneal nerve seen in the years 2000–2007 are set out in Table 8.33. This excludes tidy wounds, and iatrogenous injuries occurring during operations upon varicose veins. Fifty of these lesions were iatrogenous. Conduction block was rare. The behaviour of Tinel’s sign provided reliable information about the likelihood of spontaneous recovery. Serial examinations were performed in 90 nerves and the findings Table 8.33 200 lesions of the common peroneal nerve by level, depth, and energy transfer. Depth of lesion Level of lesion Energy transfer Conduction block
4
Buttock
19
Massive
20
Axonotmesis
99
Thigh
38
High
100
Neurotmesis 97 Knee and leg 143 Low 80 1. There were 50 (25%) iatrogenous lesions, 25 of which (12.5%) were associated with arterial or other ischaemic injury. 2. There were 50 (25%) cases of arterial injury or compartment syndrome. 3. Twenty seven (13.5%) patients presented with severe neuropathic pain. 4. The number of low energy injuries seems rather high, but this includes most of the iatrogenous lesions.
368
Surgical Disorders of the Peripheral Nerves
recorded by radiation of sensation into the superficial or the deep peroneal nerves, that is 180 nerves in all. The sign remained static at the level of lesion in 75 nerves. At operation a rupture or other unfavourable lesion was revealed. Decompression of a nerve which had not been ruptured often converted a static into a progressing Tinel sign. One example is illustrated in Fig. 5.34. A progressing Tinel sign, where percussion distally over the regenerating axons evoked stronger sensations than those induced by percussion over the level of lesion was recorded in 105 nerves. This was followed by recovery of function in 93 of them. The advancing Tinel sign was misleading in 12 nerves, 11 of these related to the deep peroneal nerve; in all of these ischaemic fibrosis of the muscles in the anterior compartment became evident. Seventy two superficial and deep peroneal nerves were extricated from scar or entrapment or were set free from displaced bone. The results are rather encouraging although 11 of the 12 poor results from neurolysis were seen in the deep peroneal nerve. There was useful recovery in the stretch lesions in continuity in which the perineurium had not been ruptured (Table 8.34). Case Report: A young woman, who was a considerable athlete, dislocated her knee during a long jump. Her distal pulses were lost, they returned after urgent reduction. The tibial nerve recovered within days but the common peroneal palsy remained complete. The nerve was explored 8 months later. At operation a 5 cm segment was stretched and fibrosed. There was no conduction across this lesion. Clearly, the posterior tibial plateau had smashed against the nerve in a true antero-posterior dislocation. Repair was not done because time was against us, and we did no more than release the nerve from scar. The patient, and ourselves, were agreeably surprised by the early onset of recovery. Function was normal at 15 months after operation. It seems that, in this case, the operation of ‘neurolysis’ or extrication contributed to the recovery as it was in Horsley’s case (1899) (Fig. 8.67).
Fig. 8.67 The common peroneal nerve 8 months after it had been crushed by the tibia in anteroposterior dislocation. The Tinel sign was static. There was no conduction across the lesion. There was recovery after neurolysis (axonotmesis).
Table 8.34 Results of operation in 114 compound lesions of the common peroneal nerve 2000–2007. Repairs 78 nerves Neurolysis Graded by the 72 superficial and deep peroneal divisions of the common peroneal nerve in 36 patients Good
30
Good
53
Fair
8
Fair
7
Poor 40 Poor 12 1. A result was good in 11 of the 22 nerves repaired within 14 days of injury, fair in 3 more and poor in 8. 2. The result was good in 8 of the 32 nerves repaired after 3 months had elapsed; it was poor in 20 of these. 3. 11 of the 12 poor results from neurolysis were seen in the deep peroneal nerve. 4. “Tidy” wounds, including 23 iatrogenous lesions, are excluded.
Fig. 8.68 A sore was caused by an ankle foot orthosis applied to insensate skin. This 38 year old man required 5 days of intravenous antibiotics for septicaemia.
Forty of the 78 repairs of the whole common peroneal nerve failed but a good result was seen in 11 of the 22 nerves repaired within 14 days of injury and there was a fair result in three more. Twenty of the 32 repairs performed after an
Compound Nerve Injury
interval of 3 months failed. Useful recovery into either the deep peroneal (1 case) or the superficial peroneal nerve (7 cases) was seen in eight patients. Repair of the superficial division by graft and transferring the nerve to lateral gastrocnemius to the deep division was successful in three out of five cases of severe ruptures with long gaps between the nerve stumps. A foot drop is disabling and the surgeon must take direct charge of post operative care and rehabilitation. The conventional ankle/foot orthosis is a deplorable instrument (Fig. 8.68). Below knee amputation was unavoidable in two cases where the appliance was prescribed in older patients with complete lesions of the sciatic nerve, in whom complete erosion of the plantar skin led to osteomyelitis and we have been informed of two other such cases. Six of the 200 patients studied during the period 2000–2007 required inpatient treatment for spreading cellulitus or even septicaemia and 11 more required prolonged outpatient treatments for pressure sores. Other and better appliances are available but all too often it is left to the patient to find them for his or herself.
References Adams DB (1982) Wound ballistics: a review. Milit Med 147:831–835 Aids to the examination of the peripheral nervous system. O’Brian M (ed) (2000), 4th edn, Elsevier, London Alberty RE, Goodfried G, Boyden AM (1981) Popliteal artery injury with fracture dislocation of the knee. Am J Surg 142:36–40 Allieu Y, Gomis R, Asencio G, Mekhaldi A (1986) Sciatic paralysis following hip surgery. In: Peripheral nerve repair and regeneration, vol 4, Liviano Press, Padua, pp 27–31 Badger FG (1959) Fractures near or involving joints of the lower limb. In: Ruscoe C, Badger FG, Sevitt S (eds) Modern trends in accident surgery and medicine. Butterworth & Co, London, pp 176–195 Barnes R (1954) Causalgia: a review of 48 cases. In: Seddon HJ (ed) Peripheral nerve injuries by the Nerve Injuries Committee of the Medical Research Council. Medical Research Council Special Report Series 282, London. HMSO, pp 156–185 Barnett HG, Connolly ES (1975) Lumbosacral nerve root avulsion: report of a case and review of the literature. J Trauma 15:532–535 Barros D’Sa AAB (1982) A decade of missile-induced vascular trauma. Ann R Coll Surg Engl 64:37–44, Hunterian Lecture. Delivered at the Royal Army Medical College on the 30th April 1980 Barros D’Sa AAB (1992) Arterial injuries. In: Eastcott HHG (ed) Arterial surgery, 3rd edn. Churchill Livingstone, Edinburgh, pp 355–413 Barros D’Sa AAB, Harkin DW (2005) Pathophysiology of acute vascular insufficiency (Chapter 2). In: Barros D’Sa AAB, Chant ADB (eds) Emergency vascular and endovascular surgical practice, 2nd edn. Hodder, Arnold, London, pp 19–27 Bigliani LU, Craig EV, Butters KP (1991) Fractures of the shoulder. In: Rockwood CA, Green DP, Bucholz RW (eds) Fractures in adults, 3rd edn. Lippincott, Philadelphia, PA, pp 871–1020 Birch R, Achan P (2000) Peripheral nerve repairs and their results in children. Hand Clinics 16:579–597 Birch R, Raji ARM (1991) Repair of median and ulnar nerves. J Bone Jt Surg 73B:154–157
369 Birch R, St Clair Strange FG (1990) A new type of peripheral nerve lesion. J Bone Jt Surg 72B:312–313 Birnstingl M (1982) Vascular injuries. In: Watson-Jones R Fractures and joint injuries, 6th edn. JN Wilson Ed. Churchill Livingstone, Edinburgh, pp 211–241 Blakey CM, Biant LC (2008) Transection of the common peroneal nerve during harvesting of tendons for cruciate ligament reconstruction. A case report. J Bone Jt Surg 90A:1567–1569 Blakey CM, Biant LC, Birch R (2009) Ischaemia and the pink, pulseless hand complicating supracondylar fractures of the humerus in children. J Bone Jt Surg 91B:1487–1498 Blom S, Dahlback LO (1970) Nerve injuries in dislocation of the shoulder joint and fractures of the neck of the humerus. Acta Chir Scand 136:461–466 Bonnard C, Anastakis DJ, van Melle G Narakas AO (1999) Circumflex and suprascapular nerves. J Bone Jt Surg 81B:212–217 Bonnel F (1989) Anatomie du plexus brachial chez le nouveau-né et l’adulte. In: Alnot JY, Narakas A (eds) Les Paralysies du Plexus Brachiale, Monographies du Groupe d’Étude de la Main. Expansion Scientifique, Paris, pp 3–13 Bonney G (1963) Thrombosis of the femoral artery complicating fracture of the femur. J Bone Jt Surg 47B:344–345 Bottomley N, Williams A, Birch R, Noorani A, Lewis A, Lavelle J (2005) Displacement of the common peroneal nerve in postero-lateral corner injuries of the knee. J Bone Jt Surg 87B:1225–1226 Bowden REM, Scholl DH (1954) Recovery of nerves. In: Peripheral nerve injuries. Medical Research Council, Special Report Series No. 282, Part 2, pp 16–24 (Chapter1) Brown PW (1970) The time factor in surgery of upper extremity peripheral nerve injury. Clin Orthop Rel Res 68:14–21 Brunelli G, Vigasio A (1986) Lumbar brachial plexus surgery. In: Peripheral nerve repair and regeneration, vol 4, Liviano Press, Padua, pp 21–25 Buncke HJ, Buncke GM, Valauri FA (1991) In: Meyer VE, Black MJM (eds) Microsurgical procedures: Hand and upper limb series, vol 7. Churchill Livingstone, Edinburgh, pp 1–8 Burton AC (1951) On the physical equilibrium of small blood vessels. Am J Physiol 164:319–329 Calvet J, Leroy M, Lecroix L (1942) Luxations de l’Épaule et lesions vasculaires. J Chir 58:337–346 Camp SJ (2010) Iatrogenous lesions of the spinal accessory nerves, J Bone Jt Surg (92B) in press Carlstedt T (2007) Central nerve plexus injury. Imperial College Press, Covent Garden. London Carrel A (1902) La technique opérative des anastomoses vasculaires et la transplantation des viscères. Lyon Méd 98:859–864 Cavanagh SP, Birch R, Bonney G (1987) The infraclavicular brachial plexus: The case for primary repair. J Bone Jt Surg 69B:489 Chant DBS (2005) Emergency vascular services in the UK. In: Barros d’Sa AAB, Chant ADB (eds). Emergency vascular and endovascular surgical practice, 2nd edn, Hodder, Arnold. London Section 1A, pp 3–6 Chen CW, Ch’en YC, Pao YS (1963) Salvage of the forearm following complete traumatic amputation: case report. Chin Med J 82:632 Chen ZW, Meyer VE, Kleinert HE, Beasley RW (1981) Present indications and contraindications for replantation as reflected by long-term functional results. Orthop Clinics N Am 12: 849–870 Clarke RA (1959) Principles and technique of wound surgery. In: Clarke RA, Badger FC, Sevitt S (eds) Modern trends in accident surgery and medicine. Butterworth, London, pp 107–122 Cobbett JR (1969) Free digital transfer. J Bone Jt Surg 51B:677–679 Coene LNJEM (1985) Thesis: Axillary nerve lesions and associated injuries. University of Leiden, the Netherlands Comtet JJ, Herzberg G, Alnaasan I (1993) Biomechanics of the shoulder and the scapulothoracic girdle. In: Tubiana R (ed) The hand
370 volume 1V, WB Saunders, Philadelphia, PA, pp 99–111 (Chapter 8) (English translation) Cooper GJ, Ryan JM (1990) Interaction of penetrating missiles with tissues: some common misapprehensions and implications for wound management. Brit J Surg 77:606–610 De Laat EAT, Visser CPJ, Coene LNJEM, Pahplatz PVM, Tavy DLJ (1994) Nerve lesions in primary shoulder dislocations and humeral neck fractures. A prospective clinical and EMG study. J Bone Jt Surg 76B:381–383 DeBakey ME, Simeone FA (1946) Battle injuries of the arteries in WWII. Ann Surg 123:534–579 DeBakey ME, Simeone FA (1955) Acute battle incurred arterial injury. In: Surgery and world war two, vascular surgery, Medical Department U.S. Army, U.S Government Printing Office, Washington, DC, pp 60–148 Delorme E (1915) The treatment of gunshot wounds of nerves. Brit Med J 1:853–855 Donaghy M (2005) Lumbosacral plexus lesions. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy. Elsevier, Philadelphia, PA, pp 1375–1390 Dunkerton MCD, Boome R (1988) Stab wounds involving the brachial plexus. J Bone Jt Surg 70B:566–570 Eadie DGA (1979) Post-traumatic ischaemia. J Bone Jt Surg 61B: 265–266 Eardley WGP, Stewart MPM (2010) Early management of ballistic hand trauma. J Am Acad Orthop Surg 18:118–126 Eardley WGP, Ramasamy A, Clasper J, Stewart MPM (2009) The primary treatment of war wounds, personal communication Eastcott HHG, Blaisdell FW, Silver D (1992) Acute ichaemia. In: Eastcott HHG (ed) Arterial surgery, 3rd edn. Churchill Livingstone, Edinburgh, pp 31–54 Eaton RG, Green WT (1972) Epimysiotomy and fasciotomy in the treatment of Volkmann’s ischemic contracture. Orthop Clin North Am 3:175–186 Ekholm R, Adami J, Tidermark J, Hansson K, Tõ¨rnkvist H, Ponzer S (2006) Fractures of the shaft of the humerus. J Bone Jt Surg 88B:1469–1473 Ellis H (1958) Disability after tibial shaft fractures. J Bone Jt Surg 40B:190–197 Emadian SM, Lee WW (1996) Axillary artery pseudoaneurysm and axillary nerve palsy: delayed sequelae of anterior shoulder dislocation. Am J Emerg Med 14:108–109 Epstein HC (1980) Traumatic dislocation of the hip. Wilkins & Wilkins, Baltimore, MD Finney LA, Wulfman WA (1960) Traumatic induced lumbar nerve root avulsion with associated traction injury to the common peroneal nerve. Am J Roentgen Rad Ther Nuclear Med 84:952–957 Flaubert AC (1827) Mémoire sur plusiers cas de luxations dans les efforts pour la réduction ont été suivis d’accidents graves. Rép Gén d’Anat et de Physiol Pathol 3:55–79 Fleming A (1915) On the bacteriology of septic wounds. Lancet ii:638–643 Foster RJ, Swiontkowski MF, Bach AW, Sack JT (1993) Radial nerve palsy caused by open humeral shaft fractures. J Hand Surg 18A: 220–222 Gallen J, Wiss DA, Cantelmo N, Menzoin JO (1984) Traumatic pseudoaneurysm of the axillary artery: Report of three cases and literature review. J Trauma 24:350–354 Gay JR, Love JG (1947) Diagnosis and treatment of tardy paralysis of the ulnar nerve. J Bone Jt Surg 29:1087–1097 Gibbons KJ, Soloniuk DS, Razack N (1990) Neurological injury and patterns of sacral fractures. J Neurosurg 72(6):889–893 Gilbert A, Masquelet AC, Hentz VR (1990) Les lambeaux artérieles pédiculés du membre supérieur. Expansion Scientifique Française Godina M (1986) Early microsurgical reconstruction of complex trauma of the extremities. Plast Reconstr Surg 78:285–292
Surgical Disorders of the Peripheral Nerves Goldie BS, Powell JM (1991) Bony transfixion of the median nerve following colles fracture. Clin Orthop Rel Res 273:275–277 Gopal S, Majumber S, Batchelor AG, Knight SL, Deboer P, Smith RM (2000) Fix and flap: The radical orthopaedic and plastic treatment of severe open fractures of the tibia. J Bone Jt Surg 82:959–966 Gousheh J (1995) The treatment of War injuries in the brachial plexus. J Hand Surg 20A:S68–S76 Gousheh J, Arasteh E, Beikpour N (2008) Therapeutic results of sciatic nerve repair in Iran/Iraq war casualties. Plast Reconstr Surg 121:878–886 Green NE, Allen BL (1977) Vascular injuries associated with dislocation of the knee. J Bone Jt Surg 59A:236–239 Gruson KI, Moed BR (2003) Injury of the femoral nerve associated with acetabular fracture. J Bone Jt Surg 85A(3):428–431 Gumina S, Postacchini F (1998) Anterior dislocation of the shoulder in elderly patients. J Bone Jt Surg 79B:540–543 Gustilo R, Anderson JT (1976) Prevention of infection in the treatment of one thousand and twenty five open fractures of long bones. J Bone Jt Surg 58A:453–458 Guthrie CC (1912) Blood vessel surgery and its applications (reprint 1959; Harbison SP, Fisher B (eds)), University of Pittsburgh Press, Pittsburgh, PA Hallock GG (1992) Fascio-cutaneous flaps. Blackwell, Oxford, London Hallowell (1761) Cited by Lambert 1844. In: Erichson JE. Observations on aneurysm Hamdan FB, Jaffar AA, Ossi RG (2008) The propensity of the common peroneal nerve in thigh level injuries. J Trauma 64:300–303 Helm AT, Watson JS (2002) Compression of the brachial plexus in a patient with false aneurysm of the axillary artery as a result of anterior shoulder dislocation. J Shoulder Elbow Surg 11:278–279 Hewitt RL, Smith AD, Drapanas T (1973) Acute traumatic arteriovenous fistulas. J Trauma 13:901–906 Holden CEA (1979) The pathology and prevention of Volkmann’s ischaemic contracture. J Bone Jt Surg 618:296–299 Holmes JC, Hall JE (1978) Tardy ulnar nerve palsies in children. Clin Orthop Rel Res 135:128–131 Holstein A, Lewis GR (1963) Fractures of the humerus with radial nerve paralysis. J Bone Jt Surg 45A:1382–1288 Horn JS (1959) Peripheral nerve injuries. In: Clark R, Badger FG, Sevitt S (eds) Modern trends in accident surgery and medicine. Butterworth & Co, London, pp 305–319 Horsley V (1899) On injuries to peripheral nerves. The Practitioner. Old Series vol 63, new series vol 10:131–144 Horwitz MT (1939) The anatomy of (A) the lumbosacral nerve plexus – its relation to variation of vertebral segmentation and (B) the posterior sacral nerve plexus. Anat Record 74:91–107 Huang K, Pun WK, Coleman S (1998) Entrapment and transection of the median nerve associated with greenstick fractures of the forearm. J Trauma 44:1101–1102 Hughes CW (1958) Arterial repair during the Korean War. Ann Surg 147:555–561 Hughes CW, Bower WF (1961) Traumatic lesions of peripheral vessels. Charles C Thomas, Springfield, Illinois Huittinen V-M (1972) Lumbosacral nerve injury in fracture of the pelvis. A post mortem radiographic and patho-anatomical study. Acta Chir Scand Suppl 429:7–43 Huittinen V-M, Slätis P (1972) Nerve injury in double vertical pelvic fractures. Acta Chir Scand 138:571–575 Hull RB, Rob CG (1997) Amputation. In: Ballistic trauma clinical relevance in peace and war. Arnold, London, Sidney, Auckland, pp 149–159 (Chapter 12) Hunter J (1793) In: Palmer (ed) Works of John Hunter. Longman, London (1835) Issack PS, Toro JB, Buly RL, Helfet DL (2007) Sciatic nerve release following fracture or reconstructive surgery of the acetabulum. J Bone Jt Surg 89A(7):1432–1437
Compound Nerve Injury Jancke EJ (1958) Late structural and functional results of arterial injuries primarily repaired. Surgery 43:175–183 Jiang D, Yu X, An H, Liang Y, Liang A (2002) Hip and pelvic fractures and sciatic nerve palsy. Clin J Traumatol 5(6):333–337 Kadiyala RK, Ramirez A, Taylor AE, Saltzman CL, Cassell MD (2005) The blood supply of the common peroneal nerve in the popliteal fossa. J Bone Jt Surg 87B:337–342 Kim DH, Murovic JA (2008) Lower extremity nerve injuries. In: Kim DH, Midha R, Murovic JA, Spinner RJ (eds) Kline and Hudson’s nerve injuries, 2nd edn. Saunders Elsevier, Philadelphia, PA, pp 209–278 (Chapter 11) Kim DH, Murovic JA, Tiel RL, Kline DG (2004a) Penetrating injuries due to gunshot wounds involving the brachial plexus. Neurosurg Focus 16:1–6 Kim DH, Murovic JA, Tiel RL, Kline DG (2004b) Management and outcomes in 318 operative common peroneal nerve lesions at the Louisiana State University Health Sciences Center. Neurosurgery 54:1421–1429 Kinmonth JB, Hadfield GJ, Connolly JE, Lee RH, Amoroso EC (1956) Traumatic arterial spasm: Its relief in man and in monkeys. Brit J Surg 44:164–169 Kirkup JR (1963) Major arterial injury complicating fracture of the femoral shaft. J Bone Jt Surg 45B:337–343 Kitsis CK, Marino AJ, Krikler SJ, Birch R (2003) Late complications following clavicular fractures and their operative management. Injury 34:69–74 Kleinert HE, Kasdan HL, Romero JL (1963) Small blood vessel anastomosis for salvage of severely injured upper extremity. Plast Recon Surg 52:301 Klenerman L (2007) The evolution of the compartment syndrome since 1948 as recorded in the JBJS(B). J Bone Jt Surg 89-B: 1280–1282 Kline DG (1989) Civilian gun shot wounds to the brachial plexus. J Neurosurg 70:166–174 Kline DG, Hudson AR (1995) Nerve injuries. WB Saunders, Philadelphia, PA Kline DG, Kim D, Midha R, Harsh C, Tiel R (1998) Management and results of sciatic nerve injuries: a 24 year experience. J Neurosurg 89:13–23 Komatsu S, Tamai S (1968) Successful replantation of a completely cut off thumb. Plast Recon Surg 42:374 Lambert S (2005) Further opinion on Shao et al 2005, Radial nerve palsy. J Bone Jt Surg. Doi:10.1302/0301-620X.87B12.17720 Lamerton AJ, Bannister R, Withrington R, Eastcott HHG (1983) Claudication of the sciatic nerve. Brit Med J 286:1785–1786 Lang EM, Borges J, Carlstedt T (2004) Surgical treatment of lumbosacral plexus injuries. J Neurosurg Spine 1:64–71 Larsen LB, Barfred T (2000) Radial nerve palsy after simple fracture of the humerus. Scand J Plast Reconstr Surg Hand Surg 34:363–366 Laverick MD, Barros D’Sa B, Kirk SJ, Mollan RAB (1990) Management of blunt injuries of the axillary artery and the neck of the humerus. J Trauma 30:360–361 Leclerq DC, Carlier AJ, Khuc T, Depierreux L, Lejeune GN (1985) Improvement in the results in sixty-four ulnar nerve sections associated with arterial repair. J Hand Surg Second Int Suppl 10A:997–999 Lefrac BA (1976) Knee dislocation. Arch Surg 111:1021–1024 Leffert RD, Seddon HJ (1965). Infraclavicular brachial plexus injuries. J Bone Jt Surg 47B: 9–22 Letournel E, Judet R (1981) Fractures of the acetabulum. Elson RA (trans). Springer Verlag, New York Lexer E (1913) Ideale aneurysmaoperation und getässtransplantation. Dtsch Med Wochenschr 39:725 Lin J, Shen PW, Hou SM (2003) Complications of locked nailing in humeral shaft fractures. J Trauma 54:943–949
371 Lindsey D (1980) The idolatry of velocity or lies, damn lies and ballistics. Editorial J Trauma 20:1068–1069 Lister G (1988) Emergency free flaps. In: Green DP (ed) Operative hand surgery, 2 edn, vol 2. Churchill Livingstone, Edinburgh, pp 1127–1149 Lister G, Scheker L (1988) Emergency flaps in the upper extremity. J Hand Surg 13A:22–28 Livani B, Belangero WD, Castro de Medeiros R (2006) Fractures of the distal third of humerus with palsy of the radial nerve. J Bone Jt Surg 88B:1625–1628 London PS (1967a) A practical guide to the care of the injured. E & S Livingstone, Edinburgh London PS (1967a) Repair of blood vessels ibid: pp102–109 London PS (1967b) Arterial injury in shoulder dislocation. Ibid: p 197 London PS (1967c) The hip. Ibid: pp 319–337 Luce EA, Griffen WO (1978) Shotgun injuries of the upper extremity. J Trauma 18:487–492 MacCarty CS (1951) Two-stage autograft for repair of extensive damage to sciatic nerve. J Neurosurg 8:319 Makins GH (1919) On gunshot injuries to the blood vessels: Founded on experience during the Great War 1914–1918. Wright, London Malt RS (1964) Replantation of severed arms. J Am Med Assoc 189:114 Mathes SJ, Nahai F (1979) Clinical atlas of muscle and myocutaneous flaps. CV Mosby, St Louis Matsen FA (1975) The compartmental syndrome: A unified concept. Clin Orth 113:8–14 McCready RA, Procter CD, Hyde GL (1986) Subclavian axillary vascular trauma. J Vasc Surg 3:24–31 McGregor IA, Jackson IT (1972) The groin flap. Brit J Plas Surg 25:3–16 McQueen MM, Court-Brown C-M (1996) Compartment monitoring in tibial fractures. J Bone Jt Surg 78B:99–104 Medical Research Council (1944) Arterial injuries, By the Vascular Injuries Subcommittee of the MRC War Wounds Committee, HMSO, London Medical Research Council Special Report Series No. 282 (1954) Peripheral nerve injuries (Seddon HJ: ed) HMSO, London Mellor SG, Easmon SF, Sanford JP (1997) Wound contamination and antibiotics. In: Ryan JM, Rich NM, Dale RF, Morgans BT, Cooper GJ (eds) Ballistic trauma, pp 61–71 (Chapter 5) Merle M, Amend P, Cour C, Foucher G, Michon J (1986) Microsurgical repair of peripheral nerve lesions: a study of 150 injuries of the median and ulnar nerves. Periph Nerve Rep Regen 2:17–26 Meyer VE (1982) Langzeit result nach Nervenrekonstrucktion. Bereich des Plexus. lumbalis - Ein Fallbericht. Hand Chirurgie Mikrochirurgie Plastische Chirurgie 14:11–13 Mikami Y, Nagano A, Ochiai N, Yamamoto S (1997) Results of nerve grafting for injuries of the axillary and suprascapular nerves. J Bone Jt Surg 79B:527–531 Montgomery AS, Birch R, Malone A (2005) Entrapment of a displaced common peroneal nerve following knee ligament reconstruction. J Bone Jt Surg 87B:861–862 Moossy JJ, Nashold BS, Osborne D, Friedman AH (1987) Conus medullaris root avulsions. J Neurosurg 66:835–841 Morris GC, Beall AC, Roof WR, DeBakey ME (1960) Surgical experience with 220 acute arterial injuries in civilian practice. Am J Surg 99:775–780 Mullett H, Babu J, McMahon B, Burke P (1998) Vascular complications of anterior dislocation of shoulder with associated brachial plexus injury – a case report. J Med Sci 167:196 Murphy JB (1897) Resection of arteries and veins injured in continuity. End-to-end suture. Experimental and Clinical Research. Med Rec 51:73–88
372 Narakas AO (1993) Paralytic disorders of the shoulder girdle In: Tubiana R (ed) The hand, vol 4, WB Saunders, Philadelphia, PA, pp112–125 (Chapter 9) Nash E, Soudry M, Abrahamson J, Mendes DG (1984) Neuropraxis secondary to haemorrhage in traumatic dislocation of the shoulder. J Trauma 24:546–547 Neiman R, Maiocco B, Deeney VE (1998) Ulnar nerve injury after closed forearm fractures in children. J Paed Orthop 18:683–685 Niall DM, Nutton RW, Keating JF (2005) Palsy of the common peroneal nerve after traumatic dislocation of the knee. J Bone Jt Surg 87B:664–667 O’Brien BM (1975) Microsurgery in the treatment of injuries. In: McKibben B (ed) Recent advances in orthopaedics. Churchill Livingstone, Edinburgh, pp 235–279 Oberlin C, Bastian D, Gréant P (1994) Les Lambeaux pédiculés se couverture des membres, Expansion Scientifique Française, Paris Ochiai N, Nagano A, Mikam V, Yamamoto S (1997) Full exposure of the axillary and suprascapular nerves. J Bone Jt Surg 79B:532–533 Omer GE (1974) Injuries to nerves of the upper extremity. J Bone Jt Surg 56A:1615–1624 Omer GE (1982) Results of untreated peripheral nerve injuries. Clin Orth Rel Res 163+:15–19 Omer GE (1991) Nerve injuries associated with gunshot wounds of the extremities. In: Gelberman RH (ed) Operative nerve repair and reconstruction. JB Lippincott, Philadelphia, PA, pp 655–676 Omer GE (1998) Peripheral nerve injuries and gunshot wounds. In: Omer GE, Spinner M, Van Beek AL (eds) Peripheral nerve problems, 2nd edn, WB Saunders, Philadelphia, PA, pp 398–405 (Chapter 41) Omer GE, Eversmann WW (1994) Peripheral nerve problems In: Burkhalter WE (ed) Orthopaedic surgery in Vietnam, Office of the US Army Surgeon General and Center of Military History, Washington DC Orcutt MB, Levine BA, Root MD, Sirinek KR (1983) The continuing challenge of popliteal vascular injuries. Am J Surg 146:758–761 Osborne AWH, Birch R, Munshi P, Bonney G (2000) The musculocutaneous nerve: Results of 85 repairs. J Bone Jt Surg 82B:1140–1142 Ouellette EA (2000) Gunshot wounds. In: Gupta A, Kay SPJ, Scheker LR (eds) The growing hand, 1st edn, Mosby, ISBN OT 234 21331, pp 717–724 Owen R, Tsimboukis B (1967) Ischaemia complicating tibial and fibular shaft fractures. J Bone Jt Surg 49b:268 Owen-Smith MS (1981) High velocity missile wounds. Edward Arnold, London Panas J (1878) Sur une cause peu connue de paralysie du nerf cubital. Arch Gén Méd 2:5 Parkes AR (1945) Traumatic ischaemia of peripheral nerves, with some observations of Volkmann’s ischaemic contracture. Brit J Surg 32:403–414 Pasila M, Faroma H, Kiviluoto O, Sundholm A (1978) Early complications of primary shoulder dislocations. Acta Orthop Scand 49:260–263 Pasila M, Faroma H, Kiviluoto O, Sundholm A (1980) Recovery from primary shoulder dislocations. Acta Orthop Scand 51:257–262 Patel J, Turner M, Birch R, McCrory P (2001) Rupture of the axillary (circumflex) nerve and artery in a champion jockey. Br J Sports Med 35:361–362 Pellegrini VD, McCollister Evarts C (1991) Compartment syndromes. In: Rockwood AC, Green DP, Bucholz RW (eds) Fractures, vol 1, 3rd edn. J B Lippincott, Philadelphia, PA, pp 390–416 Platt H (1928) On the peripheral nerve complications of certain fractures. J Bone Jt Surg 10:403–414 Pollock RC, Birch R (1999) Complete transposition of the radial nerve complicating an open fracture of the shaft of humerus. Injury 30:623–625 Pollock FH, Drake D, Bovill EG, Day L, Trafton PG (1981) Treatment of radial neuropathy associated with fractures of the humerus. J Bone Jt Surg 63A:239–243
Surgical Disorders of the Peripheral Nerves Porter MF (1967) Arterial injuries in an accident unit. Brit J Surg 54:100 Pringle JH (1913) Two cases of vein grafting for the maintenance of a direct arterial circulation. Lancet 1:1795 Prudnikov OE (1994) Lésions simultanées de la coiffe des rotateurs et du plexus brachial. Rev De Chirurg Orthop 80:602–609 Puri R, Clark J, Corkery PH (1985) Axillary artery damage following closed fracture of the neck of the humerus. Injury 16:426–427 Räf L (1966) Double vertical fractures of the pelvis. Acta Chir Scand 131:298–305 Ragland R, Moukok D, Ezaki M, Carter PR, Mills J (2005) Forearm compartment syndrome in the new born: A report of 24 cases. J Hand Surg 30A:997–1103 Raju S (1979) Shotgun arterial injuries of the extremities. Am J Surg 138:421–425 Ramachandran M, Birch R, Eastwood D (2006) Clinical outcome of nerve injuries associated with supracondylar fractures of the humerus in children. J Bone Jt Surg 88B:90–94 Ramage D (1927) The blood supply to the peripheral nerves of the upper limb in man. J Anat 61:198 Ramasamy A, Eardley WGP, Stewart MPM, Etherington J, Birch R (2009) Ballistic nerve injuries. Proceedings: Combined Services Orthopaedic Society. Reilly MC, Zinar DM, Matta JM (1996) Neurologic injuries in pelvic ring fractures. Clin Orthop Relat Res 329:28–36 Rich NM, Baugh JH, Hughes CW (1970) Acute arterial injuries in Vietnam: 1000 cases. J Trauma 10:359–369 Rich NM, Jarstfer BS, Greer TM (1974) Popliteal artery repair failure: causes and possible intervention. J Cardiovasc Surg 13:340–351 Rich NM, Hobson RW, Collins GJ (1975) Traumatic arteriovenous fistulae and false aneurysms: a review of 558 lesions. Surgery 78: 817–828 Ring D, Chin K, Jupiter JB, Boston MA (2004) Radial nerve palsy associated with high-energy humeral shaft fractures. J Hand Surg 29A:144–147 Rockwood CA, Thomas SC, Matsen PA (1991) Subluxations and dislocations about the gleno-humeral joint. In: Rockwood CA, Green DP, Bucholz RW (eds) Fractures, vol 1. JB Lippincott, Philadelphia, PA, pp 1021–1179 Rodgers B, Pearce R, Walker R, Bircher M (2009) Incidence and outcome of neural injuries following acetabular fractures. British Orthopaedic Association Scientific Meeting, Manchester, September Rodgers B, Pearce R, Walker R, Bircher M (2009) Incidence and outcome for neural injuries following pelvic fractures. British Orthopaedic Association Scientific Meeting, Manchester, September Roganovic Z (2004) Missile-caused ulnar nerve injuries: outcomes of 128 repairs. Neurosurgery 55:1120–1129 Roganovic Z (2005a) Missile-caused median nerve injuries: results of 81 repairs. Surg Neurol 63:410–419 Roganovic Z (2005b) Missile caused complete lesions of the peroneal nerve and peroneal division of the sciatic nerve: results of 157 repairs. Neurosurgery 57:1201–1212 Roganovic Z, Mesovic S, Kronja G, Savic M (2007) Peripheral nerve lesions associated with missile induced pseudo aneurysms. J Neurosurg 107:765–775 Samardzic M, Grujicic D, Milninkovic ZB (1990) Radial nerve lesions associated with fractures of the humeral shaft. Injury 21:220–222 Secer HI, Daneyemez M, Tehli O, Gonul E, Izci Y (2008) The clinical, electrophysiologic and surgical characteristics of peripheral nerve injuries caused by gunshot wounds in adults: a 40 year experience. Surg Neurol 69:143–152 Seddon HJ (1954) Nerve grafting. In: Peripheral nerve injuries by the nerve injury, Committee of the Medical Research Council, MRC Special Report Series 282, London, HMSO, pp 402–403 Seddon HJ (1964) Volkmann’s ischaemia. Brit Med J 1:1587–1592 Seddon HJ (1975a) Neurovascular injury. In: Surgical disorders of peripheral nerves, 2nd ed. Churchill Livingstone, Edinburgh, pp 89–111
Compound Nerve Injury Seddon HJ (1975b) Common causes of nerve injury. In: Surgical disorders of peripheral nerves, 2nd edn. Churchill Livingstone, Edinburgh, pp 67–88 Seddon HJ (1975c) Results of repair of the nerves. In: Surgical disorders of peripheral nerves, 2nd ed. Churchill Livingstone, Edinburgh, pp 303–314 Shah JJ, Bhatti NA (1983) Radial nerve paralysis associated with fractures of the humerus. Clin Orth Rel Res 172:171–176 Shao YC, Harwood P, Grotz MRW, Limb D, Giannoudis PV (2005) Radial nerve palsy associated with fractures of the shaft of the humerus: a systematic review. J Bone Jt Surg 87B:1647–1652 Shaw JL, Sakellarides H (1967) Radial nerve paralysis. J Bone Jt Surg 49A:899–902 Shepard GH (1980) High-energy low velocity close range shotgun wounds. J Trauma 20:1065–1067 Shergill G, Birch R, Bonney G, Munshi P (2000) The radial and posterior interosseous nerves: Results of 260 repairs. J Bone Jt Surg 83B:646–649 Sherrill JG (1913) Direct suture of brachial artery following rupture, result of traumatism. Ann Surg 58:534–536 Siegel DB, Gelberman RG (1991) Peripheral nerve injuries associated with fractures and dislocations. In: Gelberman RH (ed) Operative nerve repair and reconstruction. J B Lippincott, Philaadelphia, PA, pp 619–633 Sonneveld GJ, Patka P, van Mourik JC, Broere G (1987) Treatment of fractures of the shaft of the humerus accompanied by paralysis of the radial nerve. Injury 18:404–406 Spilsbury J, Birch R (1996) Some lesions of the circumflex and suprascapular nerves. J Bone Jt Surg 73B:Suppl. 1.59 Stahl S, Rozen N, Michaelson M (1997) Ulnar nerve injury following midshaft forearm fractures in children. I. J Hand Surg 22B:788–789 Stenning M, Drew S, Birch R (2005) Low energy arterial injury at the shoulder with progressive or delayed nerve palsy. J Bone Jt Surg 87B:1102–1106 Stewart M, Birch R (2001) Penetrating missile injuries. J Bone Jt Surg 83B:517–524 Stewart MPM, Kinninmonth A (1993) Shotgun wounds of the limbs. Injury 10:667–670 Strange FG StC (1947) An operation for pedicle nerve grafting. Brit J Surg 34:423–425 Subbotitch V (1913) Military experiences of traumatic aneurysms. Lancet ii:720–721 Sumner DS, Eastcott HHG, Rich NM (1992) Arteriovenous fistulae. In: Eastcott HGG (ed) Arterial surgery, 3rd edn. Churchill Livingstone, Edinburgh, London, pp 521–559 Syed AA, Williams HR (2002) Shoulder disarticulation: A sequel of vascular injury secondary to a proximal humeral fracture. Injury 33:771–774 Thompson VP, Epstein HC (1951) Traumatic dislocation of the hip: A survey of 204 cases covering a period of 21 years. J Bone Jt Surg 33A:746–778 Tinel J (1917) Nerve wounds. Ballière Tindall & Cox, London. Authorised translation Rothwell F. Revised and edited by Joll CA
373 Tsao BE, Wilbourn AJ (2003) The medial brachial fascial compartment syndrome following axillary arteriography. Neurology 61:1037 Tsuchihara T, Nemoto K, Arino H, Amako M, Murakami H, Yoshi Zumi Y (2008) Serial nerve grafts for long defects of the femoral nerve after resection of retroperitoneal schwannoma. J Bone Jt Surg 90B:1097–1100 Vascular Surgical Society (2001) The provision of emergency vascular services November 2001. Prepared by the Vascular Society of Great Britain and Ireland VSS, November 2001 Venouziou A, Gougoulias N, Dailiana Z, Varitimidis S, Malizos KN (2007) Surgical treatment of radial nerve lesions associated with fractures of the humeral shaft. J Hand Surg 32E:43–44 Verga M, Peri di Caprio A, Bochiotti F, Bruschi S, Petrolati M (2007) Delayed treatment of persistent radial nerve paralysis associated with fractures of the middle third of the humerus. J Hand Surg (Eur) 32E:529–533 Vichard P, Tropet Y, Landecy G, Briot JF (1982) Parlaysies radials contemoraines des fractures de la diaphyse humérale. Chirurgie 108:791–795 Wadsworth TG (1977) The external compression syndrome of the ulnar nerve at the cubital tunnel. Clin Orth Rel Res 124:189–204 Wadsworth TG, Brooks DM (1982) Neurological disorders. In: Wadsworth TG (ed) The elbow. Churchill Livingstone, Edinburgh, p 71 Wajcberg E, Thoppil N, Patel S, Fernandez M, Hale De Fronzo Cersosimo E (2006) Comprehensive assessment of post ischaemic vascular reactivity in Hispanic children and adults with and without diabetes mellitus. Paediatr Diab 7:329–335 Watson-Jones R (1930) Primary nerve lesions in injuries of the elbow and wrist. J Bone Jt Surg 12:121–140 Watson-Jones R (1936) Dislocation of the shoulder joint. Proc R Soc Med 29:1060–1062 Weglowski R (1925) Uber die getässtransplantation. Zentrabl Chir 52(ii):2241–2243 Wilbourn AJ (2005) Brachial plexus injuries In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn, Elsevier Saunders, Philadelphia, PA, pp 1339–1373 (Chapter 55) Williams WW, Twyman RS, Birch R, Donell ST (1996) The posterior triangle and the painful shoulder: Spinal accessory nerve injury. Ann R Coll Surg 78:521–525 Woodhall B, Beebe GW (eds) (1956) Peripheral nerve regeneration, US Government Printing Office, VA Medical Monograph Washington DC Woodruff RE (1898) The causes of the explosive effects of modern small calibre bullets. New York Med J 67:593–601 Wright AE (1915) A lecture of wound infections and their treatment. Proc R Soc Med 9:1–72 Wynn Parry CB (1981) Rehabilitation of the hand, 4th edn. Butterworth, London Young JZ (1993) First evidence of axonal transport. In: Dyck PJ, Thomas PK, Griffin JW, Low PA, Poduslo JF (eds) Peripheral neuropathy, 3rd edn, p 2 Zachary RB (1954) Results of nerve sutures In: Seddon HJ (ed) Peripheral nerve injuries. Medical Research Council, Special report series No. 282, pp 354–387
9
The Closed Supraclavicular Lesion
‘There must be a beginning of any good matter, but the continuing to the end, until it be thoroughly finished yields the true glory’ Drake to Walsingham Types of intradural (preganglionic) injury. Mechanisms of injury. Anatomy; functional distribution; blood supply. Lesions of the spinal cord. Work of George Bonney; definition and prognosis. Epidemiology. Diagnosis: clinical; clinical significance of pain; electrodiagnostic; imaging. Strategies of repair: the incomplete lesion; the complete lesion; bilateral lesions. Techniques of repair. Results: methods; definitions. Repair of avulsed ventral roots. Recovery of function by nerve and by patients. Recovery in children and in the older patient. Relief of pain by repair; relation to the innervation of muscle; more effective in early repair. Return to work; retraining; interval before return. Repair in preganglionic injury: Jamieson and Bonney; early experimental and clinical work; developed and extended by Carlstedt; Carlstedt achieves functional success with ventral rot reimplantation. ‘As with many other operable lesions of the nervous system, there is an unpardonable delay between the time the diagnosis is established and resort to operation. Subcutaneous injuries of the plexus must be regarded as serious injury either with or without operative interference. The number of spontaneous recoveries is exceedingly small and it is extraordinary how many patients have been allowed to progress to a hopeless state without operation and how many are not operated on until months have elapsed. In view of the gravity of the lesion, of the resulting disability and of the occasional intractable neuralgias, surgeons should insist on resort to operation, if not immediately, at least as soon as it can be demonstrated with reasonable certainty that spontaneous recovery is improbable. An excellent working rule is that of Sherren (1906) that all subcutaneous injuries which give reactions of degeneration at the end of 10 day should be submitted to operation’. Frazier and Skillern 1911
Injuries to the brachial plexus are the worst of all peripheral nerve lesions because of the frequency of associated injury to the spinal cord and the common complication of severe pain. In many cases the damage lies between the dorsal ganglion and the spinal cord. To these injuries (Bonney 1954) applied the term preganglionic and this designation will be used interchangeably with intradural to signify rupture or avulsion of roots within the spinal canal. The fact of intradural injury was clearly recognised by Klumpke (1885), Horsley (1899), and Thorburn (1900) and it was described by Frazier and Skillern (1911) who exposed the spinal cord of their patient by hemi laminectomy: ‘on the left lateral aspect of the canal corresponding to the level of the 6th cervical and the 1st thoracic vertebrae there were two defects in the dura, each as large as a 10 cent piece. On the left side both the anterior and posterior roots of the 6th, 7th and 8th cervical nerves were absent. The roots evidently had been torn completely from the cord’. We shall consider, in this chapter, the closed traction lesion of the supraclavicular brachial plexus. Open injuries and the
closed traction lesions predominantly affecting the infraclavicular brachial plexus have been described in Chap. 8. The preganglionic injury is the most common lesion of spinal nerves because the union of the root with the spinal cord is indeed the weakest mechanical link in the long chain between the central nervous system and the periphery. The lesion was intradural in about one half of the 7,500 spinal nerves exposed at operation in 1,500 patients since 1966; the incidence of intradural lesion is highest in C7. We recognise two types of intradural injury: intradural rupture peripheral to the transitional zone (TZ) and avulsion central to it (Chap. 2), of which the former may be the more common (Jamieson and Bonney 1979, Birch and Bonney 1998, Schenker 1998, Schenker and Birch 2001). The lesion may be confined either to the ventral or the dorsal roots, (Bonney 1977, Privat et al. 1982). The extent of displacement of the dorsal root ganglion and the level of ruptures of the dura varies. These patterns are recognised in a classification set out in Fig. 9.1 and some examples are illustrated in Figs. 9.2 and 9.3.
R. Birch, Surgical Disorders of the Peripheral Nerves, DOI: 10.1007/978-1-84882-108-8_9, © Springer-Verlag London Limited 2011
375
376 Fig. 9.1 Classification of preganglionic injury. Type A: avulsion central to the transitional zone. Type B1: avulsion distal to the transitional one, the dura torn within the canal. Type B2: the dural sleeve is intact as far as the mouth of the foramen. Type B3: rupture of roots within the spinal canal without displacement of the dorsal root ganglion or rupture of the dura. Type B4: avulsion confined to dorsal (as shown) or ventral root.
Surgical Disorders of the Peripheral Nerves
Type A
Type B1
Type B2
Type B3
Type B4
9.1 Mechanisms of Injury Flaubert (1827) described a fatal outcome following late reduction of dislocation of the shoulder by manual traction applied by eight students. Flaubert recorded a Bernard-Horner syndrome long before this was formally recognised and he also described an affection of the cord. At necropsy a rupture of C5 was found with avulsion of C6, C7, C8 and D1. The subclavian artery was ruptured and bleeding from the torn radicular vessels led to a haematoma within the spinal canal. Horsley (1899) in order to study the mechanism of damage dropped cadavers on their heads from the ceiling of his laboratory at University College and then went on to dissect the brachial plexus. He found that the brachial plexus was vulnerable when the head was violently forced in one direction whilst the shoulder was forced in the opposite direction. Barnes (1949) pointed out that with the limb adducted the brunt of injury is borne by the upper roots of the brachial plexus, whereas with the arm abducted and extended the lower roots are more vulnerable, a mechanism which causes preganglionic injury to C8 and T1 with ruptures of the cords of the plexus and axillary artery. Leffert (1985), drawing on Stephen’s model (1934), suggests with the arm at 90° of
abduction the axis of force passes through C7; with the arm adducted the first rib concentrates the force on to the lower roots but if a weight falls on the shoulder then the upper roots suffer most. If the arm is abducted then the coracoid acts as a pulley and force is transmitted along the line of the cords to fall chiefly onto the lower roots of the plexus. Preganglionic injury with displacement of the ganglion into the posterior triangle of the neck was attributed by Mansat et al. (1979) to a lateral or peripheral traction force. Sunderland (1974) suggested that avulsion without displacement of ganglion was caused by an axial force on the spinal cord during severe lateral flexion injuries to the cervical spinal column so that the roots are sheared off from the cord. There is one common element underlying closed supraclavicular traction lesion and that is the violent distraction of the forequarter from the head, neck and chest so that the angle between the head and shoulder is opened. One important physical sign showing this is the presence of linear abrasions on the chin and on the face with corresponding abrasions and bruising at the tip of the shoulder (Fig. 9.4). We have seen complete avulsion of the brachial plexus from weights falling on the shoulder. A 48 year old seamstress who was walking along the road and was struck on her right shoulder by a
The Closed Supraclavicular Lesion
Fig. 9.2 Total avulsion. Note the variation in level of rupture of the dura and extent of displacement of dorsal root ganglia.
Fig. 9.3 Intact C5 and C6; central avulsion of C7 and C8.
collapsing wall. Intradural injury of C4-T1 was confirmed at operation. Other patients have provided a clear history of the shoulder colliding with an immovable object and complete preganglionic injury has been seen in pedal cyclists or motor cyclists whose speed at the time of accident was 25 mph or less but whose shoulder struck a post or kerbstone.
377
Fig. 9.4 Linear abrasions in the neck indicate separation of the forequarter from the trunk. The abrasion at the tip of the shoulder marks the point of impact against a road side kerb. Rupture of C5, preganglionic C6 to T1.
Case Report: A 19 year old rugby player went in for a low tackle, his shoulder struck the opponent’s shin and he experienced severe lightning like, shooting pain extending from the shoulder to the tip of the thumb. Initially the whole arm was paralysed but he soon regained function in the forearm and hand. He continued to experience burning pain along the radial aspect of the forearm with occasional shooting bursts of pain. Paralysis was confined to those muscles supplied by the 5th and 6th cervical nerves and there was no Tinel sign. At operation preganglionic injury of C5 and C6 was confirmed. Semple (2008), one of the pioneers of brachial plexus surgery in the United Kingdom, has advised us about several of his cases. In one instance two motor cyclists were playing ‘chicken’ by riding at high speed towards one another on the central white line of a road. One patient was knocked unconscious and came off his machine, the other remained conscious and remained on his bike. Both suffered serious injuries to the brachial plexus. Semple comments: “in the
378
instant prior to a collision it is an unavoidable reflex for a motor cyclist to tense both arms in full extension, and probably also tense the scalene muscles. At impact the shoulder joint, scapula, shoulder girdle are all projected violently posteriorly, with frequent disruption of the sterno-clavicular joint’. Altaf F. (2009, pers. comm.) has provided information from his study of seven patients with injuries caused by tackles in rugby or American football. Fourteen spinal nerves were avulsed, one was ruptured, and there were stretch lesions in continuity in 16 more. Altaf found that the pattern of nerve lesion was related to the posture of the neck and the forequarter at the moment of impact. In one 21 year old man, a video recording showed him running at full speed with the trunk flexed and the head extended towards another player who was running forward in front of him. He attempted a sideways tackle onto the other player so that there was impact between the hip of the opponent and the supraclavicular region of the patient. This caused a downward depression of the shoulder and flexion of his neck to the contralateral side. He sustained a total plexus lesion and experienced very severe pain which started about 20 min after his injury. There was a Bernard Horner sign. Intra-operative findings confirmed an intradural injury of the roots of C5 to T1. Several mechanisms may operate in an individual injury as Narakas (1993) pointed out: ‘the victim may hit the oncoming vehicle, then the road, and will slide upon the ground to be stopped by the kerb, a post, a tree….’
9.2 Anatomy Leffert (1985) has discussed the topographical anatomy of the brachial plexus in detail, referring particularly to the work of Ker (1918) and Walsh (1877), who performed over 400 dissections finding abnormalities in the formation of the plexus in some 2% of cases. Other important works include those of Harris (1904) and Hovelaque (1927). The course and relations of the brachial plexus are illustrated in Chaps. 1 and 7. We shall discuss here some anatomical features that have particular significance for the surgeon embarking on repair. Herzberg et al. (1989) demonstrated that C5 and C6 are protected against traction by the transverse radicular ligaments and by proximal branches such as the dorsal scapular and rami to the phrenic nerve and nerve to serratus anterior. The vulnerability of the lower nerves is enhanced by the increasing obliquity of the roots within the spinal canal and the upward course of T1 where it traverses the inner face of the first rib.
9.2.1 Course in the Neck The anterior primary rami enter the posterior triangle of the neck between scalenus anterior and scalenus medius. The
Surgical Disorders of the Peripheral Nerves
first thoracic nerve passes upward round the neck of the first rib, behind the pleura and behind the vertebral artery and the first part of the subclavian artery. The formation of the trunks of the brachial plexus is fairly consistent. C5 and C6 form the upper trunk, the middle trunk is a continuation of C7, the lower trunk from C8 and T1. These lie in front of one another rather than side by side, with the subclavian artery passing antero-medially. The phrenic nerve crosses C5 to pass anteromedially on the surface of scalenus anterior. The transverse cervical and greater auricular nerves wind round the posterior border of the sterno mastoid no more than 1 cm cephalad, the spinal accessory nerve emerges from deep to the sterno cleidomastoid about 5 mm further cephalad. A significant branch from C4 to C5 or to the upper trunk is encountered in between 2% and 3% of operated cases. We have seen the suprascapular nerve arising directly from C4 in three cases whilst the musculocutaneous nerve so arose in two patients. We have not encountered T2 innervating skeletal muscle in the upper limb and have seen one case where T1 did not contribute to the plexus. Three significant nerves pass from the brachial plexus within the posterior triangle. (1) The nerve to serratus anterior is formed by rami from C5 to C6 which pass deep to scalenus medius. C4 contributes to this nerve in one-third of cases, C7 in about two thirds. (2) The dorsal scapular nerve (C5) may be conjoined with the ramus to the nerve to the serratus. It leaves C5 within the foramen lying posterior to the main trunk. (3) The suprascapular nerve passes away from C5, or from the proximal part of the upper trunk, a finger’s breadth above the clavicle passing laterally and then posteriorly through the suprascapular notch. The divisions of the brachial plexus lie deep to the clavicle and their display in a scarred field can be particularly tedious. The posterior division of the upper trunk is consistently larger than the anterior; this is true also for the middle trunk. In some 10% of cases there is no posterior division of the lower trunk. The formation and relations of the three cords are variable and indeed their designations somewhat misleading. Immediately inferior to the clavicle the posterior cord lies lateral to the axillary artery, the medial cord behind, the lateral cord in front. The cords assume their appropriate relations about the axillary artery deep to pectoralis minor. Miller (1939) found considerable variation in 9% of 480 dissections of the neurovascular axis. Most commonly the axillary artery lies anterior to the three cords and median nerve. The branches of the posterior cord, the largest of the three trunks, are consistent: in sequence they are the subscapular, the thoraco dorsal and the circumflex (axillary) nerves. The branches of the medial cord too are usually predictable, the medial pectoral nerve, the medial cutaneous nerve of the forearm, and then the division into the medial root of the median nerve and the ulnar nerve. Not infrequently the ulnar nerve arises as two or three branches. The greatest variation
The Closed Supraclavicular Lesion
in formation of trunk nerves is found within the lateral cord. In about 10% of cases the musculocutaneous nerve arises more distally than usual, springing directly from the lateral cord as two or three branches or even from the median nerve itself. Sometimes the highest of these branches enters the coraco brachialis muscle no more than 2 or 3 cm below the coracoid process. The lateral root of the median nerve may arise as two or three branches and in some cases it appears as a branch of the musculocutaneous nerve. It is useful to be aware of these variations of the formation of the plexus and of its relation to the great vessels because they can cause difficulties during urgent exploration.
379
brachio radialis, extensor carpi radialis longus (ECRL), and sometimes contributes to pronator teres, supinator, and flexor carpi radialis. The seventh cervical nerve gives widespread innervation throughout the limb: in those unusual instances where it alone is damaged the patient will note a rather diffuse loss of function throughout the upper limb often without complete anaesthesia and sometimes without paralysis of any significant muscle group. It consistently supplies latissimus dorsi. The eighth cervical nerve innervates the extensors of the digits and of the thumb in at least one third of cases; the first thoracic nerve innervates these muscles and also the medial head of triceps in at least 10%. It is useful to bear these variations in mind in clinical examination of the injured plexus and whilst interpreting neurophysiological evidence.
9.2.2 Micro Anatomy 9.2.3.1 Blood Supply The arrangement of nerve fibres within their bundles has been extensively analysed by, amongst others, Mansat (1977), Narakas (1978), Slingluff et al. (1987) and Bonnel (1989). There is general agreement that the total number of myelinated fibres (MNF) in the brachial plexus in the adult is between 120,000 and 150,000 and that fully 25% of these innervate the shoulder girdle and gleno humeral joint. The 5th cervical and first thoracic nerves contain the least number of MNF, between 15,000 and 20,000. The eighth nerve is usually the largest containing about 30,000. Harris (1904) suggested that the proportion of motor fibres was highest in C5 and then C8 and was least in C7 and D1. He thought that the sensory contribution was greatest in C7, then C6 and then C8. The segregation of nerve fibres is established well proximally so that the suprascapular nerve can be traced to an anterior bundle within C5 at the level of the anterior tubercle. One bundle, controlling the radial extensors of the wrist, can be consistently displayed by stimulation within the posterior division of the upper trunk. One patient lost wrist extension after removal of a massive solitary neurofibroma from the axilla. The tumour arose from only one bundle in the posterior cord which could not be preserved. The transfer of one bundle from an intact spinal nerve to a nearby avulsed ventral root or to a peripheral nerve is permitted by this segregation.
9.2.3 Functional Distribution The cutaneous innervation is illustrated in Chap. 5. There are, however, frequent and important variations. The variations of supply amongst the spinal nerves within the brachial plexus are more important than pre or post fixation especially so for C7, C8 and T1. The fifth cervical nerve regularly controls extension, abduction and lateral rotation of the shoulder. The sixth cervical usually innervates biceps, brachialis, lateral head of triceps,
Abdulla and Bowden (1960) and Herzberg et al. (1996) showed that the blood supply of the brachial plexus arises through vessels from the subclavian and vertebral arteries. Important branches pass from the vertebral artery to C5, C6, and the more proximal cervical nerves. Branches from the costocervical trunk provide a rich supply to T1 and C8. Two major arteries arise from the thyrocervical trunk: the suprascapular, which which runs behind the clavicle in front of the brachial plexus and the subclavian vein into the supraspinous fossa, and the superficial cervical artery which crosses anterior to the phrenic nerve and the scalenus anterior becoming bound down to the upper trunk of the brachial plexus by the pre vertebral fascia. Here the artery is between 2 and 3 mm in diameter and it is very useful in vascularising a free ulnar nerve graft. The dorsal scapular artery may arise from the third part of the subclavian artery to pass between the upper and middle trunks. It is a cause of substantial bleeding in a scarred field. In at least one third of cases the superficial cervical artery and the dorsal scapular vessel arise from the thyrocervical trunk as the transverse cervical artery. The great vessels and their branches are particularly at risk in severe closed traction lesions. The arteries arising from the thyro cervical trunk must be preserved during operations in cases when the subclavian artery has not been repaired. They have become the lifeline to the upper limb. The vertebral artery lies anterior to the brachial plexus and is in close relation to C5 and C6. It supplies radicular vessels to the spinal nerves which are particularly abundant about C7 and C8. These pass to the anterior spinal artery (Chap. 2). A Brown-Séquard Syndrome is common in complete preganglionic injuries perhaps by impairment of the blood supply of the spinal cord (Chap. 3). Bleeding following rupture of the radicular vessels may be profuse: we have seen, in one case of avulsion, compression of the trachea from bleeding of the vessels about C7 so severe as to demand urgent tracheostomy (Flannery and Birch 1990a). Intradural injury may cause
380
ischaemia of the cord, and the effects of this may be apparent almost immediately or after delay. It is this risk of ischaemia which presents the greatest single obstacle to the reimplantation of avulsed spinal nerves into the cord.
9.3 Lesions of the Spinal Cord Unrelated cord injury: Grundy and Silver (1984) and Akita et al. (2006) describe series of combined injury to the brachial plexus and to the thoracolumbar spinal cord. The outcome was generally poor, chiefly because of delay in diagnosis and in treatment. We have treated nine patients with complete paraplegia associated with severe injuries to the brachial plexus who made it clear that protraction at the thoraco scapular joint, adduction and medial rotation at the shoulder, extension of the elbow and extension of the wrist were particularly important functions. This advice has been reinforced by many patients who have not suffered an associated injury to the spinal cord and it led to a considerable change in our approach away from earlier preoccupations with lateral rotation at the shoulder and flexion of the elbow. Cord lesion at level of injury: The intradural injury is effectively a ‘longitudinal’ injury of the spinal cord (Carlstedt 2007a,b). The central nervous system is directly injured in
Surgical Disorders of the Peripheral Nerves
avulsion lesions and this affect is compounded by interruption of flow through the radicular vessels. The evidence may be rather subtle; no more than a hint of increased tone in the ipsilateral lower limb and disturbance of thermal pinprick and light touch sense in the contralateral lower limb. A partial Brown Séquard syndrome was identified in 13 of 249 patients operated between 2000 and 2004, an incidence of 11.8% in those patients who had sustained an intradural injury to at least three spinal nerves (Fig. 9.5). Cord lesion from haematoma and displacement: Flannery and Birch (1990b) described two patients in whom acute compression of the cervical spinal cord was caused by haematoma. Both patients experienced severe causalgia in the lower limbs but perineal sensation was preserved. Case Report: A 19 year old man sustained multiple injuries in a fall and he developed acute renal failure from rhabdomyolysis. The plexus lesion was complete at first and there was weakness and sensory loss in the lower limbs. Bladder function was lost but he retained bowel control and erectile function. When he was seen 6 weeks after the injury there was clonus in both lower limbs accompanied by loss of vibration, temperature and tactile sensibility. The lower limbs were red and dry. There was causalgia, so that he was quite intolerant to physical examination or even the weight of the bed clothes on his legs. Pain was eased by his sitting with his feet and legs in a bucket of cold water. A contrast
Fig. 9.5 MR scan showing deviation of the cord by haematoma and a change in the cord signal in preganglionic injury C5, C6 and C7.
The Closed Supraclavicular Lesion
enhanced CT scan revealed displacement of the cervical cord away from the side of the lesion (Fig. 9.6). There was no injury distally. By 3 months his pain had resolved and there was considerable recovery in the cord lesion with return of bladder control. By 1 year power and sensation in
Fig. 9.6 One of the earliest CT myelograms at St Mary’s Hospital (1981) showing displacement of the cord away from the side of lesion by haematoma.
Fig. 9.7 Wasting of the left upper limb and partial lesion of the cord 10 years after intradural injury of C6 to T1. The level for warm sensation is marked in green, that for pinprick sensation in black.
381
both lower limbs was normal. Recovery of the brachial plexus was incomplete. Late onset of post ischaemic cord lesion: Two cases of post traumatic syrinx within the spinal cord have been seen. These patients reported no symptoms until at least 3 years had passed from the injury. In case 2 of the series published by Nordin and Sinisi (2009) it seemed that the cord lesion was related to deposits of haemosiderin. Case Report: A 27 year old man sustained an isolated injury to his left upper limb in a motor cycle accident in April 1995. He suffered intradural injury of C6 to T1. Two months later grafts were placed between the post ganglionic rupture of C5 and the distal stumps of C5 and C6. Two years later intercostal nerves were transferred to the radial nerve in an attempt to relieve severe pain in the back of his hand. This operation was successful and he returned to work as an engineer and barman. Four years after his accident he described sudden onset of severe shooting pain into his left arm. An MR scan showed irregularity of the spinal cord with deposits of haemosiderin. Ten years after his initial injury he noticed weakness in his left lower limb and decreased sensation, particularly thermal sensation, in his right lower limb. At this time examination revealed generalised weakness and wasting of the left lower limb. Reflexes were brisk bilaterally; there was ankle clonus and his left plantar response was up going. The abdominal reflexes were preserved (Fig. 9.7).
382
The sudden onset of his symptoms and the cyclical course was attributed to an ischaemic cause rather than progressive distortion of the cord by fibrosis. Mechanical causes of late onset cord lesion: Thoracic disc protrusion is rare. In the case reported by Flannery et al. (1990) symptoms did not develop until 8 years after the patient’s initial injury. The patient, by now 25 years in age and in full time employment presented with an incomplete Brown Séquard syndrome, with a sensory level at T4 and a diagnosis of a lesion of the spinal cord secondary to the original avulsion injury was made. This was revised to a protrusion of the intervertebral disc between T2 and T3 by imaging. The cervico dorsal spine was exposed using the transclavicular approach and sequestrated disc material removed. A cortico cancellous graft was inserted between the bodies of the vertebrae. His condition stabilised and there was some improvement in tactile sensation and in temperature sensibility. Spasticity of the left lower limb persisted but the bulk of his quadriceps muscle had improved. By 2 years his gait appeared normal. The case is reminiscent of that described by Chesterman (1964) in which spastic paraplegia caused by a sequestrated thoracic intervertebral disc was successfully treated by the anterolateral approach. The cord may become tethered or even herniated into the defect of the dura. Penfield’s (1949) patient sustained a severe injury to the brachial plexus at the age of 14 years. He presented 37 years later, with increasing weakness and spasticity of both lower limbs. Nine years later his symptoms were so incapacitating that operation was performed. The spinal cord was exposed through a lower cervical laminectomy. It was displaced to the right at the level of the first thoracic segment where it passed in a curve out into the intervertebral foramen and back again. Penfield wrote: ‘when it (the dura) was opened the operator was shocked to find no spinal cord at all’. The cord was decompressed and it was permitted to return to the mid line by division of the scarred root remnants of T1. There was some improvement. Da Silva et al. (2003) described a similar case. In their patient new symptoms developed 19 years after the injury to the brachial plexus in which the lower spinal nerves on the right side were avulsed from the cord. The patient presented with wasting of the small muscles of the left hand. The cord was exposed by laminectomy at T1 and T2 and Da Silva commented that there was: ‘attenuation with outpouching of the spinal cord through a 6 mm diameter dural defect’. The cord was tethered by arachnoid bands to the dural ring and it was freed. There was significant recovery. The first patient described by Nordin and Sinisi (2009) illustrates the late onset and the slow but inexorable progression of defect in a similar situation. Their patient was aged 23 and he had sustained preganglionic injury of C7,
Surgical Disorders of the Peripheral Nerves
C8 and T1with post ganglionic ruptures of C5 and C6. He retrained as an engineer and returned to full time work becoming a considerable athlete practicing running and cycling. Fifty-four months after his injury he developed severe, intermittent, pain in the left upper limb, and loss of temperature sense in the left lower limb. By 7 years weakness in his right lower limb had progressed so that he could no longer run and had difficulty walking. He noted diminution of light touch sensation in the left lower limb. By 13 years he had developed upper motor neurone weakness in the right lower limb. Pin prick and temperature sense were reduced in his left leg. The threshold for cooling was 10°C on the sole of the left foot compared with 1.4°C on the right. Pinprick and temperature sense were diminished below T4 on the left side. Vibration sense was diminished in both lower limbs, with a threshold of 24 units on the right and 18 units on the left (normal < 10). However joint position sense was normal. Magnetic resonance scan showed cavitation of the spinal cord at C7 with displacement to the right side where it was tethered in the foramen of C7 (Fig. 9.8). A CT angiogram showed that the carotid and vertebral arteries were intact, and dynamic flexion/ extension magnetic resonance scan of the cervical spine revealed no instability. The patient declined operation but remains under review. It is very important to search for even subtle signs of disturbance of the spinal cord at the time of injury. The later development of symptoms must be taken seriously and the patient fully investigated to ascertain cause. It is important always to remember the possibility of rupture or thrombosis of the vertebral artery particularly in cases where replantation of the spinal nerves is contemplated (Fig. 9.9). Carlstedt (2007b) describes a patient who sustained a complete intradural injury of the right brachial plexus. There was deep bruising in the posterior triangle of the neck and fractures of the transverse processes of C6, C7 and T1 as well as fractures of the first and second ribs on the side of injury. Operation was performed 2 days after the injury and the vertebral artery was found blue and thrombosed, probably from the intimal tear. After operation the patient was unable to move his right lower limb but this rapidly recovered over the ensuing 24 h. When he became able to walk it was clear that there was disturbance of dorsal column function, with signs of a rotatory nystagmus to the right and a positive Romberg’s sign. Magnetic resonance angiography confirmed a complete occlusion of the right vertebral artery. The patient made a complete recovery apart from the lesion to the brachial plexus. Carlstedt emphasises that magnetic resonance angiography is a valuable investigation in the early days after injury and it should be used in any patient in whom there is the slightest suspicion of damage to the vertebral or other main arteries in the neck.
The Closed Supraclavicular Lesion Fig. 9.8 Wasting of right arm and of muscles in the right lower limb 13 years after preganglionic injury of C7, C8 and T1. Sensory levels marked are (top to bottom) warm, cold, pinprick and light touch. The MR scan (right) shows deviation of the spinal cord (By courtesy Editor Journal of Bone and Joint Surgery [British]).
Fig. 9.9 MR angiogram of vertebral artery thrombosis in complete preganglionic injury with fractures of transverse processes.
383
384
Surgical Disorders of the Peripheral Nerves
9.4 The Evolution of Our Policy of Treatment and the frequent rupture of subclavian or axillary arteries is a of Closed Traction Lesions of the Brachial powerful indication for repair, not only of the artery, but also of the associated nerve lesion. As Magalon et al. (1988) say: Plexus There is a long history of surgical endeavour in repair of the brachial plexus, and there have been many false dawns (Fig. 9.10). Expectations have been raised only to be dashed by events and this no doubt accounts for the alternation between optimism and pessimism which marked earlier writings. Over the last 40 years we have followed a policy of early exploration and repair especially in cases with arterial injury. An accurate diagnosis is essential in determining prognosis; all reasonable attempts should be made to improve that prognosis. This enables the patient to start the difficult, prolonged, and, at times, painful process of rehabilitation. Most of these lesions are caused by great violence and potentially life threatening associated injuries to the head, the chest and the viscera, and damage to the spinal column and the spinal cord are common. These must always take priority over the injury to the nerves. Fractures of the long bones are not a contra indication to urgent exploration of the plexus
Fig. 9.10 Repair of the upper trunk of the brachial plexus using the supraclavicular nerves with plasma clot suture at operation in 1952 by Donal Brooks and George Bonney.
‘emergency nerve surgery is technically easier and the overall results are better; if combined vascular and nerve injuries are involved immediate emergency surgery is mandatory’. Probably the first reported cases of suture of closed traction lesion of the brachial plexus was performed by Thorburn of Manchester in 1896. He described this case in 1900. The patient, a mill girl, was aged 16 years, and 7 months previously had been injured at work. The limb was completely paralysed. Thorburn, by clinical examination, made a diagnosis of a severe injury to the brachial plexus in the posterior triangle of the neck: ‘the plexus has been torn across or hopelessly contused in the region named, but its roots have not been torn away from the spinal cord. On these grounds a plastic operation was regarded as possible’. At operation he excised a massive neuroma of the upper trunk and approximated the ends with fine silk sutures. Some recovery was noted at 7 months. Ultimately his patient regained powerful flexion of the elbow and of the wrist. There is no doubt that other surgeons, including Sherren, had performed repair of the brachial plexus during these years but Thorburn’s paper stands out by the clear analysis of the evidence which led him to accurate diagnosis and by the detailed analysis of recovery of motor and sensory function 4 years after the operation. Foerster (1929) repaired 64 lesions of the brachial plexus during the First World War and Robotti et al. (1995) say this of Foerster: ‘in his treatise of neurology, he outlined a rational and comprehensive treatment for brachial plexus injuries’. Puusepp (1931) and Tavernier(1932) are credited by Alnot (1988) and by Narakas (1988) with the first successful grafts. Scaglietti (1947) reported findings and results in a series of several hundred patients. Davis et al. (1947) described a series of 47 cases commenting that: ‘the outlook for recovery of function was much improved with prompt surgical repair’. Lurje (1948) performed different types of nerve transfers for irreparable lesions. Seddon (1954) described the outcome of five repairs by graft which were performed in Oxford in 1944 and 1945. The work of the early pioneers and of those who followed them is admirably set out by Leffert (1985), by Narakas (1988) and by Robotti et al. (1995). However there was also a widely held pessimistic view about the value of operation because of the extreme technical difficulty of operating in the late case where normal structures are replaced by a mass of scar, which is worse after arterial injury or sepsis, and because of the great difficulty in forming an accurate diagnosis. These difficulties have not gone away: even now we encounter occasional neglected injuries where it is impossible to form an accurate diagnosis in spite of the modern advantages of imaging, intraoperative neurophysiological monitoring and all the experience gained and the lessons taught by those who went before. Few things
The Closed Supraclavicular Lesion
are more disheartening than to come to the field where the normal tissue planes are replaced by scar tissue with the consistency of concrete in which it is impossible to define a rupture or even an avulsed spinal nerve. At least concrete does not bleed.
9.5 The Definition of Pre and Postganglionic Lesion: Prognosis for Recovery During the 1950s a series of investigations by Bonney and his colleagues defined pre and post ganglionic injury and developed physiological methods for distinguishing between the two lesions. The prognosis for recovery after preganglionic injury and through ruptures ‘in continuity’ was established. The proposition was that damage to the proximal limb of the axon of the dorsal root ganglion would not affect the distal axon or its myelin sheath. The first experiments (Bonney1954) studied the axon reflexes evoked by an intradermal injection of histamine and by the exposure of the digits to cold water in a series of patients with injuries to the peripheral nerves, to the brachial plexus, and after cervicodorsal sympathectomy. The investigations were prompted by observations made by Seddon: firstly, the occasional finding of myelinated axons in the peripheral nerves of the affected limb; secondly the occasional operative finding of an apparently undamaged plexus, and thirdly, the occasional demonstration of ganglion cells in biopsy specimens of the plexus (Bonney 1954). It was shown that there were two types of traction lesion of the brachial plexus: in the first type, the site of damage was distal to the intervertebral foramen and there was degeneration of all peripheral axons; in the second, the roots were avulsed from the cord so that the peripheral axons arising from cells in the dorsal root ganglion did not degenerate. The clinical picture of motor and sensory paralysis would certainly be similar in both types, but axon reflexes would be absent in the first type whilst remaining present in the second. These methods of diagnosis were further scrutinised in a study of 19 patient with complete lesions (Bonney 1959) in whom the clinical diagnosis was confirmed or modified by the axon reflex, by exploration of the brachial plexus and by the examination of nerve biopsies. In one of these patient (Case 27) a diagnosis was made of post ganglionic lesion of the 5th and 6th cervical nerves with pre ganglionic lesion of C7, C8 and T1. The response to histamine on the radial side of the forearm was negative, but the cold vaso dilation response in the index and little fingers and the response to histamine on the ulnar side of the forearm were positive. The arm came to amputation and portions of the skin and of the nerves were examined histologically. Nerve fibres could be demonstrated in all the areas of skin showing positive
385
axon responses. They were absent from areas of skin where the axon response was negative: ‘sections from the nerves showed some loss of nerve fibres, but in all numerous intact myelinated fibres were present’. Bonney’s investigation established that there was no recovery in nerves which had been avulsed and very little through the hard lesions in continuity which are ruptures reconnected by fibrous tissue (Figs. 9.11 and 9.12). Recovery through less severe lesions in continuity was often complicated by: ‘cross innervation causing simultaneous contraction in two muscles of opposing actions. The muscles most commonly affected in this way were the triceps and biceps, and the triceps and pectoralis major’. All patients experienced pain and it remained severe in 11 of the 12 patients with no, or only trivial, recovery. The presence of a BernardHorner syndrome was a bad sign but it did not appear in two of the patients until some 2–3 weeks after injury. In the third investigation Bonney and Gilliatt 1958 demonstrated persisting conduction in the afferent fibres of peripheral nerves in cases of pre ganglionic injury. It was shown that the finding of sensory action potentials (SAP’s) from the superficial radial nerve at the wrist implied preganglionic injury of C6, whilst SAP’s measurable from the middle and ring fingers indicated preganglionic injury of C7. Intradural injury of C8 was likely if SAP’s could be recorded from the little finger. This investigation provided further support for the concept of pre and post ganglionic injury and it established the validity of physiological investigations in distinguishing between the two. The measurement of persisting peripheral sensory conduction stimulated a profound and lasting interest in conduction in the central pathways, between the spinal nerve, the spinal cord and the brain. The axon reflex is maintained probably by the smallest myelinated and unmyelinated C fibres and the histamine and cold water response tests have generally been superseded by quantitative sensory testing (QST) and by measurement of cholinergic (sympathetic) sweating. Studies of large fibre function have, of course, been combined with electromyography for more than half a century but evaluation of the central motor pathways has more recently been supplemented by the development of transcranial magnetic evoked potential studies (see Chap. 5). The Dorsal Primary Ramus: The dorsal primary ramus is the first somatic branch of a spinal nerve, and useful evidence is acquired by examining it. Bufalini and Pescatori (1969) showed a reasonable co-relation between denervation of the erector spinae muscles and intradural injury of C5 and of C6, but the overlap of segmental supply of these muscles does cause difficulties in interpreting the precise root injury. Celli and Rovesta (1987) became dissatisfied with the results in 234 patients operated between 1971 and 1982 and introduced electromyographic examination of paravertebral muscles combined with preoperative and
386
Surgical Disorders of the Peripheral Nerves
Fig. 9.11 Untreated preganglionic injury C5 to T1 with rupture of the subclavian artery at the age of 9 years. The appearance of the limb 30 years later.
intraoperative studies of sensory evoked potentials. Selective avulsion of a ventral or a dorsal rootlet was detected in eight of 47 spinal nerves confirming the findings of Bonney (1977), Alnot et al. (1981) and Privat et al. (1982) who, by hemi laminectomy, had recognised the particular vulnerability of the ventral roots.
Somatosensory evoked potentials (SSEP): The first records of potentials from the sensory cortex evoked by stimulation of the median nerve at the wrist were made by Dawson (1947) and by Dawson and Scott (1949) at the National Hospital, Queen Square. Jones (1979) provided the first detailed analysis of peripheral, spinal and cortical
The Closed Supraclavicular Lesion
Fig. 9.12 Elevation of the shoulder and flexion of the elbow 20 years after untreated rupture of C5 and C6.
sensory evoked potentials in patients with traction lesions of the brachial plexus. Twenty-six patients were investigated. The three most prominent potentials with a peak latency of less than 25 ms were measured in conscious patients. These were (1) N9, maximal over the clavicle, a potential which reflects the mixed nerve action potential through the brachial plexus, (2) N13 which is measured over the cervical vertebrae and is thought to arise from the cervical spinal cord or possibly the brain stem, and (3) N20, which is recorded through electrodes on the scalp overlying the parietal cerebral cortex. The accuracy of prediction of the nature of the lesion was confirmed at operation carried out by Antonio Landi in eight patients. Landi et al. (1980) recorded cortical evoked potentials through scalp electrodes evoked by stimulating the spinal nerves in the posterior triangle and compared pre and intra operative somato sensory evoked potentials (SSEP’s) with surgical findings in 15 patients. Sugioka et al. (1982) from Osaka provided a detailed analysis of pre and intra operative neurophysiological evidence compared to the findings at operation and they concluded that: ‘to confirm the extent and degree of the nerve lesion in
387
brachial plexus injuries, an intraoperative SEP and NAP recording is very useful, practical, and also indispensible’. They were right. The technique was further explored by Jones (1987) Sugioka (1984) Sugioka and Nagano (1989) and Hetreed et al. (1992). From 1977 the regular use of somato sensory evoked potentials before operation and of intraoperative recordings from the spinal nerves to the central nervous system became established practise at the Royal National Orthopaedic Hospital through the initiative of Antonio Landi, Stephen Copeland, Christopher Wynn Parry and Stephen Jones but it was always difficult to reproduce regular and reliable records in the ‘noisy’ theatres at St Mary’s Hospital where the senior author continued to use the histamine and cold water tests, with electromyography and conduction studies in the outpatient department. Physiological investigations of the peripheral nerves cannot be interpreted before Wallerian degeneration has occurred and so are unsuitable in the examination of lesions within the first few days of injury. The revival of interest in the repair of closed lesions of the brachial plexus in the 1960s and 1970s is often ascribed to the ‘advent of microsurgery’. It is more likely that the increasing numbers of these injuries from the rise in the use of motor cycles and the establishment of reliable methods to distinguish between the pre- and postganglionic injury enforced further developments. The first repair of a closed traction lesion was done at St Mary’s Hospital in 1962 and 41 repairs were performed by 1975. Fifty-one more operations were performed to establish diagnosis, frequently by exposure of the cervical spinal cord through hemi laminectomy (Jamieson and Bonney 1979; Jamieson and Hughes 1980). Results following repair in complete lesions were poor and for a period of about 4 years in the early 1970s George Bonney generally performed operations to establish diagnosis and prognosis. It is pleasing to acknowledge the inspiration for renewed effort which was provided by Algimantas Narakas. George Bonney became acquainted with Narakas through the initiative of James Scott, at that time house surgeon, and now Editor of the Journal of Bone and Joint Surgery. In the early 1970s James Scott won the Pulvertaft prize and used this to visit Narakas in Lausanne. He saw the pioneering work and described it at one of the clinical conferences held in the Orthopaedic Department at St Mary’s Hospital, at which cases of brachial plexus palsy were being discussed. So began a long and fruitful friendship. Millesi’s first publication appeared in 1968, and that from Narakas and Verdan in 1969. In 1977 Millesi described 56 patients with a 5 year follow up and Narakas reported 107 patients of whom 60 had been followed for more than 2 years. Other surgeons throughout the world made important contributions. In the USA these included amongst others Lusskin et al. (1973); Leffert (1985) who had earlier worked with Seddon, and Kline et al. (1978) who reported repairs of
388
the plexus using the posterior subscapular approach. From Japan came Tsuyama et al. (1968); Tsuyama and Hara (1973) who refined methods of investigation and who perfected techniques of intercostal nerve transfer. Significant developments came from a number of surgeons in France, (Allieu 1977; Alnot et al. 1981; Magalon et al. 1988; Privat et al. 1982; Sedel 1979). The work of Morelli was reported by Raimondi and Morelli (1987) in a paper relating the outcome in 456 operated patients between 1974 and 1986. Petrov (1973) published his long term outcome studies and Santos Palazzi (1971) contributed considerable experience.
9.6 Epidemiology Incidence – Associated Injuries: Goldie and Coates (1992) asked fellows of the British Orthopaedic Association for information about patients with traction lesions in 1987. They received replies to 55% of requests, 430 in all, and uncovered 328 patients admitted with complete or partial lesions. The lesion was complete in 26.2%. There was a wound in 5.1%, the subclavian artery was ruptured in 5.5%, and 40% of these patients had other major injuries. Rosson (1987) studied 102 motor cyclists treated at St Mary’s Hospital between 1966 and 1986. Preganglionic injury was found in 56.6% of the spinal nerves. The patients were aged, on average, 21 years and the dominant limb was injured in 65. Fifty-one (50%) of patients sustained significant or severe injuries to the head, the chest, the viscera or other limbs. At 1 year after injury more than two thirds of these patients remained in significant or severe pain and over one third were still unemployed. In 1988 Rosson made a further contribution by studying what he termed an epidemic among young motor cyclists. One hundred and six consecutive patients operated in the course of 2 years were reviewed. Ninety-one were aged less than 25 years, 43 held only a provisional license and one third of their machines had an engine capacity of 125 cc or less (Rosson 1988). Midha (1997) analysed the incidence of plexus lesions amongst patients with multiple injuries and he found that this was 4.2% amongst the motor cycle injuries but only 0.67% amongst those injured in motor vehicle accidents. Narakas (1993) saw a marked increase in such injuries after the enactment of laws compelling helmet use amongst motor cyclists in Switzerland in the middle of 1981. He had operated on 211 cases before that time and recorded that 29 patients (13.7%) had suffered intradural injury of four or five spinal nerves and that the incidence of arterial rupture was 16%. He compared these findings with those from 108 patients operated after 1982. In these there were no less than 33(31%) with intradural injury to four or five spinal nerves. The subclavian or axillary artery was torn in 36 patients. This
Surgical Disorders of the Peripheral Nerves
apparent paradox may be explained by the survival of some unfortunates who would have perished in earlier years. However, our own findings suggest that the severity of the injury has somewhat diminished over the years. Tables 9.1 and 9.2 set out the cause and summarise the associated injuries in the most recent group of patients studied, that is in the 249 cases operated in the years 2000–2004. Table 9.1 Cause of injury in 249 patients. 2000–2004 (%). Motor cycle, rider or pillion passenger
175
70.3
Car driver or passenger
27
10.8
Cyclist
16
6.4
Pedestrian
6
2.4
Falls
7
2.8
Weights falling onto patient
6
2.4
Contact sports
5
2
Ski, toboggan, mountaineering
4
1.6
Traction from machinery 3 1.2 35 patients were aged 41 years or more, of these, 17 were 51 years or more in age Table 9.2 Associated injuries in 249 patients, 2000–2004. Head and Spine Region Number Head Spinal cord Cervical spine
%
69
27.7
2
0.8
23
9.2
Pelvis
9
3.6
Thoracic spine
9
3.6
Lumbar spine
1
0.4
Fracture and/or Dislocation upper limb Ipsilateral
114
45.8
Both
9
3.6
Opposite
9
3.6
Ipsilateral
42
16.9
Both
10
4.0
8
3.2
Chest wall
73
29.3
Lung
59
23.7
lower limb
Opposite Chest wall and viscera
Heart Abdominal viscera Pelvic viscera
1
0.4
16
6.4
1
0.4
Arterial injury Subclavian
25
10.0
Vertebral
2
0.8
Other
3
1.2
The Closed Supraclavicular Lesion
389
Less than one-half (41.3%) of the 1,245 spinal nerves sustained intradural injury. This is less than the incidence of 56.6% recorded by Rosson (1987) and lower than that recorded in 300 more patients operated between the years 1989–1993 (55.1%). The proportion of cases with intradural injury to C5-T1 in Rosson’s series was 18.7%, it was 17.2% in the cases operated 1989–1993 and it had fallen to 7.2% in the most recent group of patients. Patterns of referral: Over the years many patients have been referred to us urgently and in the most recent group of 249 patients nearly 90% of them were referred by orthopaedic surgeons who contacted us within 7 days of injury in 98 cases and in another 27 during the second week after injury. Forty-one more patients were referred in the 3rd and 4th weeks following injury. Operation had to be deferred beyond the preferred time, that is within 7 days of injury, in 41 patients because of associated injuries so that 52 of the repairs were performed within 7 days, 25 during the 2nd week and a further 31 between 15 and 28 days. Many patients have good cause to be grateful to the acumen and the interest shown by these referring surgeons. The severity of the injury was not recognized in 18 patients (17.4%) and an overoptimistic prognosis for spontaneous recovery explained the delay in 27 more (11%). A first exploratory operation proved to be misleadingly optimistic in six patients (2.4%) and it is only right to point out that we were responsible for some of these errors.
9.7 Diagnosis Adequate clinical examination will lead the clinician to an accurate diagnosis about the extent and the level of injury in most cases. The best time to search for clinical evidence is on the day of injury or as soon after that as possible and this may entail a visit to the referring hospital. The cause of injury: Evidence from the patient or witnesses about the violence of the injury and the axis of application of force to the damaged limb is of the highest importance. A description of the shoulder being violently arrested by an object, stone or tree or kerb or vehicle whilst the body is flying through the air is associated with severe stretching of the structures in the posterior triangle of the neck (Fig. 9.13). However it is important to emphasise that once the diagnosis of an injury to the brachial plexus has been made it should be put to one side whilst the systematic and systemic examination is made of the whole patient. Significant injuries may be missed even in the very best run Accident Departments. In the case of a young cyclist who suffered a massive injury to the right forequarter resuscitation was successfully performed within one University Hospital before he was transferred to a second where the
Fig. 9.13 Bruising and swelling of the left shoulder, neck and upper arm a few hours after a motor cycle injury. The subclavian artery was ruptured, there was a complete preganglionic injury.
subclavian artery was successfully stented. He was then transferred to us deeply sedated and ventilated. He was weaned off supportive ventilation 3 weeks later when it became clear that he had paraplegia from a fracture of T10. Rupture of the ipsilateral hemidiaphragm may be confused with phrenic palsy. The severity of an injury to the lung may not be apparent at first examination and it is all too easy to overlook a fracture or dislocation of the carpus or tarsus. It is important always to seek for evidence of affliction of the cord, and to search for signs of increased tone or undue briskness of reflexes within the lower limbs. An MR scan of the head and of the whole of the spine is advisable when there is the slightest suggestion of an upper motor neurone lesion or other abnormality of the central nervous system and fractures of the spine and pelvis may require further elucidation by fresh radiographs and CT scan. Pain: Severe pain within the paralysed and anaesthetic upper limb indicates a very serious injury to the spinal nerves. The first element is constant, and it is described as crushing
390
Surgical Disorders of the Peripheral Nerves
or burning or as intense pins and needles which are experienced chiefly in the forearm and hand. Two thirds of conscious patients who experienced this pain did so on the day of injury. The superimposed lightning shoots of pain coursing into the dermatome of a spinal nerve signifies preganglionic injury to that nerve. More than one half of conscious patients experienced this pain on the day of injury; in others lightning pain became apparent at intervals ranging from 14 h to more than 4 weeks after injury (Tables 9.3–9.5, Fig. 9.14). We might contrast this with the case of multiple cranial nerve palsies described by Smith (1995). His patient, a motorcyclist, was knocked off his machine, sustaining a head injury, a fracture of left clavicle and a crush fracture of the eighth thoracic vertebral body. The left side of the neck was bruised, swollen and there was a friction burn from the strap of the helmet which had been wrenched from his head. At no time did he experience severe pain. He presented at 2 months with left sided palsies of the spinal accessory (anterior to sternocleidomastoid), hypoglossal, superior laryngeal and cervical sympathetic nerves. The level of lesion was clearly established by the site of the burn: the prediction of a good
Table 9.3 Onset and nature of pain in 198 patients, 2000–2004. Interval (days) Constant Shooting Totals 0–7
147
112
259
8–14
12
16
28
15–28
18
16
34
over 28
32
48
80
Total 209 192 401 15 patients in coma after the accident are excluded. They all reported pain on awakening. Some patients describe new types of pain during recovery.
prognosis, for this was a crush injury, not a traction lesion, was confirmed by later progress to full recovery. Inspection: Inspection of the limb often provides critical evidence about the nature of the injury to the nerves. Linear cuts passing from the face towards the tip of the shoulder show how the limb has been distracted from the torso. This supposition is confirmed by being shown a fractured or battered motor cycle helmet. Deep bruising on the point of the shoulder confirms that the limb was violently arrested whilst the patient was airborne. Deep bruising in the posterior triangle of the neck is a most important sign indicating rupture of the prevertebral muscles and possibly rupture of the subclavian artery. An increasing swelling in the posterior triangle of the neck indicates either collection of spinal fluid from avulsed nerves or an expanding haematoma from ruptured vessels or both. Linear bruising in the arm suggests rupture of a nerve trunk or of a main artery below the clavicle. Examination: The patient who is not unconscious will, of course, be in pain and in distress. Even so it should be possible to find loss or impairment of sensation of the skin above the clavicle indicating injury to the cervical plexus. This is associated with intradural injury of C5 and C6 (Fig. 9.15). Examination of muscles is not easy in this situation but it should be possible to ascertain whether trapezius and serratus anterior are working. Amongst the 72 patients with intradural injury to C5, C6 and C7 operated in the years 2000–2004 serratus anterior was paralysed in 65, 28 had a phrenic palsy and C4 was involved in 23 patients. Whilst a Bernard Horner sign is not an absolute indication of intradural lesion to C8 and T1 (Bonney 1959; Hentz and Narakas 1988) it is a strong indicator. Tinel’s sign: This sign is invaluable in the early detection of ruptures. The method of detection and the reliability of the sign have already been described in Chap. 5. It is important
Table 9.4 The six most common qualities of pain described by 198 patients, 2000–2004. Interval (days) Burning Crushing Electrical Lightning
Bursting
In a Vice
Total
0–7
65
60
47
34
16
10
232
8–14
3
3
4
3
1
2
16
15–28
4
6
7
4
4
1
26
11
8
17
14
4
Over 28
Total 83 77 75 55 25 Many patients describe more than one quality of pain; others use the term “constant” or “shooting”. Table 9.5 Distribution of pain by region and/or by dermatome in 198 patients, 2000–2004. Interval (days) Whole Limb Shoulder/Arm Forearm Hand C5
1
55
14
329
C6
C7
C8
T1
Total
0–7
16
22
32
51
6
33
40
18
7
225
8–14
1
1
1
3
1
4
4
4
3
22
15–28
2
3
3
4
3
8
5
3
31
Over 28
2
4
10
10
2
11
14
8
5
66
68
9
51
66
35
18
344
Total 21 30 46 Most patients experience pain in more than one region or dermatome.
The Closed Supraclavicular Lesion
391
Fig. 9.14 Intense lightning like pain was felt in the dermatomes of C5, C6 and C7 in this patient on the day of injury. These three nerves were avulsed from the spinal cord.
to advise the patient that percussion in the posterior triangle of the neck may be painful, at times extremely so and they should be asked to indicate into which regions they experience radiation of intense pins and needles. When these extend down the outer aspect of the arm and the proximal forearm then rupture of C5 is likely; when they extend to the lateral aspect of the forearm and the thumb then a similar lesion of C6 may be anticipated. Percussion over a ruptured seventh cervical nerve induces sensations into the dorsum of the hand, and over a rupture of the lower trunk, sensations down the inner aspect of the forearm and into the little finger. It may be necessary to percuss over the foramina of a cervical vertebra to elicit the sign. Sensations evoked by percussion which extend to the outer aspect of the shoulder and the upper part of the arm signify a lesion of C4, and not of C5. Absence of the sign in a complete, deep lesion accompanied by pain suggests intradural injury for that spinal nerve. It is difficult to over state the significance of a strong positive Tinel sign in detecting a rupture of the spinal nerve within the posterior triangle; the stronger the sign the more likely are the chances of finding a robust stump. It should be
Fig. 9.15 Sensory loss in a case of preganglionic injury C5, C6, C7 and C8 with involvement of C4. The ipsilateral hemi diaphragm was paralysed.
emphasised that Tinel’s sign is detectable in a conscious patient on the day of injury. The clinical evidence usually permits an accurate diagnosis about the spinal nerves involved and also about the level of lesion (Table 9.6). There are pitfalls. In earlier years, before the regular use of studies of conduction for the central as well as the peripheral limb of the spinal nerve, a few examples of intradural rupture without displacement of the spinal nerve were encountered. In others the finding of a nerve which had clearly been stretched but was apparently in continuity led to the hope of useful recovery. In these, the extent of traction upon the intraspinal component was underestimated and recovery was poor and often complicated by co contraction. Case Report: An 18 year old electronics student, who was right handed, was riding pillion on a motor cycle which collided with another vehicle at 40 mph. He was in a coma for 4 weeks but ultimately made a complete recovery from his
392
Surgical Disorders of the Peripheral Nerves
Table 9.6 Diagnosis of rupture or preganglionic injury in 780 spinal nerves (in 200 patients), 2000–2004. Nerve Rupture Preganglionic Clinical Operation Intra NPI Final Clinical Operation
Intra NPI
Final
C5
134
149
135
133
66
34
43
C6
95
98
86
87
102
91
77
83
C7
27
36
29
35
134
131
113
120
C8
5
9
7
9
111
109
93
97
T1
5
10
8
9
101
93
45
Total 266 302 265 273 514 469 Intra NPI = intraoperative conduction studies. A clinical diagnosis of intradural injury was made in 58 spinal nerves which were shown to be intact or recovering.
head injury. He sustained a bilateral lesion of the brachial plexus which was initially complete on the right and which affected C6, C7 and C8 on the left. His case was thoroughly investigated in the referring hospital with NPI, CT myelography and MR Scan. These suggested that recovery would be complete on the right side and that there would be considerable recovery on the left. We saw him at 3 months because of slow progress and he pointed out that he was aware of pain in the left upper limb as soon as he awoke. There were intense shooting pains into the dermatomes of C6 and C7 and a constant pain in the hand and forearm which he described as ‘a row of nails being whacked into the forearm and hand, with intense and throbbing pins and needles’. The pain was an ‘early warning system’ for cold or illness. The left brachial plexus was explored. C6, C7 and C8 appeared to be intact but there was no SSEP. These were undisplaced preganglionic injuries (Type 5). Stimulation of C5 evoked strong responses in the muscles of the shoulder but also in the biceps. Stimulation of T1 evoked powerful activity in the long flexor and small muscles but also in the extensor muscles of the fingers and thumb. He had returned to his studies by 6 months from injury, and he rose high, coming to a senior responsible position. His pain abated at 12 months from is injury. The seventh cervical nerve never recovered in the right upper limb and the defect of extension of the digits was corrected by simple muscle transfer. On the left side recovery was confined to C5 and T1. He regained adequate extension at the elbow in which the lateral head of triceps was reinnervated, presumably, by C5 and hand function was improved by flexor to extensor transfer.
79
86
396
429
within 48 h of injury not only can the central sensory pathways be studied but also the integrity of the nerves distally. These techniques permit the evaluation of the integrity of the proximal stump in a rupture, they enable some mapping of the bundles within the distal stump and they may detect a second more distal lesion or the loss of conduction from impending or actual critical ischaemia. Conduction studies are useful in gauging the extent of axonotomy in lesions in continuity by studying responses evoked by stimulation of the nerve proximal to the lesion and noting the response distally, and by stimulating the nerve distal to the lesion and recording the central response. There are pitfalls. Proximal conduction block may be found in nerves subjected to the effects of haematoma. Intraoperative recording is sensitive to scarring about the exposed spinal nerve and sensory recording may not detect isolated intradural injury of the ventral roots alone. It is important to remember during operations performed within 2 or 3 days of injury that avulsion of ventral roots may not be recognised because the spinal nerves continue to conduct: our own observation suggest that, after avulsion or a rupture, a motor response to stimulation of the distal stump persists for at least 2–3 days. The potentials recorded through electrodes placed over the scalp or the neck evoked by stimulation of a ruptured spinal nerve do not directly measure the potential of that stump to produce new axons; the method is not quantitative (Fig. 9.16). Chen (1998) found a high failure rate of grafts of ruptured spinal nerves showing an abnormal SSEP (Fig. 9.17). In spite of these limitations we believe that these methods are essential elements during exploration, especially in the urgent case, and they form the basis of a classification of injuries of the spinal nerves which is set out in Table 9.7. That classification also uses evidence acquired by radiological and imaging techniques.
9.7.1 Conduction studies: The Current Situation
9.7.1.1 Radiographs and Imaging
We have found that intraoperative studies of motor and sensory conduction are essential in emergency or urgent exploration of the brachial plexus. If the operation is performed
It is possible to infer a good deal about the state of the roots within the canal from plain radiographs (Fig. 9.18). Roaf (1963) and Yeoman (1968) showed that tilting of the spine
The Closed Supraclavicular Lesion
393
Fig. 9.16 Good traces from spinal nerves at the foramen showing characteristic wave forms. The amplitude of C6 is consistently larger than that from C5: there is a similar difference between C8 and T1. The trace from C7 is characteristically triphasic. A trace from uninjured intercostal nerve (T4) is shown.
away from the side of injury and opening of the intervertebral spaces is associated with very severe traction lesions. Fracture or fracture and dislocation of the first rib often signify a very serious injury to C8 and T1, and one often
associated with arterial injury. Indeed, we have exposed two cases where the subclavian artery was entrapped between the fragments of the rib. A displaced fracture from the transverse process of C6 or C7 may be difficult to detect but it almost
394
Surgical Disorders of the Peripheral Nerves
Fig. 9.17 Examples of abnormal traces in Type 4 lesions of spinal nerves.
certainly indicates that the spinal nerve has been torn out from the spinal canal pulling with it the bony attachments of the radicular ligaments. Myelography: In early days we used oil based media, introducing the needle under general anaesthesia, screening
it through the cervical lesion and withdrawing it before proceeding to operation. The limitations of myelography were revealed by Davis et al. (1966). Nagano et al. (1989) studied 90 patients by myelography using water soluble agents and by exploration. An influential classification of the
The Closed Supraclavicular Lesion
395
Table 9.7 Characteristics of lesions of the spinal nerve. Type of lesion Tinel’s sign Conduction between spinal nerve and spinal cord
Peripheral conduction
Conduction across lesion
CT myelography MRI
Appearance
1. Intact
Absent
Intact
Intact
Not applicable – no lesion
Normal
Normal
2. Recovering stretch
Absent or weak
Intact
Intact or diminished
Intact or diminished
Normal
Bundles intact. Epineurium stretched or even torn
3. Rupture
Strong
Intact
Absent
Absent
Normal
Clear separation of stumps (early cases). Good architecture of proximal stump
4. Rupture with intradural component
Present, but weaker than in type 3
Diminished
Absent
Absent
Usually normal
Clear separation of stumps. The proximal stump abnormal, even close to foramen
5. Intradural with no displacement of DRG
Absent
Absent
Sensory conduction preserved
N/A
Separation of roots may be seen
Normal(early). Atrophy, sometimes gray-yellow color (late)
Present if dorsal root intact
Sensory conduction preserved
N/A
Separation of root(s) may be seen
Normal or mild atrophy
Absent 6. R upture or avulsion of dorsal or ventral root
N/A Clear abnormality DRG Visible, with Absent Sensory 7. Intradural with Absent with CSF Leak the ventral and conduction displacement dorsal roots preserved of DRG The timing of peripheral conduction studies is critical. Motor conduction can be detected for up to 4 days after rupture or intradural injury. Sensory conduction persists for up to 7 to 10 days after rupture and indefinitely after intradural injury.
Fig. 9.18 Complete preganglionic injury. Left: The cervical spine is tilted away from the side of injury, there is widening of the intervertebral spaces and fractures of the transverse processes and the first rib. Right: Rupture of the right hemidiaphragm was initially attributed to phrenic palsy.
396
appearances was proposed. Absence of the filling defects which indicate an intact nerve root sheath was associated with preganglionic injury in 96.5% of nerves examined. Diagnosis of the lesion of C5 and of C6 and of isolated ventral root rupture was difficult. Myelography was extensively used at the Royal National Orthopaedic Hospital where Denis Stoker provided images of quite outstanding quality, often performing the investigation out of hours on the evening before urgent operation. Computerised Tomographic Scan: David Sutton introduced computerised tomographic (CT) scanning with contrast enhancement at St Mary’s Hospital in 1981 and the first of these investigations were presented at the International Meeting held at the hospital during that year. Marshall and De Silva (1986) went on to compare the accuracy of myelography and enhanced CT scan against the operative findings in 16 patients. CT scan with contrast enhancement was a good deal more accurate than standard myelography, especially for C5 and C6. In some of their cases the dorsal and ventral roots could plainly be seen. The myelogram proved to be accurate in 37.5% of the nerves examined; it was 75% with enhanced CT. They pointed out that the slices used at that time, which were 4 mm thick, were too coarse to regularly identify the roots. Carvalho et al. (1997) and Oberle et al. (1998) recorded quite a high correlation between the findings from enhanced CT and those displayed at operation. Tavakkolizadah et al. (2001) thought that one major advantage of this investigation lies in the detection of the isolated root avulsion and the demonstration of a residual stump. Avulsion central to the transitional zone was indicated by a small pit in the cord, confirming the earlier reports from Hayashi et al. (1998) and Hems et al. (1999) but this finding can be detected only when the investigation is done early after the injury (Fig. 9.19).
Fig. 9.19 CT myelography. Left: avulsion of the ventral root of C7 on the left side with a defect in the cord. There is a suggestion of a remnant of the dorsal root. Right: preganglionic injury of left C6 roots. There is a remnant of the ventral root.
Surgical Disorders of the Peripheral Nerves
Magnetic Resonance Imaging: The role of magnetic resonance imaging was studied by Hems et al. (1999). It does not distinguish between ruptures and stretch injuries in the posterior triangle. A number of abnormalities associated with preganglionic injury were described: (1) oedema or lateral displacement of the cord, (2) the absence of roots in the foramen and the replacement of the root shadow by haemorrhage or by scar, (3) meningocoele and (4) denervation of the posterior spinal muscles. Changes in the muscle signal could not be detected within 15 days of injury. Post ganglionic injury was suggested by swelling of the nerves on the T1 weighted image and by an increased signal intensity on the T2 weighted image (Fig. 9.20). Slice thickness is a limiting factor. Ochi et al. (1994) used axial and oblique planes to detect root avulsion at C5 and C6 achieving an accuracy of 73% for C5 and 64% for C6 although Carvalho et al. (1997) found that the investigation was rather less reliable. Nakamura et al. (1997) evaluated magnetic resonance myelography, in which a heavily T2 weighted sequence is combined with fat suppression to produce a myelogram like sequence. A multiple intensity protection algorithm is applied to the data volume, and the image can be rotated to produce oblique views. Nakamura used the classification of Nagano and recorded an accuracy of 98% in the detection of meningocoeles and of 92% in the root avulsions. Hayashi et al. (1998) studied 500 nerves by enhanced MR Imaging with ‘enhanced intradural nerve roots’. There is no doubt that the quality of the most recent MR scans provided by our colleague Dr. Asif Saifuddin (RNOH) are extremely good and it is likely that further refinements will displace CT myelography. However, the inability to assess perfusion of the cervical spinal cord remains a serious problem. Imaging is particularly important in the detection of an occult injury to the spinal cord or to the spinal column and
The Closed Supraclavicular Lesion
397
Fig. 9.20 MR Imaging. Above: muscle denervation signals involve serratus anterior and there is elevation of the left hemi diaphragm in a case of intradural injury C5, C6 and C7. Below: the scan suggests rupture of the dorsal root of C6 without displacement of the dorsal root ganglion.
colleagues in referring hospitals are only too willing to carry out adequate studies before patients are transferred. Early myelography is no longer used. It is unpleasant and potentially hazardous in those who have suffered a head injury. MR scanning reveals bleeding within the spinal canal and
displacement of the spinal cord. The investigation is essential in any patient presenting with signs of cord lesion. CT scan remains the investigation of choice for the investigation of fracture or fracture dislocation and MR or digital subtraction angiography are required in suspected arterial injury.
398
Ultrasonography: It is likely that ultrasonography will prove to be a most valuable investigation in the detection of interruption of bundles within the nerve trunks in the posterior triangle or below the clavicle but the investigation will have to be done as early as possible before tissue planes are obliterated by fibrosis (Chap. 5). Unfortunately we have very little experience with this method. Endoscopy: Our experience with this method is set out in Chap. 7.
9.7.2 The Operation The objects of operation include: (1) the confirmation or modification of a clinical diagnosis and (2) improving the prognosis by as extensive a repair as is reasonably possible. Operation is but the first step in the prolonged and arduous course to rehabilitation. The arguments in favour of an urgent approach in those cases where there are reasonable grounds to suspect rupture of avulsion include: 1. The biological imperative. The sooner that reconnection is made between the neurones in the anterior horn and in the dorsal root ganglion and the peripheral tissues the better (Chap. 3). 2. It is far easier to detect a rupture of a spinal nerve within a few days of injury than it is weeks or months later, when a mass of fibrous tissue mimics a lesion in continuity and may block display of partial or complete avulsion. 3. Ruptured nerves recoil deep to the clavicle or even further and it is usually possible, in urgent operations, to reduce the gap between stumps to 3 or 4 cm. 4. Exploration within 48 h of injury permits mapping of the distal faces using the nerve stimulator. Distal, second, lesions can be detected. 5. Repair of an avulsed root, or intradural repair by replantation of avulsed nerves is generally possible only in the early days after injury. The methods of exposure and of repair have been described in Chap. 7. Some points of technical interest are restated. 1. The transverse supraclavicular incision is preferable to the ‘trap door’ approach. It is easily extensile and heals kindly. 2. The clavicle should only be cut to enable urgent access to the subclavian artery. It may be already fractured. 3. Every effort must be made to find the proximal stumps and it is not always easy to find a short proximal blood stained stump. Rolling the finger across the transverse process and palpation at the mouth of the intervertebral foramen may reveal a soft cord-like structure. The proximal part of C5 may be defined by working anterior to the phrenic nerve.
Surgical Disorders of the Peripheral Nerves
4. It is important always to examine the distal stumps, after pulling them back from their displaced position. Demonstration of the dorsal root ganglion is absolute proof of avulsion but the level of rupture of the ventral root may open the opportunity for direct repair. It is not uncommon to find C6 avulsed whilst the fifth cervical nerve has been sheared from its junction with the trunk. It is not rare to find a similar lesion affecting C8 and T1. 5. It is possible to exaggerate the difficulties of recognising a rupture versus a lesion in continuity and also in determining the level of section of a ruptured stump preparatory to grafting. It is far easier to detect rupture with retraction of the bundles, especially within an intact epineurium in cases exposed within 48 h. It is not difficult to resect nerves back to a recognisable architecture and the level of section is helped by determining the level at which conduction is detectable both centrally and distally, always remembering the possibility of a conduction block from haematoma. Only rarely have we found it necessary to resect more than 5 mm of the proximal or distal stump in fresh cases and even less than this when preparing the tips of the ventral roots. 6. At least three grafts are necessary to satisfy C5 and up to six are required for the other spinal nerves. 7. Haemostasis is tedious but essential. We use small pieces of absorbable sponge and sometimes flood the bed of the graft with fibrin clot glue. Repair of the divided fat pad and omohyoid muscle helps to stabilise the grafts and possibly aids revascularisation.
9.8 Some of the Techniques Used for Repair Nerve Transfer: At least one nerve was found ruptured in more than three quarters of patients and for these grafts remain the mainstay of repair. Narakas (1988) described the evolution of nerve transfers, their indication and results pointing out the discrepancy in fibre counts between the donor and recipient nerves. The smallest of the spinal nerves, C5, contains between 15 and 20,000 myelinated nerve fibres; the robust and versatile spinal accessory nerve which is so often used effectively in nerve transfer contains, at the most, 1,500. Nerve transfers complement grafts, they do not replace them, nor should they be seen as the only alternative in multiple or complete preganglionic lesion for this evades the admittedly difficult question of intradural repair. Addas and Midha (2009) express two important reservations in their excellent review of nerve transfers: first, ‘nerve transfers tend to take the surgeon away from exploring the injury site, the brachial plexus, which carries the potential for surgeons to not even offer an anatomic nerve reconstruction, even in situations when these are perfectly
The Closed Supraclavicular Lesion
appropriate’. Next, this policy may lead to surgeons acquiring less experience with the exposure of the brachial plexus and becoming: ‘increasingly unfamiliar with the detailed anatomy and intra operative electrophysiological assessment of the lesion’. The vascularised ulnar nerve graft (VUNG): The pedicled vascularised ulnar graft was used in the early 1970s, in suitable cases of the brachial plexus in which a hopeless prognosis for C8 and T1 had been established. The operation was tedious and it fell to Jamieson, in 1977, to propose the idea of taking the ulnar nerve from the forearm with the ulnar artery and veins and transferring this as a free vascularised tissue transfer. We were concerned about taking a more major forearm artery and in three cases repaired the ulnar artery with a vein graft, a procedure adding 2 h to the time of operation in those days. Later Jamieson suggested the superior ulnar collateral vessel as a pedicle, a suggestion which fell on stony ground until one of us, in 1981, during the second stage of a pedicle operation cut through the already prepared suture line. The remedy lay to hand, in the superior ulnar collateral vessel proposed by Jamieson. So was born a second type of VUNG. The advantages were considerable. The operation was undoubtedly easier, the nerve trunk unbranched and the vessels usually of adequate size (2 mm) but not essential for the perfusion of the limb. The ulnar nerve could be elevated on this pedicle from its origin to the wrist; preparation of two vascularised strands was quite straightforward and with care three could be so prepared. As we shall see, the promise of the VUNG was only rarely fulfilled (Birch et al 1988), and we now think that the indications for it are narrow, namely for the severely scarred bed with a long gap, or as a means of bringing healthy nerves from the contra-lateral brachial plexus to re-innervate the upper limb or to innervate free muscle transfers. Reinnervation of the avulsed ventral root (VR): Over the years we have gazed helplessly at hundreds of avulsed spinal nerves not realising there was an opportunity here to reinnervate the ventral root. The first operation was done in 1992. Case Report: A 27 year old woman suffered complete lesion of the brachial plexus in a motor cycle accident. At exploration 4 days later we found ruptures of C5 and C6 and avulsions of C7, C8 and T1. C5 and C6 were grafted to the upper trunk: three grafts from C6 were passed to the sensory component of C7 after excision of the dorsal root ganglion. The dorsal scapular nerve was transferred to the VR of C7 and the accessory nerve to the VR of C8. At 20 months she was free of pain. There was limited recovery through the repaired C5 and C6 nerves. The shoulder was stable and the biceps reached MRC Grade 3. Recovery through the repaired ventral roots was more impressive: triceps and latissimus dorsi MRC Grade 4; wrist flexors MRC Grade 3; and long finger flexors MRC Grade 2. There was
399
light touch sensation in the palm of the hand and in the index and middle fingers. It seems that most of this patient’s recovery was due to the transfer to the VR of C7 and C8. The roots of the spinal nerves contain far less connective tissue than the peripheral nerves. Of course, this is one of the reasons why they are so vulnerable to traction. However this feature opens the prospect of reinnervation of elements with a dense concentration of motor axons. It is possible that the specificity of Schwann cells related to the large efferent axons actively encourages the re-entry of new motor axons. Most repairs were performed within a few days of injury for only then can the dorsal root ganglion be displayed with ease the ventral root separated from the spinal nerve and the displaced spinal nerve be approximated to the donor. There are three methods of repair: direct; extra plexal, usually by transfer of the accessory nerve to the ruptured VR; and intraplexal, by transfer of the VR to a ruptured stump or to a bundle within an adjacent intact spinal nerve. Direct Repair of the Ruptured Ventral Roots: On occasion the ventral root is sheared off the spinal nerve where they meet just distal to the dorsal root ganglion. In four such cases the proximal stump of the ventral root was seen using the small joint arthroscope and it was possible to place a graft onto the stump and connect it to the distal stump. The grafts were short and segments of the avulsed dorsal roots were used. Some regeneration was attributed to this repair in all four patients; and in two of them there was useful recovery into the long flexor muscles of the digits and the small muscles of the hand.
9.9 Strategies of Repair The most important distinction lies between those cases where some roots are intact or recovering from those where all roots are damaged. The prognosis for overall function is very different.
9.9.1 The Upper Lesion: Rupture or Avulsion of C5, C6 (C7) with Intact (C7) C8, T1 This is the most favourable lesion because there is useful hand function and it is in this group that all the endeavours of the last 40 years have borne the greatest fruit. It is common: there is usually at least one rupture of the upper nerves (Tables 9.8 and 9.9). Early repair, within a few days of injury, either by graft, by transfer or both can achieve such useful results that it is possible to argue that the problem of treatment in this group is now solved. This is a bold claim and there are of
400
Surgical Disorders of the Peripheral Nerves
Table 9.8 Patterns of injury in 301 consecutive operated supraclavicular lesions, by number of patients 1989–1993. Complete lesions: pre-and postganglionic injury
a
148 cases
Ruptures upper nerves C5 (C6,C7)
83
Intradural lower nerves (C6,C7, C8) T1 Ruptures middle nerves (C6) C7 (C8)
5
Intradural above and below Ruptures lower nerves C8, T1
1
Intradural upper nerves C5, C6, C7 Total intradural C5-T1
52
Incomplete lesions: some roots intact
153 cases
Damage C5, C6 (C7)
117
Recovering or intact (C7) C8, T1 Damage C6, C7, C8
b
23
Recovering or intact C5, T1 Damage C7, C8, T1
13
Recovering or intact C5, C6
Table 9.9 (a) Lesions of C5, C6 and C7 in 108 consecutive repairs. 1993–1994. (b) The nerve injury in the 51 cases of C5, C6 and C5, C6 and C7 lesion. (a) Total of repair of brachial plexus (graft or transfer or both)
108
Repairs of C5 and C6 (graft or transfer or both)
28
`Repairs of C5, C6 and C7 (graft or transfer or both)
23
(b) Intact – recovering
Rupture
Intradural injury
C5
0
38
13
C6
0
29
22
C7
26
16
9
course many failures and disappointments but there are reasons for these which can be avoided and chief amongst them is delay. The aims of surgical treatment include restoration of control of the thoraco-scapular and gleno-humeral joints and of elbow flexion; if need be extension of the wrist and of sensation within the median territory of the hand. We have seen (Chap. 5) that abduction and lateral rotation at the gleno humeral joint is provided by the suprascapular nerve and its muscles and that the deltoid provides power, endurance, and extension. Accessory to suprascapular transfer is particularly valuable in this pattern of injury but distal rupture of the nerve or rupture of the rotator cuff must be excluded. Patterson et al. (1990) studied 44 patients with lesions of the upper part of the brachial plexus in whom the suprascapular nerve was repaired either by transfer to the spinal accessory nerve or by graft. Shoulder function was decidedly better when the transfer was used. Some examples of remarkably good function followed
Fig. 9.21 Examples of repair by graft. (a) A 48 year old man. Right sided lesion. Rupture of C5, C6. Repair delayed to 4 weeks because of severe chest injury. Function at 60 months. This patient was away from work for 6 weeks, and is now running his own company. (b) A 31 year old man. Right sided lesion. Rupture of C5, C6 and C7 repaired at 20 days. Recovery for the graft of C7 was unusually good but that for C5 and C6 was marred by co contraction and subsequent transfer of pectoralis minor proved necessary for elbow flexion.
grafting ruptures of C5 and C6 combined with this transfer. These patients demonstrated a fluent range of movement at the glenohumeral joint without the co contraction which is so common after conventional grafting (Figs. 9.21a, b, and 9.22a–d, and 9.23a,b and 9.24a-c).
9.9.1.1 Preganglionic Injury C5 and C6 There is no doubt that conventional transfers including accessory to suprascapular, ulnar to nerve to biceps and nerve to
The Closed Supraclavicular Lesion
a
401
b
c
d
Fig. 9.22 Examples of ruptures and avulsions of the upper nerves repaired by graft and conventional transfers. (a) A 22 year old man: Left sided lesion. Rupture of C5, preganglionic injury C6. Repair at 4 days: accessory to suprascapular transfer and graft of the upper trunk. The result at 2 years. (b) A 22 year old man: Left sided lesion. Rupture of C5, preganglionic injury C6 and C7. Operation at 10 days from injury: accessory to suprascapular nerve and branch of dorsal scapular nerve to serratus anterior, graft of C5 to upper trunk. Function at 18 months. (c) A 26 year old woman: Right sided lesion. Rupture at C5, preganglionic
injury at C6 and C7. Accessory to suprascapular transfer, branch of dorsal scapular to nerve to serratus anterior, graft of C5 to the lateral cord and intercostal T3 and deep branch of T4 and T5 to circumflex nerve. Function at 26 months. (d) A 20 year old man: Right sided lesion. Rupture at C5, preganglionic injury at C6, C7 and C8. Operation at 4 days from injury: accessory to suprascapular transfer, graft of the upper trunk from C5, transfer of intercostal T3 to nerve to serratus anterior, of intercostal T4 to a nerve to triceps and of intercostal T6 to the nerve to ECRB. Function at 70 months.
402 Fig. 9.23 Examples of preganglionic injury to the upper nerves treated by conventional transfers. (a) A 24 year old man: Right sided lesion: preganglionic injury C5, C6 and C7. The extensor muscles of the digits were innervated by C8. Operation was delayed to 6 weeks because of his chest injury: accessory to suprascapular nerve, ulnar to nerve to biceps and intercostal T3 and T4 to nerve to serratus anterior. Function at 30 months after transfer of flexor carpi ulnaris to extensor carpi radialis brevis. Note the persisting Bernard Horner sign. (b) A 32 year old man: Left sided lesion: preganglionic C5, C6, C7, C8 recovering T1. Operation at 14 days: accessory to suprascapular nerve, ulnar to nerve to biceps, intercostal T3 to nerve to serratus anterior, intercostal T4 to thoracodorsal nerve, intercostal T5 and T6 to nerve to triceps. Function at 1 year. A subsequent flexor to extensor transfer (using FDS) improved hand function. This man returned to work before his operation. He was a long-distance runner and noticed that pain was worsened by dehydration or salt depletion.
Surgical Disorders of the Peripheral Nerves
a
b
triceps to circumflex often achieve a very gratifying outcome. (Leechavengvongs et al. 2006). However it should be remembered that C7 does not always contribute a ramus to the nerve to serratus anterior and loss of the dorsal scapular nerve is not inconsequential. We would now consider transfer of the VR of C5 to the accessory nerve and transfer of the VR of C6 onto one bundle within the intact C7.
9.9.1.2 Preganglionic Injury C5, C6 and C7 This is very serious. There is paralysis of the thoracoscapular, the glenohumeral and the thoracohumeral muscles. The loss of cutaneous sensation is extensive. Variations in the distribution of C8 are particularly important, as it usually innervates the medial head of triceps and also, in about 30% of cases, the extensor muscles of the digits. Reinnervation of serratus anterior and of the adductor and medial rotator muscles of the shoulder is essential. There is much to be said for transfer of the VR
of C5 onto the spinal accessory nerve and of the VR of C7 onto an intact bundle within C8. We found that ulnar to biceps nerve transfer restores elbow flexion in about 70% of patients with this injury compared to the intercostal nerve transfers which have a success rate of about 50%. The intercostal nerves and the medial cutaneous nerve of forearm are available to reinnervate the lateral root of the median nerve. It may be wiser to defer these supplementary transfers until after Wallerian degeneration has occurred in the motor axons so that it is possible to distinguish between those which are conducting to muscle from those which are not. Extension of the digits and of the wrist is reliably restored by later flexor to extensor transfer.
9.9.1.3 Preganglionic C5, C6, C7, C8, Intact T1 These patients have extremely poor function. Serratus anterior is paralysed. There may be some weak activity in the sternal head of pectoralis major and in the medial head of
The Closed Supraclavicular Lesion
triceps. About 10% of patients are able to extend the digits and the wrist through the activity of extensor pollicus longus and extensor indices but the hand collapses when they attempt to grasp an object. The hand is insensate. There are a number of possibilities. Thomas Carlstedt (2007a) has achieved significant improvement by reimplanting the avulsed spinal nerves into the spinal cord in patients with this
403
injury (see below). Alternatively, the VR of C5 is transferred to the spinal accessory nerve, that of C7 to one bundle within T1, whilst the intercostal nerves are available for the lateral cord. The medial cutaneous nerve of forearm is also available for transfer of sensory fibres into the territory of the median nerve. There is no doubt that repair of the ventral root has improved the outlook in these patients.
b
a
Fig. 9.24 Examples of ventral root (VR) repair in lesions of the upper nerves. (a) A 51 year old man: right sided lesion: rupture C5, preganglionic C6, C7. Operation at 6 days: accessory to suprascapular transfer, the VR of C6 transferred to the proximal stump of C5, grafts from proximal C5 to distal C5 and to the dorsal component of C6. C7 was undisplaced (type 5 lesion) Function at 24 months. The power of elbow flexion was MRC Grade 5, that of the infraspinatus MRC Grade 3. Note the persisting partial Bernard Horner syndrome. (b) A 31 year old man:
c
left sided lesion: rupture C5, preganglionic C6, C7. Operation on the day of injury: accessory to suprascapular, the VR of C6 and C7 were transferred to the anterior face of proximal C5, the rupture of that nerve was grafted. Function at 28 months. (c) A 23 year old man: right sided lesion. Rupture C5, preganglionic C6 and C7. Operation at 5 days: accessory to suprascapular, graft of C5, VR of C6 and C7 were transferred to the anterior face of C5. Function at 24 months by which time he was back in full time work in the construction trade.
404
9.9.2 The Lower Lesion: Intact C5, C6 (C7), Rupture or Avulsion C8, T1 The shoulder girdle, the shoulder and the elbow are intact, prono-supination is preserved and there is extension of the wrist with normal sensation for the thumb, middle and index finger. There are opportunities for useful palliation, which is fortunate, for recovery following repair of the lower roots of the plexus, whilst more promising than 30 years ago, remains uncertain and on the whole, poor, except when the repair is performed within days of injury.
9.9.3 The Middle Group Recovery of C5(C6) and T1 (C8), rupture or avulsion (C6) C7 (C8). This pattern is less common in adults than in the birth lesion of the brachial plexus. Intradural injury to C7 induces a generalised loss of power throughout the limb and is often attended by severe pain. Reinnervation of the VR of C7 is particularly valuable. There are other odd patterns with isolated avulsion or rupture of one or more roots, which, broadly speaking fall into one or other of the above categories.
9.9.4 The Complete Lesion This can be divided into those in which some nerves are ruptured in whom modest function can be expected from grafts, and those in which all nerves are avulsed (Fig. 9.25a–d). In these, nerve transfer offers paltry mitigation and it is for these that the only real prospect of useful function lies in replantation. We shall discuss that later. However, prompt repair of ruptures can achieve striking recovery. Dickson and Biant (2010) studied two patients in whom the entire plexus had been repaired, the first patient was followed for 10 years after operation and the second for 15 years. Case Report: A 23 year old right handed nurse sustained multiple injuries to the left upper and lower limbs in a road traffic accident. She was riding pillion on a motor cycle. Mr. Gareth Thomas (Derriford Hospital, Plymouth) recognised the injury to the left brachial plexus and treated the fractures of the left humerus, radius and ulna by open reduction and internal fixation. He found that the left radial nerve was stretched and contused but not ruptured. We saw her 5 days later when she was in intense pain which could not be controlled with high doses of opiates. The pain commenced immediately after the injury and she described a constant burning pain throughout the hand. There were also superimposed ‘lightning like shocks’ of shooting pain from the elbow along the radial side of the forearm to the thumb. These shooting pains occurred
Surgical Disorders of the Peripheral Nerves
approximately 12 times in every hour. There was bruising and linear abrasion in the posterior triangle of the neck. There were strong painful Tinel signs with radiation of intense paraesthesae to the radial aspect of the forearm (C5) to the dorsal and palmar aspect of the hand (C7, C8) and to the medial aspect of the forearm (T1). A myelogram confirmed a clinical diagnosis of ruptures of C5, C7, C8 and T1 with avulsion of C6. Operation was performed on the following day, 6 days after her injury and good stumps were identified at C5, C8 and T1. C6 was avulsed. The quality of the stump at C7 was poor. SSEP’s from C5, C8 and T1 were normal in amplitude and form but that from C7 was abnormal in form and diminished in amplitude. The whole plexus was repaired using 20 grafts, the spinal accessory was transferred to the suprascapular nerve and suprascapular nerves to the distal stump of C7. The condition of the stump of C7, allied to the proven second lesion of the radial nerve in the arm offered scant prospect of recovery of the extensor muscles of forearm and hand. Pain remained severe for 15 months and then began to improve at the same time as she noted recovery of muscle function, shooting pain had disappeared by 18 months after injury, by 96 months she was pain free. Recovery into the muscles about the shoulder was detectable by 8 months, and into the biceps and flexor digitorum superficialis by 13 months. Activity in the thenar muscles was detectable at 36 months. At 96 months after injury she had recovered a full range of shoulder movement with powerful flexion and extension at the elbow and powerful flexion of the wrist and digits. Transfer of flexor carpi ulnaris to extensor carpi radialis brevis, at 98 months, successfully restored wrist extension (Fig. 9.26). The results of quantitative sensory testing (QST) are outlined in Table 9.10. These showed no recovery of sensation in the hand nor any recovery of cholinergic (sympathetic) sweating, at 15 months. By 46 months the thresholds to Von Frey hairs had improved and thermal thresholds were now demonstrable although elevated. By 96 months the von Frey thresholds had improved still further and were recorded as grade 3 in C5, grade 5 in C6, grade 4 in C7 and grade 3 in T1. By now she had accurate localisation to the digit, and there was recovery of sweating. Case Report: A 28 year old man suffered a complete lesion of the brachial plexus on his left (dominant) side in a bicycle accident. In addition he sustained open fractures of the ipsilateral radius and ulna, and of the carpus and metacarpus. A diagnosis of a complete lesion to the brachial plexus was made by Mr. James Robertson (Southampton), who treated the fractures before referring the patient. He was seen at 5 days. He was in severe pain which had commenced instantly on injury, this was constant, burning and crushing and it was felt in the forearm and hand. There was no shooting pain. There was a strong Tinel sign for C6. A clinical diagnosis of post ganglionic lesion was made, probably without pre ganglionic injury because of the nature of his pain. Operation
The Closed Supraclavicular Lesion
a
405
c
b
Fig. 9.25 Examples of repair in cases of one or two ruptures, the other nerves having been avulsed. (a) A 32 year old man: Right sided lesion: rupture C5, C7, preganglionic C6, C8 and T1. Operation at 3 days: accessory to suprascapular, the upper trunk grafted from C5, the middle and lower trunks from C7. Function at 28 months. The repair of the middle trunk restored useful extension at the elbow and weak flexion and extension at the wrist (MRC grade 2). (b) A 32 year old man: Right sided lesion. Rupture of C5, preganglionic C6, C7, C8 and T1. Accessory to suprascapular, the upper and middle trunks were grafted from C5. Function at 30 months. He is in full time employment and an active sportsman. (c) A 34 year old man: Left sided lesion. Rupture of C5, preganglionic injury C6, C7, C8 and T1. Operation at 4 days: accessory to suprascapular, graft from C5 to the upper trunk and transfer of
intercostal T3, T4, T5 to the ulnar nerve. Function at 16 years. Finger flexion returned at 14 years, and it was associated with breathing. He rehabilitated himself, and joined the Queen Elizabeth College in Leatherhead. He returned to work at 24 months and is now in full time employment as a senior manager. (d) A 16 year old girl: Right sided lesion. Rupture C5, preganglionic C6, C7, C8, T1. Operation at 21 days, delayed because of other injuries: graft of C5, VR of C6 transferred to C5, accessory to VR of 7. There was great improvement in her pain by 3 months and the planned intercostal to ulnar transfer was cancelled. Recovery was more impressive through the ventral roots than it was through the graft: adduction, medial rotation, extension of elbow and of wrist with weak flexion of the fingers came through C7 (see over).
406
d
Fig. 9.25 (continued)
Surgical Disorders of the Peripheral Nerves Table 9.10 Quantitative sensory testing at 46 months. Normal Warm threshold Cold threshold Monofilament values °C (<3.9°C) °C (<2.6°C) thresholds (<3) C5
4.4
2.5
4
C6
14.7
3.6
6
C7
8.0
5.5
4
C8
-
-
8
muscle function at 9 months after operation and this coincided with improvement in his pain. By 24 months pain had disappeared. He had a full range of movement at the shoulder and there was recovery of biceps, the wrist flexors and finger flexors. At 5 years there was neither wrist extension nor recovery into the small muscles of the hand and there were significant contractures at the metacarpal and interphalangeal joints of the fingers because of persisting paralysis but also because of fibrosis from the original injury. At 15 years (180 months) he had recovery to MRC grade 5 in the muscles of the rotator cuff, deltoid, biceps, triceps, wrist and finger flexors (Fig. 9.27). Abductor digiti minimi was graded at MRC 2 and the opponens pollicis at MRC grade 3. The findings from QST are outlined in Table 9.11. Dickson and Biant comment that the rate of recovery was faster in the large myelinated efferent fibres than in the Ad, and C fibres. In both patients continuing improvement in power, coordination and cutaneous sensation was detectable for up to 10 years. The outcome was marred, in both patients, by second level injuries, to the radial nerve in the nurse and to the hand in young man. These patients, and there are many others, illustrate what can be achieved by urgent diagnosis and urgent repair. The first surgeons recognised the severity of the injury in both patients and ensured that they were given the best possible chance of some recovery.
9.9.5 Free Functioning Muscle Transfer (FFMT)
Fig. 9.26 Left sided lesion. Rupture C5, C7, C8 T1, avulsion C6. Function at 96 months after repair in a nurse aged 28 at the time of injury. Wrist extension was regained by subsequent transfer of FCU to ECRB (below).
was performed 6 days after the injury and a rupture of all five spinal nerves was confirmed. The condition of the stump at C5 was rather poor, there was a leak of CSF here and the SSEP, although reproducible, was small in amplitude and abnormal in form. The whole plexus was repaired using 20 grafts. The spinal accessory nerve was transferred to the suprascapular nerve. He noted the first signs of recovery into
Akasaha et al. (1991) used free muscle ‘transplantation’ reinnervated by intercostal nerves to regain not only elbow flexion but also wrist extension and this outstanding achievement was extended by Doi et al. (1995) who used two free muscles to restore hand grasp. Hattori et al. (2005) used this method to regain important elements of hand function in three children with complete avulsion. Most of our own cases were operated in conjunction with Professor Roy Sanders and Mr. David Evans (London) (Chap. 13). The technique offers a real prospect for reliably regaining extension of the wrist and digits, essential for hand function and which are so difficult to achieve through nerve repair. It is important to plan ahead and to reserve such potential ‘motor’ nerves as the spinal accessory and intercostal nerves at the first operation.
The Closed Supraclavicular Lesion
407
Fig. 9.27 Right sided lesion. Rupture C5,C6, C7, C8 and T1. Function 180 months after repair of complete lesion in a 28 year old man. Hand function was compromised by severe open fracture/dislocation of the carpus and metacarpus sustained at the original injury.
Table 9.11 Quantitative sensory testing at 180 months. Normal Warm threshold °C Cold threshold °C values (<3.9°C) (<2.6°C)
Monofilaments threshold (<3)
Pinprick sensation
Vibrations threshold (V) (<10 V)
C5
2.9
4.5
9
“Blunt”
Shoulder – 15
C6
5.0
8.6
8
“Tingling”
Elbow – 48
C7
5.0
3.2
10
“Tingling”
Wrist – 42
C8
7.4
2.8
14
“Tingling”
Thumb – 32
T1
10.3
6.3
2
“Tingling”
Middle finger – 35 Little finger – 42
Sweating was found to be 29 and 30 units in the right and left palms respectively. At the fingers he was able to sense movement but not direction.
9.9.6 The Bilateral Lesion We have treated three cases of complete bilateral lesion. Bilateral, but incomplete, lesions have been seen in 14 patients. It is perhaps in the complete bilateral lesion that there is a place for contra lateral transfer, in which the post ganglionic ruptures within the more severely damaged limb are transferred to the less severe damaged limb by interposed grafts. Nerve transfers may offer some palliation for the
worst limb. The effect on the regenerative capacity of the central nervous system from bilateral avulsions is profound. Case Report: A left handed mechanic sustained a bilateral complete lesion of the brachial plexus from a motor cycle accident which was accompanied by a head injury with fractures of the left scapula and first and second ribs. He regained consciousness the next day and became aware of constant burning pain with severe pins and needles in both hands. At about 2 weeks shooting pains developed, which radiated
408
from the elbow into the hands. The pain was symmetrical in distribution and intensity. There was never any clinical evidence of disturbance of the long tracts. Extensive and repeated radiological and neurophysiological investigations were done which suggested a bilateral intradural injury to C7, C8 and T1. There was no radial pulse in the left upper limb and although the general condition of that limb was good the evidence pointed to a second level lesion caused by the fracture of scapula. His pain was terrible. We saw him 5 months after the injury when his medication included 140 mg morphine daily. Cannabis helped his pain. At this time a clinical diagnosis of rupture of C5 and C6 was made on the basis of Tinel signs and activity in serratus anterior. The Tinel signs were weaker on the left. The operation was conducted with Marco Sinisi at the Royal National Orthopaedic Hospital and it lasted for eight and a half hours. The clinical diagnosis was confirmed, of type 3 ruptures of C5 and C6 on the right, and type 4 ruptures on the left. Circulation to the left upper limb was maintained by branches from the thyro cervical trunk. The left upper limb was reinnervated by transfer of the accessory nerve to the nerve to biceps and transfer of intercostals T3, T4 and T5 to the median nerve. The right brachial plexus was repaired using 180 cm of graft which included the medial cutaneous nerve of forearm, superficial radial nerve and lateral cutaneous nerve of forearm from both sides. The left ulnar nerve
Fig. 9.28 A 22 year old left handed mechanic with complete bilateral lesion. The appearance 18 months after operation which was performed after an interval of 5 months. Although his pain was relieved there was no significant recovery of function.
Surgical Disorders of the Peripheral Nerves
was also used. It was not possible to use this as a vascularised graft. The nerve was stripped it of its adventitia to improve revascularisation. The right accessory was transferred to the suprascapular nerve, the right upper trunk was repaired by grafts and then three strands of the ulnar nerve were transferred from the left C5 and C6 by a retropharyngeal path and distributed over C7, C8 and T1. The right intercostal nerves were preserved as possible motors for later free muscle transfers. He experienced very considerable improvement in his pain at 12 months after operation, at about the same time that he noted some activity in the proximal muscles of both upper limbs and the ability to sense stimuli in the arms and hands. Electromyography confirmed nascent reinnervating units in both biceps muscles and more advanced reinnervation in right pectoralis major and teres muscles. There was, at that time, evidence of cutaneous sensibility restricted to the dermatome of C5 on the right side, where he had elevated, but recordable, thresholds to temperature and also pin prick sense. He did not regain useful power in the months that followed. Power in the thoracohumeral muscles on the right side reached a grade of MRC 3, and MRC Grade 2 was recorded for both biceps muscles. There was palpable activity in the triceps and in the flexor muscles in the right upper limb. His pain relief was maintained (Fig. 9.28). One striking feature in this case was the considerable improvement in his pain coincident with reinnervation of
The Closed Supraclavicular Lesion
409
muscles. However, repair in a lesion of this severity so late as 5 months ruled out realistic prospects of recovery of function. Had the operation been performed within 7 days of the injury, and there was no reason it could not have been, then it is likely that he would have regained some elements of useful recovery at least in the right upper limb. As it is reinnervation of muscles secured an improvement in pain which could not have been expected by recourse to such palliative methods as FFMT.
which included a bias towards function at the shoulder and elbow at the expense of the hand, a bias in favour of motor recovery at the expense of sensation, and the lack of any assessment of pain. We incorporated his system into our own and information about more than 600 cases was entered into the expanded document between 1988 and 1998. The record was modified in 1998 to reflect the revised classification of injuries to the spinal nerves and also to expand collection of data about pain, in particular, the onset, distribution, qualities, and the response to treatment.
9.10 Results 9.10.1 Methods
9.10.2 Definitions
Over the years the documents for recording the progress of patients have been modified and extended. Throughout, pain and rehabilitation have been central to these studies. Our late colleague Mark Dunkerton studied 154 cases of closed supraclavicular lesion operated on between 1975 and 1983 and he developed a system of assessment some elements of which are set out in Table 9.12. He allotted scores to different segments of the limb by the ability to perform tasks and then scored the limb as a whole. In 1986 Algimantas Narakas outlined a system or measuring recovery. He was well aware of defects of the system
It may be helpful to define the terms that are used.
Table 9.12 Method of assessment of outcome (Drawn from Dunkerton (1984) and much abbreviated). 1
Patient details
including limb dominance and occupation
2
Accident details
including analysis of force expended on forequarter
3
Clinical findings
including directly related injury distant injury noting endocrine abnormality and Brown-Séquard syndrome
4
Ancillary evidence
including physiological; histamine; SAP; radiological; myelography
5
Findings at operation
Action
6
Progression of recovery
for – sensation, muscle power, temperature sense, by individual spinal nerve
7
Usefulness of limb
shoulder – holding objects against side; reaching above head elbow – supporting objects hand and forearm – writing
8
Overall limb function
graded by such functions as brushing or combing hair; opening a door; taking food to mouth; holding objects.
SCORE 0 – 10. First published in first edition of Surgical Disorders of the Peripheral Nerves 1998.
1. Repair of a neural ‘element’ means repair of a ruptured spinal nerve, an avulsed ventral root or a peripheral nerve. 2. A repair is graded good if at least one useful function is restored which is defined as the ability of the patient to move the limb in one axis of the joint through at least three quarters of normal range and against resistance. The grading is more severe for proximal muscles; an MRC grade 3 in serratus anterior or deltoid is of little use whereas a grade 3 in the extensor or flexor muscles of the digits is so. An MRC Grade 2 in the small muscles of the hand improves balance of the digits. A fair result is accorded to repairs in which there is undoubted regeneration, one in which muscles reach an MRC grade of between two and three. A poor result means that there has been no regeneration. The distinction between fair and good is an important one because of the fact of pain relief associated with some regeneration which does not reach a useful level. 3. One postganglionic rupture may be used to repair a number of elements and grafts may be passed to two or even three spinal nerves distally. These are considered separately so that it is possible for a repair of a ruptured C5 to achieve a good result by grafts passed to C6 but a poor one through those passed to distal C5. 4. All functions that might conceivably be regained by the repair have been studied. For C8 and T1 this includes extension and flexion of the digits, abduction and extension of the thumb, and small muscle function. 5. The recovery of individual functions was attributed to the different repairs by the disposition of the grafts or transfers, matched where possible by the muscle responses evoked by stimulation of the distal stumps, by the tempo of recovery, by the progression of Tinel’s sign, by neurophysiological investigations and, where appropriate, by QST. 6. Recovery of sensation within the hand was graded good if the patient could accurately localise to the digit without over reaction and was able to recognise hot, cold, and pinprick stimuli.
410
Surgical Disorders of the Peripheral Nerves
Table 9.13 Results of intercostal nerve transfers. Recipient Nerve Number of Good (MRC Grade 4) operations Number %
Fair (MRC Grade 3) Number %
Poor (MRC Grade 2 or less) Number %
Nerve to serratus anterior
92
76
83
10
11
6
7
Circumflex
19
6
32
6
32
7
33
Nerve(s) to triceps
50
15
30
25
50
10
20
Nerve to extensor carpi radialis brevis (interposed graft necessary)
23
6
26
8
35
9
39
Thoracodorsal or lateral or medial pectoral
24
10
42
7
29
7
29
Lateral cord/ 236 80 34 51 musculocutaneous Operations designed to relieve pain or to improve cutaneous sensation are excluded. Percentages are rounded up to the nearest whole number.
22
105
41
The record is reasonably complete in just over 1,000 patients in whom repairs were carried out after 1976 who were followed from 3 to 30 years. Contact was lost with 310 patients, and in 287 more the records were incomplete. The outcome from conventional nerve transfers and those following repair of the avulsed ventral root are considered first before describing recovery of function in different groups of patients. This is followed by an analysis of pain and of rehabilitation.
Table 9.14 Transfer of intercostal nerve to medial cord or derivatives: 46 cases. Results Number % Light touch sensibility without over reaction in little and ring fingers.
7
15
11
24
12
26
FDP MRCS Grade 4, intrinsic muscles MRC Grade 2 “Protective” sensation little and ring fingers. FDP and FCU MRC Grade 3 Some sensation, FDP MRC Grade 2
9.10.3 Conventional Nerve Transfers
No recovery 16 Percentages are rounded up to the nearest whole number.
The results of 950 nerve transfers are set out in Tables 9.13– 9.17. Nerve transfers for the relief of pain are considered later in this chapter. We offer some conclusions;
9.10.4 Repair of Avulsed Ventral Roots
1. One of the most valuable and successful of all nerve repairs is transfer of one or two deep divisions of an intercostal nerve to the nerve to serratus anterior. 2. Transfer of the spinal accessory nerve to the suprascapular nerve is especially valuable in lesions confined to C5 and C6 and when used in combination with grafts of those two nerves it is common to see patients regaining a full range of abduction and lateral rotation. 3. Ulnar to nerve to biceps transfer is reliable and this principle can be extended to reinnervate the nerve to extensor carpi radialis brevis using one bundle from either the median or the ulnar nerve and to the repair of the suprascapular nerve or an avulsed ventral root by transfer to a bundle within an adjacent spinal nerve. 4. Just over one half (52%) of patients with lesions of C5, C6 and C7 regained elbow flexion after intercostal transfer. Only one-third (34%) did so in complete lesions and the operation usually failed in patients with a partial Brown Séquard syndrome.
35
Messrs Staton Phillips, Max Horwitz and Sanjay Patel (2009) have kindly provided details of their analysis of 111 repairs by transfer of avulsed ventral roots. The results of the first 70 cases were presented to the Annual meeting of the British Society for Surgery of the Hand in London in 2005. The results are summarised in Table 9.18. The findings are summarised. 1. The repair of 89 ventral roots was followed by recovery of 166 functions, including protraction of the scapula by reinnervation of serratus anterior in cases of avulsion of the upper three nerves. No useful recovery could be seen in 31 of the repairs. 2. Repair of the VR of C7 was followed by the return of powerful adduction and medial rotation in 35 of the 49 roots and this was usually accompanied by about 20° of forward flexion at the shoulder. Useful extension of the elbow was seen in 29 of the 49 repairs of C7 and useful extension of the wrist occurred in 10.
The Closed Supraclavicular Lesion
411
Table 9.15 Holt’s method (1994) measuring function in the suprascapular nerve and used in the analysis of accessory to suprascapular transfer. Grade Active abduction Active lateral rotation MRC grade of power Gain by Narakas points Excellent
>120°
>60°
Supraspinatus 5 Infraspinatus 4 or better
9
Good
> 60°
>30°
Supraspinatus 4 Infraspinatus 3
7 or 8
Fair
>30°
>10°
Supraspinatus 3 Infraspinatus 3
From 4 to 6
Poor <30° <10° Supraspinatus 2+ or less Infraspinatus 2+ or less 3 or less The Grade fair is offered if the patient requires abduction greater than 90° without active lateral rotation or if he or she regains lateral rotation greater than 45° without active abduction. Table 9.16 Accessory to suprascapular transfer. Results 156 Cases operated 1986–1994 Partial lesions Complete lesions 96 nerves 58 nerves
184 Cases operated 1995–2005 Partial lesions 184 nerves
Excellent
16
4
44
Good
21
6
54
Fair
37
15
46
Poor 22 33 40 The results are graded by the system developed by Holt (1994) who analyzed the first 156 cases.
Table 9.17 Other nerve transfers for elbow flexion 128 patients. Ulnar to Nerve to Biceps (106 patients)
Accessory to Musculocutaneous (22 patients)
Power of Elbow Flexion by MRC Grade 4 or better 45
9.11 Recovery of Function by Patients
18
3 or better 43 2 or less 18
4
Table 9.18 Results, by functions restored, in 89 ventral root repairs, 2000–2004. M4 or M3 < M4 M2 M1 or M0 better Shoulder
Elbow
Abduction
0
0
1
0
Lateral rotation
0
0
1
0
Adduction and medial rotation
59
13
4
12
Flexion
6
3
3
4
29
12
11
25
Flexion
2
3
3
41
Extension
8
3
2
25
Flexion
0
4
8
33
Extension
0
1
0
29
Extension Wrist Fingers
3. Repair of the VR of C6 was followed by recovery of flexion at the elbow against gravity in nine patients, and by powerful flexion in six of them. Six of these patients regained some adduction at the shoulder through recovery of the clavicular head of pectoralis major. 4. Only rarely was any recovery into the small muscles of the hand recorded after repair of the VR of C8 and T1. 5. The outcome was not affected by the method of repair. The success of some repairs based on poor quality proximal stumps (Type 4) was surprising. 6. Seven VR were repaired in patients aged more than 40 years. Four of these regained two or more functions. 7. We consider that this is the method of choice in preganglionic injury to C7. These repairs can be supplemented by passing grafts or the supraclavicular nerves to the dorsal component of C7 after excision of the dorsal root ganglion (Fig. 9.29a, b).
Intrinsics 0 1 2 42 C7 accounted for 64 of the M4 or better results. C6 responsible for elbow flexion. Powerful adduction and medial rotation is associated with about 20° of forward flexion at the shoulder.
Dunkerton (1984) studied the results of grafts of 125 spinal nerves in 85 patients who were operated between 1975 and 1983. He excluded repairs using the vascularised ulnar nerve graft. Grafts for ruptures of C5 and of C6 (38 patients) were rather successful, with restoration of shoulder control and elbow flexion in 17 cases, of elbow flexion alone in 18 cases and no recovery at all in three patients. The results for repair of C7 proved particularly disappointing. One third recovered elbow extension, only one patient regained extension of the wrist. Repair of C8 and T1 was generally followed by significant recovery into the flexor muscles of the forearm but only rarely by recovery into the small muscles of the hand. There was no difference in outcome between repairs carried out within 1 month of injury and those performed at longer intervals but exploration within 3 weeks was more reliable in predicting recovery through lesions in continuity. The vascularised ulnar nerve graft: One hundred and twenty-eight spinal nerves were repaired in 65 patients by this method in the years 1975–1984. Elbow flexion was recovered in 51 of the 65 patients, and some hand function was regained in 14 but it proved to be significant in only three cases (Table 9.19).
412
a
b
Fig. 9.29 Examples of ventral root repair. (a) A 22 year old woman. Left sided injury. Ruptures C5, C6, preganglionic injury C7, recovery C8 and T1. Operation at 12 days: accessory to suprascapular, graft of C5, C6, the VR of C7 was transferred to the proximal stump of C6. Function at 40 months. Recovery through the VR of C7 was more impressive than through the graft of C6. She regained adduction, and extension of elbow and wrist. Pectoralis minor transfer was used to supplement her weak biceps. (b) A 21 year old woman. A right sided injury. Rupture C5, C6, preganglionic C7, recovery of C8 and T1. Operation at 5 weeks: accessory to suprascapular, graft of C5 and C6, the VR of C7 was transferred to the anterior face of C5 and grafts were passed to the dorsal component of C7 after excision of the DRG. Function at 36 months.
Case Report: A 17 year old girl was knocked down by a car whilst crossing the road, sustaining ruptured of C5, C6 and C7 with avulsion of C8 and T1. Operation was performed 3 days later and the repair included accessory to suprascapular transfer, and the use of four strands of VUNG from the proximal stumps of the ruptured spinal nerves to the upper and middle trunks. Intercostal nerves T3, T4 and T5 were
Surgical Disorders of the Peripheral Nerves Table 9.19 Vascularised ulnar nerve graft in 65 patients operated between 1975 and 1984 followed for 2 or more years. Number of Number Number of Number of patients of patients strands or spinal nerves vascularized repaired graft (112 in all) (128 in all) 1
22
1
27
2
25
2
30
3
16
3
7
4
2
4
1
transferred to the medial head of the median nerve. At 9 years after operation the patient had regained good shoulder function (deltoid and adductors MRC grade 5), a full range of extension and flexion at the elbow (biceps MRC grade 5; triceps, MRC grade 4) and 100° of prono-supination. The power of extension of the wrist, digits and thumb was MRC Grade 4, and that of wrist flexion was MRC Grade 3. The patient recovered a functional pinch grip between her thumb and index finger from the intercostal transfer and she was able to distinguish hot from cold throughout the hand. She could localise with accuracy to each digit except for the little finger. Her total gain from the repair amounted to about 30 points on the Narakas system, of which about 10 came from nerve transfers (Fig. 9.30a). Two other patients recovered worthwhile function within the hand, one, another adult aged 17 years who was operated at 4 days after injury and the third,a child aged 3 who was operated at 8 weeks from injury. In both three spinal nerves were repaired using three or four strands of VUNG. In nine more patients digital flexor muscles power reached MRC Grade 3+ or better with ‘protective’ sensation in the median territory of the hand (Fig. 9.30b, c). It seemed, on the whole, that the technique only rarely achieved its purpose. A number of patients were presented at the international meeting on the brachial plexus at St Mary’s Hospital in 1982 where Laurent Sedel pointed out that the results were scarcely any better than those following conventional grafting. The operation was abandoned in about 1986 for three reasons. The first was that some patients continued to have significant pain in spite of recovering useful function. Next, we realised that the central assumption upon which the technique was based, that the prognosis after avulsion or even rupture of C8 and T1 was hopeless was incorrect; instead every reasonable effort should be made to reinnervate these two nerves. Thirdly, there was renewal of interest in reanimating the shoulder which was often ‘bypassed’ by the vascularised graft. Results of repair in 360 patients operated between 1990 and 1996: Four hundred and three adult patients with closed traction lesion of the supraclavicular plexus were operated in the 7 years between 1990 and 1996. Seventy-three patients were lost to follow up. The number of ‘elements’ repaired in
The Closed Supraclavicular Lesion
the 360 patients available to study was 898. The results are set out in Table 9.20. The evidence that delay is harmful is clear. Seven patients regained elements of hand function all of whom were operated within 14 days. Age is no barrier: relief of pain and useful recovery was seen in six of eight patients aged more than 60 years who were operated within 2 weeks of injury. Two of these patients presented with intense pain. Kato et al. (2006) studied 148 patients in whom at least one spinal nerve had sustained a preganglionic injury. The a
Fig. 9.30 Examples of recovery after vascularised ulnar nerve graft. (a) Right side lesion in a 17 year old girl: Four strands of VUNG were used. Result at 8 years. (b) Right sided lesion in an 18 year old man. Rupture C5, C7, preganglionic C6, C8 and T1. Operation at 6 months: the upper trunk grafted from C5, C7 grafted to the middle and lower trunks by three strands of VUNG. Function at 24 years after operation. This man undertook his own rehabilitation. He recovered elbow flexion
413
average of the number of nerves so injured was 3.2. The chief purpose of this investigation was to study the response of pain to repair and these findings are related later in this chapter. Kato et al. also reported the results of repairs in 284 spinal nerves in 137 patients. Eighty five of these had been studied prospectively, they underwent operation in the years between 1986 and 1991. A further 63 patients were operated between 1999 and 2001. There was a clear relation between outcome and delay before repair (Table 9.21). There is some b
11 years after operation and extension of the wrist 3 years after that. (c) Right sided lesion in a 23 year old man. Rupture C5, C6, preganglionic C7, C8, T1. Operation at 6 days: accessory to suprascapular, repair of upper and middle trunks by VUNG. Function at 27 years. The power of elbow extension was only MRC grade 2 and he missed this function more than any other. He returned to fulltime work 6 months after his accident.
414
Fig. 9.30 (continued)
overlap, and the results in 15 of the patents studied by Kato are also included in Table 9.22. Results in 228 patients operated 2000–2004: Two hundred and sixty patients operated on mainly by one of us (RB) between 2000 and 2004 were studied to define more closely the effects of delay before repair and also to measure the outcome after the use of different methods of repair. Quantitative sensory testing and measurement of postganglionic sympathetic sweating was used to measure recovery in the small fibres. Eleven patients were lost to follow up, urgent amputation proved necessary in three patients with arterial injury, and in seven more the planned repair was abandoned because extensive post ischaemic fibrosis. Eleven patients, with particularly severe lesions were considered for reimplantation of avulsed spinal nerves. The results in the remaining 228 patients are described. Five hundred and eighty-five elements were repaired, 377 by grafts from 265 ruptures of spinal nerves, 177 by conventional nerve transfer and 89 by ventral root reinnervation. The average number of preganglionic lesions was 2.5. There was no preganglionic injury in 30 patients. The results have been prepared for all repairs and then by excluding the ventral root repairs as most of these were performed within the first 14 days (Table 9.22). Some conclusions have been drawn.
Surgical Disorders of the Peripheral Nerves
1. The outcome was better when repairs were performed within 7 days of injury especially so in the presence of arterial injury. The nerves and the artery were repaired in 10 patients on the day of, or shortly after, injury. In these 22 of 30 nerve repairs were successful. Only 5 from the 36 repairs were considered good in the 13 patients who were operated after 7 days. We have already seen that ischaemia or post-ischaemic fibrosis necessitated amputation or abandoning the proposed repair in ten more patients who were excluded from this analysis. 2. Preganglionic lesions seem to exert a depressing effect upon recovery. Good results were recorded in 63% of repairs in patients without any preganglionic injury compared with 40% of all repairs in patients with one or more intradural lesion. Of the repairs carried out within 84 days of injury, 72% were good in patients with no preganglionic injury compared with 45% in those with at least one such lesion. 3. The decline in outcome with increasing delay is more marked in the grafts (Fig. 9.31). Of those performed within 7 days of the injury, 52.5% achieved a good result but the success rate for all grafts was 35.5%. No regeneration could be detected in just over one quarter of them. The success rate for conventional transfers was 42.3% and 65% in the VR repairs. It should be pointed out that the average number of functions regained through a successful graft was 3.2 compared with 1.9 for the good VR repairs and 1.6 for the good conventional transfers. The admittedly few cases of recovery of useful cutaneous sensation and motor function in the hand followed urgent grafts of ruptured spinal nerves. 4. The anticipation of spontaneous recovery in 2,066 functions through intact or recovering spinal nerves was not met in a little over one quarter (28.5%) of them. The predicted recovery of lateral rotation at the shoulder, and of extension at the elbow, the wrist and of thumb, did not occur in 40% of cases. 5. Lateral rotation at the shoulder, extension of the elbow and extension of the wrist was regained in between 30 and 40% of the grafts designed to achieve these functions compared to 60–70% of the successful transfers. 6. The data is adequate in 135patients who presented with a flail shoulder because of lesions of C5, C6 and C7. The lower nerves were also involved in 72 of these. The mean active ranges of movement at final review was 56° for forward flexion, 52° for abduction, 18° degrees for lateral rotation and 66° for medial rotation. The final Mallet score was, on average, 8.5 and the average gain for the shoulder by the Narakas point system was 4.9. 7. Useful recovery into the small muscles of the hand was noted in eight patients, five from grafts and three from VR repair. Nine patients recovered strong or useful abduction and extension of the thumb, eight of these by graft and one by VR repair.
The Closed Supraclavicular Lesion
415
Table 9.20 Results of repairs, in 360 patients, performed between 1990 and 1996 by interval and by severity of lesion. Timing of repair Functions attributed to the repair
Within 14 days No
%
To 3 months
To 6 months No%
More than 6 months
Total
No
No
No
No
%
%
%
%
Incomplete Lesion 0
8
11.8
10
15.6
8
28.6
20
58.8
46
23.7
1
10
14.7
22
34.3
8
28.8
8
23.5
48
25
2
14
20.6
20
31.2
8
28.8
6
17.6
48
25
3
20
29.4
4
6.2
2
7.2
0
0
26
13.5
4 or more
16
23.5
8
12.5
2
7.2
0
0
26
13.5
Total
68
64
28
34
194
Complete Lesion 0
15
16.9
18
41.9
8
36.4
8
66.7
49
29.5
1
15
16.9
10
23.2
4
18.2
2
16.6
31
18.6
2
19
21.3
12
27.8
8
36.4
2
16.6
41
24.6
3
23
25.8
2
4.6
2
9.1
0
0
27
16.2
4
6
6.7
1
2.3
0
0
0
0
7
4.2
5
4
4.9
0
0
0
0
0
0
4
2.4
6 or more
7
7.8
0
0
0
0
0
0
7
4.2
Total
89
43
22
12
166
Overall totals (%) Functions
14 days
0
23
3 months 14.6
28
6 months 26.2
16
More than 6 months 32
28
60.9
95
25.1
1
25
15.9
32
29.9
12
24
10
22
79
20.5
2
33
24.1
32
29.9
16
32
8
17.6
89
22
3 or more
76
48.6
15
13.6
6
12
0
0
97
25.2
157 107 50 46 This excludes patients with planned, late operations designed to relieve pain. Incomplete lesions: at least one nerve intact or recovering. Regeneration did not restore useful function in 91 patients and in 38 of these there was no detectible regeneration.
Table 9.21 Outcome of repair in 137 patients – by delay before operation (Drawn from Kato et al. 2006). Interval Between Injury and Repair Number of Patients Good Fair Number (%) Number
(%)
Poor Number
(%)
11
18.3
Early: within 1 month
60
34
56.7
15
25
Delayed: 1–3 months
28
11
39.3
11
39.3
6
21.4
Late: up to 6 months
27
7
25.9
13
48.1
7
25.9
11
50.0
8
36.4
Neglected: more than 6 months 22 3 13.6 A “good” result means that the patient regained at least one function Ninety (31.7%) of the nerve repairs were graded “good,” and 140 (41.7%) were graded “fair”
8. Seventy-two (56%) of the 127 musculotendinous transfers were successful. Eighty-three of these were designed to restore extension of the wrist, the fingers and abduction and extension of the thumb in patients with intact or recovery in C8 and T1. Of the twenty-five muscle transfers for elbow flexion, 19 enabled the patient to flex the elbow against gravity and some resistance. In nearly all of these there had been some recovery into the elbow flexor muscles. The muscle transfer effectively improved function by one MRC grade.
9.11.1 Age An increasing number of patients coming up to or even past normal retirement age have been treated and perhaps now is as good a time as any to dispel the myth that nerve injuries will not recover after a certain age. Indeed, the response of the nervous system to age suggests that there may be increasing vulnerability to pain, because of the diminishing threshold to noxious stimuli.
416
Surgical Disorders of the Peripheral Nerves
Table 9.22 Results of repairs in 585 elements in 228 patients operated between 2000 and 2004 – by interval between injury and operation. Average number of Interval in Number of Results of repairs Results (excluding ventral Average number functions regained in days patients root repairs) of elements repaired each patient in each patient Good/Total % Good/Total % 0–7
52
114/175
65.1
86/40
61
3.4
5.4
8–14
25
41/72
57
21/45
46.7
2.9
3.8
15–28
31
48/87
50.1
34/73
46.6
2.8
3.3
29–56
32
25/74
33.8
21/68
30.1
2.3
1.6
57–84
31
31/67
46.2
30/65
46.2
2.2
1.8
85–112
16
13/35
37.1
12/33
36.4
2.2
1.9
113–182
22
8/34
23.5
7/33
21.2
1.5
1.1
More than 182
19
12/41
29.3
11/39
28.2
2.2
1
228 288/585 49.2 222/496 44.8 The average numbers of repairs for each patient was 2.6 The average number of functions regained in each patient was 2.9; the total of functions regained was 658
Fig. 9.31 The exception to the rule! A 26 year old man. Preganglionic C5, rupture C6. Operation 26 months later: accessory to suprascapular, the upper trunk grafted from C6. Function at 33 months.
There is a widely held assumption that nerve injuries in children do better than in the adult. We used to hold to it. The falsity of that assumption is nowhere better seen than in the outcome of the complete closed traction lesion in children. This should come as no surprise because it is clear that the cell bodies in the anterior horn and the dorsal root ganglion are more vulnerable to proximal axonotomy in the immature nervous system. Furthermore the disturbance of growth consequent upon denervation and the deformity provoked by muscular imbalance is particularly severe in the younger child. Recovery of cutaneous sensation is certainly better than in the adult but this may not be true for recovery of muscle function. No child aged less than 15 years complained of the classical pain of avulsion injuries (Fig. 9.32a–c). The results following repair in patients aged over 45 years is set out in Table 9.23.
9.12 Relief of Pain by Repair Between 1986 and 1991 Margaret Taggart (1998a) studied prospectively the cases of 324 patients who were followed for intervals ranging from 30 to 108 months after operation. This work was stimulated by her observation that patients who never had an operation at all seemed to experience more severe pain than those who did have an operation and that where there was recovery of muscular function then relief of pain was more likely. Berman, Taggart and colleagues (1995) presented their findings in 116 of these patients with proven root avulsions. The whole plexus was avulsed in 34% of the patients; there was a mixture of pre and post ganglionic injury in the remainder. In all cases repair by nerve transfer and, where possible, by graft had been performed. All patients experienced pain and this was graded as shown in: Table 9.24. A
The Closed Supraclavicular Lesion
strong concordance between the PNI scale and the linear visual analogue scale of 0–10 was confirmed by Kato et al. (2006). It was severe in 88% of patients at some time during their course. Pain began within 24 h of the injury in 62%: the mean time to onset was 13 days (range 0–180 days); pain was at its most intense at, on average, 6 months from injury. Paroxysmal pain was never experienced in the absence of constant pain, and the intensity of the pain was closely related to the extent of differentiation of the spinal cord. It was exceptionally severe in cases of avulsion of C4–T1. No close relation between the depth of pain and the extent of associated multiple injuries was evident.
417
A measurable decrease in pain was noted at a mean time of about 6 months after operation. In 34% of cases pain remained moderate or severe at 3 years or more from that operation. Case Report: In 1984 a young woman with proven avulsion of C5, C6 and C7 had intercostal nerve transfer to the musculocutaneous and lateral root of the median nerves with the aim of regaining elbow flexion and some protective sensation within the territory of the hand. Accessory to suprascapular transfer was also performed. Nine months after operation she regained elbow flexion and told us that her pain had disappeared at the same time (Fig. 9.33). This experience prompted
a
b
Fig. 9.32 Results in children and in older patients. (a) Left sided lesion in a 3 year old girl. Rupture C5, C6, C7 preganglionic C8 and T1. Operation at 6 weeks: the whole plexus was repaired using four strands of VUNG. Function at 2 years. However she came to see us 18 years later and showed us that the affected upper limb was largely excluded from function because of shortening, colour change and atrophy of the hand. (b) Left sided lesion in a 9 year old girl suffered rupture of C5, C6, C7 with preganglionic C8, T1. The subclavian artery was ruptured
and the suprascapular nerve irreparably damaged. Operation at 2 weeks: graft of the whole of the plexus from C5, C6 and C7. Function 15 years later. She used her hand in all daily activities, she was a skilful puppeteer but there was considerable shortening of the limb. She found recovery in the thenar muscles useful. (c) Right sided lesion in a 70 year old man. Rupture of C5, C6. Operation at 10 days: accessory to suprascapular, graft of C5 and C6. Function at 15 months.
418
Surgical Disorders of the Peripheral Nerves Table 9.24 Peripheral nerve injury unit scale for measuring pain. Grade Severity of pain Equivalent on visual analogue scale (VAS) (0–10) 0
None
0
1 – Mild
Patient is aware of pain but leads a normal life
1–3
2 – Moderate
Work or study sometimes interrupted by severity of pain. Sleep usually possible with medication
4–6
3 – Significant
Pain regularly disturbs or prevents sleep, and interferes with work or study
7–8
9–10 Continuous disturbance of all aspects of daily life and sleep The strong concordance between the PNI and the VAS scales was confirmed by Kato et al. 2006. 4 – Severe
Fig. 9.32 (continued)
us to use intercostal transfer for the treatment of intractable pain even in patients where useful motor recovery could not be expected. The outcome in 19 patients in whom intercostal nerve transfers were performed more than 1 year from injury with the sole purpose of pain relief was published in 1996 (Berman et al. 1996). Seven of these patients had avulsion of five spinal nerves; six had avulsion of four nerves; four had avulsions of three nerves; and two had avulsion of two nerves. In five patients limited repair by graft was possible in addition to the intercostal transfer. T3, T4 and T5 were transferred to the lateral cord or its derivative branches in nine cases, and to the medial cord or its branches in ten cases. The choice of the recipient nerve was made on the basis of the location of the pain in the hand and the extent of recovery from either nerve graft or intact root. The results were startling and unexpected: 16 of the 19 patients reported significant relief of pain at an average of 8 months after intercostal transfer. Before Table 9.23 Effect of age on recovery: 38 patients with complete lesions and partial lesions operated during 1988–2004; aged 45 years or more at the time of injury. Recovery Number of Patients Useful
25
None 13 Useful recovery is defined as restoration of one or more functions.
Fig. 9.33 Left sided lesion in a 28 year old woman: preganglionic C5 and C6. Operation at 8 weeks: accessory to suprascapular transfer, and intercostal T3 and T5 with deep branches of T4 to musculocutaneous nerve. Function at 22 months; she could abduct to 120°. Her pain disappeared with recovery of elbow flexion at 9 months.
The Closed Supraclavicular Lesion
operation 18 patients had severe pain (PNI grade 4) and one had significant pain (PNI 3). In ten patients pain dropped from severe to mild (PNI 1) and in six more it reduced from severe to moderate (PNI 2). The time between the intercostal nerve transfer and pain relief ranged from 3 days to 3 years, the average being 8 months. Case Report: A 42 year old man suffered closed traction lesion of the right brachial plexus with rupture of the subclavian artery and avulsion of C6, C7, C8 and T1. There was functional recovery from C5 into the shoulder and the elbow flexor muscles. This patient had severe pain radiating into the ulnar three fingers of the hand, and additional shooting pain radiating from the shoulder to the hand, lasting for about 2 min and occurring between 30–40 times a day. Sleep was disturbed, he gained no lasting relief from analgesic tablets or from transcutaneous nerve stimulation. Two intercostal nerves were transferred to the ulnar nerve 43 months after injury. By 3 weeks he reported his pain much improved and his sleep undisturbed. Pain had dropped from 8 to 4 on the VAS scale, (PNI 3–1) the shooting pains disappeared within the first week, and at 6 months from operation pain was insignificant. It is not easy to define the mechanisms underlying this response to a late reinnervation, one which cannot restore function, nor is it easy to reconcile these findings with the demonstration by Berman et al. (1998) that pain relief was related to the return of muscle power. Fifteen patients with avulsion injuries were followed for at least 18 months. All had repairs, by graft and by intra or extra plexus transfers. Pain was measured at direct interview using the McGill Pain Questionnaire, a verbal 10 point pain score (Melzack 1975), and by the PNI system. Power was measured by the MRC system; cutaneous sensation was assessed for light touch and prick. Return of C fibre function into the skin was measured by the capsaicin induced axon reflex and sympathetic sudomotor fibres by nicotine induced sweating. Intradermal injection of capsaicin is painful in normal skin, and it is remarkable that these patients submitted themselves to the investigation on at least two occasions! Five of the 18 patients reported significant improvement of pain at 18 months after repair. This pain relief coincided with, or preceded by a few days, the return of muscle power. There was slight relation to return of light touch, and no improvement of C-fibre function was found over the period of study. The correlation between pain relief, measured by the McGill system, and the return of function was highly significant; there was no such relation in those patients with poor or no recovery. Berman comments: ‘this decrease in pain appears to accompany or slightly precede the first sign of recovery of function. As the capsaicininduced flair did not show any signs of recovery at the time of the last follow up, the reduction in pain scores appears to be associated with returning large fibre function rather than A-delta or C-fibre function’. This is an important
419
observation, and suggests a different mechanism for pain relief from that following late intercostal transfer. Case report: A 23 year old woman sustained rupture of C5 and of C6 with avulsion of C7, C8 and T1. Repair was performed 48 h after the accident and, by combining graft with intra and extra plexus transfer using the spinal accessory nerve and intercostal nerves all the major trunks of the upper limb were, at least partially, innervated. She had severe pain, which reached its peak at about 3 months from injury. A marked improvement in her pain was recorded at 14 months from injury; this diminution occurred over the course of a few days. Two weeks later palpable activity was detected in deltoid, biceps and triceps. There was, by this time, no detectable change in cutaneous innervation, and there was no measurable change in the capsaicin-induced flare or the nicotine-induced axon reflex sweating. Kato et al. (2006) studied 148 patients, 85 of them prospectively and 63 retrospectively. The mean number of avulsed spinal nerves was 3.2. Onset of pain was immediate in 59 patients, within 24 h in 21, within 2 weeks in 35 and within 6 weeks in 18. A total of 15 patients were intubated and ventilated for up to 3 weeks after the injury. There were two patterns of pain: first, a constant pain in the insensate hand, usually described as crushing, burning and bursting; and secondly shooting pain felt always in the dermatome of the injured spinal nerve but only occurring when there was a constant feature to the pain. The most important mitigating factor was distraction, either by work, study, hobbies or conversation. Other helpful factors included warm weather and warm water. Pain was exacerbated by cold weather, cold water, stress and concurrent illness. The consumption of alcohol sometimes helped and sometimes increased the pain. There were 108 patients who took analgesic medications regularly for more than 6 months. These provided good relief in 16 patients, some relief in 55 but were of no benefit in 37. Cannabis was helpful in 30 patients and a transcutaneous nerve stimulator helped 22 out of 88 patients. Those who responded to medication were more likely to find TNS effective. Kato et al. showed that the earlier the operation is done the more likely it is that the patient will experience relief of his or her pain (Table 9.25). However, as we have already seen, recovery after repair was better in the early repairs and Kato et al. were unable to distinguish between the relief of pain brought on by recovery from that associated directly with the timing of operation. One important observation is made: ‘a few of our patients showed a dramatic improvement in their neuropathic pain immediately after operation; others showed relief from pain shortly after their return to work, suggesting that psychological factors are important. We found that the mean VAS-final score patients who returned to work (78 of 131; 59.5%) was lower than for those who did not’.
420
Surgical Disorders of the Peripheral Nerves
Table 9.25 Improvement in pain against delay before repair 148 patients studied by Kato et al. (2006). Interval Between Injury Number Worst Pain by Visual Analogue Final Pain by Visual Analogue and Repair Scale Scale Mean Median Mean Median
Mean of Improvement in PNI Scale (ranging from 0 to 4 maximum)
Less than 1 month Group 1
61
8.2 (SD0.3)
9.0 (3–10)
2.6 (SD0.3)
2.0 (0–10)
2.2 (SD 0.1)
From 1 to 3 months Group 2
29
9.1 (SD 0.2)
9.0 (3–10)
3.7 (SD0.4)
3.0 (0–10)
1.8 (SD 0.2)
From 3 to 6 months Group 3
32
8.5 (SD 0.3)
9.0 (3–10)
4.0 (SD0.5)
4.0 (0–10)
1.3 (SD 0.2)
After 6 months Group 4 26 9.0 (SD 0.3) 9.0 (3–10) 5.3 (SD 0.6) 6.0 (0–10) 1.1 (SD 0.2) The changes were statistically significant taking p = 0.05: group 1 and group 3 p < 0.01; group 1 and group 4 p < 0.01; groups 2 and group 3 p < 0.05 (Drawn from Kato et al. 2006).
Table 9.26 Initial and final pain values in 179 patients – operated between 2000 and 2004. Pain (PNI Scale) Initial Final 0
5
54
1
10
56
2
36
42
3
75
21
4
53
6
The findings from the patients operated between 2000 and 2004 support these observations although the relationship between relief of pain and the timing of operation is not so clear cut (Table 9.26). Sixty-six patients said that they had a major improvement in pain at about the time of muscular recovery; 25 noted a fair response at this time. Twenty-nine patients did not think that the operation had made any difference. Two experienced a worsening of their pain with recovery. They experienced reinnervation hypersensitivity and ‘over reaction’. Htut et al. (2006) studied pain phenomena and sensory recovery in considerable detail in 76 patients who had sustained at least two avulsions of spinal nerve roots. No repair was possible in eight of these patients. These suffered the worst pain. The relief of pain was rather better in cases where the lesion was repaired by graft and by other nerve transfers than it was in patients where spinal nerves were reimplanted into the spinal cord, but it is important to note that the severity of the injury was worse in the latter. The mean number of avulsions in the 54 patients who were repaired by graft and nerve transfer was three, it was 4.6 in those patients who underwent reimplantation (Table 9.27). There is strong evidence which shows that repair of the plexus offers the patients a good chance of easing their pain. There is a definite suggestion that the earlier the operation is done the higher the chance of pain relief. This fact, that reinnervation eases pain even when it fails to restore worthwhile function, provides the strongest indication for operation and repair by one means or another even in the most severe case.
9.12.1 Return to Work Taggart’s (1998b) figures Table 9.28 indicate that the rate of return to work after serious injury to the brachial plexus is encouragingly high but the finding that four out of five patients return to a different job indicates the importance of retraining and of systems of information about employment. More recent findings from patients operated between 2000 and 2004 appear, at first sight, similarly encouraging but there has been a significant falling away of the rate of return to work since that time. We attribute this to the collapse of rehabilitation services within the National Health Service (NHS) which has occurred over the last few years with the outstanding exception of certain specialist units (Fig. 9.34). In 1969 Dr Derek Brewerton introduced at the Royal National Orthopaedic Hospital the concept of the practice of the Hospital Employment Advisor, later known as Disablement Resettlement Officer. These positions have fallen under the sword of ‘cost benefit analysis’ and of ‘cost saving’; this serious mistake exemplifies the detachment of the NHS from the process of rehabilitation and from the principle of continuity of care. Matters are not helped by the ignorance of the principles of rehabilitation that are shown by too many surgeons in training and by those responsible for their training. The breakdown of team working between clinicians, therapists, orthotists and others is quite deplorable. Private agencies have taken over much of this responsibility. Whilst their work is often very good most of their referrals come from solicitors or insurance companies, months or even years after injury. This is far too late. Matters are not helped, either, by the attitudes expressed by some ‘expert’ medical witnesses. We have been shown reports prepared by such witnesses by some of our patients, which state: ‘this person will never work again’. The effect upon the patient, and upon the confidence and trust between those patients and those treating them is severe (Table 9.29). Kato (Kato et al. 2006) found that the level of pain was higher in those who did not return to work or study. This is a pity because return to work or to study or to the distractions of normal life, does more to ease pain than any other form of treatment.
The Closed Supraclavicular Lesion
421
Table 9.27 Response of pain to treatment (S.D). Treatment Number of patients
VAS (P)
VAS (L)
VAS (H)
McGill (S) pain score
McGill (T) pain score
Graft and transfer
54
2.7 (0.3)
1.7 (0.2)
6.3 (0.4)
17.9 (1.0)
29.4 (1.8)
Reimplantation and transfer
14
4.0 (0.6)
1.1 (0.4)
6.5 (0.4)
20.2 (1.6)
32.4 (3.2)
4.8(0.6)
1.8 (0.5)
8.9 (0.7)
22.0 (3.6)
37.8 (6.2)
No surgical repair
8
P values (Kruskal-Wallis test) 0.01 0.3 0.02 0.5 VAS visual analogue scale: P present, L lowest, H highest, S sensory, T total, McGill (S) sensory, McGill (T) total. Drawn from Htut et al. 2006.
0.4
Table 9.28 Return to work, retraining, further study, with interval before re entry minimum follow up of 36 months. Years of Study, 1986–1993 – 324 Patients Years of Study, 2000–2004 – 238 Patients 54
86
195
63
Formal retraining or return to study
81
65
Did not return to work
75
24
Same occupation (often with modification) Different occupation
9.13 Reimplantation of Avulsed Spinal Nerves 9.13.1 The First Clinical Case – George Bonney, 1977 (from the first edition of this work) The complete supraclavicular traction lesion of the brachial plexus is perhaps the most shattering of all injuries of peripheral nerve. Its high incidence is in young persons, in young men in particular; many of those injured are not educationally prepared for sedentary work; the results of treatment, though better than formerly, are still poor; many patients are subject to constant pain from early days after injury, and many of these continue for years to experience that pain. The principal cause of the poor results and of the pain is certainly the high incidence of avulsion of roots from the spinal cord. I use the term avulsion because that is the description in general use, but I shall later show that in certain cases at least, the lesion is an actual rupture of roots near the spinal cord. In this respect – that of the proximity of the lesion to the spinal cord – the supraclavicular lesion of the plexus may be regarded as one not only of the peripheral but also of the central nervous system. From that extension of the lesion flow many of the difficulties surrounding its treatment. Avulsion is more common in the lower than the upper roots. It usually affects both posterior and anterior roots of the nerve or nerves affected. The extent of movement of the avulsed roots depends on the degree of force to which the
Fig. 9.34 ‘Distraction or destruction!’ The patient’s own motto. Right sided lesion in a 21 year old man. Preganglionic injury C5 to T1 with rupture of subclavian artery. Operation at 4 days. There was one branch from C4 passing to the suprascapular nerve and another passing to the nerve to serratus anterior. The subclavian artery was repaired. No nerve repair was performed. He returned to work at 3 months rising to a senior position working with the disabled. Whilst he found the flail arm splint useful he discarded it because it made him feel disabled. Free functioning muscle transfer (latissimus dorsi) with Professor Roy Sanders was innervated by the accessory nerve.
Table 9.29 Time off work or study (in months). 201 Patients (1986–1993)
177 Patients (2000–2004)
Less than 3
19
23
3–6
44
24
7–12
60
36
13–18
32
20
19 or more
46
44
201
177
limb is subjected and of the amount of separation of the forequarter from the axial skeleton. Usually, the roots are dragged from the spinal cord, through the intervertebral foramen and into the supraclavicular region. In some they may be dragged as far as the space deep to clavicle; in others, the separation is slight enough to permit the roots to remain in the intervertebral canal. At the time of injury, the dura-arachnoid of the
422
root sleeve is torn, so that there is an outflow of spinal fluid. The amount of outflow depends on the extent of tear of the dura-arachnoid: it may be so slight as to form no more than a small collection in the intervertebral foramen. As time goes by, the collection of fluid becomes encapsulated to form a pseudo-meningocele. When they are widely displaced, the avulsed roots soon become fixed in the position of displacement by longitudinal contracture of the nerves themselves and by surrounding organisation of exudate and blood. The actual level of the neural lesion is certainly in some cases and in the case of the dorsal roots, just distal to the surface of the spinal cord. In the case of the anterior roots, it has been reported that anterior horn cells have been avulsed with the roots, although limited observations seem to show that this is an exceptional circumstance. I dare say that the usual site of rupture of anterior roots is just distal to the surface of the spinal cord. I do not doubt that more study of the spinal roots is required to determine the level of separation. If indeed there are two categories of ‘avulsion’ – separation just outside the surface of the cord and true avulsion from the substance of the cord – important questions are raised about the different degrees of damage to the cord. In the first type of lesion, the damage to the cord would be slight; but in the latter case, extensive damage would be done to all the laminae in the posterior horn. It could be, that the cases in which the separation has occurred outside the cord are those in which pain is not experienced. ‘De-afferentation’ alone may not be enough to explain the whole of the pain in these cases. Up to the present, high traction lesions of the brachial plexus have been treated where possible by graft repair of nerves ruptured outside the spinal canal, and by ‘neurotisation’ according to art and to availability of donor nerves, of the distal segments of avulsed roots. The results have certainly been an improvement on those of former methods of treatment by waiting for natural recovery, by peripheral ‘reconstruction’ and by amputation arthrodesis. In particular, relief of pain following re-innervation along a ‘neurotised’ nerve has been seen too often for this phenomenon to be discounted as incidental. The goal of many of those working in this field has for long been the re-implantation of avulsed roots. The idea is superficially attractive: in the case of many fibres of the posterior root in particular, the nerve cell is not far from its termination in the posterior horn either at the level of entry or at a few levels up or down the cord. In a simplified mode of thought, the liability to confusion between afferent and efferent axons is diminished by the separation into anterior and posterior roots. When the matter is viewed more deeply, many difficulties arise: practical difficulties; technical difficulties; difficulties concerned with the microstructure of the elements concerned. First, re-implantation is in most cases possible only within
Surgical Disorders of the Peripheral Nerves
the first 48 h after injury: after that time the avulsed roots are so much fixed in the position of displacement that they cannot be drawn back through the intervertebral foramina into contact with the spinal cord. Next, such early intervention is rarely possible: lesions of this severity are often associated with other life threatening, injuries which require treatment before neural repair can be contemplated. Thirdly, surgeons capable of re-implanting an avulsed brachial plexus are so widely scattered as to make it unlikely that a suitable patient and suitable surgeon will meet at the rendezvous. The next consideration concerns the spinal cord itself. In these cases, the cord has been damaged by the avulsion of one to five roots. The risks of interfering however marginally on this damaged cord are clear. In the case of the anterior roots, re-implantation through a posterior approach is possible only if the cord is rotated. Certainly, the cord can with care and for short periods be rotated by means of small stays inserted into the denticulate ligaments, but any prolonged rotation or any clumsy rotation will surely damage – at least – the ipsilateral long tracts. The posterior roots are near the cord represented as multiple rootlets posing particular problems in repair. The stouter usually single anterior roots are, for reasons already advanced, difficult of access by the posterior approach. Although the roots have a collagen content, they are, in comparison with the peripheral nerve, intensely fragile and susceptible to surgical trauma. As to the small scale: we are here proceeding in the transitional region between the central and the peripheral nervous systems, where the ‘matrix’ changes from one of oligodentrocytes and astrocytes to one of Schwann cells and collagen. Indeed it may be that in the case to which I shall later refer, the actual separation took place in that transitional zone. Axons are indeed continuous through this zone, though extensive arrangements take place near it; it may be that certain axons have smaller diameters in the central nervous system than they have in the periphery. There is, I think, a separation of the blood supplies of the two systems; Carlstedt and colleagues tell us that ‘no blood vessels are seen in the mantle zone’ of the transitional region. The separation of function implied by the division into anterior and posterior roots is not of course complete. There are many unmyelinated and small myelinated afferent fibres in the anterior root: there are in the anterior root of the first thoracic nerve the pre-ganglionic fibres of the autonomic system. The variation of destination of the pre-ganglionic fibres of the posterior root is extreme: many of the large myelinated fibres terminate in the first to the fourth laminae of the posterior horn; proprioceptive primary afferents terminate in the two deeper laminae. Great possibilities for confusion are present. I describe my single experience of the ‘re-implantation’ of cervical nerve roots to show that in certain cases this is
The Closed Supraclavicular Lesion
technically possible. Since that time, experimental work has been done to show that the theory of this ‘re-implantation’ was unsound; on the other hand, other experimental work has shown the validity of such re-connection in the case of the anterior roots; that end to end anastomsis or re-implantation is not essential. In 1977 my colleague Michael Laurence, then orthopaedic surgeon to Guy’s Hospital called me to see a young man who had been admitted to Guy’s with all the signs of a complete avulsion of the right brachial plexus resulting from a motor cycle accident. There were no serious associated injuries; the patient’s condition was very good. This patient was instantly transferred to St Mary’s Hospital, Paddington, where I was then orthopaedic surgeon. I confirmed Michael Laurence’s diagnosis, and made arrangements for operation. Operation, with anaesthetic given by Dr Clive Roberts, lasted 12 h. It was done by Angus Jamieson and me. Operation was started about 24 h after injury. With the patient in the left lateral position, I exposed the brachial plexus above the clavicle: the plexus had, as was expected, been completely avulsed. I marked the seventh and eighth cervical nerves and their roots. Now I exposed the posterior elements of the cervical spine through a posterior incision. and went on to expose the right side of the cervical dura by hemi-laminectomy. I opened the dura to expose the right side of the spinal cord and to show the avulsion of the roots. Now I passed fine forceps through the foramina between the sixth and seventh and seventh and eighth vertebrae and with these caught hold of sutures tied to the cuffs of dura at the junction of roots with peripheral nerves. With these sutures, the roots were drawn back into the spinal canal. Forceful traction was not needed; the roots lay easily in the canal offering themselves for re-attachment. I decided, because of the dangers already mentioned, to re-attach the posterior roots and to confine the re-attachment to the seventh and eighth nerves. With some difficulty, I repaired the dural defects at the levels of avulsion in order to lessen later traction on the repaired roots. Now, using the microscope, Angus Jamieson inspected the surface of the cord. It was clear that the rupture had taken place just distal to the surface of the cord: little stumps of the rootlets were visible. This observation may be of particular importance. With the microscope and using 11/0 sutures, Angus was able to re-attach the rootlets of the two nerves (Fig. 9.35). The dura was closed; both wounds were closed in the usual way. I did not use any external splintage after operation. Although the patient recovered easily and well from the intervention, there was no evidence of regeneration during the period of observation, which was unusually short; only 18 months. It is perhaps interesting that there was no pain during the period of observation. Evidently, re-implantation, or perhaps re-attachment, of avulsed (or perhaps ruptured) roots is possible. It is however doubtful whether any regeneration is possible along
423
Fig. 9.35 Roots of C7 and C8 drawn back into the canal and reimplanted (George Bonney 1977).
posterior, afferent roots. Experimental evidence suggests that regeneration is possible along anterior, largely efferent roots; the experience of this single case suggests that such reattachment is possible and that the hazard is not prohibitive. If such re-attachment were to be attempted, it might be preferable to approach the cord form the front, through the vertebral bodies and intervertebral discs. Not only is the cord accessible by this approach, but the line of avulsion or rupture is presented to the operator’s vision. Further, such an approach would allow the whole operation to be done through a single incision and approach, with the patient in the semisedentary position. Further work is, as I have suggested, necessary in order to attempt to confirm the level of separation or avulsion of the roots in human being after injury. The best evidence comes from inspection of the cord within 48 h of injury, but opportunities for that are rare and may be getting rarer. It may be, that more attention should be given to the proximal ends of avulsed roots found at supraclavicular exploration, in an attempt to determine whether the separation is in fact at the transitional zone between central and peripheral nervous tissue. I have many people to thank for giving me this opportunity; only myself to blame for the failure to take full advantage of it. Michael Laurence was sharp enough to recognise the opportunity; Clive Roberts saw the patient safely through a prolonged operation; operation theatre staff provided the means and Angus Jamieson did the difficult part of the procedure, adding at the same time a valuable insight into the level of separation. Sadly, the patient did not benefit, but he did not suffer damage.
9.13.2 Subsequent Work by Thomas Carlstedt This account is drawn from the first edition of this work. The patient was seen 14 years later. He was working full time in
424
Surgical Disorders of the Peripheral Nerves
senior management. He had no pain, and he related that his pain had abated quite rapidly at about 15 months after the operation, when he had become aware of new activity in the muscles of the shoulder girdle. On examination, it became clear that there had been considerable recovery into serratus anterior, pectoralis major, deltoid and some recovery into biceps. We have already outlined (Chap. 5) the extensive experimental work of Thomas Carlstedt and his colleagues and the techniques of exposure and reconnection between the spinal cord and the avulsed spinal nerves have been described in Chap. 7. Forty-five operations have been performed since Carlstedt joined the Peripheral Nerve Injury Unit in 1995. The indications, contra-indications and results have been described (Carlstedt 2007b, 2007c; Htut et al. 2006; 2007). The method may be considered in cases of complete lesion with four or five avulsions. Operation must not be undertaken in patients with rupture of the subclavian or vertebral arteries, nor in any patient showing any sign of affliction of the spinal cord. Transforaminal endoscopic examination of the cord is helpful: methods of measuring the perfusion of the cervical spinal cord are required. The possibility of risk to the damaged cord is illustrated by one case of partial Brown Séquard lesion which developed after operation in a man with rupture of the vertebral artery. He recovered. In another otherwise healthy man with avulsion of all five spinal nerves the operation was complicated by a partial anterior cord syndrome which also recovered. Regeneration into proximal muscle, notably pectoralis major, has been confirmed in all patients. Useful functions such as adduction, medial rotation and forward flexion at the shoulder and flexion and extension at the elbow have been regained in 21 patients. Some patients have done rather better than this. Case Report: A 23 year old man sustained a complete lesion to the left brachial plexus in a motor cycle accident. He experienced severe, constant and shooting pain on the same day. The diagnosis was made by Mr. Peter Magnussen (Guildford) and confirmed by one of us (RB) at operation 3
days later when C5, C6 and C7 were found avulsed. Although C8 and T1 were not displaced no SSEP could be recorded. The avulsed nerves were drawn back towards their foramina and the accessory nerve was transferred to the suprascapular nerve. Ten days later the nerves were re-explored by Thomas Carlstedt and Rolfe Birch using the posterior subscapular approach. The spinal cord was exposed by hemi laminectomy, no residual stumps of the roots of C5, C6 and C7 could be seen but there were some strands of tissue passing from the origin of the roots of C8 and T1 towards their respective foramen. Three grafts were implanted through slits in the pia mater and passed to the ventral roots of the avulsed spinal nerves. The dura was closed with a vein patch and recovery was uneventful. The first signs of recovery were noted in the proximal muscles 12 months later and these coincided with improvement in his pain. At 11 years after operation he was in full time work and free of pain. He demonstrated powerful and fluent abduction and adduction and medial and lateral rotation at the shoulder. Flexion and extension of the elbow was graded MRC grade 4. Percussion of the biceps tendon evoked a delayed biceps jerk. He also recovered flexion at the wrist and the fingers to MRC grade 3 (Fig. 9.36). Sensory recovery was confined to C5 but the fluency of movement, the recovery of joint position sense at the shoulder and the elbow with apparent recovery of the biceps tendon reflex raised the possibility of some regeneration into the deep afferent pathway. This possibility has been examined further in the following case (Carlstedt et al. 2009) Case Report: A 9 year old boy sustained complete avulsion of his right brachial plexus and he experienced severe, spontaneous, constant and shooting pain. All five spinal nerves were reconnected by interposed grafts at operation which was performed 4 weeks later. Recovery of muscles at the shoulder girdle was evident by 10 months and at the elbow by 12 –15 months by which time his pain had completely resolved. Muscle recovery in the forearm, wrist and hand was apparent by 2 years. He recovered useful function throughout the damaged upper limb and he could grip and carry objects. There was cocontraction at the shoulder and
a b
Fig. 9.36 A 23 year old man. Left sided lesion: avulsion C5, C6 and C7, type 5 lesion C8, T1. Function at the shoulder and elbow 11 years after reimplantation of C5, C6 and C7 (Courtesy of Thomas Carlstedt).
The Closed Supraclavicular Lesion
Fig. 9.37 A 9 year old boy. Right sided lesion. Avulsion C5,C,C7,C8,T1. Reimplantation of all five spinal nerves. The hand at 6 years showing useful pinch grip function with some recovery into the small muscles (Courtesy of Thomas Carlstedt).
elbow but not in the hand. At 6 years after operation no recovery of autonomic or cutaneous sensibility could be detected within the hand. The power and pinch grip was measured at 10% of the uninjured side and hand function tests (Jebson-Taylor, Sollerman) showed recovery of hand function to about one-half of the uninjured side (Fig. 9.37). The velocity of nerve conduction was maintained but amplitude was reduced particularly in the ulnar nerve. Functional MRI investigations showed that there was marked activity on the ipsi lateral side when the affected hand was used, which was not seen when the normal hand was used. Carlstedt and his colleagues suggest that: ‘the ipsilateral cortex activity is most likely to be the result of cortical plasticity caused by diminished trans-callosal inhibition, as has been noted in CNS disorders such as multiple sclerosis or spinal cord infarction’. They go on to suggest that the restored hand function ‘might rely on cortical sensory programmes established before the injury’.
References Abdulla S, Bowden REM (1960) The blood supply of the brachial plexus. Proc R Soc Med 53:203–205 Addas BMJ, Midha R (2009) Nerve transfers for severe injury. In: Spinner RJ, Winfree CJ (eds) Peripheral nerves: injuries. Neurosurgical clinics of North America, vol 20., pp 27–38 Akasaha Y, Hara T, Takahishi M (1991) Free muscle transplantation combined with intercostal nerve crossing for reconstruction of elbow flexion and wrist extension in brachial plexus injuries. Microsurgery 12:346–351 Akita S, Wada E, Kawai H (2006) Combined injuries of the brachial plexus and spinal cord. J Bone Joint Surg 88B:637–641
425 Allieu Y (1977) Exploration et traitement direct des lésions nerveuse dans les paralysies traumatiques par elongation du plexus brachial chez I’adulte. Rev Chir Orthop 63:107–122 Alnot JY (1988) Traumatic brachial plexus palsy in adults. In: Tubiana R (ed) The hand vol. III. WB Saunders, Philadelphia, pp 607–644, Chapter 60 Alnot JY, Jolly A, Frot B (1981) Traitement direct des lésions nerveuses dans les paralysies traumatique du plexus brachial chez l’adute. Orthopaedics 5:151–168 Altaf F (2009) Brachial plexus injuries in rugby and American football. Pers. comm. Barnes R (1949) Traction injuries of the brachial plexus in adults. J Bone Joint Surg 31:10–16 Berman JS, Taggart M, Anand P, Birch R (1995) The effect of surgical repair on pain relief after brachial plexus injuries. In: Association of British Neurologists Proceedings. p 44 (abs) Berman J, Anand P, Chen L, Taggart M, Birch R (1996) Pain relief from preganglionic injury to the brachial plexus by later intercostal transfer. J Bone Joint Surg 78B:759–760 Berman JS, Birch R, Anand P (1998) Pain following human brachial plexus injury with spinal cord root avulsion and the effect of surgery. Pain 75:199–207 Birch R, Bonney G (1998) Traumatic lesions of the brachial plexus. In: Birch R, Bonney G, Wynn Parry CB (eds) Surgical disorders of the peripheral nerves. Churchill Livingstone, London/Edinburgh, Chapter 9 Birch R, Dunkerton M, Bonney G, Jamieson AM (1988) Experience with the free vascularised ulnar nerve graft in repair of supraclavicular lesion of the brachial plexus. Clin Orthop Rel Res 237:96–104 Bonnel F (1989) Anatomie du plexus brachial chez le nouveau-né et l’adulte. In: Alnot JY, Narakas A (eds) Les Paralysies du Plexus Brachiale, Monographies du Groupe d’Étude de la Main. Expansion Scientifique Française, Paris, pp 3–13 Bonney G (1954) The value of axon responses in determining the site of lesion in traction lesions of the brachial plexus. Brain 77:588–609 Bonney G (1959) Prognosis in traction lesions of the brachial plexus. J Bone Joint Surg 41B:4–35 Bonney G (1977) Some lesions of the brachial plexus. Ann R Coll Surg Engl 59:298–306 Bonney G, Gilliatt RW (1958) Sensory nerve conduction after traction lesion of the brachial plexus. Proc Roy Soc Med 51:365–367 Bufalini C, Pescatori G (1969) Anterior cervical electromyography on the diagnosis and prognosis of brachial plexus injuries. J Bone Joint Surg 51B:627–631 Carlstedt T (2007a) Central nerve plexus injury. Imperial College Press, 57 Shelton Street, Covent Garden. London W2H 9HE Carlstedt T (2007b) The intraspinal plexus injury, Ibid. Imperial College Press, 57 Shelton Street, Covent Garden. London W2H 9HE, pp 7–18, Chapter 2 Carlstedt T (2007c) Outcome, Ibid. Imperial College Press, 57 Shelton Street, Covent Garden. London W2H 9HE, pp 139–166, Chapter 8 Carlstedt T, Hultgren T, Nyman T, Hansson T (2009) Cortical activity and hand function restoration in a patient after spinal cord surgery. Nat Rev Neurol 5:571–574 Carvalho GA, Nikkhari G, Mathies C, Penkert G, Samii M (1997) Diagnosis and root avulsion in traumatic brachial plexus: value and computerised tomography myelography and magnetic resonance imaging. J Neurosurg 86:69–76 Celli L, Rovesta C (1987) Electrophysiologic Intraoperative Evaluations of the Damaged Root in Traction of the Brachial Plexus. In: Terzis JK (ed) Microreconstruction of nerve injuries. WB Saunders, Philadelphia, pp 473–482 Chen L (1998) Results of grafts against SSEP. In: Birch R, Bonney G, Wynn Parry CB (eds) Surgical disorders of the peripheral nerves.
426 Churchill Livingstone, London/Edinburgh. Traumatic lesions of the brachial plexus, Chapter 9, pp 157–208 Chesterman PJ (1964) Spastic paraplegia caused by sequestrated thoracic intervertebral disc. Proc R Soc Med 57(2):87, Section of Orthopaedics pp 9–10 Da Silva VR, Al-Gahtany M, Midha R, Sarma D, Cooper P (2003) Upper thoracic spinal cord herniation after traumatic nerve root avulsion. Case report and review of literature. J Neurosurg (Spine 3) 99:306–309 Davis L, Martin J, Perret G (1947) The treatment of injuries of the brachial plexus. Ann Surg 125:647–657 Davis ER, Sutton D, Bligh AS (1966) Myelography in brachial plexus injury. Br J Radiol 39:362–371 Dawson GD (1947) Cerebral responses to electrical stimulation in man. J Neurol Neurosurg Psychiatry 10:134–140 Dawson GD, Scott JW (1949) The recognition of nerve action potentials through the skin in man. J Neurol Neurosurg Psychiatry 12:259–267 Dickson JK, Biant LC (2010) A good outcome following complete injury of the brachial plexus. J Bone Jt Surg 92B: 540–544 Doi K, Sakai K, Kuwata K, Ihara N, Kawai S (1995) Double free muscle transfer to restore prehension following complete brachial plexus avulsion. J Hand Surg 20A:408–414 Dunkerton M (1984) Measurement of function and results of repair. Unpublished work cited in Birch R, Bonney G, Wynn Parry CB (Authors) Surgical Disorders of the Peripheral Nerves 1st ed, Churchill Livingstone, London. Flannery MC, Birch R (1990a) Acute compression of the cervical spinal cord: a complication of preganglionic injury to the brachial plexus. Injury 24:247–248 Flannery MC, Birch R (1990b) A rare complication of a preganglionic brachial plexus injury. Injury 24:249 Flannery MC, Birch R, Bonney G (1990) Thoracic disc protrusion in a patient with preganglionic injury to the brachial plexus: differential diagnosis of cord compression. Injury 24:246–247 Flaubert AC (1827) Mémoire sur plusiers cas de luxations dans les efforts pour la réduction ont été suivis d’accidents graves. Répertoire Générale d’Anatomie et de Physiologie Pathologique 3:55–79 Foerster O (1929) Die therapie der Schussverletzungen der peripheren Nerven, resultate der plexus operationen. In: Foerster BO (ed) Handbuch der Neurologie von Lewandowski, vol 2. Julius Springer, Berlin, pp 1676–1691 Frazier CH, Skillern PG (1911) Supraclavicular subcutaneous lesions of the brachial plexus not associated with skeletal injuries. J Am Med Assoc 57(25):1957–1963 Goldie BS, Coates CJ (1992) Brachial Plexus Injuries – A Survey of Incidence and Referral Pattern. J Hand Surg 17B:86–88 Grundy DJ, Silver JR (1984) Combined brachial plexus and spinal cord trauma. Injury 15:57 Harris W (1904) The true form of the brachial plexus and its motor distribution. J Anat Physiol 38:399–422 Hattori Y, Doi K, Ikeda K, Pagsaligan JM, Watanabe M (2005) Restoration of prehension using double free muscle technique after complete avulsion of brachial plexus in children: a report of three cases. J Hand Surg 30A:812–819 Hayashi N, Yamamoto S, Okubo T et al (1998) Avulsion injury of cervical nerve roots: enhanced intradural nerve roots at MR Imaging. Radiology 206:817–822 Hems TEJ, Birch R, Carlstedt T (1999) The role of magnetic resonance imaging in the management of traction injuries to the adult brachial plexus. J Hand Surg 24B:550–555 Hentz V, Narakas AO (1988) The results of microneurosurgical reconstruction of complete brachial plexus palsy. Orthop Clin North Am 19:107–114 Herzberg G, Narakas A, Comtet JJ (1989) Anatomie et rapports des racines du plexus brachial dans l’espace intertransversaire. In: Alnot JY, Narakas A (eds) Les Paralysies du plexus brachial. Expansion Scientifique Française, Paris, pp 19–26
Surgical Disorders of the Peripheral Nerves Herzberg G, Narakas A, Comtet J-J (1996) Surgical approach of the brachial plexus roots. In: Alnot JY, Narakas A (eds) Traumatic brachial plexus injuries. Expansion Scientifique Française, Paris, pp 19–22 Hetreed MA, Howard LA, Birch R (1992) Evaluation of perioperative sensory evoked potentials recorded from nerve roots to the cervical epidural space during brachial plexus surgery. In: Jones SJ, Hetreed M, Boyd S, Smith NJ (eds) Hand book of spinal cord monitoring. Kluwer Academic, Dordrecht/Boston/London, pp 171–178, Published 1994 Holt M (1994) Accessory to suprascapular transfer in preganglionic lesions of the brachial plexus. Royal Australasian College of Surgeons. A Scientific Meeting, Brisbane. 1994 Autumn Horsley V (1899) On Injuries to peripheral nerves. Practitioner Old Series Vol 63, new series Vol 10:131–144 Hovelaque A (1927) Anatomie Des Nerfs Carniens Et Rachidiens Et Du Système Grand Sympathetique. Doin, Paris Htut M, Misra P, Anand P, Birch R, Carlstedt T (2006) Pain phenomena and sensory recovery following brachial plexus avulsion injury and surgical repair. J Hand Surg 31B:596–605 Htut M, Misra VP, Anand P, Birch R, Carlstedt T (2007) Motor recovery and the breathing arm after brachial plexus surgical repair including re-implantation of avulsed spinal nerves into the spinal cord. J Hand Surg 32E:170–178 Jamieson A, Bonney G (1979) Communication au Symposium sur le plexus brachial. Lausanne 1978 (abs) Int Microsurg 1:103–106 Jamieson A, Hughes S (1980) The role of surgery in the management of closed injuries of the brachial plexus. Clin Orthop 147:210–215 Jones SJ (1987) Diagnostic value of peripheral and spinal somatosensory evoked potentials in traction lesions of the brachial plexus. In: Terzis JK (Ed) Microreconstruction of Nerve Injuries. WB Saunders, Philadelphia. 463–472 Jones SJ (1979) Investigation of brachial plexus traction lesion by peripheral and spinal somatosensory evoked potentials. J Neurol Neurosurg Psychiatry 42:107–116 Kato N, Htut M, Taggart M, Carlstedt T, Birch R (2006) The effects of operative delay on the relief of neuropathic pain after injury to the brachial plexus. J Bone Joint Surg 88B:756–759 Ker AT (1918) The brachial plexus of nerves in man, the variations in its formation and its branches. Am J Anat 23:285–395 Kline DG, Kott J, Barnes G, Bryant L (1978) Exploration of selected brachial plexus lesions by the posterior subscapular approach. J Neurosurg 49:872–879 Klumpke A (1885) Contribution a l’étude des Paralysies Radiculaires du Plexus Brachial: Paralysies radiculaires Totales. Rev Med (Paris) 5:591–616 Landi A, Copeland SA, Wynn-Parry CB, Jones SJ (1980) The role of somatosensory evoked potentials and nerve conduction studies in the surgical management of brachial plexus injuries. J Bone Joint Surg 62B:492–496 Leechavengvongs S, Witoonchart K, Verpairojkit C, Thuvasethakul P, Malungpaishrope K (2006) Combined nerve transfers for C5 and C6 brachial plexus avulsion injuries. J Hand Surg 31A:183–189 Leffert RD (1985) Brachial plexus injuries. Churchill Livingstone, NewYork/Edinburgh/London/Melbourne, pp 91–120 Lurje A (1948) Le traitement chirurgical des traumatismes de la partie supérieure du plexus brachial, du type ERB. Ann Chir 127:317–328 Lusskin R, Campbell JB, Thompson WAL (1973) Post traumatic lesions of the brachial plexus: treatment by transclavicular exploration and neurolysis or autografts reconstruction. J Bone Joint Surg 55A:1159–1176 Magalon G, Bordeaux J, Legree R, Aubert JP (1988) Emergency versus delayed repair of severe brachial plexus injuries. Clin Orthop Relat Res 237:32–36 Mansat M (1977) Anatomie topographique chirurgicale du plexus brachial. Rev Chir Orthop Reparatrice Appar Mot 63:20–26
The Closed Supraclavicular Lesion Mansat M, Lebarbier P, Mansat A (1979) Mécanismes lésionnel dans le traumatisme fermés du plexus brachial. In: Michon J, Moberg E (eds) Les lésions traumatiques des Nerfs Périphériques. Expançsion Scientifique Française, Paris, pp 159–164 Marshall RW, de Silva RD (1986) Computerised tomography in traction lesions of the brachial plexus. J Bone Joint Surg 68B:734–738 Melzack R (1975) The McGill pain questionnaire. Pain 1:277–299 Midha R (1997) Epidemiology of brachial plexus injuries in a multi trauma population. Neurosurgery 40:1182 Miller RA (1939) Observations upon the arrangement of the axillary artery and brachial plexus. Am J Anat 64:143–163 Millesi H (1968) Zum problem der uberbrükung von defecten peripheren nerven. Wien Med Wochenschr 118:182–187 Nagano A, Ochiai N, Sugioka H et al (1989) Usefulness of myelography in brachial plexus injuries. J Hand Surg 14B:59–64 Nakamura T, Yabe Y, Horuichi Y, Takayama S (1997) Magnetic resonance myelography in brachial plexus injury. J Bone Joint Surg 79B:764–769 Narakas AO (1978) Surgical treatment of traction injuries of the brachial plexus. Clin Orthop Relat Res 133:71–90 Narakas AO (1988) Neurotization or nerve transfers in traumatic brachial plexus lesions. In: Tubiana R (ed) The hand vol. 4. WB Saunders, Philadelphia, pp 656–683, Chapter 62 Narakas AO (1993) Lesions found when operating traction injuries of the brachial plexus. Clin Neurol Neurosurg 95(suppl):56 Narakas AO, Verdan C (1969) Les Greffes Nerveuses. Z Unfall Berufskr 3:137–152 Nordin L, Sinisi M (2009) Brachial plexus avulsion causing Brown Séquard syndrome. J Bone Joint Surg 91B:88–90 Oberle J, Antoniadis G, Rath SA et al (1998) Radiological investigations and intra-operative evoked potentials for the diagnosis of nerve root avulsion: evaluation of both modalities by intradural root inspection. Acta Neurochir 140:527–531 Ochi M, Akita Y, Watanabe M et al (1994) The diagnostic value of MRI in traumatic brachial plexus injury. J Hand Surg 19B:55–59 Patterson M, Dunkerton M, Birch R, Bonney G (1990) Re-innervation of the suprascapular nerve in brachial plexus injuries. J Bone Joint Surg 72B:993 Penfield W (1949) Late spinal paralysis after avulsion of the brachial plexus. J Bone Joint Surg 31B:40–41 Petrov MA (1973) Lésions traumatiques du plexus brachial. Traitement chirurgical, transplantations, resultants éloignés. Chirurgie 99:924–934 Phillips S, Horwitz M, Yam A (2009) A study of 111 ventral root repairs. Personal Communication Privat JM, Mailhe D, Allieu Y, Bonnel F (1982) Précoce hemilaminectomie cervicale exploratrice et neurotisation du plexus brachial. In: Simon L (ed) Plexus Brachiale et Medecine de Re-éducation. Masson, Paris, pp 66–73 Puusepp L (1931) Die peripheren nerven. Chir Neuropath 1:17 Raimondi PL, Morelli A (1987) La valutazione dei risultati nella chirugia del plesso brachiale. Rivista di Churirgia della mano 24:419–432 Roaf R (1963) Lateral flexion injuries of the cervical spine. J Bone Joint Surg 45B:36 Robotti E, Longhi P, Verna G, Bocchiotti G (1995) Brachial plexus surgery: an historical perceptive. Hand Clin 11:517–533 Rosson JW (1987) Disability following closed traction lesions of the brachial plexus sustained in motor cycle accidents. Hand Surg 12B:353–355 Rosson JW (1988) Closed traction lesions of the brachial plexus – an epidemic among young motor cyclists. Injury 19:4–6 Santos Palazzi R (1971) La microcirurgia en las lesiones de los nervios perifericos. Rev Orthop Traum 15:499–526
427 Scaglietti O (1947) Einselheiten über die Operations tecknik bei Verletzungen der Wurzeln des plexus Brachialis durch Schusswaffen. Zent 61 Neurochirg 7:129–144 Schenker M (1998) Analysis of avulsed roots in traction injury of the human brachial plexus. Thesis for MSc in Surgical Science, University College of London Schenker M, Birch R (2001) Diagnosis of level of intradural ruptures of the rootlets in traction lesions of the brachial plexus. J Bone Joint Surg 83B:916–920 Seddon HJ (1954) Nerve grafting. In: Peripheral nerve injuries by the nerve injury committee of the medical research council MRC, Special Report Series 282, London HMSO, pp 402–403 Sedel L (1979) La réparation chirurgicale des lésions traumatiques du plexus brachial. Nouv Presse Méd 8:691–693 Semple Campbell (2008). Mechanisms of injury to the brachial plexus, personal communication Slingluff CL, Terzis JK, Edgerton MT (1987) The quantitative microanatomy of the brachial plexus. In: Terzis JK (ed) Microreconstruction of nerve injuries. WB Saunders, Philadelphia, pp 285–324 Smith IC (1995) Multiple cranial nerve palsies in a motorcyclist caused by the helmet. Injury 26:405–406 Stephens JH (1934) Brachial plexus paralysis. In: Codman EA (ed) The shoulder. IG Miller/Brooklyn, New York Sugioka H (1984) Evoked potentials in the investigation of traumatic lesions of peripheral nerves and the brachial plexus. Clin, Orthop and Rel Res 184: 852–92 Sugioka H, Nagano A (1989) Électrodiagnostic dans l’Évaluation des Lésions par Élongation du Plexus Brachial. In: Alnot J-Y, Narakas A (eds) Les Paralysies du Plexus Brachiale. Monographies du Groupe d’Étude de la Main, vol 15. Expansion Scientifique Française, Paris, pp 123–129 Sugioka H, Tsuyama N, Hara T, Nagano A, Tachibana S, Ochiai N (1982) Investigation of brachial plexus injuries by intraoperative cortical somatosensory evoked potentials. Arch Orthop Trauma Surg 99:143–151 Sunderland S (1974) Mechanism of cervical root avulsion in injuries of the neck and shoulder. J Neurosurg 41:705–714 Taggart M (1998a) Relief of pain with operation. In: Birch R, Bonney G, Wynn Parry CB (eds) Surgical disorders of the peripheral nerves. Churchill Livingstone, London/Edinburgh, pp 373–404, Chapter 15 Taggart M (1998b) Delay before return to work or study. In: Birch R, Bonney G, Wynn Parry CB (eds) Surgical disorders of the peripheral nerves. Churchill Livingstone, London/Edinburgh, pp 451–466, Chapter 18 Tavakkolizadah A, Saifuddin A, Birch R (2001) Imaging of adult brachial plexus injuries. J Hand Surg 26B:183–191 Tavernier L (1932) Paralysie du plexus brachial: rupture des racines supérieures dans le creux sus-claviculaire. Greffe nerveuse. Lyon Chir 29:90–93 Thorburn W (1900) Secondary suture of the brachial plexus. Br Med J 1:1073–1075 Tsuyama N, Hara T (1973) Intercostal nerve transfer in the treatment of brachial plexus injury of root-avulsion type. In: Proceedings of the S.I.C.O.T., Excerpta Medical International Congress Series No. 291, Tel Aviv, 9–12 Oct.1972, pp 348–351 Tsuyama N, Sakaguchi R, Hara T, Kondo T, Kaminuma S, Ichi M, Ryn D (1968) Reconstructive surgery in brachial plexus injuries. In: Proceedigs of the 11th annual meeting Japan Society Of the Hand, Hiroshima, 1968 Walsh JF (1877) The anatomy of the brachial plexus. Am J Med Sci 74:387–399 Yeoman PM (1968) Cervical myelography in traction injuries of the brachial plexus. J Bone Joint Surg 50B:253–260
Birth Lesions of the Brachial Plexus
10
Some differences between the adult and the neonatal nervous systems; the lesion of the nerve; the central affect; methods of study; incidence, risk factors and natural history; neurophysiological investigations; indications for nerve operations and their results; co contraction; the chief causes of deformity; posterior subluxation and dislocation of the gleno humeral joint; methods and results of operations for reduction of the shoulder; deformities of the elbow and forearm; conclusion. The birth lesion differs from the high energy transfer closed traction injury to the supraclavicular plexus lesion in the adult in two important respects. The traction and compression forces are expended upon the nerves for hours during a protracted and difficult delivery; the response of the immature nervous system is very different from that seen in the adult. There are important differences between the immature and mature nervous systems, particularly in response to injury (see Chapters 2 and 3). Those which seem especially relevant to the birth lesion of the brachial plexus include:
that the radial extensors of the wrist are sometimes supplied by C5 in infants; it was this observation that prompted Fairbank (1913) to transfer the distal end of C5 into a transverse slit in C7 in a case of avulsion of C5 and C6. The functional distribution of the spinal nerves has been investigated by Vredeveld and his colleagues (2000 in Heerlen), who showed, by electromyographic examination
1. In the infant the spinal cord fills the cervical canal, the spinal rootlets emerging at or close to a right angle from it. 2. The transitional zone is unstable. 3. The duration of myelination is prolonged. 4. In the infant there is a higher density of conducting tissue and higher blood flow to that tissue. 5. The neurones of the infant spinal cord and dorsal root ganglion are more vulnerable to proximal axonotomy or avulsion (preganglionic) injury (Groves and Scaravilli 2005). It is not unusual to see unexpectedly good recovery of function into the hand in the most severe lesions in spite of a persisting Bernard–Horner syndrome (Fig. 10.1) Huang and his colleagues (2008) provide one possible explanation for this. Dissections in infant cadavers revealed white rami (pre ganglionic fibers) passing from C7 to the sympathetic chain in addition to those expected from C8 and T1. These workers also found that the number of white rami accompanying C8 was higher than is usually seen in adults. Immunohistochemical tests using neuronal nitric oxide synthetase confirmed that preganglionic fibers were present in the ventral roots of C7 as well as in C8 and T1. This work suggests that the functional distribution of the spinal nerves in infants is more widespread than in the adult. Harris and Low (1903) noted
Fig. 10.1 Spontaneous recovery in a group 4 lesion. The hand was paralyzed at the age of 6 months when neurophysiological investigations (Shelagh Smith) suggested a post ganglionic injury at C7, C8 and T1. Hand function was virtually normal at the age of 5 years.
R. Birch, Surgical Disorders of the Peripheral Nerves, DOI: 10.1007/978-1-84882-108-8_10, © Springer-Verlag London Limited 2011
429
430
of the biceps muscle in infants, that C7 often contributes an important component. There is, it seems, polyneural or “luxury” innervation in the neonatal muscle whereby muscle fibers are supplied by more than one neurone, a pattern of innervation which regresses to that pertaining in the adult during maturation. Gramsbergen et al. (1997) demonstrated such regression of polyneural innervation in the human psoas muscle. A similar phenomenon was demonstrated by Colon et al. (2003a) for the somato-sensory pathways. Our colleagues in Heerlen go one step further by showing that spontaneous activity in muscle fibers appears much earlier in infants than it does in the adult, indeed as early as the fourth or fifth day of life. Fibrillation potentials disappear much sooner than they do in the adult when there is reinnervation by collateral sprouting from surviving or regenerating motor axons (Colon et al. 2003b). The idea that neurophysiological investigations are of only limited value in the analysis of the birth lesion still lingers. It is still said that EMG’s “are too optimistic.” The work of Vredeveld, Blaauw and their colleagues in Heerlen explains some of the difficulties in the interpretation of the findings from neurophysiological investigations (NPI). Shelagh Smith, later joined by Peter Misra (Smith 1996, 1998) have established that it is possible to define not only the extent and the depth of lesion but also the likely extent of spontaneous recovery. Their valuable work, which underpins much of our current practice, is described later in this chapter, and also in Chapter 6. The direct physical cause of the lesion is the forced separation of the forequarter from the axial skeleton caused by obstruction at the narrowest point of the birth canal. When there is obstruction to breech delivery the separation is in an upward direction. The baby is hanged, placing the upper spinal nerves and the phrenic nerve particularly at risk. When the shoulder is forward the separation is in a downward direction so that the nerves are stretched, ruptured or avulsed as the angle between the delivered head and the obstructed shoulder widens. The matter is reviewed in an authoritative and balanced manner by Carlstedt (2006). Carlstedt outlines two mechanisms: displacement of the shoulder girdle induces a more peripheral lesion whereas forceful lateral flexion of the head and neck causes a more central lesion. Carlstedt points out mathematical models of the application of force do not reflect the surface area to which force is applied nor do they allow for the elasticity of the neonate. Metaizeau et al. (1979) studied the effects of traction on stillborn babies, The shoulders were fixed and traction was applied to the neck: C5 and C6 were always subjected to more traction than the lower nerves and a force of 20–40 kg (about 200–400 N) was required to rupture C5, C6 and C7; after this C8 and T1 become subjected to stretching. C8 and T1 are less strong so that only about one half of the force is needed to rupture or avulse them.
Surgical Disorders of the Peripheral Nerves
10.1 The Lesion of the Spinal Nerve in Birth Lesion of the Brachial Plexus (BLBP) These are classed by the method described in Chapter 9. The classification is modified in the infant by pre-operative neurophysiological investigations (NPI) and it is outlined in Table 10.1. Type 1 The nerve is normal, any lesion is no more than a transient conduction block. Type 2 The nerve shows signs of traction but the perineurium is intact. The lesion is a mixture of conduction block and axonotmesis and, in general, recovery is good. Type 3 The rupture usually appears as a fusiform swelling, firm, even hard in texture. The neuroma is at least twice the diameter of the nerve. The suprascapular nerve has been so stretched that it often passes upwards and lateral from beneath the clavicle There is conduction across the lesion and SSEP’s can usually be recorded by stimulating the nerves distal to the lesion. There is always some regeneration across a rupture in the infant and this is the salient difference between the lesion in the infant from that in the adult. Some conduction across the lesion was detectable in every one of our cases of rupture. Absence of that conduction implies a central, intradural, injury. Chen et al. (2008) examined the neuroma from ruptured spinal nerves in 28 children by light and transmission electronmicroscopy. There was always a great deal of collagen: on occasion ganglion cells were seen. The regenerating nerves were finely myelinated; on average about 40% of the normal total number of fibers traversed the lesion. This disorderly and chaotic regeneration, with poor maturation of the myelin sheath may explain Chen’s findings that measurements of conduction across the lesion are not a reliable guide to potential recovery. Pondaag et al. (2008) studied conduction across the lesion and in the main nerves by measuring intraoperative nerve action potentials (NAP) and evoked compound motor action potentials (CNAP) from damaged and normal nerves in the upper brachial plexus in 95 infants. More than 1,500 recordings were analyzed. Nerve fibers to biceps passed predominantly through C6 but supplementary pathways through C5 and C7 were always demonstrated. Although there was a correlation between NAP and CNAP recordings and the severity of lesion, the sensitivity of these investigations was relatively low. Pondaag and his colleagues conclude that intraoperative studies of conduction across lesions of the spinal nerves in the infant are not especially helpful in deciding between repair or neurolysis. Malessy and Pondaag (2009) suggest that gradual exertion of traction stretches rather than clearly ruptures the nerve in the birth lesion so that some axons always traverse the lesion. Regeneration is disorderly, with much collateral sprouting. There is a good deal of misdirection of regenerating axons, especially between the anterior and the posterior divisions which may lead to cocontraction, especially
Birth Lesions of the Brachial Plexus
431
Table 10.1 The types of lesions of spinal nerves in BLBP. Intra-operative Type Lesion Pre-operative SSEP’ at neurophysiological foramen type
Intra-operative SSEP’ stimulating distal to the lesion
Muscle response evoked by stimulation at foramen
Appearance
1
Intact
A
Normal
Normal
Strong
Normal
2
Recovering rupture
B (favourable)
Normal
>50%
Strong
Fusiform neuroma, normal nerve proximally
3
Rupture
B (unfavourable)
Normal
<50%
Weak
Hard “double humped” neuroma. Normal nerve proximally
4
Rupture with partial B (unfavourable) pre-ganglionic lesion or C of intradural elements
Diminished
Diminished or absent
Weak
Diffuse longitudinal fibrosis.
5
Preganglionic with no displacement of dorsal root ganglion
C
Absent
Absent
Absent
Sometimes atrophy of nerve at foramen
6
Incomplete pre-ganglionic or selective preganglionic of ventral or dorsal roots
B (unfavourable) or C
Diminished or absent
Diminished or absent
Weak or absent
On occasion atrophy of nerve at foramen
Absent
Absent
Absent
Dorsal root ganglion visible.
7
Pre-ganglionic with C displacement SSEP, somatosensory evoked potentials.
between pectoralis major, the deltoid and the spinatii muscles or between biceps, triceps and pectoralis major. Malessy and Pondaag suggest that the disorderly regeneration of sensory axons and their faulty differentiation causes defective development of central motor patterns. Type 4 Combined rupture and intradural injury: This is common, especially in C6 and C7. The nerves lie coiled up in the posterior triangle and are often embedded in scar. The SSEP is abnormal in form and diminished in amplitude. Some limited muscular response is present, usually confined to such proximal muscles as pectoralis major. The architecture of the prepared stump at the foramen is abnormal. Recovery through grafts placed upon these stumps usually disappoints. Type 5 The undisplaced avulsion. This is more common than in the adult and it presents a particularly difficult problem. The nerve may appear somewhat atrophic but otherwise undamaged. There is neither central nor distal conduction. Curiously some Type 5 lesions go on to useful spontaneous recovery. Gilbert (1995) found that about one half of such lesions at C5 and C6 do so; we think that the same is true in at least one half of Type 5 lesions at C8 and T1. Type 6 Selective avulsion of dorsal or ventral roots of the spinal nerve. This has been confirmed in a few cases. There may be a recordable SSEP but no muscular response; conversely, stimulation of the nerve evokes a distal muscular response but no SSEP. One 4 year old child presented with
atrophy of the little finger which was clearly insensate. The ulnar SAP was intact but EMG confirmed strong reinnervation of C8 and T1 muscles. Evidently, the dorsal root of C8 was avulsed but there was regeneration through the ventral root. Type 7 Avulsion with displacement of the DRG is obvious. The decision about whether to operate in BLBP is made even more difficult by the unquantifiable extent of intradural injury, as well as by the capacity of the neonatal nervous system to regenerate across a post ganglionic rupture, albeit in incomplete and chaotic fashion.
10.1.1 The Central Affect The long duration of myelination, so that conduction velocity approached adult levels only after 5 years, shows that the deep afferent pathways are immature and possibly unstable. Iyer (2000) describes the evolution of hand control in the normal infant. At about 4 months there is grasp for small objects by prehension; by partial thumb opposition at 5 months, and by transfer of objects between the two hands by 6 months. Full thumb opposition is not seen until 7 months. Throwing a ball begins at 18 months, building a tower with four or six blocks is seen at 24 months. Erhard and Lindley (2000) say:
432
Surgical Disorders of the Peripheral Nerves
“well integrated dominance may not occur until 8 or 9 years of age.” The interruption of afferent input into the central nervous system through the somato-sensory pathways may account for the change in limb dominance which is so common in BLBP. Yang et al. (2005) studied children aged between 5 and 9 years and they found that only 17% of children with right sided BLBP showed right-hand preference. This might reflect central plasticity; equally it might reflect the dissociation between the central nervous system and the upper limb during the critical early months of development. The possibility of developmental apraxia was explored by Brown et al. (2000) who measured compound motor action potentials (CMAP) from biceps and thenar muscles innervated by the median nerve. Records of stimulus evoked muscle twitch were compared with the interpolated twitch test to determine the extent of motor unit recruitment: “the persisting disability may not be entirely due to poor nerve fiber regeneration. Rather, in a proportion of cases, the weakness and clumsiness of the arm is the result of an inability to recruit the motor units that are available.” The findings in the oldest of the 17 cases, a 35 year old woman, were especially striking: “Despite numbers of thenar and hypothenar motor units that were only just below the expected normal ranges, she had virtually no voluntary movement of the thumb and little finger.” Noetzel and Walpow (2000) reflected on the growing evidence for “activity dependant plasticity” in the CNS. “Even in adults, representation in the sensory-motor cortex is affected by motor training, as well as by manipulations that change sensory input and affect motor performance.” Fig. 10.2 Lesion of C5, C6 and C7.
10.2 Methods of Study 10.2.1 Diagnosis This is usually straightforward. The characteristic posture of the Narakas Group 1 and 2 lesions follows unrestrained activity through the lower trunk so that the shoulder is adducted and medially rotated. The elbow is extended, the forearm pronated and the wrist flexed (Fig. 10.2). In the complete lesion the arm lies flaccid. It may be cold and “marbled”. Bernard–Horner syndrome suggests serious injury. In severe lesions a hypersensitive skin rash is sometimes seen in the dermatomes of C5 and C6. Discrepancy in the size of the digits is evident from about 6 weeks of age. A lesion of the spinal cord usually passes unnoticed until the child starts to walk, when the parents observe unsteadiness of gait and disparity in the size of the foot. This may indicate a partial Brown–Séquard syndrome. Breech lesions are frequently bilateral and often complicated by phrenic nerve palsy. The differential diagnosis includes, amongst other possibilities: fractures of the clavicle or humerus; neonatal sepsis
of the gleno-humeral joint; cerebral palsy; arthrogryposis; ischaemic cord injury and even trigger thumb. Any of these may co-exist with BLBP. Posterior dislocation of the shoulder at, or soon after birth, is frequent. Records are taken at the first and, where appropriate, at every subsequent attendance. 1. The Risk Factor Chart was developed by Margaret Taggart in 1984 (Giddins et al. 1994). Four main groups of factors are considered: those affecting the parents, those occurring during pregnancy, those occurring during labor and those relating to the child. This form is completed by the parents at their first attendance (Fig. 10.3). Hospital records outlining the circumstances attending the baby’s delivery are often scanty or nonexistent but where such records were made available to us we found that the parents’ recollections were accurate. 2. The Narakas Classification. In cases born by cephalic presentation the lesion is classed into one of the four groups proposed by Narakas (1987). His system is not applicable to the special case of the lesion associated with breech delivery and it is best used no sooner than 4 weeks after birth.
Birth Lesions of the Brachial Plexus
Fig. 10.3 The risk factor chart.
Group 1 The fifth and sixth cervical nerves are damaged. There is paralysis of supraspinatus, infraspinatus, deltoid and biceps muscles. The upper limb lies in medial rotation with the elbow extended. Group 2 The fifth , sixth and seventh cervical nerves are damaged. There is now also weakness or paralysis of triceps and also of the extensors of the wrists. The hand is clenched into a fist with flexion at the wrist. Group 3 The paralysis is virtually complete. There may be some weak flexion of the fingers at, or shortly after, birth. Group 4 The paralysis is complete. The limb is flaccid; there is a Bernard–Horner syndrome. Augusta Klumpke, an American, who was the first woman to graduate from a medical school in Paris, described a number of complete lesions and recognised the significance of the Bernard–Horner syndrome (Klumpke 1885.) We have seen only one example of a lesion confined to C8 and T1 (Figs. 10.4 and 10.5).
433
The system of Narakas provides a useful guide to prognosis. Most of the Group 1 babies go on to considerable spontaneous recovery and some, but by no means all, regain normal function. Conversely we have seen only two examples of complete recovery amongst the group 4 lesions. 3. Function at the Shoulder is recorded Using Three Systems. (a) Mallet (1972) pointed out that the upper limb was useless if the shoulder was flail or if it lay fixed in severe medial rotation: “l’attitude vicieuse fixée de l’épaule detruit toute fonction.” These are Stage 1 shoulders. Stage 5 is reserved for the normal shoulder. The system of Mallet provides a strict measure of shoulder function; a bad result in any one of the five tests downgrades the shoulder as a whole. The five are: global elevation; lateral rotation with the arm at the side; the ability to place the hand behind the back; the ability to place the hand behind the neck and the ease with which the hand is brought to the mouth, the signe du clarion or bugle sign. We started to use the Mallet system in 1988 with some modification because we saw children with untreated posterior dislocation who demonstrated remarkably good function because of the high level of recovery from the injury to the nerves. The five movements are scored on a scale in range from 1–3 (Fig. 10.6). A different colored pencil is used at each attendance and the record is signed and dated with that color. This record can be completed in no more than 1 min with nearly all children aged 18 months or above. The lowest total, 4 points, is equivalent to Mallet’s stage 1 and 2. A score of 15 is equivalent to his 4 or 5. (b) Gilbert’s method (Gilbert and Tassin 1984) was introduced in 1990 (Table 10.2). (c) Our system (Fig. 10.10a–d) measures the active and passive ranges of movement at the thoraco-scapular gleno-humeral and radio-ulnar joints. The inferior scapulo-humeral angle (ISHA) is subtended by the long axis of the humerus and the lateral border of the scapula. It measures the respective contributions to elevation of the thoraco-scapula and gleno-humeral joints and it reveals weakness of muscles acting across those joints and also contractures. At rest, with the arm by the side, the ISHA is about 30°, and in full elevation of the limb it is to 170° in the normal shoulder. Movement at the glenohumeral joint accounts for about 140° of full elevation. If the active ISHA is reduced but the passive range is full then there must be weakness of the abductor muscles (Fig. 10.7). If there is equal reduction in both active and in passive ranges then there must be contractures within the gleno-humeral joint (Fig. 10.8). Sometimes this joint functions as an ankylosis and elevation of the limb occurs at the thoraco-scapular joint. The posterior scapulohumeral angle (PSHA) is subtended by the axis of the shaft of the humerus and the axis of the spine of the
434
Surgical Disorders of the Peripheral Nerves
Fig. 10.4 Klumpke’s lesion. This successful business man had no pain, spoke three languages and acknowledged scarcely any loss of function. The myelograph suggested avulsion of all spinal nerves. QST showed normal thresholds for C5, C6, C7. There was no recovery at all in C8 and T1.
Fig. 10.5 Group 4 lesion. Useful recovery was confined to C6.
scapula. The examiner holds the arm so that it is horizontal to the ground and the child’s hand rests on the opposite shoulder. If the PSHA is wide but active medial rotation is reduced then the medial rotator muscles are defective (Fig. 10.9). If both active and passive ranges of the PSHA are diminished then there must be retroversion
of the head of the humerus or the less common contracture of the posterior capsule and the lateral rotator muscles. The range of pronation and of supination is included in the shoulder record because of the profound effect of posterior dislocation of the head of humerus upon forearm rotation. Posterior dislocation of the shoulder must be suspected when a child presents with a pronated forearm and is seemingly unable to supinate even though biceps is active. 4. Flexion and extension of the elbow is recorded by the system of Gilbert and Raimondi (1993) (Table 10.3). A negative score is possible in those children with severe flexion deformity who have very weak biceps and triceps muscles. 5. Function at the wrist and hand is recorded using Raimondi’s system (1993) (Table 10.4) which is particularly useful for grades 3–5. 6. Closer analysis of hand function requires functional assessment appropriately modified to the age of the child. Quantitative sensory testing and measurement of post ganglionic (cholinergic) sweating is reserved for the children aged 5 years or more. 7. Limb growth is recorded by measuring the length of the clavicle from the mid line to the acromioclavicular joint; the length of the vertebral border of the scapula from the superior to inferior angle and for the arm by the interval between the acromion to the olecranon and of the forearm from the olecranon to the ulnar styloid. The length and breadth of the thumb and digits can be measured by drawing the outlines of both hands and by plethysmography.
Birth Lesions of the Brachial Plexus
435
Fig. 10.6 The Mallet chart.
These records are collected onto different colored sheets. There are four forms: 1. For non operated cases (Fig. 10.10a). 2. For the nerve repair (Fig. 10.10b). 3. For operations upon the shoulder (Fig. 10.10c). 4. For those children who undergo a number of operations (Fig. 10.10d). The Mallet chart is especially helpful. Yang et al. (2005) analyzed the correlation between the actual use of the
upper limb and the different scoring systems. There was a close correlation between hand function and the Raimondi score and a similar relation obtained between the elbow score and limb preference for gross movements. However, the Mallet score strongly correlated with both gross and fine movements, demonstrating: “the first significant correlation between actual task performance and two of the three tested obstetrical brachial plexus palsy outcome measurements based on the extent of movement at different joints of the upper limb.”
436
Surgical Disorders of the Peripheral Nerves
Table 10.2 The Gilbert et al. (1993) system for measuring shoulder function. Grade Description 0
Flail
I
Elevation to 45 degrees, no lateral rotation
II
Elevation less than 90 degrees, lateral rotation to neutral
III
Elevation about 90 degrees, weak lateral rotation
IV
Elevation to 100–120 degrees, lateral rotation incomplete
V Elevation better than 120 degrees, lateral rotation full. The suffix + is added to indicate sufficient medial rotation permitting the hand to come to the body.
Fig. 10.8 Both the passive and active inferior scapulo humeral angles were reduced by cocontraction and contracture.
Fig. 10.7 Paralysis of serratus anterior in a case of breech delivery. The inferior scapular humeral angle is 120°, whilst active elevation of the arm is no more than 40°.
10.3 Epidemiology 10.3.1 Incidence Earlier confusion about the incidence of this disorder has been clarified at least in the British Isles, by the census conducted by Evans-Jones et al. (2003), who prefer the term congenital brachial palsy. Active surveillance was undertaken by the monthly reporting card system of the British
Fig. 10.9 Seven year old boy. Repair of C5 and C6 was combined with re-location of shoulder. Recovery for C7 was poor. Medial rotation is weak, in spite of de-rotation osteotomy of the shaft of humerus. The active posterior scapulohumeral angle (PSHA) was 0°; the passive PSHA was 70°.
Birth Lesions of the Brachial Plexus
437
Table 10.3 Scoring system for elbow function after Gilbert and Raimondi (1993). Flexion against gravity to 90 degrees
1
Flexion against resistance to 120 degrees
2
Flexion full range against resistance
3
Absent elbow extension
0
Elbow extension against gravity
1
Elbow extension against resistance
2
Fixed flexion deformity. 0–30 degrees
0
Fixed flexion deformity 30–50 degrees
−1
Fixed flexion deformity in excess of 50 degrees −2 The maximum score is 5. A negative score is possible where poor elbow flexion is accompanied by severe flexion deformity. We record ranges of pronation and supination in the shoulder chart. Raimondi retains these ranges in his hand system.
Table 10.4 Scoring system for hand function drawn from Raimondi (1993). Grade Description 0
Flail, or useless finger flexion; useless thumb; sensation defective or absent
I
Weak finger flexion, possibility of thumb lateral pinch. No extension of wrist or digits
II
Active extension at wrist, with passive (“tenodesis”) or weak flexion of fingers. Weak lateral pinch of thumb.
III
Strong flexion of wrist and digits. Thumb mobile with useful abduction and apposition. Intrinsic balance. Good possibilities for palliative operations to regain extension of wrist, and to improve rotation deformity of forearm.
IV
Strong flexion of wrist, digits, good small muscle function and a mobile thumb. The defects include extension of digits and restricted pronation and supination.
V
As IV, with finger extension and complete or almost complete pronation ands supination. The system is most useful for Grades III, IV, and V. Grades I and II reflect the potential for thumb adduction pinch grip.
Paediatric Surveillance Unit which invites consultant paediatricians in the United Kingdom and Republic of Ireland who are members of the Royal College of Paediatrics and Child Health to report cases in up to twelve diagnostic groups. Further information was obtained from those surgeons to whom the babies were referred. The results were validated by searching for cases notified to the Office of National Statistics and to the Irish Census Office. There were 323 confirmed cases, an incidence for the British Isles of 0.42 per 1,000 live births or one in 2,300. The collection of data was rather successful; 93% of the reporting cards were returned during the 12 months study period revealing 430 possible cases. More detailed information was sought from the clinicians reporting suspected cases, and 91% of the second questionnaires were returned. Sixty eight of the cases first reported were excluded because infants were born outside the study
period or because of duplication of reporting. The diagnosis was revised in four cases. Fifty three per cent of the infants were male; the right arm was more commonly afflicted than the left (50% versus 43%, with 4.5% not specified) and there were five bilateral cases. The injury to the brachial plexus was partial in 91% of infants (Narakas Groups 1 and 2), it was complete in 6.5% (Narakas groups 3 and 4). This work stands as one of the most important contributions to the study of BLBP over the last 10 years and we shall refer to other findings presented in the report. Foad et al. (2008) carried out a study of data bases over 3 years in the United States of America and they found an incidence of BLBP of 1.51 per 1,000 births which seems a little high. The suggestion is made that the incidence is decreasing. Blaauw et al. (2008) provide information about the incidence of the disorder in the Netherlands: “official registration institutions in the Netherlands (SIG) quoted an incidence of 1.02% in 1995, 1.34% in 1996, and 0.90% in 1997.” The incidence is certainly higher in breech deliveries. Tan (1973) in a study from Singapore, found an incidence of 0.14 per 1,000 in vertex deliveries but a figure of 24.5 per 1,000 in breech deliveries. The significance of breech delivery was further depicted by Boo et al. (1991), who found, in a survey of Malaysian patients, an incidence of 1.6 per 1,000 of all live births, rising to 8.6 per 1,000 babies born by breech. Causation The birth lesion in breech delivery is severe and often bilateral. Gilbert (1995) found a high incidence of preganglionic injury, particularly of the upper spinal nerves. Slooff and Blaauw (1995) studied 40 of their cases delivered by breech (23% of their total at that time): in 11 cases there was bilateral palsy; in 12 there was phrenic nerve palsy. C5 and C6 were damaged in 24 cases, and C5, C6 and C7 were damaged in 14 more; the lesion was complete in two cases. At least one avulsion occurred in 35 of these 40 children. The birth rate was significantly less than the mean and many infants were premature. The serious complication of phrenic nerve palsy is, again, emphasised. The characteristics of the breech lesion are well set out by Ubachs and Slooff (2001). The national Census found only 10 cases of breech lesion in the 323 confirmed cases, which suggest that obstetricians here take the matter of breech presentation rather more seriously than in some other countries (Fig. 10.11).
10.3.2 Risk Factors Our earlier report of 230 consecutive cases was presented in Giddins et al. (1994) The mean birth weight of the babies was 4.5 kg against the contemporary mean for the North West Thames region, of 3.88 kg. There was a co-relation between more severe lesions and higher birth weight. Shoulder dystocia was recorded in over 60% of deliveries. A trend was found towards the mother being heavier and shorter than the national average, and also to excessive weight gain.
438
There was no significant co-relation with social class. The later report, extending to more than 1,060 babies, comes from Tavakkolizadeh (2007) (Table 10.5). The older mother, with a high body mass index (BMI) giving birth to a large baby by instrumental delivery, presents the greatest risk. The continuing debate about elective Caesarean section is illuminated by a frequency of LSCS at 2% in the BLBP group, against the national rate of 18%. No doubt those who argue
Fig. 10.10 (a) The green chart is used at first attendance and for children where no operation is performed. (b) The yellow chart is used for nerve repair. (c) The pink chart is used for shoulder dislocation. (d) The blue chart is used for children with multiple operations. The record of shoulder and forearm movement shown in the green chart is included on the reverse side of all charts.
Surgical Disorders of the Peripheral Nerves
for “cost benefit analysis” over the interests of the individual will ignore this finding. Tavakkolizadeh confirmed the findings of the National Census which found significant associated risk factors in comparison with the normal population including: shoulder dystocia (60% v 0.3%); high birth weight and instrument assisted delivery, “and a significantly lower risk of nerve injury in infants delivered by Caesarean section” (Table 10.6).
Birth Lesions of the Brachial Plexus
439
Fig. 10.10 (continued)
b
c
440
Surgical Disorders of the Peripheral Nerves
Fig. 10.10 (continued)
Table 10.5 Significant risk factors for birth lesion of the brachial plexus (BLBP) – 1,060 cases. BLBP National Maternal age
29
26.8
Birth weight (kg)
4.23
3.47
Body mass index
26
25
Hypertension
8.8%
6.4%
Gestational diabetes
10%
1%
Episiotomy
2.9%
15%
3.9%
1.1%
Ventouse
16.1%
6.3%
Forceps
25.6%
3.7%
Caesarean section
2%
18%
SCBU
15.7%
8%
Asphyxia
21.6%
2%
Multiple injuries
0.7%
0.1%
Pregnancy problems
Presentation Breech Delivery
Neonatal problems
Duration of labor (hours) 10.8 By kind permission of Tavakkolizadeh 2007, Thesis. Fig. 10.11 Bilateral lesion from breech delivery. On the right – the lesion was complete and complicated by phrenic nerve palsy. Recovery in C5 and C6 was poor. At the age of 7 years, accessory to suprascapular nerve transfer restored some lateral rotation. On the left there was avulsion of C5 and rupture of C6. Transfer of latissimus dorsi at the age of 4 failed. The discrepancy in growth is particularly severe in the forearm and hand.
5.4
It is somewhat reassuring to see concordance between Tavakollizadah’s findings and those related by Evans Jones and his colleagues. The former inevitably includes babies who were referred because their injuries were more severe,
Birth Lesions of the Brachial Plexus
441
whilst the latter reflects data collected from the whole population. The incidence of breech delivery in the Census figures was 3%, it was 3.9% in Tavakollizadah’s study. The median birth weight in the census, 4.180 kg (national median 3.380 kg) is close to that reported by Tavakkolizadeh at 4.23 kg. Both of these studies identified children in whom no risk factor could be identified. There were 19 such cases (5.9%) in the census; Tavakkolizadeh recorded the incidence at 8.9%. This brings us to the very difficult question of intra uterine causation. Evans Jones et al. point out that: “most cases are due to trauma at delivery which is not necessarily excessive or inappropriate.” A significant contribution comes from Becker et al. (2002) who described a number of cases associated with an accessory rib or fibrous band: “narrowing of the supracostoclavicular space by anatomical variation, such as cervical ribs or fibrous bands, may be predisposing factors to brachial plexus lesions in infants even if the associated trauma is mild.” Becker’s work undoubtedly defines one congenital cause and we believe that this work could, and should, be extended by increasing use of ultrasonography in the new born. Natural History The extensive earlier writings on the extent of recovery by natural process are divided between those who find good or complete recovery in as many as 90% of cases from those who came to a rather more gloomy view, a division which probably reflects the severity of the injury Table 10.6 Significant relative risks – 1,060 cases. Dystocia
180.3
Asphyxia
10.8
Gestational diabetes
10
No caesarean
9
Birth weight >3.4 kg
8.5
Forceps – assisted delivery
6.9
No episiotomy
5.2
Breech
3.6
Not induced
2.8
Ventouse – assisted delivery
2.6
Age of mother >30
2
Hypertension By kind permission of Tavakkolizadeh 2007, Thesis.
1.4
within the populations studied. The most extensive review of earlier work is provided by Pondaag et al. (2004) who studied 1,020 articles. The criteria for inclusion in the study were as follows: the study had to be prospective; the study population had to be on a demographic basis; a follow up of at least 3 years, and assessment of outcome preferably by a predefined system. Whilst no one study was wide enough in scope to define the natural course of BLBP the evidence points to persisting residual defects in between 20–30% of children. Gjörup (1965) followed 103 patients for over 33 years, finding 22 with a poor result and 40 of those with useful recovery considered that they had some persisting significant disability. Eng et al. (1978) made detailed observations in a series of 135 children. There were persisting problems in 30% of them; the deformities of the shoulder and elbow proved resistant to conventional physiotherapy. A notable finding was the demonstration of neurapraxia in eight children with paralyzed muscles but a normal electromyogram. Hoeksma et al. (2004) followed a cohort of children through the period of recovery. They considered that the nerve lesions had fully recovered in two thirds of them. The course to recovery was studied at 6 months after birth in 276 of the 323 confirmed cases collected during the national census. “Complete” recovery was observed in 143 babies (52%); there was no recovery in six (2%). The findings for recovery of abduction at the shoulder in the 127 cases with partial recovery are rather dismaying. Abduction was full in only 17 of these infants; there was none at all in 15 and it was incomplete in 95. Bisinella and Birch (2003) prospectively studied the 74 children who attended the Peripheral Nerve Injury Unit during the period of the census, that is just over one quarter (26.8%) of the 276 census cases. The children were followed for a minimum of 2 years. There is no doubt that these infants had suffered rather more severe injuries than the census population as a whole. Eight of the children (10.8%) had complete lesions compared to 19 (7%) in the census population. Evans Jones and his colleagues recorded five cases of complete lesion complicated by Bernard–Horner syndrome (Narakas group 4); Three of these were studied by Bisinella (Table 10.7). Group 1 (28 Patients) No baby had operations on the plexus. Spontaneous recovery of biceps began at between
Table 10.7 Outcome of 74 children entered into the 12 month National Census (mean follow-up 32 months). Narakas group Full recovery Persisting defect. Operation for posterior No operation dislocation of shoulder
Operation on brachial plexus
Group 1 (n =28)
21
4
3
0
Group II (n = 38)
14
5
16
5
Group 111 (n = 5)
4
0
1
1
Group IV (n = 3)
0
0
0
3
9
20
9
Total 39 Drawn from Bisinella and Birch 2003.
442
1 and 4 months of age. Twenty one of the 28 children regained a normal or near normal upper limb. The remaining seven infants had very good elbow function but there was some deficit at the shoulder. Three of these required operation for shoulder deformity within the first 15 months. The procedures included one subscapularis lengthening for medial rotation contracture and two relocations of the shoulder. One of the posterior dislocations had probably occurred at birth. In these three operated cases the average active forward flexion and abduction was 100 degrees each (passive 150 degrees) and active rotation was to 30 degrees (passive 50 degrees) at 18 months after operation. The other four infants showing slow recovery achieved an average active forward flexion and abduction of 100 degrees each and 20 degrees of active lateral rotation. Group 2 (38 Children) In 33 of these recovery of biceps was evident between one and 6 months (mean 3 months). Nineteen (50%) did not come to any operation and all of these regained normal or near normal function. Fourteen children (36.8%) were operated for secondary pathology of the shoulder between seven and 17 months of age. Two had developed medial rotation contracture; there were seven posterior subluxations and five had posterior dislocation which probably occurred at birth. Four infants (11%) with no recovery of biceps by the age of 3 months and with neurophysiological investigations showing unfavourable degenerative lesions, were explored at a mean age of 6 months; repair of C5 and C6 was performed. In two of these presenting with medial rotation contracture subscapularis lengthening was performed at the same time as repair of the nerves. One child had exploration of the plexus at the age of 8 months because of lack of recovery even though the neurophysiological studies were optimistic. A favourable lesion was demonstrated and the nerves recovered. Group 3 (5 Children) Four patients showed recovery of biceps between 2 and 4 months of age and had an excellent outcome for hand, elbow and shoulder function. One of them was operated for posterior dislocation at 8 months of age. The brachial plexus was explored in one child and C5 and C6 were repaired. Group 4 (3 Children) All underwent nerve operation at a mean age of 4 months. In one case a neurolysis was performed and this infant had a fair recovery of hand and elbow function (Raimondi and elbow score 4), but shoulder function remained poor (Gilbert score 2). In the other two cases multiple ruptures of C5 to C7 and avulsion of C8 and T1 were found. These were repaired. Nine of the 74 children (12%) came to operation on the brachial plexus and no less than 19 (26%) for impending or actual posterior dislocation at the shoulder. Three more of the 74 children came to operation for shoulder deformity in later years after completion of the study. The outcome was considered good with a normal or near normal upper limb in
Surgical Disorders of the Peripheral Nerves
39 cases, (53%) whereas 29 (39%) had some residual deficit but a useful arm. Four infants (5%) had some function, three of these after repair of the brachial plexus. A very poor outcome was recorded in two children. The range of movement at the shoulder and of forearm rotation in the 33 children with “useful” or “some function” is set out in Table 10.8. We were rather taken aback by these results and in particular by the high incidence of posterior dislocation of the shoulder. Notwithstanding the increased proportion of more severe injuries in Bisinella’s study population against the census population, these findings reveal that recovery is by no means complete in just under one half of children and that posterior dislocation of the shoulder occurs in more than one quarter of children before the age of 32 months. Simon Kay, in a personal communication (2007), expressed his reservations about the term “full” recovery for he had found that many patients showed persisting defects in function. The evidence provided here with that which will follow goes a long way in support of Simon Kay’s thoughts on the matter. We now turn to analysis of the data assembled from the prospective study of shoulder function in 1,328 children seen between 1992 and 2007. The outcome is graded by the Mallet score. Two hundred and fifty two children were seen only once and they were discharged because their function was so good. In them, the Mallet score was 15. The results for the Mallet score have been set against the Narakas grade of lesion in the remaining 1,128 children (Table 10.9). This shows that most children remained with a perceptible and, at times, significant defect and that there was a correlation between the outcome and the severity of the nerve injury. Just over one third of the 1,128 children were given the maximum score of 15 points. It is important to remember that this score does not necessarily signify a normal shoulder. Sixty three adults with BLBP were seen because of increasing pain and deteriorating function at the shoulder. All of these patients had uncorrected posterior subluxation or posterior dislocation of the shoulder. Patients with long standing dislocation complained of a rather rapid loss of function usually in the third decade. Patients with untreated subluxation experienced, usually in early adult life, worsening pain from osteoarthritis between the anterior surface of the head of the humerus and the postero-inferior facet of the glenoid (Fig. 10.12). More detailed information is provided by the prospective records made in 860 children describing the ranges of movement at the thoraco-scapular, the gleno-humeral and the radio-ulnar joints between 1998 and 2007. Thirty per cent of these children presented with complete lesions, a much higher proportion than that seen in the census population and in the patients studied by Bisinella. These children remained under observation because the nerve injury was more severe or because operation proved necessary for shoulder dislocation or other palliation (Table 10.10). The normal posterior
150 (120–170)
5 (0–20)
Passive
40 (20–80)
60 (30–80) 60 (40–100)
100 (90–150)
Active
110 (90–150)
Passive
Inferior scapulohumeral angle
1*, useful function – 29 children; 2*, some function – four children.
60 (50–100)
2*
20 (30–80)
Active
160 (110–180)
Passive
Active
110 (90–140)
1*
Lateral rotation
Forward flexion
40 (20–40)
120 (90–170)
Passive
Posterior scapulohumeral angle
60 (40–80)
100 (80–140)
Active
Abduction
Passive
150 (100–170)
160 (90–180)
50 (20–80)
70 (20–90)
Active
70 (40–90)
80 (40–90)
Passive
Medial rotation
90
90
Active
Pronation
Passive
90
90
10 (0–30)
10 (0–40)
Active
Supination
80 (60–90)
90
Passive
Table 10.8 The mean, in degrees, of ranges of movement at the shoulder and of forearm rotation in 33 children with “some” or “useful” function (Range) (from Bisinella and Birch 2003).
Birth Lesions of the Brachial Plexus 443
444
Surgical Disorders of the Peripheral Nerves
Table 10.9 The initial and final Mallet score in 1,128 children seen in the years between 1992 and 2007. The results are given by median and (mean). The median age at the first record was 13 months. The median and (mean) length of follow up is given in months. Narakas Number Initial Final Duration group of children score score of study I
249
8 (8.1)
13 (12.8)
43.5 (53.3)
II
518
8.5 (8.6)
13 (12.5)
44 (62.3)
III
217
7 (8)
13 (12.2)
41.5 (51.2)
IV
144
7 (7.3)
11 (10.9)
55.5 (63.3)
scapulo-humeral angle is at least 70 degrees and the reduced value of this measure, evident in all groups of the children, came as a surprise. There are two possible explanations: the first is persisting retroversion of the head of the humerus, and
the second is contracture of the capsule and the ligaments about the gleno-humeral joint. There is no doubt about the persisting weakness of the medial rotator muscles in many of the children in groups 3 and 4.
10.4 Recovery in the Complete Lesion The information is reasonably comprehensive in 291 children with complete lesions, who were followed for at least 5 years between 1991 and 2004. The age, at first examination, was 20 months (mean). The final data were collected at a mean age of 83 months. The final measures of shoulder function are described above.
Fig. 10.12 Untreated complex subluxation in a 35 year old labourer who complained of deterioration in function with worsening pain.
Birth Lesions of the Brachial Plexus
445
Table 10.10 Ranges of active movement in 860 children seen 1998–2007, by degrees median and (mean) for the initial (a) and the final (b) observations (%) percentage. Narakas Group
1
2
3
4
Number of children (%)
190 (22.1)
410 (47.7)
163 (19)
97 (11.2)
Forward flexion a. b
100 (109.8) 150 (128.6)
100 (103.6) 120 (118.9)
90 (101.9) 100 (109.7)
80 (80.4) 80.(79.2)
Abduction a b
90 (87.2) 120 (108.6)
90 (87.6) 100 (105)
90 (89.8) 100 (98.1)
60 (64.8) 70 (72.8)
Lateral rotation a b
30 (29.5) 40 (41.6)
30 (27.6) 35 (37.5)
20 (22.1) 30 (37.9)
30 (35.4) 30 (34)
Medial rotation a b
90 (73.2) 90 (80.7)
90 (73.9) 90 (79.9)
80 (68.1) 90 (78.3)
70 (62.6) 80 (71.7)
Pronation a (318 children) b
90 (100.4) 90 (88.3)
90 (93.1) 90 (84.6)
90 (90.2) 90 (85.9)
90 (71.6) 70 (57.4)
Supination a (297 children) b
30 (43.5) 80 (64.8)
40 (39.9) 70 (64)
47.5 (46.8) 70 (60.9)
0 (30.2) 55 (46.2)
The passive range of the inferior scapulo humeral angle a 150 (138.6) 140 (131.7) 120 (125.1) b 150 (144.8) 150 (141.9) 150 (1238.3) The initial records were made at an average age of 20 months, the final records at an average age of 85 months.
10.4.1 Group 3: 177 Children Quite significant defects were recorded in all segments of the upper limb within this group. Just over one half were given full scores for the elbow, the wrist and the hand. No less than 395 operations were performed in these children. The plexus was repaired in 44, the shoulder was relocated in 125, and 53 more were operated because of shoulder contractures or weakness. One hundred and ninety five operations were required to improve the posture of the forearm, wrist, the fingers and the thumb. The Elbow One hundred and fifty-three children reached a score of 4 or 5 at final review, 24 were scored at 3 or less; of these, five elbows were scored 2 or less. The Hand Whilst 160 children were scored at Raimondi 4 and 5 at final attendance musculo-tendinous transfers contributed to this outcome in 50 of these. The hands of 15 children were scored 3, all of whom had undergone musculo-tendinous transfer. Two hands were scored less than 3.
10.4.2 Group 4: 114 Children Two hundred and seventy seven operations were performed in these children. The plexus was repaired in 45, 68 shoulders were relocated and 31 other operations were done for weakness or other deformity at the shoulder. The remaining operations were directed towards improvement of function at the elbow and in the forearm, wrist and hand (Fig. 10.13).
120 (126.1) 130 (125.4)
The Elbow Sixty two elbows were scored at 4 or 5, and 52 at 3 or less. Seven elbows were flail. The Hand A Raimondi score of 5 was recorded for 33 hands, and a score of 4 in 29 more. The hand was scored at 3 in 20, and less than this is 42. Eight hands were flail, with no function at all. One half of the 62 hands given a Raimondi score of 4 or 5 achieved this by spontaneous recovery. The first signs of recovery in these children included improvement in the texture, the color and the temperature of the skin possibly indicating early recovery of vasomotor tone. These changes usually preceded any active movement by some weeks. Yang and Smith (2003) have kindly provided their findings from a study of 86 children with Group 4 lesions who were treated during the years 1984–1991. The nerves were not explored in 39 of these children because they came too late or because there was clear evidence of recovery of function in the hand. Thirty two of these 39 children recovered useful or good function in the affected hand by spontaneous recovery. The Raimondi hand score was 5 in five children, 4 in 14 and 3 in 13 more; there was no detectable active movement of the wrist or the digits before 12 weeks of life in eleven of the infants and in four more there was no sign of recovery until 24 weeks had passed. An elbow score of 4 or 5 was given in 15 cases and a Mallet score of 13, 14 or 15 to nine children. Neurophysiological investigations were carried out in 10 of these children. These showed extensive axonopathy and loss of peripheral conduction in eight children, findings consistent with postganglionic injury. The preservation of sensory conduction in the distal ulnar nerve pointed to preganglionic injury in the other two children. The extent of recovery
446
Surgical Disorders of the Peripheral Nerves
Fig. 10.13 A 5 year old boy with a group 4 lesion. At the age of 7 weeks ruptures of C5, 6, 7 and C8 were repaired using 16 grafts. The avulsed T1 was repaired by intraplexual transfer and the suprascapular nerve by transfer to accessory nerve. Shoulder 3+,14; elbow 4; hand 4.
suggests that this preganglionic injury was one of conduction block, and not avulsion. Magnetic resonance imaging revealed no intraspinal lesion in the five children who were examined. These findings must deter clinicians who are inclined to repair of C8 and T1 in the absence of strong evidence that these nerves have been separated from the spinal cord, for only in these can there be no spontaneous recovery. It is in this group that neurophysiological investigations are particularly helpful for findings consistent with post ganglionic injury usually indicate that there is a strong possibility of spontaneous recovery. Table 10.11 sets out the course to
Table 10.11 Evolution of recovery in two Group 4 cases. Case 1 Deep axonopathy of dorsal and ventral root fibers
recovery in two of these children. The first case is typical of recovery in post ganglionic axonopathy. The second case is more difficult to understand; it is possibly consistent with prolonged conduction block of the dorsal rootlets. The outcome after nerve repair in 89 children with complete lesions in whom repairs were performed from 1988 to 2002 is set out in Table 10.12. It proved necessary to relocate the dislocated shoulder in 42 of them. The final outcome measures were made after relocation of the shoulder but they exclude any further gains following supplementary musculotendinous transfer. Self mutilation is uncommon. Disproportionate impairment of A-delta and C-fiber function was demonstrated in a 9 year old girl who presented because she was constantly chewing her fingers and had sustained unnoticed burns. There had been no recovery for the first 8 months of life. There was a Bernard–Horner sign. Later recovery into the shoulder and elbow was good. The digital extensors and flexors were powerful and there was intrinsic balance. Localisation and proprioception were present, she was able to recognise shapes and textures but the hand was dry. The sympathetic nerves had not recovered and quantitative sensory testing revealed very poor recovery of the finest myelinated and non myelinated fibers. Self mutilation may also be caused by a rupture confined to the dorsal roots. One 11 year old boy presented with severe trophic disturbance of the little and ring fingers although recovery of the intrinsic muscles was good. A small sensory action potential was recorded from the ulnar nerve. The sensory divisions of the second, third and fourth intercostal nerves were transferred to the ulnar nerve. The parents
Case 2
Axonopathy ventral root fibers. Possible conduction block dorsal root fibers 6
9
15
B favourable. Polyphasic and normal units
–
–
Recruitment improved
C
Isolated broad potentials with fibrillation Potentials
Moderate recruitment of polyphasic units. No spontaneous activity
Age at examination In months
2
4
10
19
3
NAP median and ulnar
0
0
0
50%
Symmetrical
EMG: C5
C
Polyphasia, nascent units
B favourable
B favourable
C6
C
Polyphasia, nascent units
B favourable
B favourable
No spontaneous activity
B favourable
C7
C
C
C8
C
C
T1
C
C
MR scan aged 8 weeks (Dr. Mary Rutherford): no intraspinal lesion. STIR signal in muscle increased. Posterior dislocation of the shoulder. Function at 21 months: shoulder 1+ (dislocated) 10, elbow 4 (false tumor biceps), hand 4. Sympathetic recovery.
C C C
MR scan aged 12 weeks (Dr. Mary Rutherford) Increased STIR signal in muscle; no intraspinal lesion. Posterior dislocation of shoulder. Function at 22 months; shoulder 1+ (dislocated) 10. Elbow 4, hand 4. Sympathetic recovery.
Birth Lesions of the Brachial Plexus
447
Table 10.12 The outcome after nerve repair in 89 children with complete lesions (1998–2002). All were followed for a minimum of 60 months. The values are given as median (mean). Narakas Group III Narakas Group IV (44 children) (45 children) First Last First Last Mallet score
5.1 (5.6)
11 (11)
5 (5.5)
9.5 (9.8)
Elbow score
1 (108)
4.5 (4.2)
1 (0.8)
3 (3.1)
Hand score 3 (3.2) 5 (4.3) 1 (1.3) 3 (2.9) The outcome includes the 42 cases in whom a dislocated shoulder was reduced, but it excludes any further improvement gained by later musculotendinous transfers.
reported improvement in the color and texture of the digits by 6 months. At 6 years from operation he was able to localise light touch with accuracy and sweating returned. There was gnosis. He now works full time as a car mechanic. Improvements in color, growth and protective sensation have been observed in six other children with a similar injury who were treated by this method. In three of these a perceptible loss of intrinsic muscle function, notably of the ability to bring the little to the ring finger which required correction by musculo-tendinous transfer. Evidently some working efferent axons were cut during the nerve transfer. It is clear that at least one third of children with BLBP are left with significant or severe defects in function within the afflicted upper limb. Poor recovery through the injured nerves accounts for about 15% of these whilst posterior dislocation of the shoulder accounts for as many as one in four. The evidence presented so far summarises mechanical defects, but what are the effects on the child’s development? A common observation made by parents is that their child is clumsy, that they are unable to run properly and that the impaired limb does not fully integrate in spite of good neurological recovery. The failure to integrate the afflicted limb into normal activities may be explained by defective recovery of the deep afferent pathways from the muscle spindles and the tendon organs and we feel that this impairment, which calls for further study, contributes to cocontraction. Impairment of function within the lower limb and defects in growth of the ipsi lateral foot raises the possibility of a partial Brown–Séquard injury. Strömbeck et al. 2007a, b recently published two valuable long term studies of 70 cases seen at the Karolinska Institute. All the patients had been examined, on average, 13 years earlier. The children (many, by now, young adults) were well adjusted, although one quarter had measurable difficulties in normal daily activities of life. Elbow function had worsened because of increasing flexion deformity. Earlier operations for the correction of medial rotation contracture at the shoulder provided lasting benefit in 27 patients. Grip strength was maintained at an average 80%. Changes in limb dominance
were similar to those described by Yang et al. (2005). In the second paper Strombeck (2007b) relates that electromyography (EMG) showed a decline in the motor neurone pool for the deltoid muscle “on the one hand representing collateral innervation and on the other, motor neurone loss.” Cutaneous sensitivity was generally good. Elevation of thresholds for cooling and for warming were correlated with more severe EMG changes. Two point discrimination in the dermatome of C8 closely co-related with functional sensibility. Strombeck and her colleagues offer the important suggestion that there is a loss of motor neurones in the central pool as the children grow. This may be a factor underlying the deterioration in shoulder function which is common in late adolescence and early adult life. Bellew et al. (2000), from Leeds, demonstrated a significant relation between defects in personal and social behavior and the extent of lesion and they demonstrated that operations, hospital stay and continuing dependence upon “health professionals” are harmful. The implications of this important paper must be clear to all engaged in the treatment of these children. “If it were done when ‘tis done, then ‘twere it well it were done quickly.” At times operation is unavoidable but it is important to do as much as is reasonably possible during one admission: so it is that we combine repair of the nerves with relocation of the shoulder or correction of medial rotation contracture at one operation; or combine correction of supination deformity with muscle transfer to rebalance the wrist, restore extension of the wrist, the digits and the thumb during the same intervention. In two group 4 children who endured a number of operations in early years, there was clear psychological distress by the time they reached early adult life, distress to such an extent that psychological and psychiatric advice was sought.
10.5 Treatment The first step is to provide the parents with information about the nature and the extent of lesion and of the likely outcome. Parents appreciate a simple diagram showing the brachial plexus and outlining the nature and level of the lesion, which is written out for them by the doctor. This is far preferable to an anonymous “hand out” of printed paper. Next the parents are advised of the risks of contracture, above all of deformity at the shoulder. We actively discourage “hand outs” of material which do not relate directly to the child being treated. It is extremely important that all those involved in the care of the child avoid giving conflicting information. Much damage can follow from ill-informed interpretation of the record of operation. It is the first duty of the surgeon to explain what was found, what was done, why it was done and what may follow. It is for this
448
Surgical Disorders of the Peripheral Nerves
Fig. 10.14 Assiduous work by her mother corrected significant fixed deformity at the shoulder in this child. The scapula is held with one hand, the other stretches out the gleno-humeral joint into elevation, lateral and medial rotation.
reason that our operating records are sent to the referring clinician and to the family practitioner and to nobody else. The parents should be advised of the diagnosis once this is known. There will often be anger, guilt and in many cases legal action for compensation. One of the advantages of separate centres for treatment is to relieve clinicians from these emotional burdens and to enable all involved to ensure that the child is given the best chance of recovery. That burden must also be shouldered by the parents. Parents are involved in the treatment of the child from the outset. Regular and gentle exercises may prevent fixed deformity (Fig. 10.14). There is no place for vigorous manipulation. If gentle stretching movements cause pain then there must be either fracture or posterior dislocation of the shoulder. The role of the therapist and attending doctor is to teach and then to monitor progress. We have seen too many examples of fixed deformity resulting from parents sitting passively at home awaiting the occasional visit by the community physiotherapist. Exercises must be performed, gently, for 2 or 3 min before every feed. Both upper limbs are worked simultaneously. For medial and lateral rotation the arms are held against the side, with the elbow flexed to 90 degrees. The forearms are then brought onto
the child’s body and then moved into full lateral rotation. Then, the upper limbs are brought gently into full elevation. This manoeuvre is repeated with the arms at 90° of abduction. The inferior SHA is maintained by holding the scapula against the chest whilst abducting the arm; the posterior SHA by holding the scapula against the ribs whilst flexing and medially rotating the arm. Much fixed deformity can be prevented by this simple regime. It is essential to warn parents about the significance of pain for this may be the first indication that the gleno-humeral joint is no longer congruent.
10.6 Supplementary Investigations 10.6.1 Neurophysiological Investigations (NPI) These methods have already been described (Chapters 5, 6) but it seems necessary to restate certain principles. The whole pathway from the periphery to the central nervous system may be examined by somato-sensory evoked potentials in which the skin of selected digits is stimulated and records are
Birth Lesions of the Brachial Plexus
449
Table 10.13 The clinical interpretation of electrodiagnostic findings. Grade NAP EMG
Lesion
A
Normal
No spontaneous activity. Reduced number of normal motor units. Increased firing rates
Conduction block
B favourable
Normal or >50% of uninjured side
Relatively good motor unit recruitment. Mixture of normal and polyphasic units suggesting some collateral re-innervation
Modest axonopathy consistent with useful recovery
B unfavourable
Absent or <50% of uninjured side
Normal units few or absent. Widespread polyphasia indicating collateral reinnervation. No spontaneous activity
Significant axonopathy. Recovery particularly poor for C5.
Extensive spontaneous activity. Sometimes poor Absent; if present recruitment of nascent or small polyphasic units indicative of preganglionic injury Drawn from Smith 1996, 1998, and from Bisinella Birch and Smith 2003. C
made through electrodes placed over the skin of the scalp or the neck. Function in the finer afferent fibers can be measured using contact heat evoked evoked potentials (CHEPS); function in the postganglionic sympathetic afferent fibers by measuring cholinergic sweating. The central pathways between the injured spinal nerve and the central nervous system are examined by measuring conduction between the spinal nerve exposed at operation and electrodes placed over the skin of the scalp or the neck. (Chapters 7, 9) The peripheral pathways and the target organs are studied by recording conduction within the peripheral nerves: of the larger afferent fibers by sensory action potentials; of the somatic efferent fibers by measuring motor conduction; and of conduction within the whole nerve by measuring compound nerve action potentials. The innervation of muscles is examined by electromyography. The extent of skin innervation is measured by Quantitative Sensory testing, but this is possible only in children aged 4 years or more. Detection of conduction across the lesion involves stimulating and recording above and below the lesion exposed at operation, and also by studies of the whole pathway. Two of these methods are particularly important in the analysis of the birth lesion. 1. Examination of Central Conduction is important in detecting the hidden intradural injury (Types 5 and 6) and also in those cases where rupture is combined with elements of intradural injury (Type 4). 2. Conduction in the Peripheral Nerve Trunk: Electro myography. We turn now to an appraisal of the methods developed and applied by Shelagh Smith and latterly Peter Misra in the examination of the peripheral nerves and the target organs. These are particularly helpful. The clinician using information provided by neurophysiological investigation (NPI)
Severe axonopathy consistent with no, or poor, recovery. Preserved NAP suggests preganglionic injury
Table 10.14 Outcome predicted by electrodiagnostic grade. Spinal Grade of lesion Outcome nerve Gilbert score Mallett score C5
A
5+
14–15
B favourable
3+
12–13
B unfavourable
2+
10–11
Elbow score C6
A
5
B favourable
5
B unfavourable
4 Hand score
C7
A
5
B favourable
5
B unfavourable 3 or 4 Grade C predicts no, or poor, recovery. Drawn from Bisinella et al. (2003).
needs to understand the method; their interpretation must always consider clinical findings. Nerve action potentials are measured from the median and ulnar nerves by stimulating at the wrist and recording at the elbow. The deltoid (C5), biceps (C6), triceps or forearm extensors (C7) forearm flexors (C8) and 1st interosseous (T1)muscles are sampled by EMG. The lesion is graded for each spinal nerve, according to the degree of demyelination and axonopathy (Table 10.13). A prediction is made about the likely extent of recovery on the basis of the neurophysiological grade (Table 10.14, Fig. 10.15a–c). NPI are usually pessimistic in the first weeks of life, when Wallerian degeneration is at its height and before there has been much in the way of regeneration. Serial NPI are restricted to cases of post ganglionic injury of C8 and of T1 and also for some lesions of C5 where there is real doubt about progress.
450
Surgical Disorders of the Peripheral Nerves
The method was tested by Bisinella (2000, and by Bisinella et al. (2003) who analyzed findings in 350 spinal nerves in 126 children with Groups 1, 2, and 3 lesions in whom elbow flexion had not recovered by the age of 3 months. Operation was done in 53 babies where poor progress was associated with NPI grades of unfavourable B or C. In these rupture or avulsion was found in at least one spinal nerve. Operation was not performed in the 73 children (199 nerves) with grades A or B lesion. Predictions for recovery were matched against outcome at a mean of 4.3 years The predictions were confirmed in 92% of 65 C6 lesions and in 96% of 45 C7 lesions. The predictions for C5 (65 nerves) were confirmed in a smaller proportion (78%) which may be explained by inability to record compound nerve action potentials for C5, and by the complication of posterior dislocation. Recovery was good in 33 of 38 children (86%) whose shoulders were congruent, and in whom grade or A or B
favourable lesions were revealed. Recovery was good in only 11 of 35 children (34%) with equivalent lesions of C5 in whom operation for the dislocated shoulder proved necessary. Recovery of shoulder abduction was apparent at a median of 4 months and in biceps at a median of 5 months. Recovery was better than predicted in five of nine B unfavourable lesions of C6. Useful elbow flexion returned in the remaining four, but there were defects in supination. Recovery exceeded prediction in two of the five B unfavourable lesions of C7 lesions whilst the other three came to muscle transfer for wrist extension. The diagnosis of prolonged conduction block (Type A), which was found in 40 nerves in Bisinella’s series, is important because these lesions are clinically indistinguishable from those which are more severe. Complete spontaneous recovery is likely and operation is unnecessary. We think that Eng et al. (1978) first made this observation in eight babies.
Fig. 10.15 Three examples of pre-operative NPI investigations.(a) C5 and C6 were explored at the age of 4 months when the findings were more favourable than were predicted. The nerves were left alone and the posterior dislocation of the shoulder was corrected at that operation. By 2 years the shoulder scores were 12, 2+; elbow 5; hand 5, whilst the active forward flexion and abduction of the shoulder was no more than 70°. Probably, the fifth nerve should have been repaired. (b) Operation, at the age of 6 months, confirmed recovering lesions for C5 and for C7 and a rupture at C6. No repair was done, and the dislocated shoulder was reduced during the same operation. Two years later, shoulder scores
were 15, 5+; elbow score was 4. The hand score was 5 and active forward flexion and abduction was measured at 160° with active lateral rotation at 45°. (c) The findings of NPI seemed particularly unfavourable but these investigations were done at the age of 8 weeks. The plexus was explored at the age of 12 weeks, when it seemed that there were signs of recovery so that no repair was done. The dislocated shoulder was reduced at that operation. By 2 years the shoulder scores were 12, 2+, the elbow score was 5, and the hand score was 5. Flexion and abduction were 70°.
Birth Lesions of the Brachial Plexus Fig. 10.15 (continued)
451
452
Surgical Disorders of the Peripheral Nerves
Fig. 10.15 (continued)
We perform neurophysiological investigations as early as possible in birth lesions caused by breech delivery and also in Group 4 lesions. The presence or absence of peripheral conduction within the first few weeks of life is helpful in distinguishing between pre and postganglionic injury. In the partial lesions associated with cephalic delivery the investigations are most informative when they are done between ten and 12 weeks of life in infants whose progress is slow. It is at this time that a reliable prediction can be made about the
likelihood of recovery. This is helpful in deciding whether to operate upon the nerves or not. There are limitations to the method, particularly so during the first 8 weeks of life because the electromyographic findings are often very gloomy. During that time the evidence of axonopathy is strong yet there may be only slight evidence of regeneration. Such early examinations may prove to be unduly pessimistic. The demonstration of significant axonopathy in C5 provides a clear indication that shoulder function
Birth Lesions of the Brachial Plexus
will be poor. On the other hand, a type B favourable or even unfavourable lesion in C6 and C7 is usually followed by recovery which is as good, if not better than that seen after repair of the damaged nerve. If the investigation in children with complete lesions points to a post ganglionic injury to C8 and T1 then clinicians must be extremely cautious about proceeding to resection and repair of the lower trunk. Other evidence about the state of affairs at the interface between C8 and T1 and the spinal cord must be sought. It is now our practice to consider exploration only in those cases where significant axonopathy involving C5 is revealed and to hold off from exploration unless a Type C lesion is shown for C6 and C7, or unless a complete preganglionic injury involving C8 and T1 is demonstrated.
10.6.2 Imaging The role of myelography and computerised tomographic (CT) scan has been considered by Gilbert (2005) and by Slooff and Blaauw (1995). We avoid myelography and CT with contrast enhancement preferring magnetic resonance imaging (MR scan). Dr Mary Rutherford (The Hammersmith Hospital) does the investigation under light sedation, and she has found increased STIR signal from denervated muscle, lesions within the spinal canal and dislocation of the shoulder. Krone (2005) suggests that ultrasonography at birth, before ossification of the skeleton, may provide valuable evidence about intradural lesions. Alas, we have no experience with this important idea.
10.7 Nerve Operations The year 1903 was a golden year for surgical endeavor with the birth lesion. In that year Harris and Low, at St Mary’s Hospital in London, described intra plexal repair by transfer of the distal stump of C5 to the intact C6 or to C7; in Manchester Thorburn (1903) extended his experience with suture of the adult plexus to that of the infant, and Kennedy, assistant surgeon to the Western Infirmary in Glasgow, described three cases of suture. He considered that the chief factor in causation was: “forcible depression of the shoulder, while the head is bent to the opposite side and rotated.” He distinguished between conduction at the neuro-muscular junction, evoked by faradic stimulation, from the direct response evoked by galvanic stimulation over the muscle belly and wrote: “if, however, after 2 months no responses can be got in the muscles with the faradic current, although of course, the galvanic current evokes good contractions, it is safer to proceed with the operation than to put off further time in the hope that recovery will eventually be the result.”
453
One baby was operated at the age of 9 weeks, the neuroma of the upper trunk was excised and the stumps directly sutured. By 9 months after operation function was good: “the child was able to abduct the arm more than 90 degrees, was able to flex the forearm almost normally, and was using the arm constantly. The inward rotation had disappeared. The most defective movement remaining was supination, but this movement in an imperfect degree was also noticed.” The child is illustrated in Fig. 2 in Kennedy’s paper, which shows a result which would gratify most present day surgeons. The early interest fell away because of poor results, because of the risk of death during or just after operation and finally from the realisation that the only children who regained no function at all were those who had been operated. Thanks to the initiative of surgeons on the Continent of Europe and in Japan in the 1970s there was a revival of effort towards surgical repair. This is often ascribed to the “advent of microsurgery.” This is surely incorrect. It is much more likely that improvement in methods of diagnosis, and in particular in the ability to distinguish between the pre- and postganglionic lesion, enabled this advance. Morelli et al. (1984) published their findings in 57 children operated between 1978 and 1984, 28 of whom had been followed for 3 years. We think it highly unlikely that this paper would be accepted for publication by most modern surgical journals, for its scope is wide, it challenges widely accepted practice and it is “merely” a clinico-pathological study of a case series. Morelli and his colleagues examined the excised tissues, and they illustrate the great disorganisation of the surviving or regenerating nerve fibers: “Una profonda anarchia di dispotsizioni delle fiber nervose presenti nel cosiddetto ‘callo nervosa’.” Their outcome measures were rather strict; three children regained near normal levels of function; 22 more recovered useful function opening out the possibilities of further improvement by musculo-tendinous transfer. Two of the cases described in this paper are particularly interesting. In the first (case 5, Fig. 10) there was rupture of C5 and C6 with avulsion of the remaining nerves. The whole plexus was repaired, basing grafts on the stumps of C5 and of C6. This child regained useful function throughout the upper limb. In the second case (case 6, Fig. 12) there was rupture of C5, C6 and C7, with avulsion of C8 and T1. The whole plexus was repaired and the child regained good function throughout the upper limb. Wrist extension was restored by subsequent transfer of pronator teres.
10.8 The Indications for Operation These revolve around the cause of the lesion, its extent, and the tempo of recovery. There must be an argument for urgent exploration in the more severe breech lesions when they are complicated by
454
phrenic palsy. We have repaired three nerves which were ruptured where they crossed the scalenus anterior in breech lesions. These babies are usually premature, small, and often so sickly that we were able to perform early exploration in only one infant. There is also an argument for urgent exploration, as soon as the infant’s general condition permits, in some of the group 4 cases. Perhaps improvements in imaging and in particular, the wider application and refinement of ultrasonography used within the first few days of life, may unmask discontinuity of spinal nerves or of bundles within those nerves, or separation of the nerve from the spinal cord by displacement of the dorsal root ganglion into the posterior triangle. The timely use of these methods in the first few days of life may resolve the perplexities about the likelihood of recovery in some of these infants. Pitt and Vredeveld (2005) have suggested that NPI, within the first month of life, will demonstrate conduction block and the extent of axonopathy. They may be right; on the other hand early electromyography may underestimate the potential for spontaneous recovery. The Evolution of the Paralysis In favourable lesions the parents report active grasp at 2 weeks. Flexion at the shoulder and elbow follows at between 6 and 8 weeks. Return of the grasp within 4 weeks virtually guarantees a useful hand. Recovery is slower and more irregular in Groups 3 and 4 lesions. Some of these children show recovery into the shoulder and arm by between three and 6 months but never regain worthwhile hand function; in others hand function returns but the shoulder remains poor. The “middle segment lesion” (Brunelli and Brunelli 1991), is not uncommon; shoulder and hand function appear before elbow control and wrist extension.
10.8.1 The Incomplete Lesion: Groups 1 and 2 Gilbert and Tassin (1984) found that failure of recovery into biceps by the age of 3 months predicted a poor outcome. Delayed recovery of elbow flexion was associated with mediocre function at the shoulder in nearly one half of the 29 children studied by Smith and colleagues (2004). Waters (1999) showed that recovery into biceps before the age of 3 months was followed by complete recovery. Return of elbow flexion after 3 months was associated with persisting defects in function at the shoulder and elbow. As Gilbert (1995) points out these cases place the clinician in a quandary. It is a very hard matter to suggest to the parents, overjoyed by this belated recovery, that the child’s future best interests might be best served by exploring and repairing the upper trunk. The view that failure of recovery of biceps by 3 months is an indication for operation is widespread. Some years ago Alain Gilbert explained his thinking about biceps recovery to one of us during a conversation at the 1994 Heerlen Symposium.
Surgical Disorders of the Peripheral Nerves
He considered that slow return of elbow flexion should be seen as an indicator of poor recovery generally, most especially at the shoulder. Since some clinicians lack skill in examination of the shoulder it seems easier for them to monitor elbow flexion (C6) as an indication of progress at the shoulder (C5). It is possible that this admirable advice has been taken too literally. Perhaps those clinicians unable to examine the shoulder in the infant should learn to do so, or altogether desist from treating them. There are a number of pitfalls lying in wait for those who rely on late return of elbow flexion: 1. Prolonged conduction block underlies prolonged paralysis in as many as 15% of spinal nerve lesions. 2. The biceps muscle may be damaged or even torn apart during difficult delivery. 3. Shoulder movement is the key to function in the upper limb and clinicians should be more concerned about poor recovery into the shoulder than with tardy return of elbow flexion. Too many cases of posterior dislocation have remained undetected for months or even years! The clinical appreciation of recovery in C5 is greatly hindered by dislocation. It may be impossible to say how much loss of movement is caused by the nerve injury and how much comes from mechanical obstruction. Significant axonopathy in C5 predicts a poor recovery of the shoulder. The Toronto school (Michelow et al. 1994, Clarke and Curtis 2001, Curtis et al. 2002) developed a scoring system which measures recovery at different segments of the upper limb. Combining the scores for return of elbow flexion with extension of the elbow, wrist, thumb and fingers provides an accurate prediction of recovery. As Kay (1998) observed: “the system of Michelow and Clarke certainly leads to a lower error rate and helps to avoid unnecessary operations.” Nehme et al. (2001) showed that the prognosis is reliably predictable by three factors: birth weight, involvement of C7 and the tempo of recovery in biceps.
10.9 The Operation The methods used for exposure and for repair of the nerves is related in Chapter 7. Scarring in the infant neck is often worse than it is in the adult. Blunt dissection is dangerous. After induction of anaesthesia, recording electrodes are attached to the skin of the scalp and neck and somato-sensory evoked potentials are recorded from the median and ulnar nerves at both wrists. The whole of the upper limb and one lower limb are prepared. The external jugular, suprascapular and transverse cervical veins are ligated cautiously because of the proximity of the subclavian vein. The phrenic nerve must be protected, for it is often deviated laterally and involved in the
Birth Lesions of the Brachial Plexus
455
neuroma of the upper and middle trunks. Phrenic nerve palsy is an extremely serious complication and we inflicted it on two occasions. In one case the nerve was ruptured by a sharp toothed retractor. The subclavian artery requires very careful handling. Double lesions have been seen, including one distal rupture of the musculocutaneous nerve, three lesions of the lateral or medial cords and two ruptures of the medial cutaneous nerve of forearm. The clavicle is drawn up and down by a nylon tape. It should not be cut.
a
b
Fig. 10.16 Three examples of intra operative conduction studies.(a) Even in clear cut ruptures, Type 3, conduction across the lesion is invariably detectable. In this case an SSEP was recorded by stimulating the musculocutaneous nerve distal to a rupture of the upper trunk, which was grafted. (b) Somatosensory evoked potentials (SSEP’s) from suprascapular nerve stimulated distal to lesion. The trace (right) was strong, and the lesion was not treated. The left trace shows abnormality in form and amplitude. The neuroma of the upper trunk was resected and grafted. (c) Examples of abnormal traces seen in Type 4 lesions: the wave form is altered, amplitude diminished, and latency increased. Repair by graft of such lesions is usually disappointing.
c
The findings for each spinal nerve are recorded and classified. Their appearance and texture are noted. Evidence of conduction is gathered by stimulating proximal to the lesion and recording from the scalp, by proximal stimulation and noting the muscular response, by stimulation distal to the lesion and recording from the scalp, again noting the muscle response. In partial lesions particular attention is devoted to the analysis of the suprascapular nerve (Fig. 10.16a–c.).
456
Surgical Disorders of the Peripheral Nerves
10.9.1 Methods of Repair Ruptures are repaired by graft but the proportions of the infant make it a tight matter to secure adequate donor nerves. The injured upper limb provides the medial and lateral cutaneous nerves of forearm and the superficial radial nerve, sufficient for two ruptures. The sural nerve is needed in more extensive repairs; sometimes both are required. In cases where rupture of the upper nerves is associated with avulsion of the lower nerves the lower trunk is reinnervated by one or two of the post ganglionic stumps (intraplexal transfer); the ventral root may be reinnervated from adjacent post ganglionic ruptures, or by transfer to the spinal accessory or a pectoral nerve. Conventional nerve transfers may be useful in type 4 lesions of C5, C6 and C7 where there is some regeneration but the integrity of the central stump is in doubt. Transfer of the spinal accessory nerve is not an innocent procedure in the infant since it leads to a measurable defect in control and growth of the scapula. A bundle from the intact ulnar nerve may be transferred to the nerves to biceps, triceps or the wrist extensors. We do not countenance transfer of the phrenic nerve. It is better to reinnervate the ulnar nerve in cases of avulsion of C8 and T1 rather than to use it as a graft for contra lateral C7 transfer. Blaauw et al. (2006) have condemned hypoglossal nerve transfer: the deficit is unacceptable and there is surprisingly poor reinnervation of the target muscles. Each case must be taken on its merits. Good function at the shoulder is the pre-requisite for a useful upper limb and every reasonable effort should be made to define the prognosis for C5. The results of palliative muscle transfers are far inferior to those seen following good regeneration for this nerve. Pre-operative neurophysiological predictions are more reliable than intra operative studies. In some cases where C5 was graded as B unfavourable before operation we recorded good SSEP’s by stimulating the suprascapular nerve distal to the lesion and chose to let well alone. This was a mistake; recovery was as predicted. Associated posterior dislocation of the shoulder adds to the complexity. Our colleagues Marco Sinisi and Peter Misra have refined methods of intra-operative analysis by recording compound motor axon potentials from deltoid and supraspinatus (as proposed by Laurent et al. 1993) and concentric needle examination of deltoid, supraand infraspinatus muscles. This helps to discriminate between loss of movement caused by dislocation and that produced by the injury to the nerve. Nerve repair is combined with relocation if the findings are gloomy. In other cases when evidence points to a more favourable outlook for the nerve, this is left alone. Relocation is followed by improvement in function which is, at times, dramatic. Selective reinnervation of the avulsed ventral root and of the dorsal component of the spinal nerve is applied wherever possible. The undisplaced avulsion (type 5) is difficult. The
Fig. 10.17 A 9 year old boy with a group 4 lesion. Rupture (type 4) of C5, avulsion of C6, C7, C8 and T1. Repair, at 12 weeks of age, included grafting from C5 to the upper trunk and to the ventral roots of C8 and T1. The spinal accessory nerve was transferred to the ventral root of C7. Subsequent extensor to flexor transfer was effective.
spinal nerves must be traced up to the foramen and every reasonable attempt made to confirm absence of central conduction. We leave these lesions alone; useful spontaneous recovery has been seen in more than one half of them. Whatever methods are used it has to be said that the outlook remains poor in cases where there is only one post ganglionic rupture, with avulsion of the other spinal nerves. A child with a flail elbow and defective control of the shoulder makes little use of a hand with limited grasp (Figs. 10.17 and 10.18).
10.9.2 Post-operative Care The integrity of the phrenic nerve must be confirmed before closure of the wound. A plaster of Paris splint is applied so
Birth Lesions of the Brachial Plexus
Fig. 10.18 Repair of a ruptured C5 with avulsion of the other nerves at the age of 10 weeks restored useful flexion at the elbow, the wrist and the fingers in this 9 year old boy who was well adjusted and who was a skilful canoeist.
that the mother can tend to the child without undue difficulty and clean the skin about the neck and chest. This plaster is retained for 3 weeks. After this the parents return to regular gentle stretching of the joints of the upper limb.
10.10 Results of Nerve Operations Carlstedt and Strömbeck (1995) published the results of a trial of operated versus non operated cases of the partial lesion. This important paper makes for rather sobering reading. Forty nine children with deep lesions of C5, C6 or C5, C and C7 were alternately allocated to the operation or non operation group. The average age at operation was around 6 months. The children were carefully followed for 3 years. These workers were unable to demonstrate decisively better results in the operated over the non operated cases. They confirmed that the onset of recovery after 6 months of age does indicate a bad prognosis but they could not define a critical cut off period and they proposed that a period of 9 months should elapse before deciding in favor of operation. Carlstedt (2007) has outlined his thinking about the impact of the resection of the neuroma upon the vulnerable
457
and immature central nervous system. The operation must inflict a second lesion upon the neurones in the anterior horn and in the dorsal root ganglion. He suggests that the neuroma should be left alone and that function might be supplemented by precise reinnervation, by nerve transfer, of two key nerves, the suprascapular and the nerve to biceps. We should not dismiss Sever’s (1925) comment: “we have had not very brilliant results from the plexus repair. Many of the plexuses that have been operated on have shown absolutely nothing in the way of functional recovery and nothing could be expected from the type of injury found at the time of operation.” Sever was writing at a time when grafting was not used regularly and when selective reinnervation of the ventral root had not been developed. Even so we must state that the only children amongst the 2,100 cases treated who failed to recover flexion of the elbow were those in whom operations and repairs had been done, apart from five children in whom the biceps muscle had been torn apart at birth. Furthermore the only child in our series who recovered no function whatsoever had gone through an extensive repair of the whole of the plexus at the age of 4 months. Laurent et al. (1993) described 24 babies who were operated because of failure of recovery by the age of 4 months. These cases were analyzed with great thoroughness. Intraoperative neurophysiological investigations were refined: “the absence of evoked response in the parietal regions in very young children may be related to the rate of myelination and lack of significant voltage potential ascending the spinal cord, influencing our selection of the C1–C2 as the evoked response area.” Motor action potentials recorded from deltoid and biceps muscle guided the choice between grafting and neurolysis and improvement was “most dramatically seen in those grafted.” Kirjavainen et al. (2007) described the outcome in 124 cases followed for an average of 13.3 years. The babies came from 1.7 million infants born between the years 1971 and 1997. Direct suture was the most common method of repair. Many of these children were followed into adult life. One third of them required either direct assistance or modification of appliances in activities of daily living; one third experienced pain in the affected limb and this included all nine cases of non union of the clavicle. The shoulder was incongruent in 16% of cases, the radio-humeral joint was incongruent in 21% more. There was a striking reversal of limb dominance away from the injured side. Persisting medial rotation contracture was recorded in 57% of the cases. One third of the repairs of the upper trunk resulted in good function at the shoulder. The high incidence of pain, (31%), is disturbing. The outcome of accessory to suprascapular transfer has been closely scrutinised by our colleagues in Heerlen and Leiden. Results are not particularly good; they are inferior to those seen in the adult. Blaauw et al. (2001) examined the outcome following accessory to suprascapular transfer in
458
119 children, all of whom had been born by breech. These were severe injuries: C5 or C6 or both were avulsed in 41 infants. Abduction exceeded 90 degrees in one third of the children and 20 degrees or more of lateral rotation was seen in two thirds. Pondaag et al. (2005), compared the range of active lateral rotation after accessory to suprascapular transfer (21 cases) with that seen after a graft of C5 (65 cases). Active lateral rotation exceeding 20 degrees was seen in only one fifth of children. On the other hand Pondaag and Malessy (2006) related important results following reinnervation of the avulsed 8th cervical nerve. Nine from 13 children treated by this method achieved a Raimondi score of 3 or better. Alain Gilbert has acquired an experience which exceeds that of any other clinician. In 2005 he described the results in 436 children who were operated between 1976 and 1995 and who were followed for at least 4 years. Shoulder function was measured by a slightly modified Mallet system. Grade 5 (normal or near normal abduction with active lateral rotation exceeding 30 degrees) was considered excellent. A Mallet grade 4, with abduction in excess of 90 degrees, and active lateral rotation of at least 20 degrees, was deemed good. A Mallet grade 3 was accorded to those children in whom active abduction lay between 30 and 90 degrees and active lateral rotation was less than 20 degrees. Grade 2 signifies a poor range of active movement. The final functional outcome is closely related to the severity of the lesion. Good or excellent function at the shoulder was found in just over 50% of the Narakas group 1 lesions, in one third of the Narakas Group 2 lesions and in just less than one quarter of the complete lesions. As Gilbert points out;” the shoulder results in complete paralysis are less satisfactory because part of the upper roots destined for the shoulder and elbow have to be sacrificed to obtain function in the hand” Gilbert (2005). Further operations were performed in rather more than one quarter of the cases of incomplete lesions. These included: correction of medial rotation contracture by subscapularis slide (20 cases) and supplementary muscle transfer by latissimus dorsi to the rotator cuff (57 cases) or by advancement of trapezius to improve abduction (seven cases). These secondary operations brought about considerable improvement so that the final analysis showed that 80% of the repairs in the Narakas group 1 lesions were considered good or excellent and 61% reached this level in the C5, C6 and C7 lesions. No less than 75% of the hands of children with complete lesions demonstrated useful function in the hand at 8 years after the repair. This impressive work confirms the importance of reinnervation of the hand but it also demonstrates the value of secondary operations of correction of deformity and supplementing the power of weak muscles by muscle transfer. The bold step of combining relocation of the shoulder with repair of the damaged nerves at one and the same time is described by Grossman et al. (2003) who combined reinnervation of the shoulder with correction of any existing shoulder deformity at the same operation. Nineteen children
Surgical Disorders of the Peripheral Nerves
were operated, they were aged between 11 and 29 months. The suprascapular nerve was reinnervated either by the accessory nerve or by an end-to-side transfer into C7. An average improvement by no less than two points on the Gilbert scoring system for shoulder function was realised.
10.10.1 Experience at St Mary’s – Royal National Orthopaedic Hospital We have repaired elements of the brachial plexus in 210 children, which represents about one tenth of all our cases. Our first 65 cases were analyzed by Chen (1998). Nerves were usually repaired by graft and conventional transfers. It was at Alain Gilbert’s suggestion that we adopted reinnervation of the avulsed 8th cervical and 1st thoracic nerves from ruptures of the upper nerves in 1995. Although some cases of selective reinnervation of the ventral root of avulsed spinal nerves had already been performed in adults, it was at Gilbert’s prompting that we adopted this technique in the birth lesion. Chen found that useful recovery occurred through most of the grafts. Twenty eight of the first 65 cases were complicated by posterior dislocation of the shoulder. We did not combine nerve repair with correction of the dislocation in those early years. Chen found that good hand function in the complete lesions was associated with spontaneous recovery in either C8 or T1. Two of Chen’s observations were particularly interesting: the first was that none of the children seemed to exhibit the severe pain so common in adults with preganglionic injury; next was the finding that recovery of sensation was better than recovery of skeletal muscle and sympathetic function, and certainly far better than that seen in the adult case. These findings prompted an investigation conducted at the Hammersmith Hospital by Praveen Anand (Anand and Birch 2002). Twenty four patients who had sustained a Narakas group 4 lesion were studied. Four of these presented in early adult life. The plexus was repaired in 20 children. Not one of the 100 spinal nerves examined was intact; 47 had been avulsed, 58 ruptured and 12 were recovering ruptures (type 2). Quantitative sensory testing was done at an average of 6 years after operation; cholinergic sympathetic function was also measured. Recovery of sensibility was far better than that seen in skeletal muscle and in the post ganglionic sympathetic nerves. Sensation was normal in all dermatomes for at least one modality tested in 16 hands. There was accurate localisation in the dermatomes of avulsed spinal nerves which had been reinnervated by intercostal nerves transferred from remote spinal segments. Useful hand function was seen in just over one half of the repaired cases but in most of these there was some spontaneous recovery through incomplete lesions of C8 or T1.
Birth Lesions of the Brachial Plexus
Case Report A 23 year old woman, who had never been operated, and in whom neurophysiological and radiological evidence indicated rupture of C5 with avulsion of C6, C7, C and T1 was insensate to all stimuli in the C6 to T1 dermatomes. Sweating in the palm of the hand was 30% of that of the contra lateral limb. Her thresholds to monofilament testing, pin prick, vibration, joint position sense and cooling were normal in the C5 dermatome but the threshold to warm perception was elevated. Case Report Operation in a 3 month old boy confirmed rupture of C5, 6 and C7 with avulsion of C8 and T1. The lesion was repaired using the ipsi lateral ulnar, the superficial radial and medial cutaneous nerves of the forearm. He was examined at the age of 5 years. Cotton wool and pin prick sensation were absent in the ulnar distribution of the hand. Vibration sensation was 12 volts (V) in the index finger but more than 50V in the little finger compared with 6 V in the contra lateral little finger. Joint position sense was present in all digits save the little finger. The threshold for warm and cool sensation was beyond test safety limits in the ulnar territory of the affected hand: warm sensation was 2°C in the median territory, cool threshold was 1.3°C in that territory. He could feel a number 5 monofilament (0.132 G) in the median territory but was unable to feel 20 (263 G) in the ulnar territory. Localisation of sensation to the thumb, the index and the middle fingers was instantaneous and perfect. Sweating was 55% of the opposite palm in the median nerve territory and 25% in the ulnar nerve territory. The thumb and index finger were about 20% shorter than in the unaffected hand whereas the little finger was 50% shorter than in the undamaged hand. There was little evidence of spontaneous collateral sprouting (Fig. 10.19). Case Report Operation in a 6 month old boy confirmed rupture of C5 with avulsion of C6, C7, C8 and T1. The lesion was repaired by transfers of the spinal accessory to
Fig. 10.19 The ulnar nerve was used to repair C5, 6 and 7: C8 and T1 were avulsed. The atrophied little finger is excluded from function.
459
Fig. 10.20 The method of repair and the likely pathways for afferent function.
suprascapular nerve, of C5 to C8 and T1 and of intercostal nerves T3, T4 and T5 to the lateral cord. At the age of 5 years he was able to localise accurately cotton wool and pin prick sensation in both hands and was able to feel a pin in the affected side. Monofilament, vibration thresholds and joint position sense were identical in both hands. The thresholds for warming were 3.1°C (C5), 0.8°C (C6), 4.1°C (C7), 3.6°C (C8), and 3.3°C (T1) on the affected side. The thresholds for cooling were 2°C (C5), 3.9°C (C6), 3.8°C (C7), 2.3°C (C8), and normal in T1. In the unaffected hand the thresholds were less than 3.4°C for warm sensation and less than 2.2°C for cool sensation. Sweating was 70% of that of the intact palm. It seems likely that afferent impulses from thumb, the index, and probably the middle fingers entered the spinal cord through the three intercostal nerves whilst afferent impulses from the little and probably ring fingers entered the spinal cord through the 5th cervical nerve as shown in Fig. 10.20. Similar findings have been observed in four other cases after transfer of intercostal nerves to the lateral cord (Fig. 10.25). Hiroshi Kono (Osaka), who was visiting our Unit at that time, was instrumental in completing a prospective study of the results of 100 consecutive repairs performed in the decade 1990–1999 (Birch et al. 2005). Some of his findings are now set out. 1. The pre-operative NPI predictions correlated strongly with the findings at operation. A type C lesion was recorded in 191 of the spinal nerves examined. Rupture or avulsion or a mixture of both was found in 177 (94%) of these. The prediction appeared to be over optimistic in six nerves but operation revealed rupture or avulsion of an adjacent nerve. Pre-operative NPI were thought unduly pessimistic in 34 nerves. These seemed less seriously
460
Surgical Disorders of the Peripheral Nerves
damaged than anticipated. None were repaired. Twenty two from the 34 recovered as had been predicted; 12 did better than had been anticipated. The lesions for each spinal nerve are set out in Table 10.15. 2. The effect of the associated dislocation of the shoulder was severe. The final mean Mallet score in children without dislocation was 12.4; it was 10.8 in those children who came to operation for relocation of the shoulder. The final Gilbert score for children without the complication of dislocation was 3.3; it was 2.5 in those who came to operation for that complication. It may be that fibrosis provoked by the deformity, by later operation and prolonged immobilisation contributed towards this difference; it is also possible that the dislocated joint interfered with the maturation and integration of the afferent pathways into the central nervous system. 3. The outcome for repair of ruptures of C5, C6 and C7 was disappointing and failed to match those seen after urgent repair of equivalent lesions in the adult.
Table 10.15 Findings at operation in 500 spinal nerves. Type of lesion C5 C6 C7 1
0
5
15
4. The final finding was unexpected: the pre-operative neurophysiological prediction closely matched the final outcome after repair. 50% of the grafts in type B unfavourable lesions achieved a good result but only 34% of the repairs of type C lesion reached this level. It is possible that the pre-operative NPI evidence pointed not only to the extent of peripheral axonopathy but also to the effect upon the neurones within the central nervous system, and upon their ability to respond and to regenerate after a severe proximal lesion. The Criteria Used for Grading the Results of Repair for Each Spinal Nerve are set out in Table 10.16. Those for the shoulder are rather strict. A good result was reserved for shoulders with active abduction of at least 120 degrees and active lateral rotation in excess of 30 degrees. A fair result means abduction ranging from 90 to 120 degrees and active lateral rotation to at least 10 degrees. The results are summarised in Table. 10.17.
C8
T1
Total
Number repaired
43
47
110
0
2
8
5
31
14
16
74
1
3
82
50
11
0
0
143
136
4
4
14
10
3
2
33
25
5 and 6
2
6
12
16
11
47
17 58
7
4
20
21
24
24
93
Total of nerves
100
100
100
100
100
500
Total of lesions
100 (8 favourable)
95 (5 favourable)
85 (31 favourable)
57 (14 favourable)
53 (16 favourable)
390 (74 favourable)
74
29
28
19
Number repaired 87 Drawn from Birch et al. 2005.
Table 10.16 Grading of results of repair by spinal nerve. Results C5 C6 Good
Gilbert 4+ or better
Full flexion/supination
Mallett 13 or better
Biceps MRC 5
237
C7
C8 and T1
Full extension wrist
Raimondi 4 or 5; digital extension and flexion – intrinsic balance. Hands used appropriately in bi-manual activity
Wrist extension MRC 3
Raimondi 3; digital flexion providing simple grasp. Hand supports or helps the uninjured hand
Less than above
Less than above, hand with no, or only vestigial function
Abduction >120° Lateral rotation >30 Fair
Gilbert 3+
Functional flexion
Mallett 11 or 12
MRC 3+. Supination to 45
Abduction 90–120° Lateral rotation 10–30° Poor
Less than above
Based on Birch et al. 2005.
Less than above
Birth Lesions of the Brachial Plexus
461
Table 10.17 Results of repair by spinal nerve. C5 C6 C7
C8 and T1
Number of unfavourable lesions
92
90
54
80
Total repairs
87
74
29
47
Good (%)
29 (33)
41 (55)
7 (24)
27 (57)
Fair (%)
42 (48)
19 (26)
10 (35)
17 (36)
Poor (%) 16 (19) Based on Birch et al. 2005.
14 (19)
12 (41)
3 (7)
1. C5. There were 87 repairs in 92 unfavourable lesions, in which four were avulsions. A good result was achieved in one third, the operation failed in 16 nerves (19%) (Fig. 10.21). 2. C6. There were 74 repairs of 90 unfavourable lesions. The failure rate was 19% (14 nerves). A good result was obtained in 41 (55%) nerves. 3. C7. There were 29 repairs in unfavourable lesions, 21 of which were avulsions. The results following repair of this nerve were particularly poor. Functional extension of the wrist and fingers was restored in only seven (24%) of repaired nerves (Fig. 10.22). Twenty five other unfavourable lesions of C7 were not repaired either because recovery was anticipated or because a higher priority was given to repairing the nerves to the shoulder or to C8 and T1. There was no demonstrable difference in the outcome between the 29 repaired lesions of C7 and the 25 lesions which were not repaired. 4. C8 and T1. Forty seven repairs from 80 (48 avulsions) unfavourable lesions were done. Only the avulsions were repaired, no repair was done in any case where evidence of recovery was demonstrated at operation. In particularly
Fig. 10.22 An 8 year old boy. Repair of C5, C6 and C7 at 14 weeks of age. He had a good result: shoulder scores 5+, 15; elbow 5, hand 5. Dislocation of the shoulder was corrected by open reduction at 22 months. By kind permission of editor. J. Bone and Jt. Surg. 87B, 1093, 2005.
severe cases where three or four nerves had been avulsed repair was usually confined to C8. The results were good in 27 repairs (57%), there was some recovery in 17 more and the operation failed in three. There were few good results following intraplexal transfer for shoulder and for elbow function perhaps because one ruptured spinal nerve was used to reinnervate two segments within the upper limb. Three out of the four cases of intercostal to musculocutaneous nerve transfers failed (Figs. 10.23–10.25).
Fig. 10.21 Six year old boy. Breech delivery. Ruptures of C5 and C6 were repaired at the age of 5 months. The shoulder was relocated at 4 years.
The phrenic nerve was damaged twice. Both nerves were repaired but one infant required plication of the hemi diaphragm. This particularly serious complication is life threatening and the utmost care is needed in defining and protecting the nerve. Two children developed pressure sores from the plaster cowl. There was no deep infection; there were two cases of superficial wound infection which responded to treatment with antibiotics. Two babies, operated on early in the series, required blood transfusion of 30 and 50ml respectively.
462
Surgical Disorders of the Peripheral Nerves
Fig. 10.23 A 10 year old boy had a group 4 lesion with rupture of C5 and 6, avulsion of C7, C8 and T1. Repair was undertaken at 4 months of age. C6 was transferred to the lower trunk and the upper trunk was reinnervated by C5. Accessory to suprascapular transfer was carried out and one graft was interposed between C5 and the distal stump of C7. Function at the shoulder was graded at 2+, 10; elbow 4, hand 4. Fig. 10.25 A 10 year old boy had a group 4 lesion with rupture of C5 and avulsion of C6, C7, C8 and T1. At 10 weeks of age C5 was grafted to the upper trunk and to the ventral root of C8 and T1, accessory was transferred to the ventral root of C7, and sensory divisions of intercostal nerves, T3 and T4, were transferred to the lateral root of the median nerve. He regained a useful grasp between thumb and index finger and is an extremely well adjusted sportsman and musician.
the suprascapular nerve and other grafts onto the postero lateral part of the posterior division of the upper trunk leaving the neuroma at C6 alone. Repair of C8 and T1 is justified only with proof of avulsion. The failure of recovery of the deep afferent pathways, particularly those from muscle spindles, may account for some disappointing results.
Fig. 10.24 A 10 year old girl had a group 4 lesion with rupture of C5 and C6, avulsion of C7, C8 and T1. Repair was undertaken at 10 weeks of age. C6 was transferred to the lower trunk, the upper trunk was reinnervated by C5 and the ventral root of C7 by transfer to the accessory nerve. Shoulder 3+, 12; elbow 4; hand 4.
These findings led us to narrow our indications for operation. Operation should be done as soon as is possible in breech lesions with phrenic palsy and also in the complete lesions where neurophysiological and radiological evidence supports a diagnosis of preganglionic injury. A pre-operative neurophysiological finding of B unfavourable or type C is certainly an indication for exploration of C5. A type B unfavourable lesion for C6 and for C7 is consistent with potentially useful spontaneous recovery. Exploration in these nerves is indicated in a type C lesion. We are no longer persuaded that repair of a rupture (Type 3) of C6 and C7 materially improves the outlook and now prefer to repair a rupture of C5 by placing one graft from the anterior face of the stump of C5 onto
10.10.2 Late Reinnervation What can be done for the child presenting at the age of 3 years or more with defects so extreme that there is little or no chance of success from musculotendinous transfer? Free muscle transfers cannot restore sensation, nor can they replace either the small muscles of the hand nor those acting at the shoulder. Eight transfers of the accessory to suprascapular nerve were done in children aged 3 years or more who showed poor shoulder function but in whom electromyography revealed activity in the supra and infraspinatus muscles. The deltoid muscles appeared to be powerful in these children; the defect lay in the muscles of the rotator cuff. The shoulders were not flail. The results were useful and similar to those obtained by Grossman et al. (2003). The place for late sensory reinnervation, in cases of selective avulsion of the dorsal roots has already been described. We must point out that reinnervation of the lower trunk was performed in three other children aged 18 months or more. There was no useful recovery in any one of these.
Birth Lesions of the Brachial Plexus
10.10.3 Cocontraction The relative success of late accessory to suprascapular transfer may be explained by bypassing the segment of disorderly regeneration and so eliminating at least some cocontraction. Cocontraction especially involves the thoraco-scapular and gleno-humeral muscles. Yagi (1984) detected some degree of cross innervation in two thirds of 342 children. Roth (1983) studied 16 cases, stimulating the nerves and recording heterogeneous motor axon reflexes (MAR) from different muscles by both surface and needle electrodes. Axons from one parent neurone may reach widely different muscles. Perhaps the abnormality rests in the final common pathway; impairment of the normal patterns of inhibition and facilitation of the antagonistic motor neurones because of poor deep afferent recovery is another possibility. Rollnick et al. (2000) successfully treated six children aged between 2 and 4 years, with severe cocontraction of biceps and triceps by injection of botulinum toxin into the triceps muscle, followed by intensive, arduous and prolonged physical therapy. Price et al. (2007) demonstrated lasting benefit by injecting Botulinum toxin Type A into the pectoralis major as an adjunct to operation for medial rotation contracture at the shoulder, and suggest that: “Because of the direct relationship between the muscle afferent input and motor cortical output, more signals can be detected from the muscles affected by birth lesion of the brachial plexus.” Curra et al. (2004) go further: Botulinum toxin Type A “affects the functional organization of the CNS indirectly through peripheral mechanisms,” possibly by reducing signals from muscle spindle afferents which could: “alter the balance between afferent input and motor output, thereby changing cortical excitability.” At times it seems selective denervation is necessary; Kawano et al. (2007) cut the nerves to biceps or triceps and reinnervated them by intercostal transfer, so overcoming cocontraction between the two.
463
1. Injuries Inflicted During Birth. Parents often describe severe bruising around the shoulder and arm after a difficult delivery. Fairbank (1913) recorded a history of difficult labor in every one of his 28 cases of subluxation of the head of the humerus: “in many the child was absolutely black at birth.” Injury at birth was responsible for the scarring, or even fibrosis of the muscles of the shoulder, especially the subscapularis and it is responsible for tearing of the biceps. Paul MacNamara (MacNamara et al. 2009) suggested the term “pseudo-tumor” of the biceps muscle which is caused by damage to the muscle at birth and his idea was analyzed by Yam et al. (2009). The lesion presents as a firm, ovoid, non tender and non contractile swelling in the biceps (Fig. 10.26). It was detected at an average age of 36 months. Eighteen cases were recognised from amongst 720 children seen between 1997 and 2002, an incidence of 2.5%. It is probably higher than this. All of the 18 children had regained good function at the shoulder and the hand by the age of 3 months. Electromyography confirmed strong reinnervation of the biceps in all cases. The diagnosis was made because of
10.11 Deformities Deformities complicating the birth lesion are very common and they are often severe. The principles underlying operations of reconstruction by musculo-tendinous transfer are described in Chapter 13. It seems that the renewed interest in repair of the damaged nerves has led to oversight, or even neglect of earlier work, particularly so with the matter of the posterior subluxation or dislocation of the gleno-humeral joint which was well described by clinicians writing before the first World War. There are some obvious causes of deformity in the birth lesions.
Fig. 10.26 Biceps “pseudotumor” in 9 year old boy. The shoulder is dislocated.
464
delayed or impaired recovery of elbow flexion and supination. Eight children regained good elbow flexion by 6 months, whilst nine more required transfer of pectoralis minor to biceps to augment elbow flexion. Biopsies from two of the affected biceps muscle showed ruptured and degenerate muscle fibers with a fibro-fatty infiltrate, suggesting previous muscle trauma. The whole muscle had been torn apart in two children. These workers say: “reinnervation of the biceps is futile if substantial muscle trauma is confirmed by the presence of a pseudo-tumor.” Dubousset (1972) illustrates a pseudo-tumor of the biceps in his Fig. 12 which he described as “le biceps en boule.” Posterior dislocation occurs at or very shortly after birth in about one in four of the cases that we have treated. One father described how the afflicted arm was pulled into abduction and then across the chest, into forced flexion with medial rotation, during a difficult vertex delivery. It is likely that dislocation occurred at this moment. The subscapularis muscle was found densely fibrosed in five children suggesting a post ischaemic or compartment syndrome lesion analogous to that described in the adult case by Landi et al. (1992). Posterior dislocation was confirmed by MR scanning in two babies at the age of 6 weeks.
Fig. 10.27 Shortening of the left clavicle and of the scapula in a 17 year old girl with a group 2 lesion, who required open reduction of complex dislocation of the left shoulder at the age of 2 years. The left clavicle is 10 cm in length (13 cm on the right) and the left vertebral border of scapula 10 cm (11 cm on the right).
Surgical Disorders of the Peripheral Nerves
2. Denervation of the Limb. There is atrophy of the denervated target organs; a striking example is provided by three of our earliest cases when the ulnar nerve was used as a graft in severe group 4 lesions. Whilst there was useful recovery for the median nerve the little and ring fingers were shortened by as much as 50% and reduced in volume by over 60%. There was atrophy of the skin and of the finger nails. Bone growth is impaired in all but the mildest cases of birth lesion. There is shortening of the clavicle, probably caused by paralysis of the deltoid, the subclavius and the clavicular head of pectoralis major. This shortening distorts the posture and the development of the scapula (Fig. 10.27). 3. Persisting Muscular Imbalance is responsible for many progressing deformities. These can be anticipated once the prognosis for the nerves is known. Poor recovery in C5 leads to an imbalance between the weak lateral and the strong medial rotators at the shoulder and this contributes to many gleno-humeral dislocations. Bad lesions of C7 cause weak medial rotation at the shoulder and supination deformity of the forearm (Fig.10.28). If C8 is also involved then ulnar deviation at the wrist and thumb in palm deformity will ensue. Poor recovery in T1 will cause an intrac-
Birth Lesions of the Brachial Plexus
465
Fig. 10.28 10 year old girl. When she was aged 14 weeks, ruptures of C5 (type 3) C6 (type 4) and of C7 (type 4) were repaired by graft. Recovery for the shoulder was good but medial rotation osteotomy of humerus and proximal advancement of the flexor origin proved necessary. Note the poor range of active supination. Recovery following repair of the two type 4 lesions was limited.
table extension deformity at the metacarpophalangeal joints because of persisting paralysis of the small muscles of the hand (Fig. 10.29). Disorderly regeneration across lesions of the upper nerves is responsible for many examples of cocontraction of the muscles acting across the thoraco-scapular and gleno-humeral joints. 4. Some Deformities are Provoked by Treatment. Overzealous manipulation of the incongruent shoulder damages the head of humerus and glenoid (Fig. 10.30). Incorrect muscle transfers replace one imbalance with another. This is often seen at the shoulder in cases of lesions of C7 where transfer to infraspinatus of latissimus dorsi and teres major compromises active medial rotation. It is important always to retain one powerful flexor of the wrist during flexor to extensor transfer. Damage to the medial epicondyle during Steindler’s elbow flexorplasty leads to dislocation of the elbow (Fig. 10.31). It must be restating that arthrodesis in the growing skeleton should never be performed until muscular imbalance has been corrected. Joints should be congruent before muscle transfers are performed.
Fig. 10.29 Extreme supination deformity in untreated group 4 lesion. The forearm is rotated through 180°. The small muscles of the hand did not recover There is fixed extension deformity at the metacarpo-phalangeal joints.
With the exception of posterior dislocation of the shoulder which should be corrected as soon as is reasonable, it seems generally best to defer musculo-tendinous transfer until the age of 5 years by which time regenerating nerves will have matured and the extent of weakness and cocontraction will be all too plain. Robert Jones (Jones 1916) described the case of a 4 year old child with paralysis of elbow flexion. The arm was supported in the sling for 18 months. Active flexion of the elbow returned. Careful, prolonged, functional splinting,
particularly at the wrist and thumb enables recovery in many children. These are used in the years before planned muscle transfer and often recovery is good enough for the operation to be canceled. These splints must be changed regularly and adjusted for comfort and growth. They are re-applied after operation, and retained during the period of post operative rehabilitation. Only rarely do children reject the splints; in these the matter is not pressed (Fig. 10.32).
466
Fig. 10.30 Closed manipulation of dislocated shoulder at the age of 2 years was done in another hospital. Radiograph of shoulder at the age of 7 years.
10.12 Posterior Subluxation (PS) and Posterior Dislocation (PD) of the Gleno-humeral Joint with Related Contractures More than 500 children with posterior subluxation or dislocation have come to operation since 1986. Bisinella’s (Bisinella and Birch 2003) findings suggest that PS/PD will occur in about one quarter of all cases of birth palsy, which in turn suggests that there will be about 70 new cases in every year in the British Isles. Whitman (1905) classified dislocations of the shoulder in young children under three heads: (1) True congenital dislocation; (2) traumatic dislocation resulting directly from injury at birth; and (3) dislocation, or rather subluxation, following on birth palsy. Whitman considered that the majority of cases fell into the third group. Fairbank’s paper of 1913 is a major contribution. He wrote: “in all typical cases of birth palsy, excepting those
Surgical Disorders of the Peripheral Nerves
in which the injury is slight and recovery rapid, and also those in which the plexus has suffered severely and the paralysis continues, at least a tendency to posterior displacement of the head of the humerus would seem to be invariably present.” In cases where C5 and C6 had been damaged there is recovery in the course of a few weeks or months so that the paralysis disappears but the movements of the arm remain far from perfect: “the arm will now be found to have assumed a new position.” Movements at the shoulder become restricted, especially lateral rotation: “it is often impossible, after placing the elbow to the side, to rotate out the humerus sufficiently to bring the forearm into the directly forward position. On adducting the arm to the side the scapula tilts upwards – i.e., the arm is fixed in slight abduction…active supination is hindered largely by the impossibility of rotating out the humerus…the limitation of movement of the arm is sufficient to seriously interfere with the use of the hand, and the disability is not lessened as it might be by use since the other hand is used whenever possible. The hand cannot be thrust forwards, palm upwards, to receive anything. I have known a boy learn to write with his left hand because of the condition of the right, though he learnt to write with the right after the subluxation of the shoulder had been corrected.” It is all here: the ease of clinical diagnosis, the affect of the dislocation upon the posture of the scapula and upon function of the upper limb, and also an explanation for the lack of active supination, all are described. Fairbank offered an operation to correct this deformity which involved osteotomy of the coracoid, division of the subscapularis tendon and anterior capsulotomy. Sever(1925) recorded the development of the secondary deformities of bone which involved the coracoid process, the acromion, the lateral clavicle and the glenoid (Fig. 10.33). Sever found no case of epiphyseal separation whereas Scaglietti (1938) reporting Putti’s work (Putti 1932), thought that epiphysiolysis was often seen and that it was the: “hallmark of a complicated obstetrical trauma of the shoulder joint.” He went on to say that: “the most constant and characteristic change is the deformation of the angle of declination.” He defined this as: “the angle so formed by the axis of the upper end of humerus in relation to the transverse axis of the lower epiphysis: it is about 20 degrees, open backward and inward. The angle of inclination is formed by the axis of the shaft of the humerus and the axis of the head and neck. It is about 130–140 degrees.” This is probably the first description of retroversion of the head upon the humeral shaft, one of the most important elements which must be corrected before secure congruent reduction can be achieved. Scaglietti went further by observing that: “the aplasia of the glenoid corresponds exactly to that of the acetabulum in cases of congenital dislocation of the hip joint.”
Birth Lesions of the Brachial Plexus
467
Fig. 10.31 Proximal advancement of the flexor origin was performed at another hospital. The medial condyle was damaged leading to dislocation.
10.12.1 Onset and Progression of the Secondary Deformities We estimate that more than 25% of dislocations occur at birth or within the first 12 weeks of life. About one half of cases develop in the first 3 years, during the period of neurological recovery. Perhaps as many as one in four develop after recovery of the nerves. This suggests that a number of agents are at work and it explains why the speed of development of the secondary skeletal changes varies from child to child. Important information about the early stages of development of the deformity have been provided by a number of valuable studies since the first edition of this work. Waters et al. (1998) studied prospectively by computed tomography (CT) and magnetic resonance scanning the shoulders of 94 children with BLBP and found subluxation in 26 of them. The most severe affections were present at birth. A classification based on the extent of subluxation and of retroversion of the glenoid was developed. Pearl and Edgerton (1998) undertook arthrography of the shoulder in 25 children before operating to correct medial rotation contracture. They found that the head of humerus was concentric in only seven. Other shoulders showed flattening of the posterior part of the glenoid leading to biconcavity of the glenoid or to the formation of a false socket. In 2003, Pearl et al. presented extensive evidence
Fig. 10.32 Poor recovery for C7 and C8. The forearm is supinated and there is ulnar deviation of the wrist and thumb. A simple orthosis was beneficial.
468
Fig. 10.33 AP and axial radiograph of complex subluxation in 14 year old boy. Note the double facet deformity of the glenoid, the trench deformity of the head, and the downward prolongation of the acromioclavicular arch. There is elongation of coracoid.
about the nature of medial rotation contracture in 87 children. An arthrogram was done in all cases. Thirty six shoulders were studied by magnetic resonance imaging and 37 by arthroscopy. There was substantial skeletal deformity in 51 of the 87 cases. The gleno-humeral joint was considered concentric in 33, the glenoid was flat in eight, and biconcave in 17. A pseudoglenoid was recognised in 26 shoulders. Whilst MR imaging provided more information about the extent of deformity of the cartilage and bone, arthroscopy revealed a progression from concentric reduction towards a biconcave glenoid which was associated with a flattened or ovoid head of humerus. The supraspinatus tendon and adjacent tissues, and the tissues within the rotator interval were important sites of contracture. There was increasing retroversion of the posterior glenoid. From Rotterdam, Sluisz et al. (2001) described the changes in 16 infants (the average age was 5.2 months) in whom recovery for the nerves was slow or poor. Magnetic resonance imaging demonstrated a normal glenoid in seven of the infants, it was convex in seven more and biconcave in three. The presence of abnormal glenoid forms, of glenoid retroversion and of humeral head subluxation increased with
Surgical Disorders of the Peripheral Nerves
age; there was a statistical difference (p = 0.05) between infants younger than 5 months and those who were older. These workers found no evidence of birth injury to the physis in any case and they suggested that the retroverted or convex glenoid may remodel satisfactorily after concentric reduction of the head of the humerus. Hui and Torode (2003) have shown that retroversion of the glenoid improves after relocation of the head of the humerus. Sixty five children were prospectively studied using CT scanning or magnetic resonance imaging of both shoulders to examine the degree of congruity of the glenohumeral joint, the extent of glenoid version and the development of deformity at the glenoid. The investigations were performed at intervals of about a year. Twenty-nine of the 65 children came to operation to reduce posterior subluxation or dislocation. The operations were performed at a mean age of 29 months; the average follow up was 43 months. The angle of the glenoid retroversion on the affected side was 30 degrees (mean) before operation and this reduced to 14.6 degrees (mean) at the last examination. The average of the difference of the angle of glenoid retroversion between the two shoulders was 20.3 degrees before operation. This had dropped to 8.8 degrees at the last investigation. The remodeling was particularly rapid in the first year after operation but Hui and Torode point out retroversion continued to improve during the period of observation and they say: “there was no significant difference in the change in glenoid version regardless of age at which the surgery was performed.” This demonstration of glenoid remodeling after reduction of the shoulder is important and encouraging. Sluisz et al. (2002) studied retroversion of the head upon the humeral shaft in 33 children, aged, on average, 22 months. The shoulders were examined by MR scan. The glenoid was concave or flat in 11 shoulders, it was convex in 12 and in 10 a bifacetal or biconcave form was apparent. In children aged more than 1 year the difference of the angle of retroversion between the affected and the normal shoulders reached statistical significance; it was measured at 29.9 degrees in the affected shoulders, against 19.6 degrees in the normal ones. Further analysis is provided by Sluisz in his thesis of 2003. In Chapter 7 he describes a prospective study of open reduction of the humeral head by shortening of the coracoid and step elongation of the scapularis tendon. It proved necessary to open the joint capsule in 15 of the 22 children. Although the active range of abduction and lateral rotation improved in all cases there was a measurable loss of active medial rotation. Eight of the 22 children developed a severe lateral rotation contracture at the shoulder. This was more common in children who had suffered a more severe injury to C7. Sluisz suggests that the shoulder should not be subjected to prolonged immobilisation after operations for reduction, that the upper limb should not be splinted in marked lateral rotation and that this operation is inadequate in shoulders with more advanced deformity. Hoeksma et al. (2003) studied 61 children
Birth Lesions of the Brachial Plexus
throughout the period of recovery. Shoulder contracture was detected in 56%, there was osseous deformity in one third, and there was association between shoulder deformity and fracture of the clavicle. Only recently, have we recognised the potential significance of shortening of the clavicle. The clavicle is the tie beam of the forequarter. Shortening of the clavicle pulls the scapula forward. The acromion is pulled downward and forward, elevating the superio-medial border of the scapula. It is more than 18 years since Copeland (1992) suggested that the inclination of the coracoid is altered by the displaced scapula so that it lies vertically. The coracoid and coraco-acromial ligament are displaced dorsally to impinge on the anterior aspect of the head of the humerus. Collins (2005) describes how the shaft of the clavicle is ossified from condensations of mesenchyme which appear between the fifth and sixth week of intrauterine life. Cartilage develops at both ends after the 45th day: “the sternal and acromial zones become true cartilage into which ossification extends from the shaft. Length increases by interstitial growth of these terminal cartilages.” There is only a small contribution from the secondary centre of the acromion, which may appear at 18–20 years. The scapula is ossified from at least eight centres, two of which arise in the coracoid process and two more in the acromion. The upper one third of the glenoid is ossified from the sub-coracoid centre at puberty. Collins says that the remainder of the glenoid is ossified from “a horse shoe shaped epiphysis, thicker at its peripheral than at its central margin.” Damage to the bones during a difficult delivery is compounded by the paralysis of muscle acting upon them. The combination of skeletal injury at birth with paralysis lasting for months or even longer may underlie the shortening of the clavicle and the consequent abnormal posture of the scapula. The acromioclavicular arch moves away from a gentle, Romanesque curve and narrows into an Early English lancet so encroaching upon the gleno humeral joint. The situation is made worse by the dislocation of the acromioclavicular joint (Fig. 10.34).
469
Fig. 10.34 Group 2 lesion. Successful correction of complex dislocation, right shoulder, at the age of 2 years. At the age of 6 years, the length of the right clavicle is 10 cm (left 13 cm). The length of the vertebral border of the right scapula is 10 cm (11 cm) and it is elevated.
10.12.2 Diagnosis and Classification It is difficult to overstate the requirement for scrupulous clinical examination and viewing of the radiographs. The clinical examination is generally more reliable in detecting early incongruency. Both shoulders must be examined simultaneously, initially with the eyes closed. Any hint of asymmetry of the head of humerus indicates an incongruent joint until proven otherwise. The posture of the limb and the awkward elevation at the shoulder with fixed medial rotational deformity tell all in the older child (Figs. 10.35, 10.37). Ultrasound scanning is nearly as good as clinical examination (Saifuddin et al. 2002). Radiographs in the antero-
Fig. 10.35 Dislocation of shoulder of 3 year old boy.
470
Surgical Disorders of the Peripheral Nerves
head deforms this ridge as it slips back into the true socket. The articulation is between the head and the false, posteroinferior facet which is lined with hyaline cartilage. 5. Complex dislocation (CD). In dislocation the head lies in a distended capsule adherent to the dorsal face of the scapula There is a hollowing out of cortical bone below and behind the glenoid. The articulation is between the hyaline cartilage of the head, the intervening capsule and the underlying cortex. In older children it is often possible to demonstrate translation of the head of humerus from subluxation to dislocation during forward flexion at the shoulder. There are three features which repay examination in every case. The first of these is the posture of the coracoid process, which is nearly always displaced posteriorly, inclined vertically and elongated. Secondly, the length of the clavicle should be measured between the sterno-clavicular and acromioclavicular joints It is always shorter on the affected side, by as much as 25% in the more severe cases. Finally, the acromioclavicular joint should be examined for subluxation or dislocation. The classification of the deformity is made first by clinical examination, refined by radiographs and confirmed or modified at operation.
10.12.3 Medial Rotation Contracture Fig. 10.36 Medial rotation contracture of shoulder in 6 year old boy.
posterior, and in the axial plane by the Stripp (1997)method confirm the diagnosis. MR scanning is reserved for cases of unusual complexity. We use the following classification. 1. Medial rotation contracture (MRC). The contour of the shoulder is normal. Passive lateral rotation beyond 30 degrees is blocked. The forearm lies in pronation. Radiographs show a congruent joint (Fig. 10.36). 2. Simple posterior subluxation (SS). The head of the humerus is prominent posteriorly. Passive lateral rotation is restricted to about 10 degrees beyond neutral. Radiographs must be examined carefully, for it is easy to underestimate incongruency. As yet, there are no secondary deformities of the acromion or glenoid. 3. Simple posterior dislocation (SD). The head of the humerus can be seen and palpated behind the glenoid. There is a fixed medial rotation contracture and the forearm lies in full pronation. Radiographs confirm the displacement and the absence of secondary osseous changes. 4. Complex subluxation (CS). The skeletal abnormality is evident on clinical and radiological examination. The glenoid often forms a double facet with the true glenoid lying above and anterior to the false socket. Both facets are lined with hyaline cartilage separated by a ridge. At operation, the
Gentle but persistent stretching exercises correct many uncomplicated medial rotation contractures in the earliest stages of the deformity but once this becomes resistant and, above all, once stretching causes pain then exercises must be abandoned. They will do more harm than good. Careful clinical examination in such cases will often reveal that the joint has become subluxated. Our experience with subscapularis recession has been bad; the deformity recurred in 56 of 86 cases. We have found that this operation damages the muscle and that it does not address the contributions from the abnormal coracoid and its ligaments. We abandoned this operation 15 years ago. Temporary paralysis of the subscapularis muscle is an attractive idea. Price et al. (2007) used botulinum toxin type A as an adjunct to the operative correction of medial rotation contracture. Those children in whom the pectoralis major was paralyzed with botulinum toxin secured a definite and a lasting benefit. All interested clinicians should read this valuable paper. Pearl et al. (2006) raised the possibility of release of the contracture using the arthroscope. This important paper showed that arthroscopic release of the anterior and inferior capsule and of the insertion of the supraspinatus tendon, combined with transfer of latissimus dorsi muscle to the infraspinatus tendon could in fact correct not only medial rotation contracture but also the earliest stages of subluxation. These
Birth Lesions of the Brachial Plexus
471
10.12.4 The Development of Our Preferred Operation for Posterior Subluxation or Dislocation of the Gleno-Humeral Joint
retroversion of the head (Fig. 10.37). However, Waters and Bae (2006) and Kirkos and Papadopoulos (1998) obtained useful improvement. We think that muscle transfer in the incongruent shoulder is a mistake but Waters and Bae (2005) showed that these transfers can improve function and slow or altogether halt, the progression of deformity. We hold to the view that muscles should only be transferred in joints which are congruent or after they have been made congruent. The recovery of the lateral rotators is adequate in most children. Open reduction is often complicated by defective medial rotation. We reserve muscle transfer to infraspinatus for cases of posterior dislocation of several years standing. In these, the glenoid is always badly deformed and the infraspinatus and the posterior capsule have been stretched even attenuated by the displaced head of humerus. The muscle may not recover even though innervation is good. The deformity is a very complex one. Putting to one side those dislocations which occur at birth or within a few days of birth and which are probably caused by the violence of delivery, it would appear that the imbalance between the
Four children with longstanding PD were treated by the operation described by Fairbank in the early 1980s (Dunkerton 1989). This method was abandoned because of the loss of medial rotation. We subsequently grasped the significance of narrowing of the posterior scapulo-humeral angle which is caused either by retroversion of the head of humerus or by the much less frequent contracture of the posterior capsule and the lateral rotator muscles. It is this narrowing which provokes the “winging” of the scapula after a stable reduction in which the retroversion of the humeral head has not been corrected. Medial rotation is possible only by protraction of the scapula around the chest wall. In the early 1990s some of these children were treated by delayed derotation osteotomy of the shaft of the humerus, through a lateral exposure. This method placed the radial nerve at risk. It did not increase the arc of rotation, it only changed it. From 1995 the osteotomy was done by an anterior approach at the same time as open reduction. The failure rate in advanced cases of complex dislocation was unacceptably high and this led to the idea of a posterior bone block. This proved unsatisfactory and was replaced by glenoplasty. The first glenoplasty (1998) was performed with Thomas Carlstedt in a boy who had already been subjected to several operations for dislocation, which had failed. The gleno-humeral joint was exposed through a posterior approach, through a very scarred field, and the posterior margin of the glenoid was elevated by osteotomes and secured in position with a bone graft. The operation succeeded in maintaining reduction of the head of humerus. We have no experience with lateral rotation osteotomy in the late case. Adult patients with traction lesions of the brachial plexus who had been treated by this method regretted the loss of medial rotation. The operation must increase
Fig. 10.37 Fourteen year old girl who had lateral rotation osteotomy for dislocation of the shoulder at the age of 7. The operation increased retroversion of the head of humerus and worsened the deformity.
workers point out that this method cannot be considered for the more advanced and established deformities. We use a short incision in the delto-pectoral groove, centred over the coracoid, which is exposed and, if required, shortened. In some cases release of the coraco-acromial ligament from the coracoid is sufficient. The ligament arises from the entire length of the lateral margin of the coracoid and its release is usually followed by an immediate improvement in lateral rotation by some 30 degrees. The rotator interval is defined and the superior one quarter of the subscapularis tendon is divided if there is any suspicion of incongruency. A plaster of Paris jacket is applied with the arm held at 45 degrees of lateral rotation for 3 weeks after this intervention.
472
Surgical Disorders of the Peripheral Nerves
Table 10.18 Incidence of posterior subluxation or posterior dislocation by Narakas group in 841 new cases of CBP seen 1995–2000. Narakas group 1 2 3 4 Total Number of new cases
302 (36%)
342 (41%)
120 (14%)
77 (9%)
841
Number of PS and PD
39 (13%)
119 (35%)
43 (36%)
10 (13%)
211 (25%)
Poor neurological recovery – excluded
2
7
4
10
23
Incomplete data – excluded
2
2
1
–
5
35
110
38
–
183
Number in study Drawn from Kambhampati et al. 2006.
powerful and active medial rotators and the weak, or even paralyzed, lateral rotators is a critical factor in causation. Evidently, the changes in the glenoid occur early. Perhaps the medially rotated head of humerus is thrust against the posterior and inferior margin of the cartilage of the glenoid which begins to deform it so that it becomes convex. At varying intervals after this the deformity of the glenoid evolves as a double facet or a double socket with the true glenoid lying above and anteriorly and articulating with the lesser tuberosity, and a postero inferior facet which soon becomes larger. The head of the humerus lies in the postero inferior facet. The head of the humerus may progress to dislocation. In some cases the head of the humerus was never in the glenoid. The displacement of the head of the humerus leads to mal development of the lateral clavicle and the acromion and even to subluxation or dislocation of the acromioclavicular joint. The abnormal position of the scapula and the dorsal displacement of the coracoid is induced by the shortening of the tie beam, the clavicle. We have seen a few cases where the condition of the subscapularis was so severely fibrosed that it suggested compartment syndrome of the muscle, probably occurring at birth. Einarsson et al. (2008) took biopsies from subscapularis muscle during operations for correction of medial rotation contracture. They found that the slack sarcomere length was shorter than normal, that linear deformation of the fiber occurred within a wider zone of sarcomere length and that there was a relative increase in stiffness. Operations which are designed to reduce a dislocated gleno-humeral joint must try to address all the abnormalities which cause or maintain that dislocation. Recovery of the nerves is usually good and transfer of muscles to enhance lateral rotation should be done with considerable caution for fear of replacing one imbalance with another. From 1995 an operation was used which addressed the contracted subscapularis, the deformity of the coracoid and of its ligaments and the retroversion of the head. This operation did not address the deformity of glenoid nor the abnormality of the acromioclavicular arch. Kambhampati et al. (2006) described the outcome of the operation in a prospective study of 183 consecutive cases of subluxation (101) and dislocation (82) between 1995 and 2000. The recovery of the nerves was good or, at least, useful in all of these children. Those cases with poor neurological recovery were excluded from the study. The
Table 10.19 Results in 183 shoulders treated by reduction by the anterior approach. Retroversion of head of humerus was corrected by medial rotation osteotomy of humeral shaft in 70 cases. Final Mallet Pre-operative Deformity Number Number score (183 of cases of failures Mallet score (183 shoulders) shoulders) (20) (183) SS
37
0
10.4
13.4
SD
24
0
7.8
12.4
CS
64
5
9.5
13.2
CD
58
15
9
12.7
I
35
3
10.8
13.6
II
110
13
9
13
Narakas group
III 38 4 8.9 Based on Kambhampati et al. 2006.
12.5
mean age at operation was 47 months and the mean follow up was 40 months. The incidence of dislocation was highest in the Narakas Groups 2 and 3, that is in those cases where recovery into the lateral rotator muscles was delayed whilst there was powerful recovery for the medial rotators. The incidence was much lower in the group 1 cases, where there was early recovery of the nerves and early rebalancing of the forces acting across the gleno-humeral joint and it was also relatively low in the Group 4 cases, in children whose shoulders were flail (Table 10.18). The results are summarised in Table 10.19. The improvement in function was most evident in the Narakas Group 1 cases and in those where the subluxation or dislocation was not complicated by advanced secondary bone deformities. The reduction failed in 20 shoulders; all of these were in the more advanced cases of complex subluxation and dislocations. The mean active global range of shoulder movement was increased by 73 degrees; the mean range of active lateral rotation by 58 degrees and that of supination of the forearm by 51 degrees. Active medial rotation was decreased by a mean of 20 degrees (Figs. 10.38 and 10.39a–d). Kambhampati summarised the radiological and operative findings in the 183 shoulders (Table 10.20). He found that overgrowth of the lateral clavicle and the acromion was associated with the duration of dislocation. The bifacetal appearance was the most common glenoid abnormality but in 22 cases the glenoid appeared small and flat and there was a
Birth Lesions of the Brachial Plexus Fig. 10.38 Active Range of Movement before operation and at final review: 183 shoulders (Based on Kambhampati et al 2006).
473
200
150
100
50
0
−50
100
Preop Lateral rotation active
final lateral rotation active
preop abduction active
very severe posterior defect. A derotation osteotomy of the shaft of the humerus was performed in 70 of the 183 shoulders. In those days we usually advised operations about the age of 15 months, at a time when neurological recovery has advanced and the child is walking, so easing post operative care for the parents. This was a mistake and we now recommend that the incongruent joint is corrected as soon as possible and certainly at the same operation during which the brachial plexus is explored and repaired. The technique of operation is now described, together with an outline of the method of glenoplasty which came into regular use after 2001. Although there have been no early cases of redislocation after glenoplasty, the long term outcome must await further study. We emphasise that retroversion of the head of the humerus in excess of 40 degrees should be corrected. Wahlström and Backman (2005) have studied the role of derotation osteotomy.
10.12.5 The Operation for Reduction of the Posterior Subluxation and Dislocation of the Gleno-Humeral Joint The skin of the shoulder is infiltrated with local anaesthetic before incision. Interscalene block of the brachial plexus is forbidden because of the risk to the spinal cord. The shoulder is exposed through a delto-pectoral incision which may be
final abduction active
preop medial rotation active
final medial preop final rotation supination supination active active active
extended to permit exposure of the anterior shaft of humerus. Although the deltoid and pectoralis muscles usually appear well innervated scarring between, and deep to them is usual. The coracoid, which is long and lies vertically, is shortened to its base after releasing the coraco-acromial (CAL) and coracohumeral (CHL) ligaments. The CAL is usually a well defined thick structure which impedes lateral rotation of the head. A step elongation of the subscapularis muscle is done preserving as much of the anterior capsule as possible. The muscle is often fibrosed, especially so in birth dislocations, or after earlier subscapularis slide. The head of the humerus is relocated. Now, retroversion of the head upon the shaft is measured. It is the angle subtended by the coronal plane of the humeral head and that between the lateral and medial epicondyles. Osteotomy of the shaft of the humerus with medial rotation of the distal fragment is done with two objects in mind: to improve stability of the shoulder after reduction; and to ensure an adequate range of medial rotation. It is indicated when retroversion exceeds 40 degrees. The shaft of the humerus is exposed between deltoid and pectoralis major, preserving the upper one quarter of the pectoralis tendon as guardian of the growth plate. A small plate is applied, the angle of rotation planned and marked by appropriate screw holes. The plate, held by one proximal screw, is rotated out of the way, and the bone divided. The distal fragment is rotated medially, and fixed. Glenoplasty is necessary when the head continues to drop out from the antero-superior true facet into the larger posteroinferior false facet after relocation and derotation osteotomy.
474
Surgical Disorders of the Peripheral Nerves
a
b
Fig. 10.39 The course to recovery after relocation of the left shoulder. Complex subluxation in an 8 year old boy with Narakas group1 lesion: first noticed at about the age of 4 years. A subscapularis slide was done at the age of 5 years at another hospital. There was early recurrence of deformity. (a) Antero-posterior and axial radiographs show overgrowth of the coracoid and downward displacement of the acromion and lateral clavicle. There is a double facet glenoid and a cone shaped head. Shoulder scores were 1+, 12. Active forward flexion and abduction was 180°; active lateral rotation was minus 40° and the passive range was minus 20°. Active medial rotation was 90°, the passive range was 110°.
Active pronation was 90° and active supination 40°. (b) Anteroposterior, axial radiographs taken 1 year after reduction. A 70° derotation humeral osteotomy was performed. (c) Antero-posterior and axial radiographs taken 5 years after reduction showing remodeling of the head of the humerus and glenoid. The humeral head shows some signs of earlier vascular change.(d) Function at 5 years from reduction. The shoulder scores were 5+, 15. Active abduction and forward flexion was 170°; active lateral rotation 30°; active medial rotation 90°; and pronation and supination 90°. The posterior scapulo-humeral angle was 90°. By kind permission editor of J. Bone Jt. Surg. 88B: 213–219.
Birth Lesions of the Brachial Plexus
475
Fig. 10.39 (continued)
Therefore, the false facet must be moulded over the relocated head of humerus. The posterior face of the scapula is displayed between the infraspinatus above and the teres minor below, between the territories of the suprascapular and the circumflex nerves. The reduced head of humerus leaves behind a mass of redundant capsule which is elevated from the posterior face of the scapula. An indentation marks the site of articulation between the head of the humerus and the intervening capsule against the posterior face of scapula. A radial incision permits identification of the posterior labrum and the edge of the hyaline cartilage of the inferior facet. Fine osteotomes are used to elevate the posterior and inferior wall of the glenoid so that it abuts the reduced head of
the humerus. The gap is wedged open with excised coracoid. The glenoid flap is moulded against the reduced head so containing it, and is held there by a small buttress plate. The flap consists of: the posterior and inferior labrum; the inferior (false) socket of the glenoid; the overlying capsule, and cortico cancellous strips “petalled” from the posterior face of the scapula. Now, teres major is transferred to infraspinatus because this muscle will have become defunctioned by long standing dislocation. Local anaesthetic is instilled into the joint before closure of the wound. The shoulder is protected with in a plaster of Paris splint for 6 weeks, in a position of 30 degrees of lateral rotation and about 30 degrees of abduction (Figs. 10.40 and 10.41).
476
Surgical Disorders of the Peripheral Nerves
Table 10.20 Radiological and operative findings in 183 shoulders. Subscapularis muscle Normal
141
Mild fibrosis
28
Severe fibrosis
14
Coracoid process Normal Tip at level of capital physis
22
Tip below capital physis
57
Acromion and lateral clavicle Normal
140
Anterior spur
38
Overgrowth of whole
5
Glenoid Normal
60
Double facet
101
Planar or severe posterior defect
22
Head Normal
138
Conoid or oval
36
Severe deformity
9
Retroversion in degrees Less than 30
70
30 to 70
88
More than 70 Drawn from Kambhampati et al. (2006).
Fig. 10.40 AP and axial radiographs 24 months after glenoplasty in a 4 year old child.
25
This operation does not address the problems caused by the deformity of the acromioclavicular arch. Satisfactory re-modeling of the glenoid is usual in children aged less than 3 years and also in less advanced deformities. There is, however, a risk of avascular necrosis of the head of the humerus which was confirmed radiologically in at least six cases in Kambhampati’s series. The incidence is probably higher than this. It is possible that this is caused by stretching of the circumflex humeral vessels as the head of humerus is re-located into full lateral rotation. These children presented with an irritable and stiff shoulder. Vigorous exercises were suspended. The pain was never severe and resolved in all cases by about 6 months from onset. The appearance of an enlarged head on the final radiograph was suggestive of earlier avascular necrosis and this may lead to impingement because the head is too big for the glenoid and it is also too big for an abnormal acromioclavicular joint. It seems that attempts to relocate a dislocated head of humerus may do more harm than good in cases with advanced secondary bone changes. These changes are not necessarily related to the age of the child nor are they always related to the duration of dislocation. Reduction proved impossible in one child aged 7 years, on the other hand a long standing posterior dislocation was successfully reduced in a 14 year old child. The chief obstacle to successful reduction in the advanced cases is the deformity of the acromio clavicular arch. It may be the case that palliative lateral rotation osteotomy without attempting reduction succeeds in such cases
Birth Lesions of the Brachial Plexus
477
Fig. 10.41 Shoulder function 5 years after reduction and glenoplasty on the right side.
because the head of the humerus is moved away from its articulation with the deformed glenoid. Nath and Paizi (2007) have developed an elegant operation to deal with the deformed acromioclavicular arch in younger children. The spine of scapula is divided, and the clavicle is divided at the junction between the lateral one third and the medial two thirds. The bone flap migrates upwards
and backwards opening out the AC arch which forms a more distant relation with the head of humerus. The arm is supported in a light orthosis. Early lateral and medial rotation is encouraged. Perhaps early elongation of the clavicle at the same time as open reduction will go some way to obviate the need for later, more difficult and more complex reconstruction. Marco Sinisi has done this in one 8 month old child
478
during relocation of complex dislocation. The periosteum was split longitudinally, then the bone divided. The fragments sprang apart by at least a centimetre within the periosteal sleeve, which was sutured. The gap ossified rapidly.
10.12.5.1 Fixed Contracture of the Inferior Scapulo-Humeral Angle In some children this contracture is so tight that the angle is diminished to as little as 30°. In some of these cases cocontraction between the teres major and latissimus dorsi and the abductors of the shoulder is evident. In others, however, there is condensation of the fascia in the axilla and of the fascia enveloping teres major and latissimus dorsi. In some children an obvious band is palpable beneath the skin. A relatively simple operation may be useful. A short incision is made, centred over the tendon of latissimus dorsi. Condensations of fascia within the axilla are released, the sheath of the teres major muscle is divided and step elongation of latissimus dorsi tendon is performed. Gentle exercises start on the day after operation and it is usual to see an improvement in appearance and a gain of between 20 and 30 degrees of active abduction.
10.12.6 Deformities at the Elbow and Forearm These are common and have been extensively studied by Aitken (1952), Chuang (2001), Joseph (2000), Hentz (2001), Zancolli (2001) amongst others. Flexion contracture of the elbow is common and can be serious. It may be prevented or diminished by gentle persistent stretching and splinting. Our preferred method of serial splinting using carefully applied plaster of Paris splints is described in Chapter 13. We have operated on 21 cases, dividing the bicipital aponeurosis and the sheath of the brachialis muscle. Sometimes there are adhesions enveloping the biceps tendon but this should not be elongated nor should the capsule of the elbow joint be opened. A modest improvement, with a gain of about 30°, is usual. It is important to continue with serial splinting after the operation. Open reduction of the dislocated head of the radius failed in three cases. It may be better to await skeletal maturity before excision of the head of the radius if that is causing pain; excision of the head of the growing radius is a disastrous error.
10.12.6.1 The Supination Deformity The range of pronation and supination is greatly influenced by the shoulder posture. When the joint is dislocated active
Surgical Disorders of the Peripheral Nerves
supination is lost and there is nearly always quite dramatic improvement in active supination after the shoulder has been reduced. A fixed supination deformity is particularly common in severe lesions of C7 and C8 and it is usually associated with weakness of extension of the wrist, ulnar deviation at the wrist and poor abduction of the thumb ray. Most of our cases had developed a fixed deformity. Anderson (2000) describes the role of the interosseous membrane: “we have found in cadaveric investigations that the interosseous space widens at its maximum in a supination of 20 degrees (20 mm) and reduces to 14 mm in maximum supination and to 10 mm in maximum pronation.” The association between the supination deformity and other deformities within the upper limb in severe birth lesions was examined by Yam et al. (2009). Forty two cases were identified amongst 696 children whom we saw between 1997 and 2004, an incidence of 6.9%. There were no cases in Narakas Group 1. The deformity was first recorded at an average age of 5 years. The forearm lay in just under 90 degrees of supination. The passive range of pronation was diminished in most of these children. The brachial plexus had been explored and repaired in 27 of the 42 children; recovery was consistently poor in C5 and in C7. Recovery in C8 andT1 was inconsistent. No less than 32 of the 42 children had come to operation for posterior dislocation of the shoulder and in 20 of these medial rotation was weak. The fixed flexion deformity at the elbow was 30 degrees (median). Poor extension of the wrist and of the digits was recorded in 30 hands; in 18 there was ulnar deviation at the wrist and in 19 more there was a thumb in palm deformity. Yam and Fullilove propose that to the first insult to the 7th cervical nerve is added the complication of posterior dislocation of the shoulder. This leads to a pronation posture of the forearm. After reduction of the shoulder, and after recovery of the biceps, the supination deformity is unmasked. This becomes fixed because of contractures at the proximal and distal radio-ulnar joint and also because of contracture of the interosseous membrane. The children were treated by pronation osteotomy combined with muscle transfers to enhance extension and radial deviation of the wrist. Transfer of extensor carpi ulnaris to the abductor pollicus longus was particularly useful. The radius was divided distal to the insertion of pronator teres tendon; it was never necessary to divide both of the forearm bones. The result was generally highly appreciated by parents and the children and hand function improved by between 1 and 3 points on the Raimondi scale. However, there was a high rate of recurrence (40%). The family was advised of this possibility and therefore of the prospect of repeating the operation as the child approached maturity. Yam and Fullilove suggest that the brachio radialis acts as a supinating force after rotating the radius and they suggest that combining pronation osteotomy with re-routing of the brachio radialis muscle, as proposed by Ozkan et al. (2004) may diminish the risk of
Birth Lesions of the Brachial Plexus
recurrence in those cases of fixed deformity where re-routing of the biceps tendon must fail. This complex field is discussed by Allende and Gilbert (2004) who operated in 94 cases of supination deformity. The association with poor recovery in C7, C8 and T1 is described. The indications for re-routing the biceps tendon are defined: this operation is preferred when there is a full passive range of pronation without dislocation of the head of the radius, and it will succeed in cases of mild fixed deformity if it is combined with release of the interosseous membrane. Osteotomy of the radius is recommended for the fixed rigid forearm.
10.13 Conclusion Whilst it is true there have been considerable advances in the understanding of the birth lesion of the brachial plexus it is not always easy to demonstrate that there have been major improvements in the outcome since the work of Kennedy, of Fairbank and Putti. It is clear that nerve repair may improve function in the upper limb in cases of BLBP, particularly so when avulsed nerves are reinnervated. A good shoulder is the foundation for function in the upper limb in a growing child and every reasonable effort must be made to reinnervate the shoulder muscles and to prevent or to treat, medial rotation contracture, posterior subluxation and posterior dislocation as soon as they are detected. Many of these children require review into adult life but hospital admissions and attendances must be kept to an essential minimum to diminish the effects upon the child and the family. Passing the parcel from one “specialist” to another will not do.
References Aitken J (1952) Deformity of the elbow joint as a sequel to Erb’s obstetrical paralysis. J Bone Joint Surg 34B:352–365 Allende CA, Gilbert A (2004) Forearm supination deformity after obstetric paralysis. Clin Orth Rel Res 426:206–211 Anand P, Birch R (2002) Restoration of sensory function and lack of long-term chronic pain syndromes after brachial plexus injury in human neonates. Brain 125:113–122 Anderson GA (2000) The child’s hand in the developing world. In: Gupta A, Kay SPJ, Scheker LR (eds) The growing hand, 1st edn. Mosby, ISBN OT 234 21331, Chapter 114 Becker MHJ, Lassner F, Bahm J, Igianni G, Pallua N (2002) The cervical rib. A pre disposing factor for obstetric brachial plexus lesions. J Bone Joint Surg 84:740–43 Bellew M, Kay SPJ, Webb F, Ward A (2000) Developmental and behavioural outcome in obstetric brachial plexus palsy. J Hand Surg 25B:49–51 Birch R, Ahad N, Kono H, Smith S (2005) Repair of obstetric brachial plexus palsy. Results in 100 children. J Bone Joint Surg 87B: 1089–95
479 Bisinella G (2000) Le lesioni ostetriche del plesso brachiale a lente guaragine neurologica: nuove indicazione al trattamento Tessi di Specializzazonie. Universita degle Studi di Padova. Anno Aacademica 1999–2000 Bisinella G, Birch R (2003) Obstetric brachial plexus lesion: a study of 74 children registered with the British Surveillance Unit. J Hand Surg 28B:40–45 Bisinella G, Birch R, Smith SJM (2003) Neurophysiological predictions of outcome in obstetric lesions of the brachial plexus. J Hand Surg 28B:148–152 Blaauw G, Slooff ACJ, Muhlig S (2001) Results of surgery after breech delivery. In: Gilbert A (ed) Brachial plexus injuries. Martin Dunitz, United Kingdom, pp 217–224 Blaauw G, Sauter Y, Lacroix CLE, Sloof ACJ (2006) Hypoglossal nerve transfer in obstetrical brachial plexus palsy. J Plast Rec Surg 59:474–478 Blaauw G, Muhlig RS, Vredeveld JW (2008) Management of brachial plexus injuries. Adv Tech Stand Neurosurg 33:201–31 Boo NY, Lye MS, Kanchanala M, Ching CL (1991) Brachial plexus injuries in Malaysian neonates: incidence and associated risk factors. J Tropic Paeds 37:327–330 Brown T, Cupido C, Scarfone H (2000) Developmental apraxia arising from neonatal brachial plexus palsy. Neurology 55:24–30 Brunelli GA, Brunelli GR (1991) A fourth type of brachial plexus lesion: the intermediate (C7) palsy. J Hand Surg 16B:492–494 Carlstedt T (2006) Mechanisms behind closed brachial plexus injuries. Clin Risk 12:16–17 Carlstedt T (2007) Timing of operation. Discussion at the second British Brachial Plexus Club Meeting, Birmingham (Chairman Ruth Lester) Carlstedt T, Strömbeck C (1995) Surgical versus conservative treatment of obstetrical brachial plexus palsy: a preliminary study. In: Vastamaki M (ed) Current trends in hand surgery, Elsevier, Amsterdam pp 255–259 Chen L (1998) Results of 65 repairs of OBPP. Cited in: Birch R, Bonney G, Wynn Parry CB Surgical disorders of the peripheral nerves. Churchill Livinsgtone, Edinburgh, p 221 Chen L, Gao SC, Gu YD, Hu SN, Xu L, Huang YG (2008) Histopathological study of the neuroma-in-continuity in obstetric brachial plexus palsy. Plas Reconstr Surg 121:2046–2054 Chuang DCC (2001) Palliative surgery: forearm and hand deformities. In: Gilbert A (ed) Brachial Plexus Injuries Federation of European Societies for Surgery of the Hand (FESSH). Martin Dunitz, London, ISBN 1-84184-015-7, Chapter 28: 292–302 Clarke HM, Curtis C (2001) Examination and prognosis. In: Gilbert A (ed) Brachial plexus injuries. Martin Dunitz in association with the Federation of European Societies for Surgery of the Hand, London, Chapter 12, pp 159–172 Collins P (2005) Ossification of the clavicle and scapula. In: Chief. Standring S (ed) Grays anatomy, 39th edn, Elsevier Ltd, Amsterdam pp 818–822 Colon AJA, Vredeveld JW, Blaauw G, Slooff ACJ, Richards R (2003a) Extensive somato-sensory innervation in infants with obstetrical brachial palsy. Clin Anat 16:25–29 Colon AJ, Vredeveld JW, Blaauw G, Zandvoort JA (2003b) Spontaneous muscle fibre activity appears early in cases of obstetric brachial plexopathy. Muscle Nerve 28:515–516 Copeland S (1992) Shoulder dislocation and the role of the coracoid process. Discussion. The Reading International Shoulder Meeting. University of Reading. Organised by S. Copeland of the Royal Berkshire Hospital Curra A, Trompetto C, Abbruzzese G, Berardell A (2004) Central effects of botulin toxin type A: evidence and supposition. Mov Disord 19(supple 8):60–64 Curtis C, Stephens D, Clarke HM, Andrews SD (2002) The active movement scale: an evaluation tool for infants with obstetrical brachial plexus palsy. J Hand Surg 27A:470–478
480 Dubousset J (1972) Séquelles du coude et de la main Figure 12. p. 41 Paralysie Obstétricale du plexus brachial. Symposium sous las direction de J Mallet Rev Chir Orthop 58:139–151 Dunkerton MC (1989) Posterior dislocation of the shoulder associated with obstetrical brachial plexus palsy. J Bone Joint Surg 71B:764–6 Einarsson F, Hultgren T, Ljung B-O, Runesson E, Friden J (2008) Subscapularis muscle mechanics in children with obstetric brachial plexus palsy. J Hand Surg 33E:507–512 Eng GD, Koch B, Smokvina MD (1978) Brachial plexus palsy in neonates and children. Arch Phys Med Rehabil 59:458–464 Erhard RP, Lindley SG (2000) Functional development of the hand. In: Gupta A, Kay SPJ, Scheker LR (eds) The growing hand, 1st edn. Mosby, St. Louis, MO, pp 71–82, 2000 ISBN OT 234 21331 Evans-Jones G, Kay SPJ, Weindling AM, Cranny G, Ward A, Bradshaw A, Hernon C (2003) Congenital brachial palsy: incidence, causes and outcome in the United Kingdom and Republic of Ireland. Arch Dis Fetal Neonatal 88:F185–F189 Fairbank HAT (1913) Subluxation of shoulder joint in infants and young children. Lancet I:1217–1223 Foad SL, Mehlman CT, Yin J (2008) The epidemiology of neonatal brachial plexus palsy in the United States. J Bone Joint Surg 90A: 1258–64 Giddins GEB, Birch R, Singh D, Taggart M (1994) Risk factors for obstetric brachial plexus palsies. J Bone Joint Surg 76B(suppl II, III):156 Gilbert A (1995) Indications et résultats de la chirurgie du plexus brachial dans la paralysie obstétricale. In: Alnot JY, Narakas A (eds) Monographies du groupe d’étude de la main. Expansion Scientifique Française, Paris Gilbert A (2005) Obstetrical Paralysis. In: Tendon, nerve and other disorders. Surgery of disorders of the hand and upper extremity series. Taylor and Francis, London, pp 277–302, Chapter 10 Gilbert A, Raimondi P (1993) A system of evaluation of elbow flexion. Presented to the international meeting on obstetric brachial plexus palsy, Heerlen, 1993 and cited In: Surgical disorders of the peripheral nerves, 1st edn. Authors: Birch R, Bonney G, Wynn Parry C. Churchill Livingstone Edinburgh, London Gilbert A, Tassin JL (1984) Réparation chirurgicale de plexus brachial dans la paralysie obstétricale. Chirurgie (Paris) 110:70–75 Gjörup L (1965) Obstetrical lesion of the brachial plexus. Acta Neurol Scand 42(suppl 18):9–38 Gramsbergen A, Ijkema-Paasen J, Nikkels PGJ, Hadders-Algra M (1997) Regression of polyneural innervation in the human psoas muscle. Early Hum Dev 49:49–61 Grossman JAL, Price AE, Tidwell MA, Ramos LE, Alfonso I, Yaylalli (2003) Outcome after late combined brachial plexus and shoulder surgery after birth trauma. J Bone Joint Surg 85B:1166–1168 Groves MJ, Scaravilli F (2005) Pathology of peripheral neurone cell bodies. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier, Philadelphia, PA, pp 683–732, Chapter 31 Harris W, Low VW (1903) On the importance of accurate muscular analysis in lesions of the brachial plexus and the treatment of Erb’s palsy and infantile paralysis of the upper extremity by cross-union of nerve roots. Br Med J 2:1035–1038 Hentz VR (2001) Palliative surgery: elbow paralysis. In: Gilbert A (ed) Brachial plexus injuries federation of european societies for surgery of the hand (FESSH). Martin Dunitz, London, pp 261–274, ISBN 1-84184-015-7, Chapter 26 Hoeksma AF, Steeg AMT, Dijkstra P, Nellisen RGHH, Beelen A, Jong BAD (2003) Shoulder contracture and osseous deformity in obstetrical brachial plexus injuries. J BoneJoint Surg 85A:316–322 Hoeksma AF, Steeg AM, Nellssen RG, van Ouwerkerk WJ, Lankhorst GJ, de Jong BA (2004) Neurological recovery in obstetrical brachial plexus injuries: an historical cohort study. Dev Med Child Neurol 46:76–83 Huang YG, Chen L, Gu YD, Gu GR (2008) Histopathological basis of Horner’s syndrome in obstetric brachial plexus palsy differs from that in adult brachial plexus injury. Muscle Nerve 37:632–637
Surgical Disorders of the Peripheral Nerves Hui JP, Torode IP (2003) Changing glenoid version after open reduction of shoulders in children with obstetric brachial plexus palsy. J Paediatr Orthop 23:109–113 Iyer VG (2000) Developmental maturation of the nervous system. In: Gupta A, Kay SPJ, Scheker LR (eds) The growing hand, 1st edn. Mosby, St. Louis, MO, New York, pp 47–52, ISBN OT 234 21331 Jones R (1916) On suture of nerves, and alternative methods of treatment by transplantation of tendons. Brit Med J 1:641–643 Joseph B (2000) Elbow problems in children. In: Gupta A, Kay SPJ, Scheker LR (eds) The growing hand, 1st edn. Mosby, New York, pp 769–782, ISBN OT 234 21331 Kambhampati SLS, Birch R, Cobiella C, Chen L (2006) Posterior subluxation and dislocation of the shoulder in obstetric brachial plexus palsy. J Bone Joint Surg 88B:213–219 Kawano K, Nagano A, Ochiai N, Kondo T, Mikami Y, Tajiri Y (2007) Restoration of elbow flexion by intercostal nerve transfer for obstetrical paralysis with co-contraction of the biceps and the triceps. J Hand Surg 32E:421–426 Kay SPJ (1998) Obstetrical brachial palsy. Rev Brit J Plast Surg 51:43–50 Kennedy R (1903) Suture of the brachial plexus in birth paralysis of the upper extremity. Br Med J 1:298–301 Kirjavainen M, Remes V, Peltonen J, Kinnunen P, Döyhiä T, Telaranta T, Alanen M, Helenius I, Niet Sovaara Y (2007) Long term results of surgery for brachial plexus birth palsy. J Bone Joint Surg 89A:18–26 Kirkos JM, Papadopoulos IA (1998) Late treatment of brachial plexus palsy secondary to birth injuries: rotation osteotomy of the proximal part of the humerus. J Bone Joint Surg 80A:1477–1983 Klumpke A (1885) Contribution a L’etude des Paralysies Radiculaires du Plexus Brachial: paralysies radiculaires Totales. Rev Med (Paris) 5:591–616 Krone S (2005) Ultrasound guided regional anaesthesia. Invited paper to Advanced Orthopaedic Anaesthesia and Critical Care. Royal National Orthopaedic Hospital, Stanmore, Middlesex UK. 16th September 2005 Landi A, Schoenhuber R, Funicello R, Rasio G, Esposito M (1992) Compartment syndrome of the scapula. Definition on clinical, neurophysiological and magnetic resonance dat. Ann Hand Surg 11(5):383–388 Laurent JP, Lee R, Shenaq S, Parke JT, Solis IS, Kowalik L (1993) Neurosurgical correction of upper brachial plexus birth injuries. J Neurosurg 79:197–203 MacNamara P, Yam A, Pringle J (2009) Biceps muscle trauma at birth – pseudo tumour formation is a cause of poor elbow flexion and supination J Bone Jt Surg 91(8):1086–1089. Malessy MJS, Pondaag W (2009) Obstetric brachial plexus injuries. In: Spinner RJ, Winfree CJ (eds) Neurosurg Clin North Am 20:1–14 Mallet J (1972) Paralysie obstétricale du plexus brachiale. Symposium sous le direction de J Mallet. Avec la collaboration de. In: Arthris M, Castaing J, Dubousset J, Fayse R, Isch Fr, Lacheretza M, Masse P, Rigault P (eds) Rev Chir Orthop 58:115–204 Metaizeau JP, Gayet C, Periat F (1979) Les Lésions obstétricales du plexus brachial. Chir Pédiatr 20:159–163 Michelow BJ, Clarke HM, Curtis CG, Zuker RM, Seifs Y, Andrews DF (1994) The natural history of obstetrical brachial plexus palsy. Plast Reconstr Surg 93:675–680 Morelli E, Raimondi PL, Saporiti E (1984). Il loro trattamento precoce. In: Pipino F (ed) Le Paralisi Ostetriche. Aulo Gaggi, Bologna, pp 57–76 Narakas AO (1987) Obstetrical brachial plexus injuries. In: Lamb DW (ed) The paralysed hand. Churchill Livingstone, Edinburgh, p 116 Nath RK, Paizi M (2007) Improvement in abduction of the shoulder after reconstructive soft tissue procedures in obstetric brachial plexus palsy. J Bone Joint Surg 89B:620–6 Nehme A, Kany J, Sales-De-Gauzy J, Charlet JP, Dautel G, Cahuzal JP (2001) Obstetrical brachial plexus palsy, predictions of outcome in upper root injuries. J Hand Surg 27B:9–12
Birth Lesions of the Brachial Plexus Noetzel MJ, Walpow JR (2000) Emerging concepts in pathophysiology of recovery of neonatal brachial plexus injury. Neurology 55:112–114 Ozkan T, Aydin A, Ozer K, Ozturk K, Durmaz H, Ozkan S (2004) A surgical technique for paediatric forearm pronation: brachioradialis re-routing with interosseous membrane release. J Hand Surg 29A:22–27 Pearl ML, Edgerton BW (1998) Glenoid deformity secondary to brachial plexus birth palsy. J Bone Joint Surg 80A:659–667 Pearl ML, Edgerton BW, Kon DS, Darakjian AB, Kosco AE, Kazimiroth PB, Burchette RJ (2003) Comparison of arthroscopic findings with magnetic resonance imaging and arthrography in children with gleno humeral deformities secondary to brachial plexus birth palsy. J Bone Joint Surg 85A:890–898 Pearl ML, Edgerton BW, Kazimiroff PA, Burchette J, Wong W (2006) Arthroscopic release and latissimus dorsi transfer for shoulder internal rotation contractures and gleno-humeral deformity secondary to brachial plexus birth palsy. J Bone Joint Surg 88A:564–574 Pitt M, Vredeveld JW (2005) The role of electromyography in the management of the brachial plexus palsy of the neonate. Clin Neurophysiol 116:1756–61 Pondaag W, Malessy MJA (2006) Recovery of hand function following nerve grafting and transfer in obstetric brachial plexus lesions. J Neurosurg 105(1 suppl Paediatrics):33–40 Pondaag W, Malessy MJS, Van Dijk JG, Thomeer RTWM (2004) Natural history of obstetric brachial plexus palsy: a systematic review. Dev Med Child Neurol 46:138–144 Pondaag W, De Boer R, Van Wijlen-Hempel MS, Hostede-Buitenhuis SM, Malessy MJ (2005) External rotation as a result of suprascapular nerve neurotisation in obstetric brachial plexus palsy. Neurosurgery 57:530–7 Pondaag W, Lieven PAJ, Van Der Veken MD, van Someren PJ, Van Dijk JG, Malessy MJA (2008) Intraoperative nerve action and compound motor action potential recordings in patients with obstetric brachial plexus lesions. J Neurosurg 109:946–954 Price AE, Ditaranto P, Yaylali I, Tidwekl MA, Grossman JAI (2007) Botulinum toxin type A as an adjunct to the surgical treatment of the medial rotation deformity of the shoulder in birth injuries of the brachial plexus. J Bone Joint Surg 89B:327–9 Putti V (1932) Analisi della triada radiosintomatica degli stati di prelussazione. Chir Organi Mov XVII:453–459 Raimondi P (1993) Evaluation of the hand, 1st edn. Presented to the International Meeting on Obstetric Brachial Plexus Palsy, Heerlen, 1993 and cited In: Surgical Disorders of the Peripheral Nerves. In: Birch R, Bonney G, Wynn Parry C (eds) Churchill Livingstone Edinburgh, London Rollnick JD, Hierner R, Schubert M et al (2000) Botulinum toxin treatment of co contractions after birth-related brachial plexus lesions. Neurology 55:112–114 Roth G (1983) Reinnervation dans la paralysie plexulaire brachiale obstetricale. J Neurol Sci 58:103–115 Saifuddin A, Heffernan G, Birch R (2002) Ultrasound diagnosis of shoulder congruity in chronic obstetric brachial plexus palsy. J Bone Joint Surg 84B(1):100–103 Scaglietti O (1938) The obstetrical shoulder trauma. Surg Gynae Obstet 66:868–877 Sever JW (1925) Obstetrical paralysis. Report of eleven hundred cases. J Amer Med Ass 85:1862–1865 Slooff ACJ, Blaauw G (1995) Aspects particuliers. In: Les Paralysies du Plexus Brachial, 2nd edn. In: Alnot J-Y, Narakas A (eds) Monographie de la Société Français de Chirurgie de la Main. Expansion Scientifique Français, pp 282–284 Sluijs JA, Van Ouwerkerk WJR, De Gast A, Wuisman P, Nollet F, Manoliu RA (2002) Retroversion of the humeral head in children with obstetric brachial plexus lesion. J Bone Jt Surg 84B:583–87 Sluisz JA (2003) Secondary deformities of the shoulder in obstetric brachial plexus lesions. Thesis “Vrijem Universiteit” Amsterdam with summary in Dutch, ISBN 90 5170 661 8
481 Sluisz JA, Van Ouwerkerk WJR, De Gast A, Wuisman P, Nollet F, Manoliu RA (2001) Deformities of the shoulder in infants younger than 12 months with an obstetric lesion of the brachial plexus. J Bone and Joint Surg 83B:551–555 Smith SJM (1996) The role of neurophysiological investigation in traumatic brachial plexus injuries in adults and children. J Hand Surg 21B:145–148 Smith SJM (1998) Electrodiagnosis. In: Birch R, Bonney G, Wynn Parry C (eds) Surgical disorders of the peripheral nerves, 1st edn. Churchill Livingstone Edinburgh, London, Chapter 19, pp 467–490 Smith NJ, Rowan P, Benson L, Ezaki M, Carter P (2004) Neonatal brachial plexus palsy. Outcome of absent biceps function at three months of age. J Bone Joint Surg 86A(10):2163–2170 Stripp WJ (1997) Special techniques in orthopaedic radiology. In: Murray RO, Jacobson HG (eds) The radiology of skeletal disrders, vol 3. Churchill Livingstone, Edinburgh, pp 1879–1940 Strömbeck C, Rehmal S, Krum Linde-Sundholm L, Sejersen T (2007a) Long term functional follow up of a cohort of children with obstetric brachial plexus palsy: I; Functional aspects. Dev Med Child Neurol 49(3):198–203 Strömbeck C, Rehmal S, Krum Linde-Sundholm L, Sejersen T (2007b) Long term functional follow up of a cohort of children with obstetric brachial plexus palsy: II: Neurophysiological aspects. Dev Med Child Neurol 49(3):204–9 Tan KL (1973) Brachial palsy. J Obs Gynae Brit Commonwealth 80:60–62 Tavakkolizadeh A (2007) Risk factors associated with obstetric brachial plexus palsy. Dissertation. University of Brighton for degree of MSc Thorburn W (1903) Obstetrical paralysis. J Obstet Gynaecol Br Emp 3:454–8 Ubachs JMN, Slooff ACJ (2001) Aetiology. In: Gilbert A, Martin Dunitz (eds) Brachial plexus injuries. Martin Dunitz, United Kingdom, pp 151–157 Vredeveld JW, Blaauw G, Slooff BA, Richards R, Rozeman SC (2000) The findings in paediatric obstetric brachial plexus palsy differ from those in older patients: a suggested explanation. Dev Med Child Neurol 42:158–161 Wahlström MP, Backman C, Birch R (2005) Treatment of lack of external rotation of the shoulder in OBP. Paper read to FESSSH Congress X. J Bone Joint Surg 30B(Suppl):P 73 Waters PM (1999) Comparison of the natural history, the outcome of microsurgical repair, and the outcome of operative reconstruction in brachial plexus birth palsy. J Bone Joint Surg 81A:649–659 Waters PM, Bae DS (2005) Effect of tendon transfers and extra-articular soft tissue balancing on gleno-humeral development in brachial plexus birth palsy. J Bone Joint Surg 87A:320–325 Waters PM, Bae DS (2006) The effect of derotation humeral osteotomy on global shoulder function in brachial plexus birth palsy. J Bone Joint Surg 88A:1035–1042 Waters PM, Smith GR, Jaramillo D (1998) Gleno humeral deformity secondary to brachial plexus birth palsy. J Bone Joint Surg 80A:668–677 Whitman R (1905) The treatment of congenital and acquired luxations at the shoulder in childhood. JAMA 42:110–115 Yagi 1 (1984) Clinical study of cross-reinnervation in obstetrical paralysis. J Jap Orth Assoc 58:761–778 Yam A, Fullilove S, Sinisi M, Fox M (2009) The supination deformity and associated deformities of the upper limb in severe birth lesions of the brachial plexus. J Bone Joint Surg 91B:511–16 Yang L, Smith SJM (2003) Pers com. Spontaneous recovery of hand function in the group 4 lesion Yang LJ, Anand P, Birch R (2005) Limb preference in children with obstetric brachial plexus palsy. Paediatr Neurol 2005 33:46–49 Zancolli EA (2001) Palliative surgery: prono-supination in obstetrical palsy. In: Gilbert A (ed) Brachial plexus injuries federation of european societies for surgery of the hand (FESSH). Martin Dunitz, London, pp 275–292, ISBN 1-84184-015-7 Chapter 27
Iatrogenous Injuries Part 1: General Considerations
11
Rolfe Birch
Incidence, delay and neglect; failure of audit. Causes: underlying disease; warning and consent; recognition; records; teaching, training and specialisation; continuity of care. The nerve lesion in hip arthroplasty. Post irradiation neuropathy. Prevention. The number of patients seen in our clinics with injuries to peripheral nerves incurred during treatment since 1979 is about 2,200 and if nerve lesions incurred during arthroplasty of the hip and after irradiation are included, that total exceeds 2,600. Every year sees an increase in the number of patients referred with iatrogenous injuries,1 an increase in the delay between the event and diagnosis and treatment and a decrease in the proportion of cases recognised and treated by, or recognised and referred by, the responsible clinician. These are alarming trends. The growing problem of the iatrogenous lesion was one major reason for undertaking the second edition of this work. The cause and effects of these accidents have been embedded throughout the whole work rather than being sequestered into one chapter so that examples of compression, crush, injection injury and ischaemia in all its forms are related in Chap. 3, the particular problems surrounding fracture work in Chap. 8 and some suggestions about avoidance during operation are set out in Chap. 7. This chapter will concentrate on injuries to main nerves incurred during operation, which represent about one half of all of the cases seen. Nerve lesions associated with arthroplasty of the hip and post irradiation neuropathy will be considered. The earlier report (Birch et al 1991) described 68 cases of partial or complete transection of a main nerve treated during the years 1984–1990 excluding injuries to cutaneous nerves and lesions associated with arthroplasty of the hip. These were equally divided between orthopaedic and other surgical disciplines. Most patients were referred by the clinician responsible with an average delay of 6 months. The later report, with the same exclusions, (Khan and Birch 1998) described findings in 621 patients seen in the years between 1991 and 1998. A total of 241 main nerves were found damaged in the 291 patients coming to further 1
he preference for the use of the term iatropathic in place of iatrogenic T was explained in the first edition. It is at the suggestion of Simon Kay that we revert to our first choice, iatrogenous, implying produced by doctors. We may be accused of pedantry but there is no harm in seeking precision.
operation, an increase of nearly fourfold over the earlier study. One hundred and seventy four patients sustained nerve lesions during orthopaedic or fracture operations and the average delay between the event and treatment had risen to 10 months. One in three of the referrals came from clinicians who were not involved in the first intervention; 34 patients were referred by solicitors. The nerves most commonly damaged were, in the upper limb, the spinal accessory, the brachial plexus, the radial and the ulnar nerves and in the lower limb the common peroneal nerve. Patients were most at risk during operations for excision of tumour or cyst, lymph node biopsy, internal fixation of fracture and varicose vein and arterial surgery. These four types of intervention accounted for 70% of all cases. Since 1998 more then 1,100 new patients have been seen with nerve injuries incurred during treatment and in many of these it was possible to identify interference by managers and others with the process of treatment, rehabilitation and referral. For many patients the concept of continuity of care has collapsed. There has been a substantial increase in the numbers of lesions incurred during ‘minimally invasive’ operations, in cases treated in Day Care Units and a new flow from Independent Sector Treatment Centres (ISTC). The increase in ischaemic lesions is alarming (Birch 2008). By 2007 the iatrogenous lesion accounted for nearly one third of all referrals to the Peripheral Nerve Injury Unit. It seems that the problem is a universal one. Kretschmer et al (2001) reported that over 17% of referrals to their Regional Neurosurgical Unit in Ulm were iatrogenous and in only a minority of cases was the referral made by the responsible clinician. The mean interval from the event to treatment was 21 months: ‘at least two thirds of patients did not undergo surgery for the iatrogenic lesion within an optimal time interval due to the delay of referral’. Kömürcü and colleagues (2005) describe 82 cases seen during the course of 10 years and emphasize the importance of an urgent response, especially in the haematoma caused by angiography: ‘delay is associated with poor outcomes in corrective surgery’. In an invited commentary Dellon (2005) wrote ‘failure to make the diagnosis of a nerve
R. Birch, Surgical Disorders of the Peripheral Nerves, DOI: 10.1007/978-1-84882-108-8_11, © Springer-Verlag London Limited 2011
483
484
injury, and failure to treat that complication of the first surgery, the iatrogenic nerve injury, is as much a cause for concern as the initial injury to the peripheral nerve’. Dellon is quite right and it seems that the possibilities of improving the patient’s situation by proper treatment are being increasingly ignored. In our most recent investigation of 194 nerve lesions in 174 patients seen consecutively, potentially significant underlying disorders in 42 cases were recognised, including endocrine or connective tissue disease, peripheral vascular disease and malignant disease. The median interval between the event and referral was 25 weeks. This interval was about 12 weeks for cases caused by nurses, ENT surgeons and anaesthetists; it was 23 weeks for cases caused by physicians, and orthopaedic or vascular surgeons but it rose to 63 weeks in cases caused by general surgeons. The delay was considerably greater in injuries inflicted during radiological procedures. The responsible doctor made the diagnosis and organised referral for 88 of 169 patients. However, many surgeons minimise disability by prompt recognition and treatment. Case Report: a 19 year old woman, 5¢4 in height and weighing 7 stones, sustained a closed fracture through the middle of the right femur. Operation was undertaken 9 days after injury; during clearance with the scissors of soft tissues overlying the upper end of the femur, the muscles of the calf were seen to twitch. The exposure was extended, and it was found that the sciatic nerve had been cut. It was repaired then and there with sutures of 6/0 prolene. A month later the operator and one of us explored the site of the lesion: the suture line was intact, and the bundles were seen to run within a few millimetres of it. When the patient was seen 2 years later there was remarkably good recovery of motor power and sensibility in the distribution of both components of damaged nerve. This surgeon not only recognised instantly the fact of damage, but also proceeded at once, to repair, and later sought advice from one with particular experience in the field. Other similarly commendably handled accidents have been followed in the Peripheral Nerve Injury clinic at the operating surgeon’s request, and in more than 150 cases the operating surgeon ensured rapid transfer so that the lesion could be rectified within a few days. Some operators are insufficiently familiar with topographical anatomy; some are unaware that nerves may be distorted or displaced by pathological changes and that there are anomalies of position and size; many are unaware of the concentration of function in the peripheral nervous tissue and of the vulnerability of that tissue relative to that of bone and muscle. The use, during operation of agents acting on the neuromuscular junction ensures that transmission is usually blocked, so that accidental section or pinching of a nerve is not signalled by a contraction of muscle. Under such circumstances too, electrical stimulation for identification becomes unreliable. Section of a main nerve does not produce the instant, obvious catastrophic effect produced by section of a main artery or vein.
Surgical Disorders of the Peripheral Nerves
The enforced revival of ‘day case’ operations in the interests of economy has too often led to operations being done by inadequately qualified persons, under conditions inimical to the maintenance of standards of care (Bonney 1992). Increasingly, it seems that surgeons do not see their patients after operation, indeed there is much pressure applied by some senior ‘medical managers’, commissioning agencies and by other managers to weaken or to abolish adequate post operative follow up. When mistakes are made by surgeons acting under such conditions, the onus is, or should be, as much on managers as on clinicians, though in practice it is always the clinicians who are obliged to justify their actions in open court. In some cases the obstacles to treatment are imposed by managers within the hospital to which the injured patient has been sent as a tertiary referral. In one such hospital a senior manager suggested that to give such a case any priority was unfair because that patient had ‘already had their bite of the cherry’. Such attitudes are not confined to managers; some of the worst cases of neglect of injuries to main nerves have come from Independent Sector Treatment Centres (ISTC) as illustrated by the following four examples. Case Report: A 50 year old fitter, in good general health, presented with a short history of pain in the knee followed by foot drop. The nerve was explored, a ganglion was seen arising from the nerve and this, with the nerve, was excised. The patient had complete common peroneal palsy and considerable pain. The patient, a good witness, advised us that the operating surgeon had said: ‘there is nothing to be done, nerves cannot regrow’. This ‘opinion’ was given to the family practitioner who wrote to us seeking further advice. The nerve was repaired at 10 months after injury. Case Report: Total knee replacement was performed in an 84 year old man, who sustained deep and very painful lesions of tibial and of common peroneal nerves. The operating surgeon took no action. The patient’s family practitioner organised good support from rehabilitation services and treated the pain by appropriate medication with success. We saw him at 15 months by which time the ulceration of the anaesthetic foot had healed and pain was under control. The patient was exceptionally strong willed, an active gardener and wine maker, had found that two pairs of sea boot socks protected his foot. There was no recovery for either nerve. Case Report: A 48 year old woman suffered transection of the femoral nerve in the course of total hip arthroplasty. Her attitude was one of resigned acceptance, a common attitude nowadays, but she did relate to us that her complaints of pain and of loss of function were dismissed by the operating surgeon. Her family practitioner sought a second opinion from a senior orthopaedic surgeon in the area, who referred her to us. The severed nerve was repaired 10 weeks after the event. Case Report: A 21 year old woman suffered transection of the spinal accessory nerve in the posterior triangle of the
Iatrogenous Injuries
neck in the course of lymph node biopsy. Her complaints of pain and loss of function were ignored. It was left to her family practitioner to arrange for palliative work with physiotherapists and she was ultimately referred to our Unit by her solicitor 50 months after the event.
11.1 Incidence and Audit The removal from the Defence Organisations of responsibility for indemnification in cases occurring in the National Hospital and Community Services, placed upon individual authorities and Trusts burdens which they were ill prepared to bear. One observer at least foresaw this hazard (Bonney 1990). Sadly perhaps, few then or later, believed his prophecies. The transfer of responsibility for indemnification also destroyed the central data bank of information on cases of medical negligence operated by the Defence Organisations, and so made the task of risk management more difficult. The possibility of a systematic analysis of the causes of action in respect of medical negligence and the identification of patients intrinsically likely to take action was destroyed. The tardy recognition of these drawbacks led, in 1995, to proposals for the formation of a Special Health Authority (The Litigation Authority) to oversee the operation of a central fund to spread the burden of indemnification. The number of cases registered these Agencies represent no more than a fraction of the whole number of iatrogenous injuries. However, important studies are available. Mrs Suzanne Collinge and Dr. Michael Saunders of the Medical Defence Union, provided carefully ‘anonymised’ information about cases concerning peripheral nerve injury either current during or settled in 1986. Of a total of 95 instances of nerve injury, the facial nerve led all the rest, with 17 cases, followed by the common peroneal (14 cases) the sciatic (9 cases), the radial, median and digital nerves (8 cases each), and the brachial plexus (5 cases). There was only one case of damage to the spinal accessory nerve during the biopsy of a cervical lymph node. In that year alone we saw five patients with lesions to the spinal accessory nerve incurred during operations of lymph node biopsy. Campbell, France and Goodwin (2002) analysed 424 medico-legal claims in vascular surgery handled by the Medical Defence Union between 1990 and 1999 and by the NHS Litigation Authority between 1995 and 1999. Nerve lesions accounted for 76 of the 244 claims arising from varicose vein surgery and for 16 of the 174 actions for arterial surgery; there were also 11 from complications of upper thoracic sympathectomy and 7 more for complications in excision of a cervical or first thoracic rib. Sixty three patients with nerve injuries complicating operations in these fields were seen in the PNI Unit during the same 10 years. McNeill and Mathewson (2006) recorded
485
62 negligence claims for the treatment of neurological disease which had been referred to the Medical and Defence Union of Scotland between 1980 and 2004. There was major permanent injury in 42 patients and 6 deaths. The procedure was deemed improperly performed in one half of cases and an error in diagnosis occurred in 26. The true figure can be reached only by meticulous audit, by a process of scrutinising practise so to improve it. The British Association of Otorhinolaryngologists (2002) provided guidelines for the treatment of the vestibular schwannoma which provides an example of such audit: O’Regan and colleagues (2007) provide another in their report on protection of the facial nerve during parotid surgery for benign disease. Audit is central to medical practice yet it seems that the process is increasingly patchy, incomplete and weakened by interference with adequate follow up of patients. Banga and colleagues (2006) found that the operating registrar saw less than one half of their cases of myringoplasty after operation. The operation failed in 30% of cases, and less than one third of the failures were seen by the operating registrar. Pollock (2007) refers to the dissolution of audit in a devastating critique of private provision in the NHS: ‘Once less than 6% of the operating budget, transaction costs rose to 12% following the introduction of the internal market into the NHS in 1991, and with the introduction of a real market will rise rapidly to approach 31%’; ‘ Some hospitals are paying 12–20% of their annual income in PFI charges. After pharmaceutical costs and administrative costs are added in, then there is little left for patient care’ … ‘in any case, the government is moving rapidly to private data and its analysis, handing over patient information to companies like Dr. Foster, so that it will be impossible to know the truth’. Pandey and colleagues (2007) revealed serious defects and errors in the work of such ‘medical benchmarking companies’ in the outcome of abdominal aortic aneurysm. It was alleged that the standardised mortality ratio, in one hospital, was 160, this in a Unit of outstanding distinction: it was in fact 67. Audit must be reliable. The study from Milburn et al (2007) is disheartening. Audit data from individual Units was compared with the data assembled by the Information and Statistics division of the Scottish Executive (ISD). Sixteen per cent of the ISD entries were duplicated and in 21% of unpaired entries the wrong consultant was recorded. It is quite wrong to think of audit and long term studies of outcome as ‘research’ and it is difficult to see any role for local research and ethical committees. The practice of seeking ethical approval for such audit was sharply rebuked by Wroblewski (2006) who criticised a paper in which it was said that: ‘ ethical approval for the follow up study was obtained and the patients were contacted’. Wroblewski writes: ‘does this mean that the authors did not follow up their patients even when they “predicted that overall, the implant would have a high rate of failure” ’. Wroblewski has
486
kindly provided copies of the letters sent by Sir John Charnley to his patients in which he sought their consent for the study of the prostheses used in total hip arthroplasty during their life and after their death so that he could remove the prostheses post mortem. It seems that audit is particularly defective in Independent Sector Treatment Centres. Oussedik and Haddad (2009) raised doubts about the performance of Treatment Centres in the provision of elective orthopedic surgery and they commented that: ‘the poor quality of data collected at ISTC’s makes their performance difficult to evaluate’. White, John and Jones (2009) studied the short term results of total hip replacements performed by visiting surgeons at an NHS Treatment Centre. The failures were sent to the authors in their NHS hospital. Twenty arthroplasties had been revised or were awaiting revision from a total of 156 arthroplasties in 136 patients, at a mean interval of 23 months: ‘The revision rate far exceeds the 0.5% 5 year failure rate recorded in the Swedish registry for the component used’. The problems surrounding the quality of elective surgery in Treatment Centres is the subject of a review by Cannon (2009) writing as immediate past President of the British Orthopaedic Association (BOA). Cannon charts the development of Treatment Centres, some run by the NHS and some by the Independent Sector and he comments: ‘however one of the most galling features to the established NHS base was that the providers of these facilities were given 40% above reference costs to deliver the service. Provider quality was to be monitored by a wide range of key performance indicators, which were management orientated matrices with little regard for actual clinical outcome’. The growing concerns about the quality of work in the Treatment Centres raised by the membership of the BOA led to a call for comparative audit by the Association of the activity and outcomes in Treatment Centres and in established NHS Units to determine complication rates, morbidity and unit cost: ‘our concerns were made known to the commercial directorate of the Department of Health, the Audit Commission and the Health Care Commission. In reply, on behalf of the Commercial Directorate, Ken Anderson, the commercial Director, made a statement of behalf of the Secretary of State for Health which made little reference to the perceived problems within the orthopaedic community. Unfortunately, he reported an overall satisfactory rate for patients at over 94%’ (House of Commons Health Committee: a statement for the Secretary of State for Health 2006). The evidence supports the statements made by Cannon and his colleagues. One of the reasons for this is the establishment of the National Joint Registry, one of several important initiatives coming from the BOA. Sibanda and colleagues (2008) described the revision rate in 80,697 knee replacements, representing about one half of the total performed in England and Wales between 2003 and 2006. No patients from Treatment Centres were included. The rate of revision
Surgical Disorders of the Peripheral Nerves
at 3 years after cemented knee replacements was 1.4%, and 1.5% for uncemented prostheses. The revision rate for unicondylar prostheses was 2.8%. These figures can be compared with those described by Kempshall et al (2009) who studied the outcome in 258 total knee arthroplasties in 224 patients, sent to an NHS Treatment centre by their NHS Trust. The results were unsatisfactory in 76 (37%) of knees, and the cumulative survival rate at 3 years (79.2%) fell far below published survival rates at 10 years. Audit must remain an integral part of the practice of medicine and it should extend, if need be, over many years. The principles of audit have been gravely damaged by senior managers driven by their political masters seeking short term solutions.
11.2 Causes Some causes underlying the increasing numbers of iatrogenous lesions will be explored. They include: possible underlying systemic causes of neuropathy; changes in education and training; subspecialisation; warning and consent, and the breakdown of continuity of care.
11.2.1 Generalised Disorders Neurological symptoms may be the first sign of a generalised disorder and this is especially so in entrapment neuropathy and nerve tumours. Case Report: A 40 year old woman presented with a swelling at the left elbow. As a child she had had two operations for acoustic neuroma and three proven schwannomas had been removed from different sites since then. There was a family history of neural tumors and of intracranial neoplasm but in spite of this the likelihood of NF2 was not considered and the likelihood of the swelling being a schwannoma was not entertained. A short incision was used to expose the swelling which was removed. It was later confirmed as a schwannoma. She developed causalgia on the day of operation which was only partially relieved by endoscopic cervical sympathectomy 2 weeks later. We were asked to see her 2 years later because of persisting severe pain expressed in the territory of the median nerve. Hand function was diminished by loss of sensation, by weakness, and also by the dryness of the skin. There was a static, painful Tinel sign at the scar. The nerve was re-explored through an adequate incision and was found densely adherent to the brachial artery. No residual tumour could be seen. External neurolysis was done and an indwelling catheter allowed the infusion of local anaesthetic for forty 8 h. Her pain was improved from PNI
Iatrogenous Injuries
Grade 3 to Grade 1 and there was some improvement in hand function Case Report: A 38 year old man, in good general health, underwent decompression of the lateral cutaneous nerve of the thigh which relieved his symptoms for 18 months. His symptoms returned and they were accompanied by a sense of heaviness in the affected (left) lower limb and feelings of numbness down the outer side of the thigh. An MR scan of the lumbar spine was normal. We saw him at 24 months after operation when weakness in the extensor muscles of the leg was evident. The tendon reflexes were preserved. A diagnosis of multiple sclerosis was confirmed by Dr. Geoffrey Schott (Queen Square) by MR scan of the brain which revealed multiple supratentorial lesions in the white matter and by examination of the cerebro-spinal fluid which confirmed the presence of oligoclonal bands. Case Report: A 38 year old healthy woman was treated, for thoracic outlet syndrome, by transaxillary resection of the seventh cervical and first thoracic ribs. On the evening after operation she experienced intense pain and by 24 h had developed a complete palsy of C8 and T1. We saw her 7 months later when she still had severe pain and a complete lesion of C8 and T1. There was no Claude Bernard Horner sign. At operation the lower and middle trunks were found enveloped within and distorted by fibrosis: ‘in appearance reminiscent of a barley sugar twist or Jacobean table leg, and there were palpable swellings extending along 6 cm within the nerve trunks’. Her pain was improved but a prolonged course of rehabilitation proved necessary to overcome the hypersensitivity and hand function was improved by musculo tendinous transfers and by sensory retraining. It was during one of these admissions that the potential significance of an elevated platelet count was recognised by our then senior house officer, Mr. Max Horwitz. A diagnosis of thrombocythaemia was confirmed and this was treated with aspirin and anegrolide, which eased her persisting circulatory symptoms. It is probable that the blood disorder was the major cause of her symptoms before the first operation and the causalgia and the palsy of the lower trunk was caused by haematoma. Serious mistakes can arise from concentration on local and peripheral manifestations to the neglect of enquiry into an underlying cause. There are many causes of generalised neuropathy; many general disorders which by their local manifestations cause local neuropathy. ‘Carpal tunnel syndrome’ may be the first symptom of rheumatoid arthritis; a peripheral neuropathy may be the first symptom of carcinoma of the lung. The clinical features of polyneuropathy, a generalised neuropathy affecting all peripheral nerve fibres, are admirably described by Thompson and Thomas (2005) and by Donaghy (2009a). Bilateral and symmetrical muscle weakness, sensory loss, loss of tendon reflexes and often some disturbance of sympathetic function are important features which can be easily demonstrate in the Outpatient Department. Some conditions causing
487
neuropathy which the operating clinician is likely to encounter include: alcoholic neuropathy; diabetes mellitus; hypothyroidism and acromegaly; connective tissue disease; malignant disease, and that strange disorder, immune brachial plexus neuropathy (neuralgic amyotrophy).
11.2.2 Alcohol It is impossible to know the incidence of symptomless alcoholic neuropathy, brought to light, perhaps, by the unexpected occurrence of a peripheral nerve lesion during the course of treatment, operative or otherwise. On the other hand, the clinician is familiar with the presenting features of established neuropathy with impairment of peripheral sensibility associated perhaps with encephalopathy and evidence of impairment of liver function. The features of alcoholic autonomic neuropathy are also well known. The alcoholic is often reticent, even deliberately evasive, about his or her drinking habits, though often enough the finding in the blood of a persistently high mean corpuscular volume gives the game away. It may well be that some of the nerve lesions ascribed to mistakes in operation represent the effect on susceptible neural tissue of trauma which would be tolerated by a healthy nerve. The work of Monforte and colleagues (1995) appears to confirm the view that alcoholic neuropathy is produced by the direct, dose related action of alcohol or its metabolites on the axon or the cell body, and that associated malnutrition is not a factor. These workers further found that one third of their patients with high alcohol intake showed electrophysiological abnormalities of the peripheral nerves. Evidently, it is as well to be particularly cautious about the handling of nerves of patients who admit to an alcohol intake of more than 100 g daily or whose mean corpuscular volume is over 100.
11.2.3 Diabetes Diabetic neuropathy usually comes to attention as an affection of sensibility beginning distally and extending proximally. The lesion is predominantly axonal (Thomas and Lascelles 1966); Llewellyn, Tomlinson, and Thomas (2005) say that the small unmyelinated fibres and the myelinated sensory fibres are the most commonly affected and describe the abnormalities that affect the blood vessels of the endoneurium, the perineurium and the epineurium. The endoneurial microangiopathy is evident before the neuropathy appears and there is a correlation between the severity of microvascular abnormality with that of the polyneuropathy. The foot is at risk from ulceration. An autonomic neuropathy may be
488
associated with a sensory neuropathy or it may present in isolation and these patients are at risk during anaesthesia. The property of nerves of diabetics that principally interests or ought to interest the operating surgeon is that of abnormal sensitivity to pressure. The Rochester diabetic neuropathy study (Dyck et al 2005) revealed an incidence of carpal tunnel syndrome in 11% of insulin dependent diabetics. The practical conclusion from all this is that the routine testing of the urine of all new patients and of all in-patients, now in many centres fallen into desuetude, remains mandatory.
11.2.3.1 Hypothyroidism and Acromegaly As Pollard (2005) remarks, ‘neurologic complications of hypothyroidism are well recognised’. Most clinicians know that ‘carpal tunnel syndrome’ may be the symptom that brings hypothyroidism to attention. However, reliable figures are lacking for the percentage of cases of carpal tunnel syndrome in which hypothyroidism is a determining factor. Nemni and colleagues (1987) examined the clinical, electrophysiological and morphological changes in four patients with ‘secondary hypothyroidism’ and polyneuropathy. They found not only changes in conductivity but also, in three cases, evidence of axonal degeneration in sural nerve biopsies. The neurophysiological changes were reversible with replacement therapy. Wise and colleagues (1995) indicated the importance of the findings of slow relaxing reflexes in the diagnosis of this condition. ‘Carpal tunnel syndrome’ may be the presenting symptom of acromegaly. (Johnston 1960). O’Duffy and colleagues (1973) found that 35 of 100 acromegalic patients suffered from carpal tunnel syndrome. Low and colleagues (1974) found a high incidence (8 of 11 patients) of generalised peripheral neuropathy, with increase in endoneurial tissue and reduction in density of myelinated fibres. The fascicles were larger in the nerves of acromegalics than in healthy nerves. The changes were independent of the presence of diabetes mellitus. We think that in all cases of entrapment neuropathy in which no local or systemic cause is apparent, tests of thyroid function (free and total thyroxine) should be done. The possibility of acromegaly ought to be considered in all cases of carpal tunnel in men without evidence of a local or other systemic cause.
11.2.4 Connective Tissue Disease Orthopaedic surgeons in particular should be interested in the association of neuropathy with rheumatoid arthritis, but the vasculitis that, at any rate in part, is responsible in that condition is common to many disorders of connective tissues. Hart
Surgical Disorders of the Peripheral Nerves
and Golding (1960) discussed 42 cases of ‘rheumatoid neuropathy’. It is interesting to find that examination of peripheral nerves in their five patients who died showed arteritis in only one and demyelination in one other. Certainly, extraneural changes are in part or wholly responsible for the nerve lesion in some cases of ‘rheumatoid mononeuropathy’. In one of Hart and Golding’s cases that was certainly so: Division of a thickened arcuate ligament sufficed to give complete and lasting relief in a case of ulnar neuropathy. Pallis and Scott (1965) considered in detail the clinical features an pathology of ‘rheumatoid neuropathy’, stressing the frequency of arterial lesions ‘like those of polyarteritis nodosa’. Dyck, Engelstad and Dyck (2005) clarified the distinction between the systemic vasculitic neuropathy affecting patients with polyarteritis, systemic lupus erythematosis, rheumatoid arthritis and other conditions, and the nonsystemic vasculitic neuropathy, in which clinically only nerves are affected. Collins and Kissel (2005) provide an extensive review of neuropathy in connective tissue diseases caused by systemic vasculitis. These are most commonly encountered in patients with rheumatoid arthritis and systemic lupus erythematosis but that they also occur in Sjögren’s syndrome, scleroderma and Behçet’s disease. Clinical studies suggest that the prevalence of peripheral neuropathies in rheumatoid arthritis is 15% a figure which rises to 25% in patients examined by electro diagnosis. Collins and Kissel say that: ‘neuropathies occur in 45% of rheumatoid vasculitis patients, indicating that 10% of all rheumatoid arthritis patients will develop a vasculitic neuropathy’. Clearly, clinicians must be aware of the possibilities of connective tissue disease presenting as a mononeuropathy and of abnormal susceptibility of peripheral nerves to pressure and ischaemia in cases with an established diagnosis of connective tissue disease. We have seen a number of patients with rheumatoid arthritis or other connective tissue diseases who sustained significant nerve lesions from well conducted steroid injections or after nerve blocks using local anaesthetics. Case report: A 63 year old man had been treated for severe rheumatoid arthritis for more than 10 years, requiring 2.5 mg of prednisolone every day. His other medications included methotrexate, atenolol, atorvastatin, folic acid, ropinirole and nicorabdil. A cerebrovascular accident 15 years previously left him with minor residual weakness on the right side. An aortic balloon was inserted before coronary artery bypass graft which was left in place for 6 days. He sustained a severe left femoral palsy which became increasingly painful. He was examined 18 months later when he complained of persisting pain and severe weakness. The power of hip flexion and of knee extension was MRC grade 2. Neurophysiological investigation (NPI) revealed that there was no sensory conduction in the femoral nerve, and whilst there was no spontaneous activity in the quadriceps muscles there were: ‘polyphasic units of long duration, mostly stable, with some rather large units recruiting early to a mildly
Iatrogenous Injuries
reduced interference pattern to 4 mV, suggesting chronic partial denervation and reinnervation changes’ (Dr. Nicholas Murray, Queen Square). In this connection, it is as well to remind clinicians operating on patients with rheumatoid arthritis that nerves in wasted limbs have less protection against external pressure than have nerves in healthier limbs. This obtains for any patient with wasting. Case report: A 48 year old man sustained a fracture of his right humerus which required open reduction and internal fixation. Poliomyelitis, at the age of 18 months, left him with severe weakness and wasting of the muscles of the right shoulder and arm and the left lower limb but function in his right hand was good. On the evening of operation he developed a complete radial and ulnar palsy which became increasingly painful. He was examined 26 months later when NPI (Dr. Nicholas Murray, Queen Square) demonstrated recovery of the radial SAP, persisting absence of the ulnar SAP whilst the ulnar NAP was considerably reduced in amplitude (1.0 mV) with a normal velocity suggesting a lesion which was partial and recovering against a background of extreme denervation. It is likely that the wasting of the muscles of the arm rendered the nerves, already damaged, more vulnerable to retraction and pressure. It should be noted that no active treatment was offered to this man until the intervention of a neurologist (Dr. Kennedy) 24 months after the event when his disability was mitigated by the provision of suitable orthoses. Carcinoma and lympho-proliferative disease: Denny Brown (1948) described sensory neuropathy caused by degeneration of cells in the dorsal root ganglion in non metastatic carcinoma of the lung. The subject was intensively studied by a group of workers at the London Hospital. Croft and Wilkinson (1969) wrote that: ‘as more examples of neuromyopathy are seen, it becomes apparent that a type of neuromyopathy may occur with any type of malignant disease’ although there was a particularly frequent association with small cell lung cancers. Mcleod (1993a) concludes that clinical neuropathy is apparent in 5% of patients with cancer, and that defects are detectable in 12% of patients by quantitative sensory testing. Neurophysiological investigation reveals abnormality in between 30% and 40%. Donaghy (2009b) distinguishes between sensory neuronopathy, sensory motor neuropathy and the neuropathy of vasculitis. A sensory motor neuropathy is the most common pattern, characterised by axonal death affecting all types of fibres. Symptoms may precede manifestation of the primary tumour by up to 15 months, and include numbness, dysaesthesiae and paraesthesiae commencing distally and advancing proximally. There is aching pain, and ataxia and orthostatic hypotension are common findings. Spies and McLeod (2005) define the specific nervous system antibodies present in paraneoplastic neurologic disease, commenting that: ‘the immune response in patients with paroneoplastic syndromes may have some
489
relevance to tumour progression. In the vast majority of patients with paroneoplastic neuropathies, the cancer is limited and in fact is often difficult to diagnose’… ‘overall, there does appear to be improvement in the course of the tumour at the expense of often devastating neurologic disease’. Similar ‘paraneoplastic’ neuropathies are seen in lymphoma, leukaemias and Hodgkin’s disease. McLeod (1993b) says that clinical evidence of involvement of the central nervous system is found in 10–25% of patients with leukaemia, from paraneoplastic neuropathy, bleeding or direct invasion. Bosch et al (2005) provide an extensive review and they refer to the high incidence of neurological complications after allogeneic bone marrow transplantation. Immune Brachial Plexus Neuropathy (Neuralgic amyotrophy): Wilbourn (1993) and Suarez (2005) give good accounts of this singular disorder. It is to Parsonage and Turner (1948) that we owe the term ‘neuralgic amyotrophy’ that so well indicates the combination of pain with paralysis and wasting. Spillane (1943) brought the condition to general notice with his description of a ‘localised neuritis of the shoulder girdle’ occurring in British, Australian, New Zealand and American troops in the Middle East and the United Kingdom. The details of 46 patients were available to him. Spillane noted the features that are now familiar to us: the linkage with infection and wounding; the onset with pain; the common affection of the serratus anterior, the spinati, the deltoid and the trapezius. There is considerable evidence suggesting an inflammatory –immune mechanism. Nerve biopsies often reveal loss of nerve fibres, perineurial thickening and perivascular epineurial or endoneurial inflammatory infiltrates (Suarez 2005). The association of neuralgic amyotrophy and its protean manifestations with precipitating events such as cold, injection, infection and operation is now well known. There is pain, often intense, followed by motor paralysis with remarkably rapid atrophy, sometimes with alteration or loss of sensibility. The muscles most commonly involved are the serratus anterior, the spinati, the deltoid, the trapezius and the latissimus dorsi and teres major. When paralysis occurs after an operation during which the nerve to one of these muscles might have been damaged, it may be very difficult, particularly in retrospect, to distinguish neuralgic amyotrophy from direct damage to the nerve. It is extremely important to ascertain whether other nerves and their muscles are affected and electromyography is helpful. It is best to be cautious about the diagnosis of neuralgic amyotrophy in cases where only one nerve is involved. It is important to remember that operation may precipitate recurrence. Case report: A 40 year old ex-paratrooper experienced intense pain of sudden onset after lifting heavy weights. He was unable to sleep for a week and he thought the pain the worst of any that he had experienced. The pain resolved spontaneously over the course of 48 h and he then noticed loss of movement at his shoulder and winging of his scapula.
490
The spinal accessory and nerve to serratus anterior were involved. Neurophysiological investigation demonstrated normal sensory conduction in the affected upper limb with copious small spiky polyphasic units of low amplitude and short duration in serratus anterior and in the lower fibres of trapezius. MR scan and examination of CSF by lumbar puncture proved normal. He did not recover and 2 years was able to abduct his arm to no more than 50°. Transfer of the sternal head of pectoralis major to the lower pole of the scapula was performed to enhance thoraco scapular stability. The risk of recurrence had been put to the patient and, alas, that risk was fulfilled. He developed very severe pain 24 h after operation which persisted for about 10 days but fortunately there was no significant further loss of muscle power and there was, ultimately, considerable improvement in function. Wilbourn (1993) says: ‘Overall, neuralgic amyotrophy has a good prognosis. A retrospective series of 99 cases found that more than 80% of patients recovered functionally within 2 years of onset and more than 90% by 4 years. In contrast, serious and permanent defect of function usually follows a deep lesion of either the accessory nerve or the nerve to serratus anterior’.
11.2.5 Warning and consent The interested reader will learn a very great deal from the work Clinical Negligence (2008) edited by Michael Powers, Nigel Harris and Anthony Barton. In this work Badenoch and Whitting (2008) refer to some of the benefits which have followed the Woolf Reforms (Woolf 1996). They refer to Lord Woolf’s conclusions that the failure to meet the needs of litigants was due to a number of factors: 1. A disproportion between costs and damages awarded 2. Delay 3. The pursuit of unmeritorious claims 4. The relatively low success rates of claimants in clinical negligence 5. The lack of cooperation between the parties in dispute They comment: ‘as all whose legal practice includes clinical negligence will attest, contested trials in the field have become almost vanishingly rare … that this is so should not be a source of surprise, since front loading has Lord Woolf’s intended effect of focussing the lawyers minds wonderfully at the outset, and well before the time when in the past the writs would fly, on the true strength and merits of the proposed allegations’. The authors define front loading as the requirement for major preparation of all proposed cases before proceedings are launched. A note of warning is sounded by John Baron, MP(2008): ‘while I welcome the act, I nevertheless believe that an opportunity was missed to
Surgical Disorders of the Peripheral Nerves
make the fact finding stage of investigation of disputes truly independent. The danger is that local NHS Trusts will act as judge and jury in their own cause. I question whether this will ensure a thorough investigation of the facts and command the confidence of patients. Time will tell’. Baron goes further: ‘most cases have been legally aided, and yet most people are not eligible for legal aid. Of course, the impoverished and the vulnerable should have access to justice, but historically vast sectors of the population have been denied the same rights’. One of us wrote to Lord Woolf in 1998 suggesting that unthinking application of the doctrine of doctrine of informed consent might carry with it two consequences: that the patient, having signed a document, might take the outcome as inevitable and that the clinician might take a sanguine view having been granted consent with acknowledgement of risk. We see many patients with injuries which no reasonable clinician could possibly countenance, accepting the event, the often intolerable delay in diagnosis and the obstacles placed before those seeking to put the matter right with a resignation worthy of Job. Doctors do have a duty of providing adequate information to their patients. In advising treatment, and in particular in advising operation, the clinician must be aware of the balance of advantage against risk, and should attempt to convey to the patient his or her assessment of that balance. This is not easy: if medicine were easy there would be little need for doctors; none for consultants. The final decision is for the patient, but the clinician should not shrink from giving a clear indication of his or her views. There are procedures which are not life saving, indeed they are scarcely therapeutic which are associated with risks so severe that we question whether they should be used. Infarction of the spinal cord caused by an interscalene block for the purpose of relieving post operative pain is too high a price. Other, and simpler methods of securing pain relief are available. It is illogical to induce a block of the brachial plexus for operations upon the glenohumeral joint from which the somatic afferent pathway chiefly enters the spinal cord through C4 and C5. Little and colleagues (2007) secured relief of pain by placing an epidural catheter into the shoulder joint after arthroscopic sub acromial decompression. Not one of their 40 patients required opiates after operation. We have already seen (Chap. 3) the catastrophic complications of interference with flow through the anterior spinal artery by spinal nerve block for the sake of diagnosis and for relief of pain. We congratulate those clinicians who have studied such cases and have published their findings but are a little surprised that instead of condemning the method out of hand they suggest instead that the patients should be informed of the risk: ‘The risks of the event are so severe that they should be included in the informed consent procedure’ (Somayaji et al 2005): ‘In our medico-legal driven system these risks should be included in the informed consent protocol’ (Brouwers et al (2001) Is this enough? We think not. It
Iatrogenous Injuries
is our impression that some patients come to accept the unacceptable and that informed consent has contributed to the increasingly common failure to treat the consequences of the event by all reasonable means. The expert witness: In earlier years it was common practice for expert witnesses considering a case to refer ‘neglected’ patients directly to us or to ensure the referral by advising the family practitioner. There have been no such referrals during the last 4 years. This is a pity, as it may lead to a lost opportunity to improve the situation of the patient/litigant/client. It is true that the treatment of such patients is often unrewarding and disappointing. It has to be said that some claimants, regrettably but predictably, seek either from the start or after an initial rebuff from the defendant, to magnify they effects of their injuries on their work and daily life. To take but one example: a 4 year old child had sustained a birth lesion of the brachial plexus in which the neurological recovery had been very good but the common complication of posterior dislocation of the shoulder had occurred. We were aware, at the time of consultation, that there was a continuing process for compensation. The parents were advised that an operation could be done which would almost certainly considerably improve function in the child’s arm and diminish disability in later life, for the secondary deformity of bone in this case was not at all advanced. The response of the parents was not one of elation rather of deep disappointment and the suggestion was made that the operation be deferred until the court case had been settled. The following example illustrates an approach which may be helpful. Case report: A 40 year old healthy female hairdresser experienced a great deal of pain after excision of a ganglion at the ankle joint. Fourteen months later, by which time a claim of medical negligence had already been made an operation was performed in another hospital in which the neuroma of the cutaneous component of the superficial division of the common peroneal nerve was excised and relocated. The operation was a reasonable one and it was recommended only after failure of a trial of other methods of pain relief. Her pain became very much worse and there was intense mechanical allodynia over the dorsum of the foot. She was seen at the request of the operating surgeon and was advised that it would be best for her to push for speedy settlement of the claim after which further action might be feasible. The patient followed this advice and once her claim had been settled the nerve was reexplored and the stumps bridged by an interposed vein graft, by now 37 months after the initial injury. There was rapid improvement in her symptoms. By 3 months after this last operation she stated that she no longer experienced pain in her foot. Expert witnesses face new threats. Singh (2008) reflects on the fate of an eminent paediatrician who had relied upon statistical information in a flawed way: ‘There was no finding that he intended to mislead the court or that he had acted in anything
491
but good faith. He had simply made a mistake. Notwithstanding that he was erased from the Register. In the successful appeal to the High Court, Collins J advanced the case that the immunity which witness has from civil suit should apply to disciplinary proceedings in respect of evidence given in a court of law, save that the immunity would not be absolute, in that it was open to a judge for whom the expert gave evidence to refer the expert to the relevant disciplinary body. This was overturned in the Court of Appeal and the situation remains that an expert witness can be referred to his/her registration body by any person, even the disappointed party in litigation. In public policy terms, this has had the greatest impact in relation to paediatricians; a Royal College of Paediatrics and Child Health survey demonstrated that paediatricians in the field are often the targets of unfounded complaints and that the number of such complaints was rising. Although over 97% of complaints were subsequently unproven, the survey identified that complaints had a profound impact of the professional and private lives of some paediatricians and had influenced their willingness to take future child protection work’. It is all too easy for clinicians to criticise reports prepared by expert witnesses who are currently engaged in this field of work, but we have seen examples where the intervention of an expert witness has tipped the balance in favour of the patients’ well being although that may mean a considerable diminution in the damages paid to the same person as a litigant. It is a pity that straightforward cases such as section of the spinal accessory nerve are not referred for repair of that nerve, an operation which is remarkably successful even after the lapse of many months. All involved in such cases might consider the advice given by Ducic, Maloney and Dellon (2005): ‘unless a standard of care can be established for surgeons operating in the cervical triangle that includes the use of loupe magnification, intraoperative electrical stimulation even with a disposable stimulator, and the use of a bipolar coagulator, then it is clear that future reconstructive operations for spinal accessory nerve injury will be required’. Failure to recognise injury: Many factors make it difficult for the clinician to recognise nerve injury at an early stage, in the accident department, when and where it should be recognised. Some of these have been described in Chap. 5. Accident officers and triage nurses must be made aware that when there is a wound across the line of a main nerve, the digits of the affected limb must be felt. If the digits in the area of supply of the nerve are warm and dry, the nerve has been cut and it must be so assumed unless and until exploration proves otherwise. It is the failure to diagnose and act appropriately that may cause the patient to lose the best chance of recovery. Regrettably, it is this failure that seems to be the most common; it is on this failure that the ‘risk managers’ must chiefly work. Case report: A 43 year old man with non union of the humerus agreed to a third operation to rectify the problem. The operating surgeon advised the patient about the possible
492
risk to the radial nerve and said: ‘I do not anticipate any injury to the nerve’. The humerus was exposed through a midline posterior approach and whilst the incision was being deepened an alert house surgeon called out: ‘Mr. Birch you have just cut the radial nerve’. So it was. The fracture was dealt with, the nerve was sutured, the patient accepted the explanation given him shortly after the operation, an adequate dynamic splint was fitted for the radial palsy and his progress was carefully followed. Recovery was good. Some lessons might be drawn from this salutary event. First, above all, the alertness of the house surgeon who saw the twitch of the fingers and thumb which were not concealed by towels. Next, it is fortunate that neuromuscular blocking agents had not been used so that the twitch of the muscle was not masked. Third is the efficacy of urgent repair and last, but by no means least, is the temperament and character of the patient and the trust between the patient and the surgeon. Case report: A woman of 56 living in a large town fell and injured her right shoulder. Because the accident department of the hospital in her town had been closed, she was taken by ambulance to that of the hospital in a large adjoining town. There she was seen, told that she was suffering from a ‘frozen shoulder,’ given two aspirin tablets and sent home. The same evening she was aware of progressive paralysis of and loss of sensibility in the limb. The next day she went 4 miles by taxi cab to the accident department in which she had originally been seen. The advice given on the first day was repeated; two more aspirins were given. Recourse was had the next day to the general practitioner, he indicated that he could not override the advice given by the hospital medical officer. One month later, on her own initiative, the patient saw an orthopaedic surgeon, who instantly recognized that there was anterior dislocation of the shoulder with a deep lesion of the posterior and medial cords of the brachial plexus. This case encapsulates for us the principal causes of clinical error and negligence: failure of knowledge, failure to observe and failure to use common sense. The medical officer should certainly have been able to recognise anterior dislocation of the shoulder in a woman of average build. He or she should, even without the benefit of knowledge, have realised that an elderly woman would not be at the trouble and expense of travelling eight miles by taxicab for advice unless something serious was amiss. Sadly, although knowledge can be imparted and the need to observe can be stressed, it may be hard to impart commonsense to those who have undergone a ‘medical education’. Records. Leigh (2006) issues essential advice to doctors: ‘Their skills are for their patients and their notes are for themselves’. The distress and damage caused to patient and clinician by legal process are so severe that all possible steps should be taken to avoid causing damage which might give rise to litigation and to mitigate the rigours of litigation when it becomes inevitable. However intelligent, alert, informed,
Surgical Disorders of the Peripheral Nerves
compassionate and communicative a clinician may be, none can hope wholly to avoid complaint; few can hope wholly to avoid legal process. The horrors of the last are very greatly mitigated if the clinician can rely on a well written, well ordered and comprehensive clinical record supplemented by properly completed charts and a well ordered nursing record. Sadly, health authorities and Trusts appear to accord a low priority to records systems; the record-keeping systems in the ‘private sector’ are often worse. The defence organisations continually exhort their members to keep good records, but the exhortations seem too commonly to be disregarded. The keeping of a good record is a matter for self-discipline and care on the part of the clinician, supported by the availability of good systems for recording and retrieval. The surgeon should not think that because he or she carries that title it is necessary in every case to propose operation; more often, the task is, rather, to dissuade a patient from such a course. It has often seemed to us, examining the papers in cases of alleged medical negligence, that some claimants at least seem very eager to submit themselves to medical examination and treatment. The records of such patients before the critical event are voluminous: in one case they ran to 600 pages. A clinician seeing such a record and a correspondingly large general practitioner record could well conclude that it would be unwise to recommend operation of election to such a patient unless there were the strongest possible indications. One might speculate that some patients secretly relish the risk inherent in all medical and surgical procedures, and continue to tempt fate until, inevitably, a mistake is made and a penalty is claimed. Certainly, all clinicians should make themselves as familiar as possible with past medical history before recommending operation to any patient.
11.2.6 Teaching and Training ‘Some undergraduate anatomy courses now contain less material than Pinnock’s Ninepenny Catechism’ (Bates 2006). In his description of the fall of the public anatomy museums and the rise of anatomy in the medical schools in early Victorian London Bates suggests that the medical monopoly of anatomy has served the subject poorly. This perception is shared by those clinicians who continue to engage in clinical teaching for undergraduate and post graduate students throughout the United Kingdom which reveals an alarming decline in the understanding of the whereabouts and the functions of vital structures. Hinduja and colleagues (2005) suggested that the changes in education and the trend to ‘problem based learning’ had made a casualty of anatomy. Pryde and Black (2006) record the decline of the teaching of anatomy in some of the great Scottish medical schools where so much of the scientific foundation of medicine was established. The
Iatrogenous Injuries
proposed withdrawal of clinical pharmacology from the undergraduate curriculum (Black 2009) can only make things worse. The extent of the problem and some potential solutions are outlined by Standring (2009). We can only say as we find but it appears that there is a growing divide in the level of knowledge of the basic sciences between graduates from different medical schools. Those from Cambridge and some of the London schools, and in particular, those from Guys, Kings and Thomas’ who have come under the tutelage of Susan Standring appear to be rather better equipped for the practice of medicine.
11.2.7 Specialisation The expansion of knowledge and the growth of specialisation, whose effects have been deliberately intensified by the operations of lawyers and the law, have created new dangers for patients. Some clinicians may have forgotten that disease does not come neatly labelled as, for instance, ‘medical’, ‘surgical’, ‘orthopaedic’ or ‘neurological’ but as a patient with symptoms and signs which have to be construed. The clinician should not permit fear of legal process to deter him or her from, at least, thinking outside his or her speciality. Surgeons regularly operating in the pelvis and about the hip are quite capable of exposing and controlling the great vessels and of exposing and, if need be, repairing main nerves. Newer interest groups should follow this example. It is absurd that a ‘shoulder’, ‘elbow’, ‘knee’ or ‘foot and ankle’ surgeon should lack ability in exposure, control and if need be repair of vital structures in those regions. Worse still is when such special interest groups use the mask of clinical governance to delay or otherwise interfere with urgent treatment. Too often still, surgeons treating limbs with combined bony and arterial damage seem content to treat the fracture and to call in the ‘vascular’ surgeon when signs of ischaemia have persisted for 12 h or more. By that time, intervention on the artery comes too late. This situation may get worse with the implementation of well intentioned recommendations from the Vascular Surgical Society of Great Britain and Ireland summarised by Chant (2005): ‘in other words, if a general surgeon in the UK now performed an operation for a vascular emergency, and were that patient to die to perhaps lose a leg unnecessarily, then that surgeon may well be open to litigation’. The only excuse for failing to act in the presence of impending or actual critical ischaemia is the presence of a life-threatening condition forbidding early operation. To wait for a surgeon competent to deal with arterial lesion is as unjustifiable as is a wait for a surgeon competent to deal with the associated lesion of bone or joint. In a combined lesion of bone and vessel, the fragments must be fixed in order to protect the arterial repair, and the artery – or artery and
493
vein – must be repaired in order to protect the extremity. A surgeon accidentally damaging an artery in the course of an operation would do well to remember the advice from Norgren (2005) ‘the value of simply compressing a bleeding point rather than resorting to hurried and blind clamping cannot be over emphasised’.
11.2.8 Continuity of Care: Timing of Operation Case Report: A 33 year old man, who had served in the paratroop brigade had operation for reconstruction of ligament injury at the knee. The operation commenced at 3 p.m. and finished at 7 p.m. No member of the operating team was on call that night. He became conscious of severe pain 2 h after returning to the ward and by 4 h, the pain was intolerable and resistant to opiates. The duty doctor who was not a member of the orthopaedic department took no action. The responsible consultant saw him early on the following (Saturday) morning and immediately took him back to theatre for decompression by open fasciotomy of all four compartments. Most of the anterior compartment was infarcted. Mauffrey (2006) warned of such occurrences: ‘The pressure imposed by government to reduce waiting times in Accident and Emergency Departments, and the reduction in training hours for junior doctors due to the EWTD (European Working Time Directive) will certainly increase negligence and malpractice’. In their important annotation about medical negligence in orthopaedic surgery, Gidwani, Zaidi and Bircher (2009) write: ‘Surgeons should not be rushed into operating on fractures at inappropriate times. When the early window of opportunity is missed it is much better to wait until the condition of the soft tissues has improved. Managers may find that keeping patients in hospital adds considerably to the cost of the treatment. However, operating on high energy fractures at the wrong time and without a sufficiently experienced surgeon is extremely dangerous. Manages should be aware of this, and clinicians must not allow themselves to be pressurised into making inappropriate decisions for financial reasons’. In ‘day case’ operating in particular, the patient should not be sent home without being seen by a clinician of adequate experience; without a record of the procedure and findings to take to his or her own doctor. General practitioners should discard the belief, real or assumed, that all hospital practitioners are all-wise and all-knowing. When evidence mounts to show that a diagnosis made in hospital is incorrect, the practitioner should be prepared to question it; to ask the clinician who made it to think again or to ask someone else. All clinicians should bear in mind the fact that patients who complain generally have something to complain about.
494
Surgical Disorders of the Peripheral Nerves
11.3 Nerve Lesions in Total Hip Arthroplasty
11.3.1 The Nerve Lesion
Total arthroplasty of the hip is a difficult operation, one in which there is very little room for error and the difficulties are more severe in cases where the joint is greatly distorted or when revision proves necessary. The complication of a painful and deep lesion of a main nerve is extremely disheartening for the patients and surgeon alike and it is important to emphasise that much can be achieved by urgent reexploration and relief of the cause. Since 1979 we have seen more than 400 nerve lesions in about 250 patients. The nerves were reexplored in 69 patients between 1979 and 2000, in 47 more between 2001 and 2007 and details of the findings in 27 urgent explorations conducted by the first surgeon have been provided. We use the transgluteal approach (Chap. 7) to explore the sciatic nerve. Neurophysiological investigations (NPI) were performed before and after operation together with studies of conduction across the lesion during operation. As a result of these secondary operations one patient developed a deep vein thrombosis requiring treatment, and another developed a superficial infection which responded to antibiotics. The incidence of nerve lesion seems to be about 1% in primary operations (Schmalzried et al 1991, Schmalzried et al 1997, Edwards et al 1987, Eftekhar 1993) although Weber, Daube and Coventry (1976) demonstrated by neurophysiological investigations before and after operation, abnormalities of conduction in about two thirds of their patients. Schmalzried and colleagues (1991) recorded nerve palsies in 3% of patients after revision arthroplasty and in 5.2% in those with development of dysplasia of the hip (DDH). In the first edition of this work we distinguished between direct causes, from knife, scissors, cement, suture or wire or crush by retractor from indirect causes such as bleeding within or around the nerve, traction, or fibrosis and entrapment (Table 11.1) This distinction is rather artificial and it does little to identify risk factors. One hundred and ten patients with 209 lesions who were seen during the years 2001–2007 have been studied in a more detailed manner and we are grateful to Mr. Alistair Davidson who conducted a scrupulous review of the extensive literature and who also contributed to the design of the form used.
Setting aside cases of complete transection or the transient conduction block from concussion or compression the lesion is invariably mixed so that some fibres are intact, others recover as conduction block or favourable degenerative lesions whilst yet others never recover. It is because of this that NPI, which are helpful in confirming the level and the extent of lesion, cannot reliably indicate prognosis. The behaviour of the Tinel sign is less reliable than it is in nerves injured by fracture or dislocation (Chaps. 5 and 8). Two features are important. First: severe pain indicates that the cause of the lesion remains active and in these cases urgent reexploration is indicated. Second: delayed onset of pain and lesion strongly suggests bleeding. Bleeding is the most likely explanation when the lesion of the nerve deepens whilst under observation and if decompression of the nerve is done within 3 h, patient and surgeon alike can confidently expect abolition of the pain and speedy recovery. After that interval it is likely that pain will be improved but recovery will probably be incomplete.
Table 11.1 The nerve lesion revealed at operation in 69 patients operated between 1979–2000. Direct
Cement burn
2
Laceration by knife or scissors
5
Entrapment by suture or wire
6
Crush by retractor
Indirect
Ischaemia ( rupture femoral vessels)
Extraneural bleeding (haematoma)
Intraneural bleed
3
Traction/tethering
31
10 1 11
11.3.2 Findings in the 110 Patients Seen Between 2001 and 2007 There were 89 women and 21 men. Twenty six patients were aged 80 years or more, 8 were 49 years or less, 78 were aged between 50 and 79 years. The distribution between sides was equal. The causes of the hip disorder and potential risk factors affecting the patient are outlined in Table 11.2. Ninety one patients were receiving active treatment for their other illnesses. The number of patients with endocrine disorders was 28 (25%) and there were 17 cases of thyroid disease. The hip was approached by the posterior route in 65 patients, by the lateral route in 39 and by the anterior approach in 4. Two patients incurred lesions during operations by limited exposure. Most prostheses were cemented; two patients had resurfacing operations. Lengthening of the affected lower limb exceeded 1.5 cm in 28 patients. Nearly all patients had some form of thromboprophylaxis and it is possible that some cases of haematoma were related to the use of anticoagulants. Spinal or nerve blocks were used in 24 patients. These may have contributed to delay in recognition of bleeding or entrapment of the nerve in six cases. The first operation was complicated by deep vein thrombosis requiring anticoagulation in nine patients, the prosthesis dislocated in six more. The nerve lesion: The distribution, the extent and the depth of lesion are shown in Table 11.3. No example of common peroneal nerve palsy at the knee was seen. The level of the lesion can usually be ascertained by examining the lateral hamstring muscles.
Iatrogenous Injuries
495
Table 11.2 110 patients seen between 2001–2007. The hip disordera
Potential risk factorsb
Diagnosis
Number
Diagnosis
Number
Osteoarthritis
85
Hypertension
36
Developmental dysplasia
10
Cardiac disease
14
Fracture
7
Peripheral vascular disorder
Rheumatoid arthritis
2
Connective tissue disease
11
Ankylosing spondylitis
1
Thyroid disease
17
Paget’s disease
1
Other endocrine disease (including diabetes mellitus)
11
Avascular necrosis
1
Liver and renal disease
2
Slipped epiphysis
1
Poliomyelitis
1
Early septic arthritis
1
Rheumatoid arthritis and fracture 1 a No potential risk factor was detected in 38 patients b There were 20 revision arthroplasties
Table 11.3 The nerve lesion by depth and extent: 211 nerves in 110 patients. (2001–2007). Nerve affected
Number
Number complete and degenerative
Number incomplete: conduction block and degenerative
Table 11.4 The syndromes in 81 patients experiencing significant or severe pain (PNI score 2 or more) 2001–2007. Pain syndrome
Immediate onset
Causalgia
13
9
22
Neurostenalgia
11
5
16
30
67
7
7
4
2
2
Post traumatic neuralgia
37
Obturator Femoral
18
9
9
Reinnervation
Lumbosacral plexus
2
1
1
Superior gluteal
6
3
3
Inferior gluteal
10
4
6
Common peroneal division of the sciatic nerve
92
70
22
Tibial division of the sciatic nerve
79
33
46
TOTAL 211 122 89 The two high lesions of the sciatic nerve are analysed with the common peroneal, tibial and inferior gluteal nerves
Pain: Eighty one patients experienced severe pain and the syndromes are set out in Table 11.4. Most patients experienced more than one pain syndrome and causalgia or neurostenalgia were frequently replaced by post traumatic neuralgia. Pain which required active treatment was associated with regeneration of the nerves in seven patients. Causalgia was almost always associated with bleeding around or within the nerve, an observation made by, amongst others, Fleming, Mickelsen and Stinchfield (1979), Johanson and colleagues (1983) and Butt and colleagues (2005). Patients in whom nerves had been encircled or pierced by suture or wire presented with severe neurostenalgia, often closely related to the posture of the limb. This pain was almost always relieved by removal of the suture
5
Delayed onset
TOTALS
or wire as in the case described by Mallory (1983) but recovery of the nerve was seen only in cases where the agent was removed within a few hours. The two syndromes of causalgia and neurostenalgia are particularly important as they signify continuing action of a noxious agent upon the nerve. So it is with the pain caused by the heat from polymerising acrylic cement (Siliski and Scott 1985, Oleksak and Edge 1992, Birch et al 1992). The likelihood of pain relief and the chance of nerve recovery are directly related to the speed with which the offending cement is removed. It is important for clinicians never to lose sight of the fact that the lesion of a nerve will deepen if the agent responsible continues its action. What was a conduction block becomes an axonotmesis or worse. The speed of deterioration is determined by the cause. Crawford, van Rensburg and Marx (2003) secured immediate relief from late onset pain and neural defect in three patients by removing a mass of wear debris impinging on the sciatic nerve. The response of pain to different methods of treatment is illustrated in Table 11.5. Most patients were treated with several methods. It should be said that no patient should be sent to a Pain Clinic until a clear diagnosis for the cause of the pain has been made. The response of pain to reexploration was often gratifying and this is one reason for considering reexploration of the nerves even in late cases (Montgomery et al 2005). The abnormalities of the nerves: The significant abnormalities displayed at operation are set out in Table 11.6. The
496
Surgical Disorders of the Peripheral Nerves
Table 11.5 Response of pain to different methods of treatment: 81 patients. 2001-2007. Method
Good
Table 11.7 The outcome for 190 nerves adequately followed (2001–2007). Good
Fair
Poor
Fair
Poor
TOTALS
Obturator
Nerve
1
1
1
TOTAL 3
Opiates
4
40
15
59
Femoral
7
4
5
16
Non steroidal anti inflammatory drugs
1
39
29
69
Superior gluteal
1
1
3
5
Gabapentin/Pregabalin
6
19
21
46
Inferior gluteal
3
3
3
9
Amitriptylene
1
23
19
43
Common peroneal division
21
23
39
83
Regional local anaesthetic block
1
3
6
10
Tibial division
37
21
16
74
TOTALS
70
53
67
190
Regional sympathetic block/ block of lumbar sympathetic chain
0
5
7
12
Re exploration of the nerve(s)
15
15
4
34
Table 11.6 Findings at operation: 47 patients. (2001–2007). Nerve encircled by suture
4
Section or partial section of nerve
6
Entrapment within joint capsule
2
Nerve immobilised by fibrosis to adjacent structures
27
Nerve entrapped within fibrosis
21
Traction lesion
8
Rupture of epineurium in nerves otherwise in continuity
9
Rupture of perineurium in nerves with apparent anatomical continuity
2
Obliteration of epineurial vessels in nerves otherwise in continuity
21
most common finding was that of tethering and immobilisation by suture, by scar between the nerve and adjacent tissues or by entrapment of the nerve within the joint capsule (33 patients) followed by strangulation of the nerve by enveloping fibrosis (21 patients). A traction lesion (8 patients) was signified by tortuous elongation of the nerve. Rupture of the epineurium was most often seen in traction lesions, obliteration of the epineurial vessels was most commonly seen in nerves strangled by scar tissue. Compound nerve action potentials traversing the lesion could be recorded in nearly every nerve save those which had been transected. The amplitude of the CNAP was usually reduced and velocity was often diminished, but these attributes did not closely correlate with the final outcome (Table 11.7). A poor grade was allocated where there was persisting pain, a common problem in the femoral palsies. Outcome: At the final review 28 patients considered that they were better than before the operation of arthroplasty. Sixty four patients had been prescribed standard foot/ankle
orthosis, and most patients had discarded these and either went without or purchased or were later provided with more effective splints. Five patients developed pressure sores and one patient who developed a deep sore in the sole of the anaesthetic foot caused by a conventional AFO came to below knee amputation. Forty four patients continued to use crutches or sticks, 18 were house bound, and 3 used wheel chairs. The causes of lasting disability included persisting pain, loss of flexion and abduction of the hip, loss of knee extension, and fixed deformity in the leg, ankle and foot. Anterior transfer of the hamstring muscles was helpful to three patients but it failed in a fourth because of weak hip flexion. Anterior transfer of tibialis posterior was performed in 12 patients with foot drop; it was effective in 6 of them. Six operations were performed to correct disorders of the hind and forefoot. Ninety seven of the prostheses were deemed technically successful. The prognostic value of neurophysiological investigations performed within 2–12 weeks of the arthroplasty is clear. Less than one quarter of patients with proven complete and degenerative lesion of the common peroneal nerve regained useful control of the ankle and toes. About one third of patients with proven complete degenerative lesions of the tibial component of the sciatic nerve recovered function in the small muscles of the foot, good balance, and sympathetic function in the sole of the foot.
11.3.3 Indications for Urgent Reexploration Chief amongst these is pain. The overwhelming, bursting or burning pain of causalgia signifies bleeding until proven otherwise. We have now followed 12 patients in whom the first surgeon re operated within 12 h, decompressed the nerve, and so ensured rapid relief of pain and , usually, recovery for the nerve except in cases where the bleeding was intraneural (Chap. 3). A valuable paper comes from Butt and colleagues (2005). Haematoma caused severe pain and sciatic palsy in six patients after primary arthroplasty. Pain and onset of lesion of the sciatic nerve were evident between 24 and 48 h.
Iatrogenous Injuries
In three patients a haematoma measuring 800–1,000 ml was evacuated from deep to the fascia lata within 48 h of operation and in these there was an immediate resolution of symptoms. In three more patients diagnosis was delayed and no action was taken and in these pain persisted and recovery was slow. Butt and colleagues make this important observation that five of their patients weighed less than 70 kg and that they may have been over anticoagulated. Lightning like, convulsive pain expressed within the distribution of the nerve trunk suggests that the nerve has been entrapped by a suture, bone fragment, bolus of cement or other agent. The pain is related to the posture of the hip so that when the femoral nerve is involved pain is eased with the hip flexed. When the sciatic nerve is involved the hip is kept in extension. We have seen ten patients where the first surgeon, recognising the significance of this pattern of pain, reoperated within 24 h and removed the offending agent. Pain relief was complete in all but one of these. Recovery of the nerve was less certain, it was more likely when the cause was corrected within 3 to 4 h. Some relief of pain was usual in the later cases but in these there was rarely useful recovery for the affected nerve. Delayed onset of nerve palsy and deepening of that palsy whilst under observation indicates continuing bleeding and is a strong indication for reexploration. We suggest that complete and deep lesions of nerves should be reexplored, particularly if the femoral nerve is involved and especially so when the approach adopted did not permit display of the affected nerve. Limb lengthening and traction: Our findings of attenuation of longitudinal blood supply and partial epineurial rupture suggest that lengthening of 2 cm or more produces a degenerative lesion which may not recover. Fleming and his colleagues (2003) measured strain of the sciatic nerve in ten patients undergoing total hip arthroplasty. Radio opaque markers were placed on the epineurium at four sites and these workers recorded an increase in strain of the nerve, or at least of the epineurium of up to 30% with the hip flexed and the knee extended. None of the nerves in this study came to any harm. These are important observations. Is there a case for instant or early revision of the arthroplasty and replacement of the components so as to relieve the strain upon the nerve in cases where the lesion is associated with lengthening of the lower limb? Edwards, Tullos and Noble (1987) did just this in a 58 year old woman who awoke with severe pain after revision arthroplasty. Pain persisted for 5 days until her femoral component was revised again to a prosthesis 6 mm shorter. Her pain resolved and she was functioning well at the 2 year follow up examination. Similarly decisive action ensured a happy outcome in the case reported by Silbey and Callaghan (1991). Sakai and colleagues (2002) reported relief of pain and full recovery of the sciatic nerve following a revision operation which was performed within 24 h of the
497
first operation. The question of intraoperative monitoring is rightly raised in cases at risk by Nercessian, Gonzalez and Stinchfield (1988) and Kennedy and his colleagues (1989). Porter and her colleagues (1989) showed that simultaneous changes in amplitude and in latency of somato sensory evoked potentials appeared to be predictive of lesions of the sciatic nerve and Satcher and colleagues (2003) have explored the use of motor evoked potentials. It may be helpful to clinicians engaged in difficult revision work to consider the potential for these investigations.
11.4 Radiation Neuropathy The most common site for radiation neuropathy is the brachial plexus of the adult woman, because of the frequency of breast cancer, the frequency with which radiotherapy is used and the position of the plexus in the field. The situation of the clinician directing radiotherapy is one of peculiar difficulty. On one hand, the dose of radiation must be enough to destroy tumour cells; on the other hand, it should not be enough nor so directed as to cause serious damage to normal tissues and in particular to nerves. The handicap of the long latency between treatment and the appearance of symptoms adds to the difficulties. To the primarily neural lesion of vasculitis with damage to both axon and myelin sheath is added the external one of compression by sclerotic tissues such as the pectoralis minor and subclavius muscles (see Chap. 3). Merle and colleagues (1988) liken these to ‘syndromes canalaires créés par la nécrose du petit pectoral et du sous-clavier’ and we cannot improve on that. Wilbourn (2005) identifies four major factors: the total dose given; simultaneous chemotherapy; overlapping fields which produce ‘hot spots’; and using fewer and larger doses (hypofractionation). In a wide-ranging paper, Bowen and colleagues (1996) consider the cases of six men who received radiotherapy to the groin and para-aortic glands (T10–L4) for testicular malignancies and who later developed neuropathy. The mean dose was 45.5 Gy; the latencies ranged from 3 to 25 years; the affection was predominantly motor; in two cases it was initially monomelic. The course of the condition was regularly one of slow but relentless deterioration. Myokymia was noted in three patients; the sural nerve action potential was obtained in five. Magnetic resonance imaging with gadolinium showed linear and focal enhancement of the lumbosacral roots within the spinal canal in two of three cases. Pathological examination of the conus and cauda equina was possible in one case: this showed a ‘vasculopathy of the proximal spinal roots, with preservation of motor neuronal cell bodies and spinal cord architecture’. The authors conclude that the affection is a radiculopathy rather than a neuronopathy.
498
In our experience of more than 250 cases, the natural course of this disorder is almost always towards steady deterioration. The initial symptom is usually that of paraesthesiae in the fingers; it is at this stage that a diagnosis of ‘carpal tunnel syndrome’ or ‘cervical radiculopathy’ is commonly made. The patient may present with pain, less commonly, muscle weakness is the initial symptom. The condition progresses to severe alteration of sensibility and motor paralysis; sometimes pain develops or progresses to become constant and severe. We have seen five cases of post irradiation thrombosis of the axillary artery in patients who experienced sudden, intense pain, and a plexopathy which became complete over the course of 24 h. Laviano (1997) has kindly provided his findings in a detailed study of 54 patients with post irradiation neuropathy affecting the brachial plexus who received treatment in the Peripheral Nerve Injury Unit over a 10 year period. The onset of symptoms ranged from during or immediately after radiotherapy, to 24 years. The most common presenting symptom was sensory disturbance (60%), pain was the presenting symptom in 30% and loss of power in one quarter. Operations were performed in 32 patients, which included biopsy, external neurolysis and occasionally epineurotomy which was supplemented by a pedicled latissimus dorsi myocutaneous flap (three cases) or a pedicled flap of omentum (three cases). Laviano recorded recurrence of the primary tumour in seven patients and the development of a post irradiation malignant peripheral nerve sheath tumour in two more. He found that pain had improved in one-half of the 32 operated patients and in one third it was possible to demonstrate some improvement in power and sensation. Malignancy after radiotherapy We have encountered five probable post irradiation tumours involving the brachial plexus and these occurred in patients who described rather rapid worsening of pain and deepening of the nerve lesion within the distribution of the affected trunk or cord. A mass was palpable in two of these, CT and MR scan confirmed the presence of a mass in all cases. Donaghy (2005) sets out the distinguishing features between radiation and tumour lumbo sacral plexopathy. In post irradiation cases pain is less important, sensory motor disturbance is often bilateral and affects the distal part of the lower limbs. Myokymic discharges are seen in about one half of patients. The detection of a mass by CT or MR scans confirms that the lesion is caused by tumour.
11.4.1 The Place of Operation Mondrup and colleagues (1990), reporting from Denmark, stated bleakly ‘Because the treatment of brachial plexopathy is unsuccessful, prevention is warranted’. Opinion is certainly divided. For years we recommended and practised exploration, biopsy, external neurolysis and occasionally epineurotomy, for confirmation of diagnosis and in the hope of giving relief. In 1963 Kiricuta in Romania drew attention
Surgical Disorders of the Peripheral Nerves
to the use of a pedicled omental flap for treatment of radionecrosis of the chest wall and of chronic oedema of the upper limb, and for reconstruction of the breast. Uhlschmid and Clodius (1978) followed the suggestion by free transplantation of the omentum with microvascular anastomosis, combined with external neurolysis, for radiation neuropathy of the plexus. They concluded, modestly enough, after reviewing the seven patients treated: ‘Als wesentlicher Erfolg muss jedoch die sofortige und anhaltende Schmerzfreiheit aller Patienten angeshen werden’. That was certainly something, even though the effect on neuropathy was probably, at the best to halt it. Later, Narakas (1984) reported the results of operative treatment for radiation-induced and metastatic brachial plexopathy in 45 cases, 15 having an omentoplasty. Narakas thought that omentoplasty was most likely to succeed if used early; he achieved a ‘functional result’ in only 20% of cases. Our experience with four cases of omentoplasty was discouraging. Progress after operation was uncomplicated, but pain persisted and there was no regression of the neurological affection. In contrast, good results were achieved in three cases in which a pedicled myocutaneous graft of the latissimus dorsi muscle was used. These operations were done in collaboration with Professor Roy Sanders. Case report: A man born in 1972 was treated at the ages of 3, 6 and 8 for ‘lesions’ of the left pectoral region later thought to be those of malignant fibrous histiocytoma. In 1992 a recurrent tumour was widely excised and the site was irradiated (60 Gy in 18 fractions over 6 weeks). Symptoms of radiation neuropathy began soon after treatment and progressed until at 6 months there was serious pain with deep affection of motor and sensory function. There was serious damage to the skin over the clavicle. Electrophysiological examination showed severe affection of the medial and posterior cords: in particular, there were frequent fibrillations and positive sharp waves in the left first dorsal interosseous muscle. At operation in 1994, 2 years after radiotherapy, an extensive neurolysis of the plexus was done, with division of the scaleni, pectoralis minor and clavicular head of pectoralis major. Stimulation of roots and recording from cords showed potentials of diminished amplitude and velocity (medial cord 25 m/s: posterior cord 10 m/s). A pedicled musculocutaneous flap of latissimus dorsi and overlying skin was inset to replace the burnt skin and to surround the plexus. There was early relief of pain; 2 years after operation there was good recovery of motor and sensory function. Results of operation: Experience with 74 traced patients who underwent operation for radiation neuropathy suggests that the likelihood of true, and lasting, relief of pain and regression of the neurological abnormality is low (seven patients). Thirty one patients obtained relief of pain and stabilisation of the neurological state; 25 obtained no relief. Recurrence of carcinoma was confirmed in nine cases; remarkably, there was substantial relief of pain in three of these. Operation on the plexus in these cases is not free of danger: in one, necrosis of the artery led to amputation; in another there was thrombosis of the subclavian artery; in one there was delayed healing with
Iatrogenous Injuries
necrosis of part of the clavicle; one patient died from cerebrovascular accident 3 months after operation. In one undoubted case of radiation neuropathy there was spontaneous remission over a period of 10 years. There are, probably, three indications for operation: (1) establishment of diagnosis and in particular the diagnosis of a radiation-induced malignant tumour of the peripheral nerve sheath; (2) Severe and persistent pain, and (3) rapid progressive loss of function. The place of operation in the first seems to us to be undeniable. As to the second: there is, it seems to us, a worthwhile chance of relieving pain in one half of cases. As to loss of function: there is, we think, a worthwhile chance of, at least, arresting deterioration in most cases. Anything more is a bonus.
11.5 Prevention of Iatrogenous Lesions That great historian, Nicholas Rodger (2004) described the rise in the status of naval surgeons thus: ‘Surgeons received a proper uniform in 1805, with the all important sword which marked them as gentlemen’. Doctors risk losing the respect and trust gained over many generations unless the causes which underlie the rise in iatrogenous injuries are defined and corrected. We tender some suggestions which fall into three groups: teaching and training; events in the Outpatient Department and in Theatre; early recognition of an impending, or actual event and the course of action to be taken.
11.5.1 Teaching and Training One simple measure is that all undergraduate and post graduate students possess themselves of a copy of Aids to the Examination of the Peripheral Nervous System. Many of their teachers might benefit from studying this work. The days are long gone when physicians could do no more than observe illness by examination of the tongue, the pulse, the faeces and the urine and offer no more than placebos. Such is the growth in interventional and endoscopic work in all branches of medicine that the requirement for anatomy teaching increases rather than diminishes. If medical schools relinquish the teaching of basic sciences underpinning medicine then other institutions must take over. One such example is Professor Susan Standring’s core surgical anatomy course at the Royal College of Surgeons of England with the support of the London Deanery. Such courses do not reach all those who need them and they are usually attended by those already possessed of insight and interest. The transfer of responsibility for teaching must of course be accompanied by the transfer of large sums of money from medical schools to those institutions carrying on the work. This is unlikely to happen. Surgeons training in endoscopic or ‘limited access’ work must never lose sight of the fact that the pathological process which they are treating will displace nerves and vessels.
499
Previous injury or operation in the region will add to this and also tether vital structures in fibrosis. The nerve may have been displaced from its normal position during a previous operation, indeed we have repaired three ulnar nerves which were transected during the course of arthroscopic work which had been transposed anteriorly during an earlier operation for fracture. A simple nerve stimulator is an essential aid in the display of nerves but this does require a modification of anaesthesia so that neuromuscular conduction is maintained. Fracture surgeons recognising a nerve palsy as a complication of skeletal injury would do well to expose the nerve and adjacent vessels during the course of operation of internal fixation. Exposure should be adequate and extensile. It is beyond comprehension that patients with complete degenerative lesions of the common peroneal nerve should be subjected to several operations for ligament reconstruction without any attempt being made to see what had happened to the nerve. To operate upon a fracture of elbow, shoulder or knee in a patients in whom flow through the axial artery has been blocked, without attempting to expose that artery and restore flow, defies reason.
11.5.2 Audit and Consent if you be constrained to use your Saw, let first your patient be well informed of the eminent danger of death by the use thereof; proscribe him no certainety of life, and let the worke bee done with his owne free will, and request: and not otherwise. Let him prepare his soule as a ready sacrifice to the Lord by earnest praiers, craving mercie and helpe unfainedly … It is no small presumption to Dismember the Image of God. (Woodall 1617)
Informed consent should extend to outlining what steps will be taken to prevent significant complications and what can be done about them should they occur. Such informed consent rest upon accurate information. There must be adequate audit at local and national levels and also by speciality of the incidence of nerve lesions complicating operations. The British Orthopaedic Association, with its affiliated societies, might introduce a suitably anonymised system, perhaps based on the ‘yellow card’ used in earlier years to notify adverse reactions to drugs or the notification card system used by the British Paediatric Surveillance Unit. Surgeons can and should do much more in preventing post operative pain by blocking nerves under vision or by infusing local anaesthetic into a joint or into the field of operation avoiding more hazardous nerve or regional blocks.
11.5.3 Conduct of Affairs The clinician must not be forced to hurry through a consultation in the Outpatient Clinic. If all goes well there the rest tends to follow smoothly. The system in use in our clinics at St Mary’s and the Royal National Orthopaedic Hospitals is
500
Surgical Disorders of the Peripheral Nerves
Table 11.8 The sequence of events from first attendance of a patient to discharge from hospital. The history: Personal and family
Questions about cardiac and cerebro vascular, endocrine and neoplastic disease.
Investigations: Carried out in the Outpatient Department at first attendance
Height, weight. blood pressure, urine test.
If decision made to admit
In young, healthy patients for short operations Hb and ECG. For older patients or long operations: ECG, Hb and full blood count, ESR, thyroid function tests, urea and electrolytes, liver function tests. For any type of operation, ascertain use of contraceptive pill in appropriately aged women. For longer operations or where considered necessary blood is grouped and saved.
In the clinic when investigations performed
The ECG trace is checked and signed as satisfactory by the requesting surgeon. If there are any abnormalities then the general practitioner is informed and asked to treat BEFORE admission for operation.
All other investigations
The results are returned to the Department where the appropriate notes are retrieved and together with the attached results are given to the appropriate surgeon to check and sign BEFORE any admission takes place. The family practitioner is advised of any abnormalities with a request to treat them and to advise the Department when all is satisfactory so that admission can be confirmed.
Consent
Preferably this is taken in the clinic at the time of the decision by the surgeon suggesting admission. At the same time consent for photographs and for permission to use tissue samples for research is taken. It may be necessary to repeat these if admission is delayed.
Other measures
First visit to physiotherapist or occupational therapist. If necessary splints are provided before operation for use after. Close liaison with the anaesthetist BEFORE admission where appropriate.
The first letter
To the referring person and ALWAYS with copy to general practitioner, and transcribed within two days of the consultation.
The operation record
Transcribed the day of operation or the day after. A clear plan is set out for post operative care, proposed readmission for rehabilitation, removal of sutures etc. next visit to the Outpatient Clinic. Specific instructions are conveyed to the Ward Staff on the day of operation. Copies sent to the referring person and to the general practitioner.
The discharge letter
Prepared by the senior house officer and it is given to the patient on discharge to take or send to her general practitioner: recording drugs necessary, timing of next appointment, and any specific requests for care.
outlined in Table 11.8. Managers must always remember that their chief function is to ensure that the patient meets with the doctor as and when necessary in the best available circumstances with adequate diagnostic support. Clinicians must insist on seeing their patients before and after operation and must institute a robust system of handing over between shifts. Of course this must involve the nursing staff on the ward and the growing tendency to conduct a ward round in the absence of the nurse in charge is not only discourteous but is asking for trouble for that patient and for that clinician.
11.5.4 Recognition and Action Increasing pain and increasing loss of sensation indicates critical ischaemia until proven otherwise. The clinician faced with a significant injury to a peripheral nerve or a vascular injury should expect immediate and wholehearted support from colleagues and from management in rectifying the situation and if need be, securing urgent referral. There is a place for Special Units receiving such cases. Their responsibilities are high. They must provide urgent advice, and ensure that the patient is either transferred immediately or seen in the next clinic, or that one of their surgeons is able to travel to the hospital where the
event took place. It is for the hospital making the request for that visit to simplify the process of ‘vetting’ by urgent conversation with that visiting clinician’s employers. There should be no waiting list for important investigations, or for treatment by therapists designed to mitigate or to palliate the consequences of the injury, and this must apply to the community services as well as to those in the hospital. Conclusion: The medical profession must accept that there is a problem, and a growing one, and it will not do to rely on the action of government, or other agencies of the state. Walker and Mead (2008) describe the role of the National Health Service Litigation Authority and set out the Authority’s aims and objectives listed in the framework document: ‘the Secretary of State’s overall aims for the Authority in administering the schemes are to promote the highest possible standard of patient care and to minimise the suffering resulting from any adverse incidents which nevertheless occur’. We have seen very little evidence of the Authority engaging in this process for the several hundreds of patients treated in our Unit over the last 5 years. It is with a sense of relief that we turn now to Part 2 of this chapter which is an extensive and authoritative review of the causes and prevention of injury to the trigeminal nerve during dental work, prepared by Professor Tara Renton.
Iatrogenous Injuries Part 2: Minimising and Managing Iatrogenous Trigeminal Nerve Injuries in Relation to Dental Procedures Tara Renton
11.6 Introduction Trigeminal nerve injury is the most problematic consequence of dental surgical procedures with major medico-legal implications (Caissie et al 2005). The incidence of lingual nerve injury has remained static in the United Kingdom over the last 30 years, however the incidence of inferior alveolar nerve injury has increased; the latter being due to implant surgery and endodontic therapy (Renton et al 2010 in press). Iatrogenous injuries to the third division of the trigeminal nerve remain a common and complex clinical problem. Altered sensation and pain in the orofacial region may interfere with speaking, eating, kissing, shaving, applying make up, tooth brushing and drinking; in fact just about every social interaction we take for granted (Ziccardi & Zuniga 2007). Usually after oral rehabilitation, the patient expects and experiences significant improvements, not only regarding jaw function, but also in relation to dental, facial, and even overall body image (Kiyak et al 1990). Thus these injuries have a significant negative effect on the patient’s selfimage and quality of life and the iatrogenesis of these injuries lead to significant psychological effects (Abarca et al 2006). One cannot discuss trigeminal nerve injuries without being aware that as the largest peripheral sensory nerve in the human body it is represented by over 40% of the sensory cortex. The trigeminal nerve or ‘three twins’ supplies the face, eyes, mouth and scalp with general sensation in three divisions (ophthalmic, maxillary and mandibular) and supplies the masticatory muscle. One must also consider that the most commonly injured trigeminal nerve branches, the inferior alveolar (IAN) and lingual (LIN) nerves are different entities in that the lingual nerve sits loosely in soft tissue compared with the IAN that resides in a bony canal. There are specific features of trigeminal nerve injuries associated with dental procedure: • Both lingual and inferior alveolar nerve injuries are closed injuries, unlike those open sensory nerve injuries seen mainly on limbs due to trauma, which avails them to immediate exploration, with no delay, and repair by ortho-
•
•
•
•
pedic or plastic surgeons. Paradoxically our profession has a ‘sit and wait’ policy for resolution of trigeminal nerve injuries unless known section has taken place. 88% of lingual nerve injuries associated with conventional lingual access third molar surgery resolve (Mason 1988; Blackburn 1990) thus lulling our specialty into a false sense of security believing that all nerve injuries get better. This misconception has also led to the assumption that most inferior alveolar nerve injuries resolve; in fact they are predominantly permanent (Hillerup J 2007). The complexity of nerve injury defies the classifications of Seddon and Sunderland. It would be difficult to injure a nerve with a drill without causing a multitude of events including: (a) direct mechanical trauma (tear, section, crush, stretch etc.), (b) neural chemical trauma due to intracellular components released during trauma; haemoglobin irritates neural tissue and (c) ischaemic injury due to entrapment within a bony canal (IAN) by continued bleeding or scar formation. Thus it is unlikely that damage to a nerve is due to a simple ‘cut’. It is more likely that these nerve injuries incorporate a combination of mechanical injuries (sectioning, stretching, crushing), chemical nerve injuries and ischaemic injuries providing a complex therapeutic challenge. The type of patient often provides additional difficulty in that they have a complication arising from elective treatment that was supposed to improve their quality of life, not detract from it. These iatrogenous injuries cause understandable distress to both patient and surgeon and the patient’s frustration is often compounded by poor management by the surgeon involved (avoidance of contact after the injury has occurred, poor consent procedures, continued reassurance that the injury will resolve over months and years rather than referring the patient to a specialist early on). Additional distress is caused in that sensory nerve injuries frequently cause pain rather than numbness. As the neuropathic area invariably involves the mouth and face the patients’ ability to eat, speak, drink, sleep, kiss, shave or apply makeup is often severely functionally compromised. Due to the chemical and neurophysical changes in 501
502
the injured sensory nerve, light touch or drafts of air can cause debilitating neuralgic pain (allodynia) or in some instances the patient might experience constant background pain. • Complaints to the General Dental Council (GDC) are predominantly related to implants and often involve IAN injury. Neuropathic pain can be very debilitating and when compounded by poor management may result in subsequent litigation. Litigation is often based on inadequate consent procedure, inadequate planning and assessment, causation of avoidable nerve injury and poor management of the patient once the nerve injury has occurred. • Current management of these nerve injuries is inadequate. There is only discussion on surgical correction and no attention to medical or counselling intervention. In part the fault rests with how we assess these patients and there is a deficiency in functional and pain evaluation and a total focus on basic mechanosensory evaluation which is not necessarily reflective of the patients’ difficulties. A recent review of publications pertaining to trigeminal nerve repair highlights that the average time from injury to nerve exploration was 16 months; this is far too late to prevent central neural changes due to altered peripheral input (neuropathic pain) (Ziccardi and Zuniga 2007). Causes of inferior alveolar nerve injury include local anaesthetic injections, third molar surgery, implants, endodontics, ablative surgery, trauma and orthognathic surgery. The inferior alveolar nerve (IAN) lesions related to third molar surgery or inferior alveolar block injections is usually temporary but can persist and become permanent (at 3 months). There are rare reports of resolution of implant related IAN lesions at over 4 years (Elian et al 2005) but these do not comply with normal reports of peripheral sensory nerve injuries. Many authors recommend referral of injuries after 6 months (Hegedus and Diecidue 2006) but this may be too late for many peripheral sensory nerve injuries to effect a recovery. We now understand that after 3 months, permanent central and peripheral changes occur within the nervous system subsequent to injury, that are unlikely to respond to surgical intervention (Ziccardi 2001). Causes of lingual nerve injury include dental local anaesthetic injections, intubation, ablative surgery and submandibular gland surgery. The most common cause of LNI injuries is third molar surgery, with a reported incidence of 1–20% temporary and 0–2% permanent (Mason 1988). Persistence of any peripheral sensory nerve injury depends on the severity of the injury, increased age of the patient, the time elapsed since the injury and the proximity of the injury to the cell body (the more proximal lesions having a worse prognosis).
Surgical Disorders of the Peripheral Nerves
In a recent study (Renton and Yilmaz 2010) of a total of 221 patients with trigeminal nerve injuries collected over 3 years and seen at the Dental Institute in King’s College Hospital, London, 38 patients presenting with trigeminal neuropathy caused by neurological disease, malignancy, multiple sclerosis, sickle cell disease, known alcoholism, injury caused by non dental trauma, orthognanthic surgery, diabetes, HIV, post herpetic neuralgia, stroke and patients on chemotherapy. The aetiology and functional status of 183 post traumatic injuries to lingual or inferior alveolar nerves were evaluated. Injury to the lingual nerve was the most prevalent type of lesion (n = 93; 52%), followed by the inferior alveolar nerve (n = 90; 47%), and the buccal nerve (n = 3; 1%). Patients were referred to us from all parts of the UK. The majority of nerve injuries were referred from specialist practitioners in secondary care trust (LNI = 50% and IANI = 32%) whereas general dental practitioners referred 40% of LNI and 51% of IANI patients. Time from injury to examination followed a skewed distribution with an arithmetic mean of 14.5 months (SD 28.0), and a median value of 8 months, range 0–430 months. Most patients were seen within a year after the injury. Injuries were regarded as being permanent if the patient had their symptoms for more than 3 months. Many of the LNI and IANI patients had permanent injuries (63.4% and 54.8%, respectively) and females were more likely to suffer from permanent nerve injury (p > 0.001). Only 12.9% and 5.4% of the LNI and IANI cases were temporary. LNI patients presented with a mean age 38.4 years (range 20–64) and IANI patients presented with a mean age 43.8 years (range 22–85). Significantly more females suffered from injured nerves (63% of LNI patients, p = 0.01 and 61% of IANI patients, p = 0.003), but there was no significant difference in the severity of affection between females and males. Although IANI patients suffered from a larger mean neuropathic area, their mean subjective function was slightly better than the LNI patients. The range of subjective function values indicated that more IANI patients suffered from hypersensitivity and possibly hyperalgesia/allodynia. The size of the extra-oral and intra-oral neuropathic area significantly correlated with the gender of only the IANI patients (p < 0.001 and p = 0.01, respectively). Third molar surgery (TMS) and local anaesthesia caused the majority of IANIs and LNIs. A more diverse range of procedures, including implant placement and endodontic treatment caused IANIs. TMS carried out under general anaesthesia resulted in significantly larger intra-oral neuropathic areas (80%) in comparison to TMS carried out under local anaesthesia amongst both groups of patients (p < 0.01). There appeared to be no significant difference between the incidences of IANI or LNI caused on the right or left side of the mouth (p > 0.05). Likewise, the cause of injury did not correlate with the permanency of the injury.
Iatrogenous Injuries
11.6.1 Signs and Symptoms Approximately 70% of all patients presented with neuropathic pain, despite the additional presence of anaesthesia and/or paraesthesiae. Seventy percent of IANI patients suffered from paraesthesiae predominantly in association with pain (47%) or numbness (48%). Fifty-nine percent of IANI patients complained of anaesthesia in combination with pain and/or paraesthesiae whilst 67% of LNI patients complained of anaesthesia. Seventy percent of LNI patients complained of pain or discomfort often in combination with numbness (49%) and or paraesthesiae (54%). A significantly greater percentage of female patients complained of evoked and spontaneous pain, paraesthesiae and anaesthesia (p < 0.05). A greater number of IANI and LNI patients reported evoked pain if their nerve injury was more than 4 or 6 months duration. However, the duration of the injury did not significantly affect the incidence of spontaneous pain. Patients who had their TMS carried out under local anaesthesia were significantly more likely to complain of evoked pain, evoked and spontaneous paraesthesiae and numbness in comparison to those patients who had their TMS carried out under general anaesthesia. Age of the IANI and LNI patients did not correlate significantly with symptoms, neuropathic area or permanency of the injury. There was also a significant reduction in the number of LNI patients reporting spontaneous paraesthesiae if they had their symptoms for more than 6 months.
11.7 Mechanisms of Nerve Injuries 11.7.1 Local Analgesic Related Trigeminal Nerve Injuries Injuries to inferior alveolar and lingual nerves are caused by local analgesia (LA) block injections and have an estimated injury incidence of between 1:26,762 and 1:800,000 inferior alveolar nerve blocks (Pogrel and Thamby 2000; Haas and Lennon 1995). More recently the incidence of nerve injury in relation to Inferior Dental Bundles (IDB) has been calculated as 1:609,000 but with a significant increase in injury rate with 4% agents (Gaffen and Haas 2009). These injuries are associated with a 34% incidence of neuropathic pain which is high when compared with other causes of peripheral nerve injury (Pogrel and Thamby 2000). However the true incidence is difficult to guage without large population surveys. The problem with these injuries that the nerve will remain grossly intact and surgery is not appropriate as one cannot identify the injured region. A recent settlement of US $1.4 million (Main USA) for lingual nerve injury caused by local
503
analgesic inferior alveolar nerve block highlights the associated disability, social repercussions and potential costs of these injuries. Recovery is reported to take place at 8 weeks for 85–94% of cases (Smith and Lung 2006). IAN injuries may have a better prognosis than lingual nerve injuries and if the duration of nerve injury is greater than 8 weeks then permanency is a risk. The management indicated is for pain management if the patient has chronic neuropathic pain (Renton & Yilmaz 2010). In the US liability claims and malpractice suits are inherent risks associated with iatrogenous nerve injury (Venta et al 1998) and the reasons for avoidance of such injury are obvious. Iatrogenous nerve lesions may produce symptoms ranging from next to nothing to a devastating affect on of quality of life. Only few studies, however, describe the range of neurosensory disturbance in terms of signs and symptoms related to impaired nerve conduction and neurogenic affliction (Hillerup and Jensen 2006), and there is a need for better standardization and documentation of sensory deficits resulting from nerve injuries and their recovery (Zuniga 2001). Due to the incidence of nerve injuries in relation to dental anaesthesia warning of patients is not considered routine and indeed in the UK these iatrogenous injuries are not considered negligent. Nerve injury due to LA is complex. There may be elements of direct mechanical trauma by the needle no matter what type of bevel or indeed the method used for LA application (Pogrel et al 2003) which has been the focus of most papers. Chemical nerve injury may also be related to specific chemical agents (Loescher and Robinson 1998) and the LA components (type of agent, agent concentration, buffer, preservative). The resultant nerve injury may be a combination of peri-, epi- and intraneural causing subsequent haemorrhage, inflammation and scarring resulting in demyelination (Pogrel et al 2003). Articaine is an amide analgesic which was introduced to dentistry in 1998, however lidocaine (also an amide analgesic) remains the gold standard in the UK. Articaine is the most widely used local analgesic in many countries for over 20 years (Oertel et al 1997; Malamed et al 2000). Articaine is said to have a number of advantages, namely; low toxicity subsequent to inadvertent intravascular injection (Oertel et al 1997) which may be due to the rapid breakdown to an inactive metabolite (Articainic acid), rapid onset of surgical analgesia (2.5 =/−1.1 min) compared with conventional Lidocaine (Rood 1978) and better diffusion through soft and hard tissue (Simon et al 1998). The conclusion drawn is that Articaine is a safe and effective local anaesthetic for use in clinical dentistry (Malamed et al 2001; Haas and Lennon 1995). There is, however, some concern with regard using Articaine for inferior alveolar and lingual nerve blocks (Pedlar 2003; van Eden and Patel 2002; Haas and Lennon 1995; Gaffen and Haas 2009). The persistent altered sensation may be due to the high concentration of the local anaesthetic; however, the technique cannot be excluded as the cause for nerve
504
injury (Haas and Lennon 1995). Another report suggests that it is the type of anaesthetic that dictates the degree of inflammatory reaction to local anaesthetic; Lidocaine being the least irritant followed by Articaine, Mepivicaine and Bupivicaine (Riberio et al 2003). The components of Septicaine only differ in the active local analgesic content and it is not yet clear whether this agent is more likely to induce permanent nerve injury. However, as stated previously, its increased efficacy when used as an infiltration may preclude the necessity of administering a block in many patients. Longstanding altered sensation or nerve pain associated with Articaine inferior alveolar nerve blocks for routine dentistry has been reported (Wynn et al 2003; van Eden and Patel 2002; Gaffen and Haas 2009). However, a double crossover study assessed the comparative efficiency of Articaine versus Lidocaine reported no significant difference in onset of action and pain experience after buccal and palatal infiltrative injections using 4% Articaine 1:100,000 adrenaline or 2% Lidocaine 1:100,000 adrenaline (Oliviera et al 2001). The product information sheets state that resolution usually takes place within 2 weeks (Septodont web site:http//www.septodont.co.uk/Articaineuk/prescribing/spc1hundred.html). As a result of these concerns rescue inferior alveolar nerve blocks using Lidocaine remains standard care. With regard to oral surgical procedures Articaine is reported to be efficient in pain relief during and post apicectomy procedures using conventional IAN block injections (Meechan and Blair 1993). Alternatives to inferior alveolar nerve blocks have been suggested for implant surgery (Heller and Shankland 2001) and it is becoming routine practice for orthodontic extraction of premolars and restorative treatment of premolars and molars in adults using Articaine local analgesic infiltrations rather than inferior alveolar nerve blocks. Thus far there is limited, but promising, clinical evidence for avoiding inferior alveolar nerve blocks using Articaine infiltrations for mandibular dentistry and oral surgical procedures using Articaine infiltrations. Prevention of LA nerve injuries is possible. It has become routine practice for paedodontic extraction of premolars using infiltrations and many practitioners are routinely undertaking restorative treatment of premolars and molars in adults using LA infiltrations rather than inferior alveolar nerve blocks. This reduces the incidence of these troublesome untreatable injuries. Some simple steps may minimise LA related nerve injuries: • Avoid multiple blocks where possible. • Avoid IAN blocks by using Articaine infiltrations only. • Avoid high concentration LA for ID blocks (use 2% Lidocaine as standard). Intra-operatively all clinicians should document unusual patient reactions occurring during application of local analgesic blocks (such as sharp pain or an electrical shock–like sensation).
Surgical Disorders of the Peripheral Nerves
Management of these LA related injuries is essentially by counselling and medication for pain if present; however, prevention is better than cure. Reassurance of the patient and giving them realistic expectations of recovery with an explanation why they are not regularly warned of this complication is suggested. Recovery rate from LN or IAN block injuries is not generally reported. However, in the authors experience, if injury persists more than 6 weeks with more than 50% of the dermatome affected, recovery is unlikely. The incidence of neuropathic pain also seems high in this cohort (Renton and Yilmaz 2010). The neuropathic pain can be managed using antiepileptic drugs if the pain is neuralgic, tricyclic antidepressants if the pain is constant and burning in nature or external LA patches if the lip is very sensitive to touch or change in temperature. The clinician must be sympathetic to the patients concerns in that all iatrogenous injuries render many patients with problems with trusting clinicians again, particularly with regard to future dental treatment and a real fear of similar problems arising on the contralateral side.
11.7.2 Implant Related Nerve Injuries The incidence of implant related inferior alveolar nerve (IAN) nerve injuries varies from 0–40% (Delcanho 1995; Rubenstein and Taylor 1997; Wismeijer et al 1997; Bartling et al 1999; Walton 2000; von Arx et al 2005; Hegedus and Diecidue 2006; Hillerup and Jensen 2006). 25% of edentulous patients present with a degree of altered IAN function, thus reinforcing the guidelines on the necessity of preoperative neurosensory evaluation. Great care must be taken when selecting the patient and possible sites for implant placement (Feifel et al 1994). Appropriate radiographic evaluation of the implant site is required. Harris (2002) have reported explicit recommendations for preoperative radiographic evaluation prior to placement of implants. Cone Beam CT Scanning now introduced to many specialist practises and dental hospitals will provide low radiation dosage and improved imaging for planning implant treatment. Several papers have drawn attention to the weakness of CT evaluation in identifying the IAN canal with poorer sensitivity and specificity compared with pantomogram radiography (Tantanapornkul et al 2007). In 15% of patients the mandibular canal was not adequately visualized, and a computed tomography (CT) scan was used to plan the implant locations (Bartling et al 1999). Many practitioners use software to assist in the planning of implants and for the identification of the IAN canal position, with the specific aim to place the implants with a safety zone of more than 2 mm from the IAN canal (Greenstein and Tarnow 2006). It may be prudent to recognise that it is the practitioner who draws in
Iatrogenous Injuries
the IAN canal for an assessment which will not be objective but merely subjective and which increasingly leads practitioners in the USA to recommend a safety zone of a minimum of 4 mm. More recently Abarca et al (2006) have highlighted the necessity for cross-sectional imaging even for surgical procedures in the symphyseal region due to unforeseen nerve injuries. Most cases of iatrogenous paraesthesiae can be prevented but not remedied. However, when this problem occurs, follow-up must be initiated quickly, since the first few hours/ days may determine the degree of nerve healing. With the specific aim of placing the implants with a safety zone of more than 2 mm in order to prevent nerve injury Fig. 11.1 many practitioners in the USA are now recommending a minimum safety zone of 4 mm (Greenstein and Tarnow 2006). Once a safety zone is identified, implants can be placed anterior to, posterior to, or above the mental foramen. Prior to placing an implant anterior to the mental foramen that is deeper than the safety zone the foramen must be scrutinised to exclude the possibility that an anterior loop is present. Clinicians must have an awareness that certain preparation drills are up to 1.5 mm longer than the placed implant. In general, altered lip sensations are preventable if the nerve and mental foramen is located and this knowledge is employed when performing surgical procedures in the foraminal area (Greenstein and Tarnow 2006). Implant burs vary depending on the manufacturer and must be understood by the surgeon because the specified length (for example, a 10-mm marking) may not reflect an additional millimetre (or up to 1.5 mm) included for drilling efficiency. When placing implants in proximity to the mental foramen, the clinician must take into consideration the anterior loop of the nerve (Walton 2000) as well as the available bone above the mental foramen, because the inferior alveolar nerve often rises as it approaches the mental foramen compared with its height in the molar region. Implant bed
505
preparation is the most probable cause of the IAN injuries in the patient cohort that the author has evaluated thus explaining the often ‘distant’ implant from the IAN canal with nerve injury but subsequent osseo-integration and bony infill. A sudden ‘give’ during preparation may be indicative of protrusion through the lingual or buccal plate but may also be associated with fracturing of the IAN canal roof which will increase the risk of haemorrhage into the canal and subsequent compression of the nerve. It will, furthermore, increase the likelihood of extrusion of preparation debris or alkalinic solutions being introduced into the canal causing potential harm to the nerve. If there is an inferior alveolar arterial or venous bleed it may be advisable not to place the implant and to wait 2–3 days to ensure no nerve damage has occurred and then place the implant in granulation tissue which should not compromise the success of the implant. However, there is no evidence to support this practice as yet. If a nerve injury is suspected, the clinician should perform a basic neurosensory examination of the neuropathic area and ascertain whether the patient experiences pain, altered sensation or numbness and document the results within the day of surgery (when the effects of the anaesthetic should have worn off). A simple telephone call 6 hours post surgery will enable the surgeon to ascertain from the patient as to whether the analgesic effects of the local analgesia has worn off and if neuropathy is present. Nazarian et al (2003) noted several modalities of implant related nerve injury which may include direct trauma, inflammation and infection are postoperative neural disturbances main causes Fig 11.2. These injuries most likely occur during preparation rather than placement and may be directly related to the depth of preparation, implant length or width (Worthington 2004). Trauma may be direct (mechanical or chemical) or indirect (haemorrhage or scarring). The use of BiOss (pH 8.4) in close proximity to the nerve bundle should be avoided. Haemorrhage induced by ‘cracking’ of the IAN canal roof that may compress and cause ischaemia of the nerve if the implant is placed with or without back up, short or long implant.
11.7.3 Management of Implant Related Nerve Injuries
Fig. 11.1 Illustrating a case with bilateral IAN injury resulting from inadequate safety zone provision.
Intra-operatively all clinicians should document unusual patient reactions occurring during implant bed preparation or placement (such as sharp pain or an electrical shock–like sensation) and IAN vessel bleed, with consideration of delaying implant placement. Post operatively the patient should be contacted after the LA has worn off. If an injury is evident then consideration should be given to removing the implant within 24 - 30h of placement. Removal later is unlikely to resolve the nerve injury.
506
a
Surgical Disorders of the Peripheral Nerves
b
Fig. 11.2 Possible aetiology of nerve injury. The nerve injury may be due to direct mechanical trauma by the preparation bur or implant (a), extrusion of debris into the canal (b), the but most likely due to haemor-
c
rhage caused by preparation which continues after implant placement and results in nerve ischaemia (c).
avoidable and potentially permanent with or without surgical intervention (Pogrel and Thamby 2000). Intraoperatively
Fig. 11.3 Illustrating implant bed after removal of implant within 24 hours with successful resolution of IANI within days.
Bone graft harvesting is also associated with IAN injuries. Again it is crucial that appropriate training, planning, assessment and training should be undertaken in order to minimise nerve injury. Avoidance of implant nerve injury is sometimes attempted by using techniques including inferior alveolar nerve lateralization and posterior alveolar distraction. However, these high-risk procedures are more likely to result in inferior alveolar nerve defect regardless of the surgeon’s experience. Prevention of implant nerve injury: The most significant issue with implant related nerve injuries are that they are
• Do NOT place implant with intraoperative bleed: place implant 2–3 days later. • Non placement of the implant: If an implant is potentially violating the canal, with a sudden give experienced during preparation then its depth could be decreased in bone by unscrewing it a few turns ‘Back up’ which may leave excessive implant exposed coronally and short of the canal or alternatively replaced with a shorter implant. However, if a bleed is identified the implant should be removed immediately (Hegedus and Diecidue 2006). The author recommends removing the implant immediately and replacing it several days later when initial healing has taken place allowing optimal neural healing. Postoperatvely • Routinely check on patient early post operatively at 6 h. • If patient has neuropathy immediately after local analgesia has worn off: −− Consider removing the implant in less than 24 h. −− Steroids and NSAIDS. Refer to specialist Home check: Routinely contact patients post operatively to ensure local analgesia has worn off. If nerve injury occurs or is suspected after the procedure, the
Iatrogenous Injuries
507
Fig. 11.4 Algorithm illustrating recommended management of IAN implant related injuries.
Implant preparation
Pain Bleed
Uneventful prep and placement At 4 hours post op Patient reports neuropathy
Adjunctive NSAIDS (Ibuprofen 600mg QDS)
Delay placement of implant 2−3 days
Prednisolone (20mg step down by 5mg daily)
Immediately remove implant (within 30 hours)
Continue to monitor patient and contact specialist
clinician must inform the patient of its existence immediately and make a timely referral to an appropriately trained micro neurosurgeon if necessary.Late removal of implant: It is evident from the patient cohort evaluated that nerve injury appears to be permanent even at weeks post injury and even with the case where the implant was removed within 24 h. If neural recovery is to be optimised the potential harmful implant must be removed very early on when there is persistent neuropathy after the LA has worn off (a 4–6 h). However this may still be too late. With patients presenting late with IAN lesion the author no longer removes the implant, a policy similar to other specialists (Pogrel A. A personal communication) as it appears to be of little value in reversing nerve damage and associated symptoms.Thus these injuries may be irreversible and places the emphasis on prevention rather than cure. • Planning – ensure more than 4 mm safety zone. • Bleed during implant bed preparation – delay implant placement. • Persistent numbness after LA has worn off – remove implant within 36 h.
11.7.4 Endodontic Related Nerve Injury Reports on endodontic (ENOO) nerve injuries which may not be limited to those teeth proximal to the IAN canal but may occur in maxillary teeth as well. The largest series reported to date was Pogrel and Thamby (2000) who reported on 61 patients with endodontic nerve injury over an 8 year period. The risk factors for endodontic inferior alveolar nerve injury include:
• • • •
Proximity of the tooth to the mandibular canal. Over instrumentation. Overfill. Chemical nerve injury (including sodium hypochlorite).
Similar to extracting mandibular teeth proximal to the IAN canal, root canal treatment (RCT) of these teeth will also have the potential for increasing the risk of nerve injury as the apex of the tooth may be adjacent or intruding into the IAN canal and any small degree of leakage or overfilling may compromise the nerve. Assessment of the proximity of the tooth apex to the IAN canal has become significantly improved with Cone Beam CT scanning (CBCT). This however has the attendant risk of additional radiation and may not provide significantly more information than a plane long cone radiograph. Any tooth requiring endodontic therapy that is in close proximity to the IAN canal should require special attention. If the canal is over prepared and the apex opened then chemical nerve injuries from irrigation of canal medicaments is possible. In addition physical injury precipitated by overfilling using pressurised thermal filling techniques can occur. Post operative RCT views must be arranged on the day of completion and identification of any RCT product in the IAN canal should be reviewed carefully. If IAN function is compromised after LA has worn off then immediate arrangements should be made to remove the over fill. The optimum pH of an endodontic medicament is as close as possible to that of body fluids, i.e., around 7.35; higher and lower pH’s are likely to cause cellular necrosis of tissues in direct contact with the medicament. The clinician must also consider the pH of some of the routinely used endodontic and related dental materials.
508
Surgical Disorders of the Peripheral Nerves
Fig. 11.5 Radiographs illustrating over fill of endodontic material into the IAN canal.
Commonly used endodontic Medicaments. •
Formocresol
pH 12.45 ± 0.02
•
Sodium hypochlorite
pH 11–12
•
Calcium hydroxide (Calyxl).
pH 10–14
• Antibiotic-corticosteroid paste (Ledermix)
pH 8.13 ± 0.01
•
Neutral
pH 7.35–7.45
•
Eugenol
pH 4.34 ± 0.05
• Iodoform paste
pH 2.90 ± 0.02
Nerve tissue is very sensitive to pH changes and chemical nerve injuries are commonly permanent and often cause severe neuropathic pain. If the patient is suffering from neuropathy after the LA has worn off and the post operative radiographs confirm that there is no radio opaque material in the canal, chemical nerve injury may be presumed. This may be an irreversible injury to the nerve and subsequent ‘swift’ removal of the RCT or tooth extraction is unlikely to result in resolution of the nerve injury. Based on the current evidence the authors recommend that the practitioner undertaking the endodontic care: 1. Preoperatively will identify teeth proximal to the IAN and take special care in preventing apical breech. 2. Will recognise and record certain events during operation including:
–– Intraoperative pain during irrigation. –– Intraoperative pain during preparation and filling. –– Inferior alveolar vessel bleed during preparation and delay filling. 3. Postoperatively. If endodontic nerve injury is suspected the post operative radiograph must be scrutinised for evidence of breach of apex and deposition of endodontic material into the IAN canal. If this is suspected the material, apex and or tooth must be removed within 48 h of placement in order to maximise recovery from nerve injury. The practitioner is well advised to routinely contact patients post operatively to ensure local analgesia has worn off. If nerve injury occurs or is suspected after the procedure, the clinician must inform the patient of its existence immediately and consider removing the overfill of endodontic material, or apicect or extract tooth if present in less than 30–48 h. If, however, there is no evidence of overfill symptoms may be caused by: • Chemical nerve injury from irrigant or endo material. • Thermal damage. • Apical inflammation (neuritis) confirmed by prescription of antibiotics. In any event once neuropathy is identified the clinician must reassure the patient, prescribe steroids (Prednisolone step down 15 mg 5 days, 10 mg 5 days and 5 mg 5 days and high
Iatrogenous Injuries
509
dose NSAIDs, 600 mg Ibuprofen), and make a timely referral to an appropriately trained micro neurosurgeon if necessary.
11.7.5 Third Molar Surgery The risks for lingual nerve injury during third molar surgery are well established. These include: • Difficulty of surgery (Renton and McGurk 2001; Valmaseda-Castellón et al 2000; Bataineh 2001; Hill et al 2001). • Age of patient (Renton and McGurk 2001). • Surgeon (Valmaseda-Castellón et al 2000; Gulicher and Gerliach 2001; Renton and McGurk 2001).
IAN injury
Endo related? Less 24-48 hours
YES Remove endo/ tooth/ material and review
NO
Leave in situ/ Reassure patient- Medication for pain
Fig. 11.6 Algorithm suggesting the management of endodontic related nerve injuries.
a
b
c
d
Lingual nerve
Fig. 11.7 Illustrating the principles of the Buccal technique. (a) The minimal buccal mucoperiosteal flap. (b) buccal bone removal. (c) position of lingual nerve (green). (d) Diagramatic representation of the underside of sectioned crown illustrating that there is no ‘breech’ of tooth material lingually, distally or mesially.
• Surgical skill (Renton and McGurk 2001). • Surgical technique – lingual flap surgery (Shepherd 2006; Gomes et al 2005; Robinson and Smith 1996; Robinson et al 1999; Valmaseda-Castellón 2000; Bataineh 2001; Pichler & Bierne 2001). A recent systematic review concluded (Pichler & Bierne 2001) that there was significant increase in lingual nerve injury when using lingual split technique with lingual flap or lingual flap compared with the buccal approach (Lingual Split [9.6% temporary 0.1% permanent], Lingual approach [6.4% temporary 0.2% permanent], Buccal approach [0.6% temporary 0.06% permanent]). Only 8 of 739 studies were eligible (including only one prospective randomised controlled study). Since this review there have been several prospective studies (Renton and McGurk 2001; Hill et al 2001) and several prospective randomised controlled studies (Valmaseda-Castellón et al. 2000; Bataineh 2001; Gargallo et al 2000; Chossegros et al 2002). The findings of these studies support the conclusion that lingual flap ‘protection’ of the lingual nerve is not necessary and is potentially harmful. Lingual nerve injuries in relation to third molar surgery: Prevention of lingual nerve injuries may be possible if under- and post-graduate trainees are trained properly using the buccal technique. There is still a reluctance to change from traditional techniques (lingual split / lingual retraction) as in some experienced hands these techniques involve minimal morbidity. However, if the buccal technique is adhered to there is no risk to the lingual nerve. Appropriate training for this technique is required though rarely undertaken in dental schools in the United Kingdom (Renton et al 2006). Avoidance of the envelope flap minimises the necessity of a long distal extension of the flap which exposes the distal bone adjacent to the third molar and thus may ‘tempt’ the surgeon to remove distal bone which would compromise the lingual nerve. The Buccal technique: Third molar surgery related inferior alveolar nerve injury is reported to occur in up to 3.6 % of cases permanently and 8% of cases temporarily. Factors associated with IAN injury are age, difficulty of surgery and proximity to the IAN canal. If the tooth is closely associated with the IAN canal radiographically; ie. superimposed on the IAN canal, darkening of roots, loss of lamina dura of canal, deviation of canal Fig. 11.8 (Howe and Poynton 1969; Rood et al 1983, 1986; Rud 1983a, b) then 20% of patients having these teeth removed are at risk of developing temporary IAN nerve injury and 1–4% are at risk of permanent injury (Howe and Poynton 1960; Rood & Noraldeen Shehab 1990; Rud 1983a, b; Renton et al 2005). Radiographic signs Fig 11.8 indicative of possible IAN risk include: • • • •
Diversion of the canal. Darkening of the root. Interruption of the canal LD. Juxta apical area (Fig. 11.9).
510
Fig. 11.8 DPT radiographs illustrating 2 cases of ‘high risk’ mandibular third molars. In both cases the lower third molar is crossing the IAN canal completely, there is darkening of the tooth roots and loss of lamina dura of the canal roof and floor.
Fig. 11.9 Juxta apical area. This new radiographic sign is a well circumscribed radiolucent area lateral to the root rather than at the apex. MRI and CT studies have elicited that this is likely to be continuity of IAN lamella with the periodontal lamina dura of the adjacent tooth.
If these plain film radiographic risk factors are identified then removal of the third molar will result in elevated risk of IANI (2% permanent and 20% temporary). The patient must be informed about this elevated risk. There is increasing evidence that Cone beam CT scanning of high risk teeth will further establish the relationship between the IAN and the roots, Fig. 11.10. In many cases the CBCT reaffirms the proximal relationship which would support planned coronectomy if appropriate, but would not change the planned treatment. However, in a few cases, despite high risk identification based on plain films, some IANs are found to be distant from the roots using CBCT which would allow for removal of the tooth rather than planned coronectomy (Frafjord and Renton 2010). Further research is required to ascertain the
Surgical Disorders of the Peripheral Nerves
risk benefits of CBCT and as to whether it is indicated for treatment planning in these high risk cases. If the tooth is in close proximity to the IAN on plain film then Cone beam CT scanning may further elucidate the relationship between IAN and tooth roots. If the tooth is non vital, or has pathology associated with it, then tooth removal has to take place and the roots should be sectioned appropriately to minimise trauma to the adjacent IAN. If the tooth is vital and the patient is not medically compromised and at increased risk of infection a coronectomy may be considered or if non-vital splitting of the roots may miminise injury if tooth if multiple rooted. Coronectomy avoids the nerve injury by ensuring retention of the roots when they are close to the canal as estimated on radiographs. The tooth must be high risk, vital and the patient must not be immune compromised and at higher risk of infection. A study of 100 patients showed that the risk of subsequent infection was minimal and morbidity was less than after the traditional type of operation (Renton et al 2005). Over a period of 2 years some apices migrated and were removed uneventfully under local anaesthetic. Dry socket incidence was similar to the surgical removal group and treated in the same way (using Corsodyl irrigation and Alvogyl past). On the premise that coronectomy reduces the risk of nerve injury this procedure has been recommended for those patients for whom there may be serious repercussions from numbness of the lip (wind instrument players, actors, singers, and others) and for those at higher risk of IAN injury. Renton et al reported that the inferior alveolar nerve was often injured by extraction of third molars, the roots of which were superimposed radiographically on the nerve canal similar to previous studies (Howe and Poynton 1969; Rood et al 1983, 1986; Rud 1983a, b). Most of these injuries were temporary but two were permanent both of whom were treated by tooth removal not coronectomy. The author found evidence that some radiographic signs may be more predictive of nerve injury than others including deviation of the canal at the apex and the presence of the juxta apical area Fig. 11.9. Five coronectomy articles report on more than a single patient. Four case series (Pogrel et al. 2004, 50 cases; O’Riordan 1997; 2004, 95 cases with 52 patients followed up; Freedman 1992, 35 cases; and Knuttson et al. 1989, 33 cases). The fifth article was a randomized controlled trial (Renton et al 2005). In all cases, coronectomy was suggested as a technique of partial root removal when Panorex imaging suggested an intimate relationship between the roots of the vital lower third molar and the IAN nerve; in these cases the tooth still needed to be removed. (Note: Cone beam CT was not available at the time the studies were conducted.) All papers suggested that the technique had merit and many practitioners regularly use the coronectomy approach in order to minimise IAN lesions. Coronectomy technique involves using the buccal approach and removal of buccal bone using a fissure bur
Iatrogenous Injuries
511
Fig. 11.10 CBCT of high risk third molar.
down to amelo-dentinal junction (crown root junction). The crown is part sectioned from the root using a fissure bur and elevated similar to the buccal approach technique Fig. 11.11. On elevation of the crown from the roots, mobilisation of the roots may occur particularly if the patient is young, female and the roots are conical (Renton et al 2005). If the roots are mobilised they must be removed. Thus the patient must be consented for coronectomy and or removal if the roots are mobilised intraoperatively. On exposure of the pulp and immobilised roots the surgeon must ensure that there is no enamel retained and the use of a rose head bur may be necessary to remove any enamel spurs. It is recommended not to touch or medicate the vital pulp. Closure of the buccal flap over the roots is achieved with 1–2 4/0 vicryl sutures. No antibiotics are recommended, just pre and post operative Corsodyl and good oral hygiene. The patient must be warned of possible ‘dry socket’ and to seek treatment if there is persistent pain or swelling. Reports of complications subsequent to coronectomy are rare. The author has had to remove roots in 2 patients of the
original 52 study coronectomies at up to 6 years post operatively. The patient must be warned of a possible second surgical intervention if complications arise. Four (8%) of our study patients reviewed at 2 years post operatively showed radiographic evidence of migration of the retained root away from the canal which may infer that if the roots do require removal at a later stage then there the risk of damage to the IAN will remain reduced. In the authors clinics we do not retreat ‘dry sockets’ or persistent infection associated with retained coronectomied roots, but prefer to remove the roots early on because we encountered two cases of temporary infection IAN neuritis (<6 weeks post operatively) associated with infected coronectomised roots. There is a need for reports on long term evaluation of coronectomy complications. The prevention of inferior alveolar nerve injuries in relation to third molar surgery must be based upon: • A clinical decision that the tooth needs to be extracted based on National Institute for Clinical Excellence
512
a
Surgical Disorders of the Peripheral Nerves
b
c
Fig. 11.11 Coronectomy technique.
(NICE) guidelines. Prophylactic surgery should only be undertaken when specifically indicated such as before radiotherapy, before chemotherapy or before IV bisposphonates (none of which are included in NICE guidelines). • The clinician must identify mandibular teeth at high risk of IAN injury based on plane radiographic features and/or cone beam CT scanning. If deemed at high risk the patient must be made aware of the increased nerve injury incidence (increased risk 2% permanent and 20% temporary) and perhaps offered alternative procedures that may reduce the risk of injury. • If the tooth is in close proximity to the IAN on plain film then Cone beam CT scanning may further elucidate the relationship between IAN and tooth roots. If the tooth is non vital, or has pathology associated with it, then tooth removal has to take place and the roots should be sectioned appropriately to minimise trauma to the adjacent IAN.
Fig. 11.12 Radiograph showing socket dressing that resulted in IAN neuropathy.
Iatrogenous Injuries
513
11.7.6 Dental Extraction of Other Teeth Proximal to IAN Canal
progress on to osteomyelitis associated with non vital bone grafts that may not be removed quickly enough in the hope that the bone graft can be saved.
The clinician should be aware that any mandibular tooth (lower 8, 7, 6, 5 or 4) in which the tooth root is crossing the IAN canal, and displays the radiographic signs is at increased risk of IAN injury and there is increased possibility of damage to the IAN on removal of the tooth. The patient must be assessed accordingly, consented and treated similarly to high risk third molar teeth.
11.7.9 Management of Trigeminal Nerve Injuries
11.7.7 Socket Medications Extraction of any mandibular tooth in close proximity to the IAN canal may expose the IAN to socket medicaments. If these are irritant to the neural tissue they can lead to chemical neuritis and, if persistent, to a neuropathy which is untreatable and often associated with neuropathic pain. There is limited availability of information on the relative alkalinity or acidity of various dental compounds used for socket medication including; Alvolgyl, Whiteheads varnish, Corsodyl and Surgicel. However, a previous study (Loescher and Robinson 1998) highlighted the relative neurotoxicity of Carnoys solution, Surgicel, Whiteheads varnish and Bismuth Iodoform paraffin paste (BIPP) and concluded that Carnoys is likely to cause permanent nerve damage and Surgicel along with Whiteheads varnish can cause temporary sensory disturbances. BIPP was the least neurotoxic. Bone wax has a neutral pH, however excessive packing or pressure can lead to nerve compression and injury.
11.7.8 Post Operative Infection Related Nerve Injuries Inferior alveolar neuritis can present as a symptom of local mandibular infection associated with a periapical abscess on a non vital tooth close to the IAN canal or as a sign of osteomyelitis. This may present as a persistent or recurrent dry socket that has required repeated socket irrigation and redressing. Suspicion should arise after the second or third dressing when accompanied by persistent pain and non response to antibiotics. Once IAN neuropathy develops this is a sign of spreading bone infection and the tooth should be removed or the socket surgically debrided immediately in order to ensure that no bone sequestrae or tooth fragments remain. More recently with the advent of bone graft surgery for implants some patient
Prevention is better than cure. As yet there is little or no consensus on policies about: • • • •
Assessment of inferior alveolar and lingual nerve injuries. Referral for management. Most appropriate management. Selection criteria for reparative surgery and timing of surgical repair. • Success outcomes of repair or other intervention. Recommendations for the development of guidelines for the assessment and management of lingual nerve injuries in order to provide an evidence base with which to improve the management of these injuries have been made by Dodson & Kaban (1997), Colin and Donoff (1992), Donoff (1995), Robinson et al (2004), Susarla et al (2005), and Susarla et al (2007). There is developing evidence that implant associated IAN injuries should be managed by removal of the implant within 24–30 h of placement in order to maximise IAN recovery (Kraut and Chahal 2002;Chaushu et al 2002; Khawaja and Renton 2009).
11.7.10 Evaluation of Trigeminal Nerve Injuries Any procedure that has an inherent nerve injury risk must be followed by ‘home check’ which ensures that nerve function has returned after resolution of LA. Most post operative visits often take place at 2 weeks and if neuropathy is present this may be a timely opportunity to refer the patient to a specialist for assessment. It may take several weeks for provision of a consultation in secondary care. In the presence of persistent nerve injury, early referral may provide an opportunity for earlier and improved care for the patient. When the patient presents with iatrogenous trigeminal nerve injury the clinician must establish whether the injury is temporary or permanent to determine whether intervention is required. This is often impossible to determine but duration and cause of injury are probably the most important determinants of how best to manage the patient. It is incorrect to assume that lingual nerve and inferior alveolar nerve injuries are similar. Only lingual nerve injuries are
514
Surgical Disorders of the Peripheral Nerves
mainly temporary with 88% of lingual nerve injuries resolving within the first 10 weeks post third molar surgery (Mason 1988; Blackburn 1990). In contrast the IAN is more at risk from a variety of dental procedures as the IAN is contained within a bony canal predisposing it to ischaemic injuries with intra-luminal bleed. Patients presenting at specialist clinics with these nerve injuries, predominantly have problematic pain or troublesome altered sensation rather than numbness. Assessment of the patient should include: • Duration of injury (weeks/months since injury). • Indentify mechanism of injury. • Symptoms: Do you have any pain or discomfort? Spontaneous pain constant? Spontaneous pain paroxysmal? Evoked pain? What is the type of pain or discomfort (dysaesthesiae)? Pins and needles = paraesthesiae? Sharp shooting = neuralgic? Burning stretching = dysaesthesiae and/or numbness)? Do you have numbness or heightened sensation or both? • Functional problems: –– Disability associated with difficulty with; eating, drinking, speaking, tasting, sleeping, tooth brushing, kissing, applying makeup or shaving. –– Interference with work ability and type of occupation are important features. –– Social problems such as embarrassment with eating or speaking in public are common. Problems with personal relationship (pain in kissing, resulting in avoidance of personal contact). • Identify the extent of injury: –– Size of neuropathic area: extent of the dermatome affected by neuropathy. –– Subjective function: normal function on contra-lateral side provides 10 out of a possible 10 on ‘crude’ repeated stimulation using closed college forceps and 0 out of 10 with no stimulation. This function is compared with the function out of 10 within matched sites in the neuropathic area. If there is heightened sensitivity then hyperaesthesia may be present. • Lingual nerve injury prediction of permanency using subjective function. • Neuropathic area
• >50% surface of the dermatome
• Subjective function score • <4/10 • Disability Lesion persisting for 8 weeks • >6/10 indicates likelihood of permanent injury
• Sensory function tests: Only routine mechanosensory tests are recommended which test mainly A beta large fibres as taste tests are unreliable. • Mechanosensory function. This is commonly tested comparing normal side function with affected area using: –– Repeated light touch (wisp of cotton wool/ paper towel). –– Sharp blunt discrimination using a dental probe repeating alternated sharp / blunt stimulation whilst asking the patient to differentiate. –– Simple two point discrimination tests can be undertaken with college forceps. Normal parameters are published, Moving point discrimination may be examined using a Sable brush no 9. • Other tests: –– Look for scarring which may be indicative of local repeated trauma. –– A diminution of the Fungiform papillae count compared to uninjured side, may indicate loss of Chorda Tympani input. –– Palpation of the lingual nerve ‘zone’ adjacent to lingual aspect of the wisdom tooth site may elicit a neuralgic sensation and may be indicative of neuroma formation and/or scarring of the lingual nerve. • Special tests. • PAIN assessment using Visual Analogue Scale –– At rest. –– Dynamic allodynia. –– Cold allodynia. • The response to Capsaicin or other specific agents known to activate specific pain receptors. • Thermo sensory: Quantitative sensory testing (see Chap. 5) measures the function of warm and cool sensitive fibres (C ad A delta fibres) and provides evidence about pain fibre function. • Biopsy is usually reserved for research purposes in these cases but it is routine practice for neurologists in diagnosis of peripheral neuropathy (see Chap. 7). Most importantly the clinician must establish the patients’ expectations. After a comprehensive assessment the possible timing of any necessary intervention may be indentified. The fundamental question is ‘What is the patient complaining of?’ It may be pain or discomfort, functional or social problems or the patient expectations of the treatment process. Normal sensation is unlikely to return to normal after 3–6 months with or without intervention. The patient also requires reassurance that the nerve injury will not increase risk of cancer or other pathology.
Iatrogenous Injuries
11.7.11 Possible Interventions and Timing The clinician must discern exactly what are he or she is trying to treat the patient for, is it poor mechanosensory function (which remains the focus of many surgical studies) or more pertinently should it be what the patient is complaining of? The patient will often complain of: • Disability associated with: –– Altered sensation, severe discomfort, pain or numbness. –– Large neuropathic area. –– Interference with eating and drinking etc. • Can’t cope! Many patients find accepting or coping with even minimal iatrogenous nerve injuries very difficult. This may be due to the unexpected nature of the injury, poor informed consent, poor post operative management of the patient and lack of information. • The planned treatment must address the patients concerns appropriately and the aims of treatment would ideally provide: –– Improved function: Treatment will NOT restore function completely in: –– Eating, drinking, speaking, sleeping. • To improve sensation: Treatment will NEVER restore normal sensation but aim to: –– Reduce neuropathic area. –– Improve general sensory (i.e. mechano-sensory function). –– Improve special sensory (i.e. taste). –– Reduce pain or altered sensation. Elevating intermittent pain to persistent pain would be a negative outcome as would causing a patient to have discomfort or pain when previously they had only anaesthesia. We have seen that the management will depend on the mechanism and the duration of the nerve injury. Many injuries have limited benefit from surgical intervention and should be managed symptomatically. Earlier intervention is required for endodontic, implant and third molar related nerve injuries as discussed. If there is a persistent large neuropathic area (>40% dermatome) then a severe nerve injury is present. If pain and or hypersensitivity are present these will often be the main precipitating factors of difficulty with daily function. These symptoms may not be best treated using surgical intervention however the patient’s inability to cope with disability and are often the driving factor for the patient seeking treatment.
11.7.12 Possible Management Tools (Table 11.9) 1. Counselling is indicated for all patients with nerve injuries caused by:
515 Table 11.9 Suggested management strategy for iatrogenic trigeminal nerve injuries. Mechanism
Duration
Treatment
Known or suspected nerve section
Immediate exploration
TMS IANI –retained roots
<30 hours
Immediate exploration
Implant
<30 hours
Remove implant
Implant
>30 hours
Treat patient therapeutically
Endodontic
<30 hours
Remove tooth / overfill
Endodontic
>30 hours
Treat patient therapeutically
TMS IANI large neuropathic area, pain and disability
<3 months
Consider exploration
TMS LNI – large neuropathic area, pain and disability
<3 months
Consider exploration
TMS IANI –
>6 month
Treat patient therapeutically
TMS LNI–
>6 month
Treat patient therapeutically
LA, fracture, ortho gnathic, other surgery
Treat patient therapeutically
2. Medical symptomatic therapy is indicated for patients with pain or discomfort and for patients with anxiety and or depression in relation with chronic pain. This includes: –– Topical agents for pain. –– Systemic agents for pain. 3. Surgical exploration: –– Immediate repair if nerve section is recognised. –– Remove implant or endo material within 24 – 30 h. –– Explore IAN injuries through socket within 4 weeks. –– Explore LN injuries before 12 weeks.
11.7.13 Reassurance Counselling/Cognitive Behavioural Therapy The clinician must consult the patient in depth, reaffirm that the nerve injury is permanent and provide reassurance and explanation. Often patients who can manage their pain but have associated functional difficulties are successfully managed by an explanation of their symptoms and informed of the permanency of their condition with realistic expectations. They can also be reassured that these injuries do not predispose them to cancer and indeed will never worsen.
516
Increasingly there is evidence for successful treatment of chronic conditions using Cognitive Behavioural Therapy (CBT). Within our field chronic temporo-mandibular pain conditions can be successfully treated and CBT or formal counselling may have an increasing role particularly for patients with injuries who are unable to cope with the consequences of their nerve injury which may be significantly impacting their social or work ability. To date there is no evidence for the value of counselling treatment for IANIs or LNIs.
11.7.14 Medical Intervention Since the altered sensation may be due to an inflammatory reaction, a course of steroid treatment may minimise inflammation and reduce damage to the nerve (oral Prednisolone 1 mg per Kg per day – maximum 80 mg) for first week with stepping down daily over the following week. This is combined with a high dose of non-steroidal anti-inflammatory medication (such as ibuprofen, 800 mg, three times per day) which should be prescribed for 3 weeks. There is minimal evidence base for steroid medication for the prevention of peripheral sensory nerve injuries (Von Arx & Simpson 1989), however it is routinely prescribed for Bell’s palsy and viral infections of the facial nerve. The majority of patients presenting with neuropathic pain require analgesia. Routine analgesics do not work for chronic neuropathic pain and in effect these patients are best treated with topical or systemic medications generally used for other chronic pain conditions. The first line medication for neuropathic pain is either Tricyclic antidepressants (amitriptyline or nortriptyline) or antiepileptics (carbamazepine, oxcarbazepine, pregabalin or gapapentin). Few patients continue their medication to control their neuropathic pain, as many patients can not tolerate the side effects. As an alternative many patients seen with extraoral IAN neuropathy with mechanical allodynia interfering with function are offered 5% Lidocaine topical patches (Khawaja and Renton 2009). To date there is no secure evidence for the treatment of IANIs or LNIs using any form of medication.
Surgical Disorders of the Peripheral Nerves
LN injury
Less 12 weeks Large neuropathic area? Poor mechanosensory function
YES
NO Reassurance Medication for pain Counseling for Functional
Surgical exploration + Review
Fig. 11.13 Algorithm illustrating treatment for Lingual nerve injuries.
A. 'Normal Nerve'. Exploration
B. Release scarring over 'normal nerve'. Decompression C. Nerve interrupted by scar tissue Neuroma in continuity (NIC) excision and re-approximation D. Nerve completely sectioned End neuromata (EN) excision and re-approximation with minimal tension
Fig. 11.14 Illustrating the possible findings on exposure of the damaged nerve. The surgical approach conventionally was extra-oral (Figure 14 and 15) with the necessary removal of the submandibular gland. More recently the intra-oral approach has become commonplace either via retromolar or floor of mouth approach.
RIGHT
LEFT
B 8
8
7
7
Floor of mouth
11.7.15 Surgical Intervention 6
Surgical intervention for repair of iatrogenous trigeminal nerve injury has remained the focus for our specialty. Being of mechanistic temperament we have focused on the technical aspects of surgery rather than a holistic perspective. A metanalysis of surgical intervention for trigeminal nerve injuries is not possible as there are no prospective randomised studies in this field (Dodson & Kaban 1999). A review of 26 studies reporting lingual nerve repair (Renton 2010 in press) highlights the variation in reported patient, management and surgical parameters. Only two
6
A 5
5 4
4 3
3 2
1
1
2
Sulcus Fig. 11.15 Illustrating placement of access incisions for A. Floor of mouth approach and B. Retromolar approach.
Iatrogenous Injuries
studies identify selection criteria for surgical repair and the mean delay from injury to repair is over 16 months. This delay is mainly due to the fact that 80% of lingual nerve injuries in relation to lingual access third molar surgery resolve at 10 weeks postoperatively (Mason 1988; Blackburn 1990) which does not necessarily reflect the resolution of IAN injuries. When assessing the criteria for surgery very few reports address any or some of the fundamental issues including: • • • • •
Why do we operate? When do we operate? What technique should be used? How do we assess the outcome? Why only surgery?
Several authors recommend referral to a micro neurosurgeon within 2 months (Nazarian et al 2003; Caissie et al 2005). Ideally the patient should be referred immediately to allow the surgeon to make their own assessment and measure the lack of functional recovery over a period of 2 months before intervening. Although considerable functional improvement is seen in many patients after surgery (Robinson 1999; Ziccardi & Steinberg 2007), regaining normal sensation is not possible; all patients who undergo surgery will have some permanent sensory deficit. Many authors tend to over optimise the suggested benefits of microsurgery (Kraut and Chahal 2002; Hegedus and Diecidue 2006). As early as 1985, Mozsary and Syers discussed guidelines for microsurgical reconstructive procedures in the treatment of inferior alveolar nerve injuries. Ruggiero 2001, Colin et al (1997) and Pogrel and Maghen (2001) reported favourable patient responses to inferior alveolar nerve repair, and all emphasized the need for repair before Wallerian degeneration of the distal portion of the inferior alveolar nerve has occurred but more importantly before chronic neuropathic pain becomes irreversible, that is before 12 weeks. Microsurgical indications for lingual nerve repair after injury incurred during third molar surgery have been reported (Zuniga and LaBlanc 1993; Zuniga and Zenn 2001;Zuniga 2001). There are no general recommendations regarding IAN injury related to implants (Khawaja and Renton 2009). Microsurgery may be indicated in the following cases; confirmed transection of a nerve, total anaesthesia of the affected area 2 months after the trauma, lack of protective reflexes on biting or burning of the tongue or lower lip 2 months after trauma with little or no improvement or dysaesthesiae (Caissie et al 2005; Pogrel 2002). Immediate exploration and necessary repair when nerve section is known or highly suspected transection of the nerve has taken place. Many authors recommend referral of injuries before 4 months (Hegedus and Diecidue 2006) but this may be too late for many peripheral sensory nerve injuries. We now understand that after 3 months permanent central and peripheral changes occur within the nervous system subsequent to
517
injury, that are unlikely to respond to surgical intervention (Zicardi & Steinberg 2007). Microsurgery involves general anaesthesia, a period of convalescence and a few weeks off work. The surgeon dissects the affected nerve and, if the damage is extensive, excises the neuroma then joins the proximal and distal portions (see below). Despite the significant shortcomings of the timing of surgery, lack of specific criteria for surgery and no standardisation of repair several authors report optimistic long term outcomes for microsurgical repair (Rutner et al 2005; Strauss et al 2006).
11.7.16 Reasoning for Early Nerve Repair Neuroplasticity is a process by which the peripheral and central nervous systems interact and changes evolve. The traditional concept of the nervous system being like an electrical wiring system is invalid. There is constant communication (chemical and electrical) between the peripheral and central nervous components and also between cell bodies and peripheral receptors within the peripheral system. If pain persists in the periphery for more than 3 months permanent changes occur within the sensory cortex. Thus treatment of nerve injuries with neuropathic pain should be undertaken within 3 months if possible to maximise any beneficial effects. Interestingly Seddon’s dictum (1943) suggested that ‘if a purely expectant policy is pursued the most favourable time for operative intervention will always be missed’. Two recent reports suggest improved holistic outcomes with surgical intervention before 3 months (Susarla et al 2007; Ziccardi & Zuniga 2007; Robinson et al 2000). The type of surgical intervention will depend on the surgical findings at the time. There are four main scenarios when exploring nerves Fig 11.13 which cannot be differentiated clinically from each other.
11.7.17 Surgical Technique Surgical findings will dictate the surgical intervention for these nerve injuries 1. ‘Normal nerve’ Exploration. On surgical exploration if the nerve is apparently uninjured then no action is indicated Fig 11.16. 2. Release scarring over ‘normal nerve’ Decompression. If on release and dissection of scar tissue in the region the nerve is found uninjured there is no indication for intervention upon the nerve. However, the release of surrounding scar tissue (external neurolysis or surgical decompression Fig. 11.17) has been reported to provide improvement for patients with IANIs and LNIs.
518
Fig. 11.16 Clinical picture illustrating the placement of the retromolar incision which involves a lingual flap extending from lower seven to canine with a distal extension over the external oblique ridge similar to that used for bilateral sagittal split osteotomy.
Fig. 11.17 On exploration the lingual nerve appears to be uninjured.
3. Nerve interrupted by scar tissue or neuroma. If on exposure of the nerve scar tissue or neuroma formation is found to interrupt the continuity of the nerve, (Neuroma in continuity [NIC] Fig. 11.18) excision of the neuroma or scar tissue and re-approximation of the ‘fresh’ nerve endings is indicated. This is performed under surgical microscopy or high magnification loupes. Anastamosis using epi-neural sutures (usually 4–8 using 6/0–8/0 either vicryl or nylon sutures) is recommended with minimal tension otherwise the tension itself may precipitate neuropathic pain. If the interneural space is large some authors recommend grafting; however, outcomes from grafting are poor and this is now rarely indicated. If approximation requires too much tension then the space is left and not grafted, this also avoids complications arising from the nerve graft site.
Surgical Disorders of the Peripheral Nerves
Fig. 11.18 On exposure and removal of scar tissue around the lingual nerve the nerve appears intact.
4. Nerve completely sectioned. If on exposure of the nerve it is found to be completely sectioned with end neuromata (EN) formation, excision of damaged endings followed by re-approximation of the ‘fresh nerve endings’ is recommended. If the gap is too large to allow anastamosis with tension then a gap is left. This more commonly occurs with IANIs due to difficulty with mobilisation within the ID canal. Even with extensive buccal corticotomy closure of gaps approaching 2 cm are difficult without tension. Surgical approach for IAN injuries will vary but the principles of findings on surgical exploration dictating the applied technique remains similar Fig. 11.20. 1. Exploration of the IAN via the socket. If the IANI is less than 6–8 weeks then assess using a pre operative radiograph for retained roots and obvious compression of the canal. If the injury presents early on the IAN can be explored via the socket. Fig 11.21. The attempted extraction took over 3 h; the patient was suffering from extreme neuropathic pain and 100% neuropathy. The exploration took place at 4 weeks with the roots removed both circumscribing the IAN. The patient made a complete recovery at 8 weeks post extraction. 2. If the IANI is of longer duration with a healed socket the preferred surgical approach may be using a lateral corticotomy Fig. 11.22. Some authors have recommended full lateral corticotomy with the removal of buccal plate from lower 8 region to mental foramen to facilitate mobilisation of the IAN. However very few reports using this method suggest improved outcomes.
11.7.18 Medico Legal Issues With regard to lingual nerve injuries related to third molar surgery most patients recover normal sensation without treatment but those with permanent deficits are often have severe
Iatrogenous Injuries
519
a
b
Fig. 11.19 Exploration reveals the lingual nerve to be interrupted by scar tissue / neuroma which on excision allows for anastamosis of the nerve.
IAN injury Recent surgery <4 weeks
Large neuropathic area? Poor mechanosensory function Poor function / pain? YES Surgical exploration Explore through socket
NO Reassess < 12 weeks Improvement?
Persistent large neuropathic area Poor function
Fig. 11.20 Suggested algorithm for management of IANIs.
disability, as indicated by the high proportion of lawsuits in such cases (Ventä et al 1998). More than half of lawsuits are associated with lack of preoperative informed consent for implant surgery and most were associated with premolar implants (Chaushu et al 2002). In another study 20% of patients underwent microsurgery for ablation of a neuroma, reanastomosis or neural decompression. In this study legal proceedings were initiated by 33 (20%) of the 165 patients and the patients who initiated lawsuits were younger, were more likely to have experienced anaesthesia and were more likely to have needed microsurgery (Caissie et al 2005). Interestingly, very few patients with permanent IAN injury due to orthognathic surgery or trauma present with significant complaints and this may in part be due to the clear pre-surgical consent and information along with the significant perceived benefits of the surgery.
Surgical exploration Via corticotomy
Increasingly complaints relating to nerve injury received by the General Dental Council (GDC) and American Dental Association (ADA) are implant related. Of claims made against American Dental Insurance companies (Fortress and OMSNIC) 34% of patients are unhappy with the anaesthetics and 24% are related to nerve injury. 24% of Oral surgery dental implant claims have an average payment of $89,000 per patient while 37% of the general dental implant claims had an average payment of $63,000 (Estabrooks, L personal communication 2009). Implant nerve injuries average payouts are higher than the average payout for IAN injury related to third molar surgery. Implant cases in the USA demonstrate increased claims against oral surgeons compared with general dentists; this may reflect the increased complexity of cases and the greater volume of dental surgery undertaken by oral surgeons.
520
a
Surgical Disorders of the Peripheral Nerves
b
c
Fig. 11.21 (a) Illustrates retained roots associated with severe IAN nerve injury at 3 weeks. (b) Appearance of IAN on removal of roots and debris. (c) Picture showing both roots of lower left 8 after removal which were perforated by the IAN.
a
b
Fig. 11.22 Intra-operative picture illustrating lateral corticotomy and exposure of IAN.
Iatrogenous Injuries
11.7.19 Improved Consent All patients must have realistic expectations of likely outcomes and be warned of Inferior Alveolar Nerve (IAN) or Lingual Nerve Injury (LNI) injury when appropriate. Assessment of risk must be undertaken in order to appropriately advise the patient with regard to alternative treatment plans and include this possibility in the consent forms (Nazarian et al 2003). The information should be explicit to ensure that the patient is aware that nerve injury may cause altered sensation (numbness, pain or troublesome altered sensation) that could be intermittent or constant, temporary or permanent. The patient must also be warned that the neuropathic area may affect all or part of the IAN dermatome; extra- and intra-orally (whole of skin and vermilion of lip and chin on each side and all lower quadrant teeth and associated buccal gums) or LN dermatome (whole side of tongue and lingual gums on the affected side). It may be important for clinicians to perform a neurosensory examination of nerve function prior to treatment to determine whether there is a pre-existing altered sensation as up to 24% of patients with edentulous mandibles may present with IAN neuropathy (Walton 2000). If the tooth is high risk (crossing both IAN canal LD on plane film) then the patient should be advised of increased risk of nerve injury and offered alternative surgical techniques that may minimise nerve injury.
11.7.20 Improved Management of These Injuries Commonly the patient’s anger and frustration due to the iatrogenous injury is compounded by poor immediate management by the clinician involved. After causing the injury many patients complain that the treating clinician refuses to even communicate with the patient or remains in denial about the injury. Furthermore, particularly in secondary care, the patients are reviewed for many months or even years, by consecutive junior staff providing them with unrealistic false hope and reassurance that their nerve injury will resolve. Earlier recognition and referral of trigeminal injuries is fundamental in the improvement in treating these patients.
11.7.21 Can We Prevent These Injuries? This article highlights some of the potential pitfalls in causing nerve injury and potential strategies as to how to prevent them. Prevention is better than cure.
521
Nerve lesions may result from administration of inferior alveolar nerve block, from chemical injuries related to alkalinic compounds used near nerve tissue or from bone harvesting close to the mental nerve or in the retro-molar region. The most desirable outcome after nerve injury is spontaneous return of normal sensation. The likelihood of this occurring depends on both the severity of the injury, the age of the patient and the nerve involved. The outcome of nerve injury will depend on multiple factors such as duration and mechanism of injury, appropriate case selection and treatment planning. When nerve injury occurs it is imperative that the surgeon recognise the injury immediately and advise the patient appropriately. Many injuries can be prevented (Worthington 2004; Walton 2000) through better patient selection, planning and execution of procedures. In addition patient management can often be improved by informed consent based on risk assessment and improved post operative care with early referral for nerve injuries. Empirically, many patients seen in the authors specialist clinics have significant psychological distress in association with these iatrogenous injuries particularly when the injury proves troublesome or exhibits a painful neuropathy resulting in limited daily and social function. This situation is often compounded by lack of prior informed consent and poor post operative management by the practitioner subsequent to the nerve injury. There is a need for research into the psychological effects of these iatrogenous injuries and the benefits of non surgical interventions for these patients. In summary the author hopes to have summarised several strategies to assist the practitioner in preventing trigeminal nerve injuries whilst at the same time reaffirming that there is no ‘silver bullet’ in treating these patients.
References Abarca M et al (2006) Neurosensory disturbances after immediate loading of implants in the anterior mandible: an initial questionnaire approach followed by a psychophysical assessment. Clin Oral Invest 10:269–277 Badenoch J, Whitting J (2008) The trial. In: Powers MJ, Harris NH, Barton A (eds) Clinical negligence, 4th edn. Tottel Publishing, Haywards Heath, Chap 23, pp 699–717 Banga R, Douglas SA, Mirza S, Gibbin KP (2006) Do trainees follow up their operative cases? An audit of myringoplasty follow up. Ann Roy Coll Surg Eng 88:171–173 Baron JMP (2008) Foreward. In: Powers MJ, Harris NH, Barton A (eds) Clinical negligence, 4th edn. Tottel Publishing, Haywards Heath, pp V–VI Bartling R, Freeman K, Kraut RA (1999) The incidence of altered sensation of the mental nerve after mandibular implant placement. J Oral Maxillofac Surg 57:1408–1410 Bataineh AB (2001) Sensory nerve impairment following mandibular third molar surgery. J Oral Maxillofac Surg 59(9):1012–1017, discussion 1017. Review
522 Bates AW (2006) Dr. Khan’s museum: obscene anatomy of Victorian London. J Roy Soc Med 99:618–623 Birch R (2008) Iatrogenous lesions of nerves and arteries in the leg and foot. Foot Ankle Surg 14:130–137 Birch R, Jessop J, Scott G (1991) Iatrogenic injuries of peripheral nerves. J Bone Joint Surg 73B:280–282 Birch R, Wilkinson MCP, Vijayan KP, Gschmeissner S (1992) Cement burn of the sciatic nerve. J Bone Joint Surg 74B:731–733 Black J, PRCS Eng (2009) Withdrawal of clinical pharmacology from the undergraduate curriculum. Personal Communication Blackburn CW (1990) A method of assessment in cases of lingual nerve injury. Br J Oral Maxillofac Surg 28(4):238–245 Bonney G (1990) The function of the defence societies. In: Powers MJ, Harris NH (eds) Medical negligence. Butterworths, London, pp 44–45 Bonney G (1992) Pitfalls in “day–case” operating. J Med Def Union 8:4–5 Bosch EP, Habermann TM, Tefferi A (2005) Peripheral neuropathy associated with lymphoma, leukaemia and myeloproliferative disorders. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy. Elsevier, Philadelphia, Chap 111, pp 2489–2504 Bowen J, Gregory R, Squier M, Donaghy M (1996) The post–irradiation lower motor neuron syndrome. Neuronopathy or radiculopathy? Brain 119:1429–1439 British Association of Otorhinolaryngologists – Head and Neck Surgeons Clinical Practice Advisory Group (2002) Clinical effectiveness, guidelines, acoustic neuroma (vestibular schwannoma). BAO-HNS document 5: Spring 2002 Brouwers PJAM, Kottiuk EJBL, Simon MAM, Prevo RL (2001) A cervical anterior spinal artery syndrome after diagnostic blockade of the right C6 nerve root. Pain 91:397–399 Butt AJ, McCarthy T, Kelly IP, Glynn T, McKoy G (2005) Sciatic nerve palsy secondary to haematoma in primary total hip replacement. J Bone Joint Surg 87B:1465–1467 Caissie R, Goulet J, Fortin M, Morielli D (2005) Iatrogenic paresthesia in the third division of the trigeminal nerve: 12 years of clinical experience. J Can Dental Assoc 71(3):185–190 Campbell WB, France F, Goodwin HM (2002) Medicolegal claims in vascular surgery. Ann Roy Coll Surg Eng 84:181–184 Cannon SR (2009) Quality of elective surgery in treatment centres. Editorial J Bone Joint Surg 91B:141–142 Chant DBS (2005) Emergency vascular Services in the UK. In: Barros d’Sa AAB, Chant ADB (eds) Emergency vascular and endovascular surgical practice, 2nd edn. Hodder, Arnold, London, Sect 1A, pp 3–6 Chaushu G, Taicher S, Halamish-Shani T, Givol N (2002) Medicolegal aspects of altered sensation following implant placement in the mandible. Int J Oral Maxillofac Implants 17(3):413–415 Chossegros C, Guyot L, Cheynet F, Belloni D, Blanc JL (2002) Is lingual nerve protection necessary for lower third molar germectomy? A prospective study of 300 procedures. Int J Oral Maxillofac Surg 31(6):620–624 Colin Cornelius CP, Roser M, Ehrenfeld M (1997) Microneural reconstruction after iatrogenic lesions of the lingual nerve and the inferior alveolar nerve. Critical evaluation. Mund Keifer Gesichts Chir 1:213–223 Colin W, Donoff RB (1992) Restoring sensation after trigeminal nerve injury: a review of current management. J Am Dent Ass 123: 80–84 Collins MP, Kissel JT (2005) Neuropathies with systemic vasculitis In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy. Elsevier, Philadelphia, Chap 105, pp 2335–2405 Crawford JR, Van Rensburg L, Marx C (2003) Compression of the sciatic nerve by wear debris following total hip replacement: a report of three cases. J Bone Joint Surg 85B:1178–1180 Croft PB, Wilkinson M (1969) The course and prognosis in some types of carcinomatous neuromyopathy. Brain 92:2–8
Surgical Disorders of the Peripheral Nerves Delcanho RE (1995) Neuropathic implications of prosthodontic treatment. J Prosthet Dent 73:146–152 Dellon AL (2005) Invited discussion: management strategies for iatrogenic peripheral nerve lesions. Ann Plast Surg 54:140–142 Denny–Brown D (1948) Primary sensory neuropathy with muscular changes associated with carcinoma. J Neurol Neurosurg Psychiatry 11:73–87 Dodson TB, Kaban LB (1997) Recommendations for management of trigeminal nerve defects based on a critical appraisal of the literature. J Oral Maxillofac Surg 55:1380–1386, discussion 1387 Donaghy M (2005) Lumbosacral plexus lesions. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy. Elsevier, Philadelphia, pp 1375–1390 Donaghy M (2009a) Polyneuropathy. In: Donaghy M (ed) Brains diseases of the nervous system, 12th edn. Oxford University Press, Oxford, Chap 21, pp 541–602 Donaghy M (2009b) Toxic and environmental disorders. In: Donaghy M (ed) Brains diseases of the nervous system, 12th edn. Oxford University Press, Oxford, Chap 5, pp 141–164 Donoff RB (1995) Surgical management of inferior alveolar nerve injuries (Part 1): the case for the early repair. J Oral Maxillofac Surg 53:1327–1329 Ducic I, Maloney CT, Dellon AL (2005) Reconstruction of the spinal accessory nerve with autograft or neurotube. Two case reports. J Reconstr Microsurg 21:29–33 Dyck PJ, Dyck PJB, Engelstad J (2005) Pathological alterations of nerves. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, Chap 32, pp 733–829 Dyck PJB, Engelstad J, Dyck PJ (2005) Microvasculitis. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, Chap 106, pp 2405–2414 Edwards BN, Tullos HS, Noble PC (1987) Contributory factors and aetiology of sciatic nerve palsy in total hip arthroplasty. Clin Orth Rel Res 218:136–141 Eftekhar NS (1993) Neurological complications. In Eftekhar NS (ed) Total hip arthroplasty. Mosby, Philadelphia, pp 1555–1568 Elian N et al (2005) Unexpected return of sensation following 4.5 years of paresthesia: case report. Implant Dent 14(4):364–367 Feifel H, Reidiger D, Gustorf-Aeckerle R (1994) High resolution computed tomography of the inferior alveolar and lingual nerves. Neuroradiology 36:236–238 Fleming P, Lenehan B, O’Rourke S, McHugh P, Kaar K, McCabe JP (2003) The strain of the human sciatic nerve during movement of the hip and knee. J Bone Joint Surg 85B:363–365 Fleming RE, Mickelsen CB, Stinchfield FE (1979) Sciatic paralysis. A complication of bleeding following hip surgery. J Bone Joint Surg 61A:37–39 Frafjord R, Renton T (2010) Coronectomy a review. J Oral Surg 3:1–7 Freedman GL (1992) Intentional partial odontectomy: report of a case. J Oral Maxillofac Surg 50:419–421, discussion 421 Gaffen AS, Haas DA (2009) Survey of local anesthetic use by Ontario dentists. J Can Dent Assoc 75(9):649 Gargallo-Albiol J, Buenechea-Imaz R, Gay-Escoda C (2000) Lingual nerve protection during surgical removal of lower third molars. A prospective randomised study. Int J Oral Maxillofac Surg 29:268–271 Gidwani S, Zaidi SMR, Bircher MD (2009) Medical negligence in orthopaedic surgery. J Bone Joint Surg 91B:151–156 Gomes AC, Vasconcelos BC, de Oliveira ED, da Silva LC (2005) Lingual nerve damage after mandibular third molar surgery: a randomized clinical trial. J Oral Maxillofac Surg 63:1443–1446 Greenstein G, Tarnow D (2006) The mental foramen and nerve: clinical and anatomical factors related to dental implant placement: a literature review. J Periodontol 77(12):1933–1943 Gulicher D, Gerliach KL (2001) Sensory impairment of the lingual and inferior alveolar nerves following removal of impacted mandibular third molars. Int J Oral and Maxillofacial Surg 30:306–312
Iatrogenous Injuries Haas DA, Lennon DA (1995) 21 year restrospective study of reports of paraesthetsia following local anaesthetic administration. J Can Dent Assoc 61:319–330 Harris D, Buser et al (2002) European Association for Osseointegration. E. A. O. guidelines for the use of diagnostic imaging in implant dentistry. A consensus workshop organized by the European Association for Osseointegration in Trinity College Dublin. Clin Oral Implants Res 13:566–570 Hart FD, Golding JR (1960) Rheumatoid neuropathy. Br Med J I: 1594–1600 Hegedus F, Diecidue RJ (2006) Trigeminal nerve injuries after mandibular implant placement–practical knowledge for clinicians. Int J Oral Maxillofac Implants 21(1):111–116 Heller AA, Shankland WE (2001) Alternative to inferior alveolar nerve block anaesthetic for placing mandibular dental implants posterior to the mental foramen. J Oral Implantol 27:127–133 Hill CM et al (2001) Nerve morbidity following wisdom tooth removal under local and general anaesthesia. Br J Oral Maxillofac Surg 39(6):419–422 Hillerup S (2007) Iatrogenic injury to oral branches of the trigeminal nerve: records of 449 cases. Clin Oral Investig 11(2):133–142, Epub 2006 Dec 22 Hillerup S, Jensen R (2006) Nerve injury caused by mandibular block analgesia. Int J Oral Maxillofac Surg 35(5):437–443, Epub 2005 Dec 15 Hinduja K, Samuel R, Mitchell S (2005) Problem based learning: is anatomy a casualty? Surgeon 3:84–87 House of Commons Health Committee (2006) Independent sector treatment centres: a statement for the Secretary of State for Health; cited in Cannon SR (2009) Howe GL, Poyton HG (1960) Prevention of damage to the inferior dental nerve during the extraction of mandibular third molars. Br Dent J 109:355–363 Johanson NA, Pellicci PM, Tsaris P, Salvati EA (1983) Nerve injury in total hip arthroplasty. Clin Orth Rel Res 179:214–222 Johnston AW (1960) Acroparaesthesiae and acromegaly. Br Med J 1:1616–1620 Kempshall PJ, Metcalfe A, Forster MC (2009) Review of Kinemax knee arthroplasty performed at the NHS Treatment centre, Weston– Super–Mare. J Bone Joint Surg 91B:229–233 Kennedy WF, Byrne TF, Majid HA, Pavlak LC (1989) Sciatic nerve monitoring during revision total hip arthroplasty. Clin Orth Rel Res 264:223–227 Khan R, Birch R (2001) Iatropathic nerve injuries. J Bone Joint Surg 83B:1145–1149 Khawaja N, Renton TA (2010) Case Studies illustrating the management of trigeminal neuropathic pain using topical 5% Lidocaine patches. Accepted JOS June 2010 Khawaja N, Renton TA (2010) Case studies on implant removal influencing the resolution of inferior alveolar nerve injury. Br Dent J 206(7):365–370 Kiricuta I (1963) L’emploi du grand épiploon dans la chirurgie du sein cancéreux. Presse Médicale 71:15–17 Kiyak HA, Beach BH, Worthington P, Taylor T, Bolender C, Evans J (1990 Spring) Psychological impact of osseointegrated dental implants. Int J Oral Maxillofac Implants 5(1):9–61, PMID: 2202671 [PubMed - indexed for MEDLINE] Knuttson K, Lysell L, Rohlin M (1989) Postoperative status after removal of the mandibular third molar. Swed Dent J 13:15–22 Kömürcü F, Zwolak P, Klepetko–Benditte H, Deutinger M (2005) Management strategies for peripheral iatrogenic nerve lesions. Ann Plast Surg 54:135–139 Kraut RA, Chahal O (2002) Management of patients with trigeminal nerve injuries after mandibular implant placement. J Am Dent Assoc 133(10):1351–1354 Kretschmer T, Antoniadis G, Braun V, Rath SA, Richter AP (2001) Evaluation of iatrogenic lesions in 722 surgically treated cases of peripheral nerve trauma. J Neurosurg 94:905–912
523 Laviano M (1997) Post irradiation brachial plexopathy. A study of 54 patients. Personal Communication Leigh MAMS (2006) Consent– an event or a memory? A judicial review. J Bone Joint Surg 88B:8–10 Little K, Pillai A, Fazzi U, Storey N (2007) Local anaesthetic infusion with elastomeric pump after arthroscopic subacromial decompression. Ann Roy Coll Surg Eng 89:410–413 Llewellyn JG, Tomlinson DR, Thomas PK (2005) Diabetic neuropathies. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, Chap 85, pp 1951–1992 Loescher AR, Robinson PP (1998) The effect of surgical medicaments on peripheral nerve function. Br J Oral Maxillofac Surg 36(5): 327–332 Low PA, McLeod JG, Turtle JR, Donnelly P, Wright RG (1974) Peripheral neuropathy in acromegaly. Brain 97:139–152 Malamed et al (2000) A comparison of articaine HCL and lidocaine HCL in pediatric dental patients. Ped Dent 22:307–311 Malamed SF, Gagnon S, LeBlanc D (2001) Articaine Hydrochloride: a study of a new amide local anaesthtetic. J Am Dent Assoc 132:177 Mallory TH (1983) Sciatic nerve entrapment secondary to trochanteric wiring following total hip arthroplasty. Clin Orth Rel Res 180: 198–200 Mason DA (1988) Lingual nerve damage following lower third molar surgery. Int J Oral Maxilofac Surg 17:290–294 Mauffrey C (2006) Is patient care affected by the recent changes in junior doctor’s training? Ann Roy Coll Surg Eng Suppl 88:58–60 McLeod JG (1993a) Para–neoplastic neuropathies. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 3rd edn. W B Saunders, Philadelphia, pp 1583–1590 McLeod JG (1993b) Peripheral neuropathy associated with lymphomas, leukaemias and polycythaemia vera. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 3rd edn. W B Saunders, Philadelphia, pp 1591–1598 McNeill A, Mathewson B (2006) Negligence claims for treatment of neurological disease: cases from the medical and defence union of Scotland. Scot Med J 51:24–25 Meechan JG, Blair GS (1993) The effect of two different local anaesthetic solutions on pain experience following apicectomy. Br J Dent 175:410–413 Merle M, Duprez K, Delandre D, Bour C, Pap F, Duprez L (1988) La neurolyse microchirurgicale des plexites brachiales postradiothérapiques. Chirurgie 114:421–423 Milburn JA, Driver CP, Youngson GG, King PM, MacAuley E, Krukowski ZH (2007) The accuracy of clinical data: a comparison between central and local data collection. Surgeon 5:275–278 Mondrup K, Olsen NK, Pfeiffer P, Rose C (1990) Clinical and electrodiagnostic findings in breast cancer patients with radiationinduced brachial plexus neuropathy. Acta Neurol Scand 81: 153–158 Monforte R, Estrach R, Valls-Sole J, Nicholas J, Vallilta J, UrbanoMarquez A (1995) Autonomic and peripheral neuropathies in patients with chronic alcoholism. Arch Neurol 52:45–51 Montgomery AS, Birch R, Malone A (2005) Sciatic neurostenalgia: caused by total hip arthroplasty, cured by later neurolysis. J Bone Joint Surg 87-B:410–411 Narakas AO (1984) Operative treatment for radiation induced and metastatic brachial plexopathy in 45 cases, 15 having omentoplasty. Joint Dis Orth Inst 44:354–375 Nazarian Y, Eliav E, Nahlieli O (2003) Nerve injury following implant placement: prevention, diagnosis and treatment modalities. Refuat Hapeh Vehashinayim 20(3):44–50, 101 Nemni R, Bottachi E, Fazio R, Mamoli A, Corbo M, Camerlingo M, Galardi G, Erenbourg L, Canal N (1987) Polyneuropathy in hypothyroidism: clinical electrophysiological and morphological findings in four cases. J Neurol Neurosurg Psychiatry 50: 1454–1460
524 Nercessian OA, Gonzalez EG, Stinchfield FE (1988) The use of somatosensory evoked potentials during revision or re-operation for total hip arthroplasty. Clin Ortho Rel Res 243:138–142 NICE third molar guidelines Website: www/nice.org.uk/nicemedia/ pdf/wisdomteethguidance.pdf Norgren L (2005) Speciality related iatrogenic vascular injuries. In: Barros D’Sa AAB, Chant ADB (eds) Emergency vascular and endovascular surgical practice, 2nd edn. Hodder, Arnold, London, pp 487–492 O’Duffy JD, Randall RV, MacCarty CS (1973) Median neuropathy (carpal tunnel syndrome) in acromegaly. Ann Intern Med 78: 379–383 Oertel R, Rahn R, Kirch W (1997) Clinical pharmokinetics of Articaine. Clin Pharmokinet 33:417 Oleksak M, Edge AJ (1992) Compression of the sciatic nerve by methyl methacrylate cement after total hip replacement. J Bone Joint Surg 74B:729–730 Oliviera PC, Volpato MC, Ramaccatio JC, Ranalli J (2001) Articaine and Lignocaine efficiency in infiltration anaesthesia: a pilot study. Br Dent J 197:45–46 O’Regan B, Bharadwaj G, Bhopal S, Cook V (2007) Facial nerve morbidity after retrograde nerve dissection in parotid surgery for benign disease: a 10-year prospective observational study of 136 cases. Br J Oral Maxill Surg 45:101–107 O’Riordan B (1997) Uneasy lies the head that wears the crown. Abstract Brit Ass Oral Maxillofac Surg 35(3):209, June O’Riordan BC (2004) Coronectomy (intentional partial odontectomy of lower third molars). Oral Surg Oral Med Oral Pathol Oral Radiol Endod 98(3):274–280 Oussedik S, Haddad F (2009) Further doubts over the performance of treatment centres in providing elective orthopaedic surgery. J Bone Joint Surg 91:1125–1126 Pallis CM, Scott JJ (1965) Peripheral neuropathy in rheumatoid arthritis. Br Med J 1:1141–1147 Pandey VA, Kerle MI, Jenkins MP, Wolfe JHN (2007) AAA benchmarking by Dr Foster: a cause for concern. Ann Roy Coll Surg 89:384–388 Parsonage MT, Turner JWA (1948) Neuralgic amyotrophy. Lancet I:973–978 Pedlar J (2003) Prolonged paraesthesia. Br Dent J 194:18 Pichl JW, Beirne OR (2001) Lingual flap retraction and prevention of lingual nerve damage associated with third molar surgery: a systematic review of the literature. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 91(4):395–401, Review Pogrel MA (2002) The results of microneurosurgery of the inferior alveolar and lingual nerve. J Oral Maxillofac Surg 60(5):485–489 Pogrel MA, Lee JS, Muff DF (2004) Coronectomy: a technique to protect the inferior alveolar nerve. J Oral Maxillofac Surg 62(12):1447–1452 Pogrel MA, Maghen A (2001) The use of autogenous vein grafts for inferior alveolar and lingual nerve reconstruction. J Oral Maxillofac Surg 59:985–988; discussion 988–993 Pogrel MA, Schmidt BL, Sambajon V, Jordan RC (2003) Lingual nerve damage due to inferior alveolar nerve blocks: possible explanation. J Am Dent Assoc 134:195–199 Pogrel MA, Thamby S (2000) Permanent nerve involvement resulting from inferior alveolar nerve blocks. J Am Dent Assoc 131:901–907 Pollard JD (2005) Neuropathy in diseases of the thyroid and pituitary glands. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, Chap 88, pp 2039–2050 Pollock A (2007) The case against private provision in the NHS. Ann Roy Coll Surg Eng 89:338–339 Porter SS, Black DL, Reckling FW, Mason J (1989) Intraoperative cortical somatosensory evoked potentials for detection of sciatic neuropathy during total hip arthroplasty. J Clin Anaesth 1(3):170–176
Surgical Disorders of the Peripheral Nerves Powers MJ, Harris NH, Barton A (2008) Clinical negligence, 4th edn. Tottel Publishing, Haywards Heath Pryde FR, Black SM (2006) Scottish Anatomy Departments: adapting to change. Scot Med J 51:16–20 Renton T, Hankins M, Sproate C, McGurk M (2005) A randomized controlled clinical trial to compare the incidence of injury to the inferior alveolar nerve as a result of coronectomy and removal of mandibular third molars. Br J Oral Maxillofac Surg 43(1):7–12 Renton T, McGurk M (2001) Valuation of factors predictive of lingual nerve injury in third molar surgery. Br J Oral Maxillofac Surg 39(6):423–428 Renton T, Thexton A, Crean SJ, Hankins M (2006) Simplifying the assessment of the recovery from surgical injury to the lingual nerve. Br Dent J 200(10):569–573, May 27, discussion 565 Renton TA, Yilmaz Z (2010) A review of 253 post traumatic trigeminal nerve injuries. J Orofac Pain (in press) Renton T, Yilmaz Z, Avedesu D, Meechan J (2010) Trigeminal nerve injuries in relation to Local Anaesthesia. Brit Dent J (in press) Riberio PD Jr, Sanches MG, Okamoto T (2003) Comparative analysis of tissue reactions to anaesthetic solutions using histological analysis in subcutaneous tissues of rats. Anaesth Prog 50:169–180 Robinson PP, Loescher AR, Smith KG (1999) The effect of surgical technique on lingual nerve damage during lower 3rd molar removal by dental students. Eur J Dent Educ 3(2):52–55 Robinson PP, Loescher AR, Yates JM, Smith KG (2004) Current management of damage to the inferior alveolar and lingual nerves as a result of removal of third molars. Br J Oral Maxillofac Surg 42(4):285–292, Review Robinson PP, Smith KG (1996) Lingual nerve damage during lower third molar removal: a comparison of two surgical methods. Br Dent J 180(12):456–461, Jun 22, Review Rodger NAM (2004) The command of the ocean. A naval history of Britain, 1649–1815 Allen Lane. An imprint of Penguin Books in association with the National Maritime Museum, p 527 Rood JP (1978) The organisation of inferior alveolar nerve and its relation to local analgesia. J Dent 6:305–310 Rood JP, Nooraldeen Shehab BA (1990) The radiological prediction of inferior alveolar nerve injury during third molar surgery. Brit J Oral and Maxillofacial Surg 28:20–25 Rubenstein JE, Taylor TD (1997) Apical nerve transection resulting from implant placement: a 10-year follow-up report. J Prosthet Dent 78:537–541 Rud J (1983a) Third molar surgery perforation of the inferior dental nerve through the root and lingual bone. Tandlaegebladet 87:659–667 Rud J (1983b) Third molar surgery: relationship of root to mandibular canal and injuries to inferior dental nerve. Tandlaegebladet J 87:619–631 Ruggiero SL (2001) Surgical management of lingual nerve injuries. In: Microneurosurgery of the trigeminal nerve. Atlas of Oral Maxillofac Surg Clins Nth Am 9:13–21 Rutner TW, Ziccardi VB, Janal MN (2005 Aug) Long-term outcome assessment for lingual nerve microsurgery. J Oral Maxillofac Surg 63(8):1145–1149 Sakai T, Sugano N, Fujii M, Nishii T, Ohzono K, Yoshikawa H (2002) Sciatic nerve palsy after cementless total hip arthroplasty. Treatment by modular neck and calcar shortening. J Ortho Sci 7: 400–402 Satcher RL, Noss RS, Yingling CD, Ressler J, Ries M (2003) The use of motor–evoked potentials to monitor sciatic nerve status during revision total hip arthroplasty. J Arthroplast 18:329–332 Schmalzried TP, Amstutz HC, Dorey FJ (1991) Nerve palsy associated with total hip replacement. J Bone Joint Surg 73A:1074–1080 Schmalzried TP, Noordin S, Amstutz HC (1997) Update on nerve palsy associated with total hip replacement. Clin Orth Rel Res 344: 188–206
Iatrogenous Injuries Septodont website:http//www.septodont.co.uk/Articaineuk/prescribing/ spc1hundred.html Shepherd J (2006) Lingual nerve retraction increases the risk of temporary lingual nerve damage during mandibular third molar surgery. Evid Based Dent 7:47 Sibanda N, Copley LP, Lewsey JD et al. (2008) Revision rates after primary hip and knee replacement in England between 2003 and 2006. PLoS Med 5:179 Silbey MB, Callaghan JJ (1991) Sciatic nerve palsy after total hip arthroplasty: treatment by modular neck shortening. Orthopedics 14:351–352 Siliski JM, Scott RD (1985) Obturator nerve palsy resulting from intrapelvic extrusion of cement during total hip replacement. Report of four cases. J Bone Joint Surg 67A:1225–1228 Simon MA, Vree TB, Gielen MJ, Booij MH (1998) Comparison of the effects and disposition kinetics of articaine and lidocaine in 20 patients undergoing intravenous regional anaesthesia during day case surgery. Pharm World Sci 20:88 Singh P (2008) UK Medical Defence Organisations. In: Powers MJ, Harris NH, Barton A (eds) Clinical negligence, 4th edn. Tottel Publishing, Haywards Heath, Chap 7, pp 173–200 Smith MH, Lung KE (2006) Nerve injuries after dental injection: a review of the literature. J Can Dent Assoc 72(6):559–564. Review Somayaji HS, Saifuddin A, Casey ATH, Briggs TWR (2005) Spinal cord infarction following therapeutic computed tomography-guided left L2 nerve root injection. Spine 30:E106–E108 Spies J, McLeod JG (2005) Paraneoplastic neuropathy In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy. Elsevier, Philadelphia, Chap 110, pp 2471–2488 Spillane JD (1943) Localised neuritis of the shoulder girdle. Lancet 2:532–535 Standring S (2009) A new focus on anatomy for surgical trainees. J ANZ Surg 79:114–117 Strauss ER, Ziccardi VB, Janal MN (2006) Outcome assessment of inferior alveolar nerve microsurgery: a retrospective review. J Oral Maxillofac Surg 64(12):1767–1770 Suarez GA (2005) Immune brachial plexus neuropathy In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy. Elsevier, Philadelphia, Chap 102, pp 2299–2308 Susarla SM et al (2005) A comparison of patient satisfaction and objective assessment of neurosensory function after trigeminal nerve repair. J Oral Maxillofac Surg 63(8):1138–1144 Susarla SM, Kaban LB, Donoff RB, Dodson TB (2007) Functional sensory recovery after trigeminal nerve repair. J Oral Maxillofac Surg 65(1):60–65 Tantanapornkul W et al (2007) A comparative study of cone-beam computed tomography and conventional panoramic radiography in assessing the topographic relationship between the mandibular canal and impacted third molars. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 103(2):253–259 Thomas PK, Lascelles RG (1966) The pathology of diabetic neuropathy. Q J Med 35:489–502 Thompson PD, Thomas PK (2005) Clinical patterns of peripheral neuropathy. In: Dyck PJ, Thomas PK (ed) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, Chap 45, pp 1137–1161 Uhlschmid G, Clodius L (1978) Eine Neue Anwendung des frei Transplantation Omentums. Der Chirurg 49:714–718 Valmaseda-Castellón E, Berini-Aytés L, Gay-Escoda C (2000) Lingual nerve damage after third lower molar surgical extraction. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 90(5):567–573
525 van Eden SP, Patel MF (2002) Prolonged paraesthesia following inferior alveolar nerve block using articaine. Brit J Oral Maxillofac Surg 40:519–520 Ventä I, Lindqvist C, Ylipaavalniemi P (1998) Malpractice claims for permanent nerve injuries related to third molar removals. Acta Odontol Scand 56(4):193–196 Von Arx DP, Simpson MT (1989) The effect of dexamethasone on neurapraxia following third molar surgery. Br J Oral Maxillofac Surg 27(6):477–480 von Arx T, Häfliger J, Chappuis V (2005) Neurosensory disturbances following bone harvesting in the symphysis: a prospective clinical study. Clin Oral Implants Res 16(4):432–439 Walker S, Mead J (2008) The role of the National Health Service Litigation Authority. In: Powers MJ, Harris NH, Barton A (eds) Clinical negligence, 4th edn. Tottel Publishing, Haywards Heath, Chap 9, pp 215–227 Walton JN (2000) Altered sensation associated with implants in the anterior mandible: a prospective study. J Prosthet Dent 83(4): 443–449 Weber ER, Daube JR, Coventry MB (1976) Peripheral neuropathies associated with total hip arthroplasty. J Bone Joint Surg 58A: 66–69 White SP, John A, Jones SA (2009) Short term results of total hip replacement performed by visiting surgeons at an NHS treatment centre. J Bone Joint Surg 91B:1154–1157 Wilbourn AJ (1993) Brachial plexus disorders. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 3rd edn. W B Saunders, Philadelphia, pp 911–950 Wilbourn AJ (2005) Brachial plexus lesions. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, Chap 55, pp 1339–1374 Wise MP, Blunt S, Lane RJM (1995) Neurological presentations of hypothyroidism: the importance of slow-relaxing reflexes. Proc Roy Soc Med 88:272–274 Wismeijer D, van Mass M, Vermeeren J, Kalk W (1997) Patient’s perception of sensory disturbances of the mental nerve before and after implant surgery: a prospective study of 110 patients. Br J Oral Maxillofac Surg 35:254–259 Woodall J (1617) The surgions mate. Lisle, London, p 62 Woolf H (1996) Access to justice: final report to the Lord Chancellor on the civil justice system in England and Wales. HMSO, London Worthington P (2004) Injury to the inferior alveolar nerve during implant placement: a formula for protection of the patient and clinician. Int J Oral Maxillofac Implants 19(5):731–734 Wroblewski BM (2006) Letter to editor. J Bone Joint Surg 88B:839 Wynn RL, Bergman SA, Meiller TF (2003) Paraesthesia associated with local anaesthetics:a perspective on Articaine. Gen Dent 51: 498–501 Ziccardi VB, Steinberg MJ (2007) Timing of trigeminal nerve microsurgery: a review of the literature. J Oral Maxillofac Surg 65(7): 1341–1345, Review Ziccardi VB, Zuniga JR (2007) Nerve injuries after third molar removal. Oral Maxillofac Surg Clin North Am 19(1):105–115, vii. Review Zuniga JR (2001) Surgical management of trigeminal neuropathic pain. In: Microneurosurgery of the trigeminal nerve. Atlas of Oral Maxillofac Surg Clins Nth Am 9:59–74 Zuniga JR, LaBlanc JP (1993) Advances in microsurgical nerve repair. J Oral Maxillofac Surg 51:62–68 Zuniga JR, Zenn MR (2001) Principles of microsurgery. Oral Maxillofac Surg Clins Nth Am 9:331–342
12
Pain
Neuropathic pain. Definitions. Pain mechanisms: The Gate Theory. Events in the wound; in the proximal axon and dorsal root ganglion; in the spinal cord; following preganglionic injury. The Rôle of the Sympathetic system. Introduction of com plex regional pain syndromes in classification. Clinical classification: complex regional pain syndrome Type I (reflex sympa thetic dystrophy). Diagnosis. Treatment: principles; drugs and other measures short of operation; indications for operation. Characteristics of, and response to treatment in: causalgia; neurostenalgia; post traumatic neuralgia. Pain syndromes after traction injuries of the brachial plexus. Natural history; non operative treatment; response to reinnervation of the limb; spinal cord stimulation; operation on the spinal cord.
Pain is the chief reason why patients attend doctors. To them it signifies harm or damage; to doctors it is a symptom of a pathological process which can be treated only after diagnosis has been made. The International Association for the Study of Pain (Merskey 1986) defined it as “an unpleasant sensory and emotional experience, associated with actual or potential tissue damage, or described in terms of such damage.” This definition recognises pain as a subjective experience, accepting that it often occurs in the absence of any tissue damage (or at least none that doctors can detect). It distinguishes activity in specific cutaneous receptors and their central projections from the experience of pain: a distinction clearly made by Sherrington in 1903. Wall (1985) considered pain as a state akin to hunger or thirst in which action is imperative: behaviour must be changed to prevent further damage (and this includes psychological damage). There are sensory and affective dimensions to pain and these can be disassociated. Rainville and colleagues (1999) modulated the experience of pain by hypnosis in volunteers who were subjected to heat pain stimulus. Unpleasantness (affect) was reduced by suggestion, whereas intensity of the sensation much less so. In one experiment the heart rate was monitored, this correlated with unpleasantness, but not with intensity. Ploner and colleagues (1999) described a 57 year old man who had suffered a right sided post central stroke with infarction of the SI and SII cortices. He lost “sensory-discriminative” pain whereas “motivational – affective” pain was preserved and Ploner et al. commented: “we were able to demonstrate, for the first time in humans, the representation of the sensory- discriminative pain component and first pain sensation in the lateral pain system. In contrast, pain affect and the ability to detect painful stimuli do not, in principle, require integrity of these
structures.” Price et al. (2003) restate earlier concepts of exteroceptive and interoceptive pain systems. The exteroceptive system enables the protective behaviour of escape and avoidance whereas the interoceptive system leads to homeostatic behaviour, quiescence with guarding of injured body regions and autonomic responses promoting recuperation and healing. There is an extensive discussion of the pathways involved. Schott (2004) reflects further on these concepts. The difference between these two pain pathways will be apparent to anyone who has experienced, on the one hand, the pain of fracture or dislocation in a limb and on the other, cardiac pain or pain from abdominal viscera. Severe damage may not be painful – at least initially. Uxbridge, at Waterloo, saw rather than felt his injury: “By God sir, I have lost my leg”; the Duke of Wellington’s response is an early example of a therapeutic approach of acceptance “By God sir, so you have.” Uxbridge’s only comment during the operation was that the surgeon’s knife was not very sharp and he asked someone to look at the mangled leg after it had been amputated to reassure him that it could not have been saved. During that same battle Lord Fitzroy Somerset was wounded by a musket ball which smashed his right elbow. He walked back to a cottage used as a field hospital to have his arm cut off between the shoulder and the elbow. He did not even murmur, saying only to the surgeon who tossed the arm away: “Hey, bring my arm back. There is a ring my wife gave me on the finger” (Hibbert 1998). Hibbert also records an early example of rehabilitation. Wellington temporarily replaced Somerset with another one armed officer as an assurance to him that his disability would not prevent his returning to his position when he was recovered. Perhaps the most elevating theme of Kirkup’s excellent book
R. Birch, Surgical Disorders of the Peripheral Nerves, DOI: 10.1007/978-1-84882-108-8_12, © Springer-Verlag London Limited 2011
527
528
on the history of amputation (Kirkup 2007) is the strength of the human spirit for he describes so many cases where limbs were removed without any form of anaesthetic, during which operations the patients exhibited extraordinary calm and patience. Beecher (1946) reported that 70% of casualties from the Anzio landings experienced no pain while Dav and colleagues (1995) observed that soldiers severely wounded in battle showed a higher threshold and tolerance for thermal pain than those lightly injured during training. Nathan (1977) described how the memory of pain from injury incurred years before was revived by manipulation of an amputation stump and he commented that “it is a general and essential feature of the behaviour of the central nervous system to preserve the effects of stimulation after stimulation has ceased.” Of itself, pain is not necessarily harmful; rather, it is essential for survival. In individuals with congenital insensitivity to pain caused by a group of hereditary sensory neuropathies, it is often found that small fibres in the dorsal root are absent or reduced in number, there are no small neurones in the DRG and the dorso-lateral fasciculus (Lissauer’s tract) is absent (Nagasako et al. 2003, Klein and Dyck 2005). The life of those affected can be short if trophic changes are not prevented and secondary infections left untreated. The progressive mutilation of the hands and the feet are reminiscent of leprosy, in which the micro bacteria preferentially colonise Schwann cells associated with small myelinated and non myelinated fibres. The relatively painless mutilation of joints described by Charcot in syphilis is seen in some patients with chronic arthritic disease and diabetic neuropathy. We are concerned with neuropathic pain, caused by injuries to peripheral nerves and by preganglionic injuries of the brachial and lumbo sacral plexuses. Some neuropathic pain states are exceptionally severe. Many are characterised by: persistence; intractability; disproportion between the extent of lesion and the severity of pain; and by the fact that it is no longer necessary for the survival of the organism. It is associated with sensory or motor disturbance and often with sympathetic dysfunction. The positive sensory symptoms described by patients are usually so clear that diagnosis of the cause and some understanding of the subsequent patho-physiological events is possible from the history alone. The recognition of neuropathic pain after inadvertent injury to a nerve during operation should be straightforward. It is associated with disturbance of sensation and power, and is expressed in the distribution of the nerve, remote from the site of operation. The recognition of neuropathic pain after a fracture or dislocation is more difficult. The injury often induces numbness or diffuse pins and needles through the injured limb for the first hour or two, or, the pain from the skeletal injury may mask the underlying nerve pain. On the other hand neuropathic pain is sometimes so severe that the patient is scarcely aware of the skeletal injury. Spontaneous sensory symptoms which are usually painful, perceived in the limb, the hand or the foot are significant and should not be overlooked. Such
Surgical Disorders of the Peripheral Nerves
spontaneous symptoms indicate that an injury to a nerve has occurred and that the cause may still be active. There should be no doubt about the severity of the situation should the pain worsen and become associated with deepening of the nerve defect. In this situation the surgeon must recognise the likelihood of critical ischaemia of the nerve or the limb. There is a vogue for measuring pain by visual analogue scales, a system which too often gives a spurious sense of objectivity. We have been interested to hear patients describe their pain following accidental damage to, say, the sciatic nerve at the hip as immeasurably greater than the arthritic pain that drove them to operation. Their records often show that the arthritic pain was accorded a VAS of the maximum of 9 or 10. Taggart (1998) elaborated on the 10 point scale to include disturbance of work or study, of social life and of sleep and this has been simplified into the Peripheral Nerve Injury scoring system which is as follows. 1. M ild or none. The patient is aware of pain but leads a normal life. VAS 0–3. 2. Moderate – able to work, but pain sometimes so severe the patient has to take time off VAS 4–6. 3. Significant - able to sleep but cannot work or study not enjoy hobbies. VAS 7–8. 4. Severe - continuous disturbance of daily life, of work, of study and of sleep. Visual analogue scale (0–10) 9–10. Kato et al. (2006) found a strong concordance between the VAS linear scale and the PNI scale. We have also used the McGill questionnaire (Htut et al. 2006). It is wise to take all reasonable steps to establish diagnosis and to treat accordingly before consigning the patient with neuropathic pain to a Pain Clinic or dismissing him or her as a psychological misfit. In all cases of chronic pain more or less resistant to treatment, the clinician is likely to attach to this patient the label “psychological elaboration” or “low pain threshold,” sometimes in order to disguise ignorance or to salve his or her own feelings of inadequacy. It is worth recalling that without the higher cerebral functions pain would hardly be so common an event in our lives as it is. The greatest of our countrymen put the matter succinctly 400 years ago: “there’s nothing either good or bad but thinking makes it so.” We should now consider some of the terms in general use.
12.1 Nocicipient Sherrington (1903) recognised that the skin was provided with a set of nerve endings which responded to harmful events. They may properly be termed algesic if they project to the sensorium, but in preparations such as the spinal dog when a thermal stimulation of the foot pad provokes a flexor withdrawal response, no psychical mechanism is possible. The term nocicipient is preferable. Nociceptors are those
Pain
neural structures which detect the existence of a noxious event: nociceptive pathways or tracts inform the mind – brain of the event which may there be perceived as pain. We offer the following definitions of the terms. 1. P araesthesiae – spontaneous abnormal sensations (from para (para): “near close against almost”. 2. Dysaesthesiae – unpleasant, spontaneous abnormal sensations (from duV:(dys) bad, unpleasant. See Medea’s complaint h rhu pod h dnothniz ei on elpidaz (Alas, I the unhappy one, once hoped…). 3. Hyperaesthesia - increased sensitivity to a stimulus which is not normally painful. 4. Hyperalgesia – increased perception of a stimulus which is normally painful. 5. Allodynia – the perception of a stimulus which is not normally painful as a painful event. 6. Hyperpathia - a state of exaggerated, prolonged and very painful perception of stimulation. It is important to distinguish between the spontaneous symptoms of paraesthesiae and dysaesthesiae, which arise from injured axons without external stimulation from evoked symptoms such as allodynia, which signify that fast conducting mechanosensor fibres are conveying impulses which are being interpreted as pain. The spread of spontaneous and evoked sensory symptoms indicates that there is sensitisation of other neurones in the dorsal horn. There is much that is unsatisfactory about these terms which were precise enough to their originators but are all too often misapplied. Both Gowers (1886) and Déjerine (1901) considered that hyperalgesia and hyperaesthesia were the same thing. “Therefore hyperaesthesia does not result from augmentation of tactile facilities but from the tendency towards rapid transformation of tactile sensations into painful sensations and towards an exaggeration of painful sensibility: it is synonymous of hyperalgesia” (Déjerine 1901). There are obvious examples in everyday practice of these different types of sensory disturbance. Patients with entrapment or other irritative lesions of peripheral nerves volunteer sensations of cold water or (for the select) cold champagne trickling down underneath the skin. These are paraesthesiae. Sometimes these spontaneous sensations have an unpleasant quality, they are described as if there are ants crawling under the skin. These are dysaesthesiae. Patients recovering from suture of peripheral nerves experience cold intolerance which is at times crippling. Cold water feels as a block of ice. This is hyperalgesia. The exquisite sensitivity of sunburned skin to warm stimuli is another such example. Lewis (1936) described this “a profuse and widespread area of hyperalgesia appears in many normal subjects at the point of faradic stimulation or tiny crush of the skin. This hyperalgesia lasts for several or many hours and may be accompanied by a little smarting of the skin.” The patient who cannot tolerate light touch on the afflicted skin is describing allodynia. More severe injuries to
529
proximal nerves brings on a state of constant racking pain, often described as burning; the part cannot be examined and moved only with difficulty. This is hyperpathia. Allodynia is one of the most important of clinical signs in medicine and it must be sought for and interpreted with precision. It is evoked by applying stimuli which are normally not painful but which the patient interprets as pain and it can be found only when there is some innervation remaining in the skin. Allodynia signifies that mechano receptor and other fibres have begun to signal pain because of events at their terminals, in their parent cell bodies, and in the second order neurones of the dorsal horn. Although it may seem reasonable to use the term for the overreaction so often seen in the earlier stages of regeneration after nerve repair we restrict it to pain induced by gentle stimulation of the skin after injury to a nerve. Allodynia may be dynamic, when it is elicited by moving touch, by a draught or breeze, or by contact with a sheet or clothing or it may be static when it is elicited by pressure. There is also warm or cool allodynia, when a normally non painful warm or cool stimulus is perceived as pain. There may be a paradoxical interpretation of a cool stimulus as one which is painfully hot or vice versa. The extension of allodynia beyond the distribution of the injured nerve is common. Hyperpathia is an unsatisfactory term because it is so often misapplied. It was introduced in 1927 by Foerster to signify the cutaneous hyperalgesia and spontaneous pain after nerve injury. It was re-defined by Noordenbos (1959) as the unpleasant response to a noxious or non noxious stimulus that it usually delayed, overshoots the normal and has a prolonged after reaction. The description by Nathan (1983) is particularly good: “there is hyperpathia in which all cutaneous stimuli cause the same sensation, penetrating fiery pain which spreads from the point stimulated; … the pain and hyperpathia are not confined to the site where they originated nor to the territory of the peripheral nerve where they started.” Sweet (1984) wrote: “the subcommittee on Taxonomy of the International Association for the Study of Pain ... defines hyperpathia more sharply as a “painful syndrome, characterised by delay, over reaction and after sensation to a stimulus, especially repetitive stimulus, all as originally described by Foerster” We reserve the term hyperpathia for that deep seated, burning and poorly localised pain extending beyond the distribution of the injured nerve, which is evoked by the palpation of the muscles of the limb and which is so common in ischaemia. This interpretation recognises hyperpathia as a phenomenon equivalent to cutaneous allodynia but one which is conveyed through the deep afferent pathways. Deafferentation pain is a term appropriately used when the injury has interrupted the pathway between cell bodies in the dorsal root ganglion and those in the dorsal horn. After all, any lesion severe enough to inflict interruption upon the axons must lead to deafferentation and to use the term in these situations renders it virtually meaningless. Post herpetic neuralgia is partly related to the death of the cell bodies
530
in the dorsal root ganglion, it is associated with reduction of the density of nerve endings in the skin (Oaklander 2001), and it is one example of true deafferentation pain.
12.1.1 The Gate Theory In 1965 Melzack and Wall (Melzack and Wall 1965)proposed a theory of pain mechanisms which provoked a great deal of critical interest. Wall restated the theory in 1978 embracing criticisms about the role of the substantia gelatinosa, the extent of pre-synaptic inhibition on nociceptor afferents in the dorsal horn, and the known fact of stimulus specificity of nerve fibres. It has been summarised (Wall 1978). 1. “ information about the presence of injury is transmitted to the central nervous system by peripheral nerves. Certain small diameter fibres (A-delta and C) respond only to injury while others with lower thresholds increase their discharge frequency if the stimulus reaches noxious levels. 2. Cells in the spinal cord or the fifth nerve nucleus which are excited by these injury signals are also facilitated or inhibited by other nerve fibres which carry information about innocuous events. 3. Descending control systems originating in the brain modulate the excitability of the cells which transmit information about injury. Therefore the brain receives messages about injury by way of a gate control system which is influenced by: (1) injury signals; (2) other types of afferent impulse and (3) descending controls.” It is remarkable how many of the assumptions implicit in the gate theory have been proven since it was first proposed. Central descending inhibitory pathways exist, pre-synaptic inhibition of afferent fibres within the dorsal horn has been shown. Disinhibition of neurones within the dorsal horn following interruption of peripheral afferents is established (Wall and Devor 1981). Their facilitation by increased input of nociceptors either from their terminals in sensitive skin or from regenerating sprouts within the neuroma has been demonstrated. Some success has followed the application of the principles underlying the gate theory in the treatment of patients. Two examples are particularly impressive. First is the successful application of transcutaneous nerve stimulation or implanted electrodes in the treatment of pain following some peripheral nerve lesions. Next is the dramatic improvement in pain that we have seen in many patients with severe traction lesions of the brachial plexus coincident with reinnervation of muscle and the deep afferent pathways. In most of these cases relief of pain preceded any significant cutaneous re-innervation.
Surgical Disorders of the Peripheral Nerves
It must be admitted that many aspects of neuropathic pain cannot be understood within the framework of the gate theory. The instantaneous onset of pain in some cases of causalgia or of traction lesion is one such difficulty. Indeed the pain of avulsion injury is more consistent with smashing the gate than lifting it. However, there is one central tenet which has been a source of constant encouragement for those treating patients with severe neuropathic pain. It is that the initial cause of an injury to a peripheral nerve and the continuing action of the responsible agent should be put right; and that reasonable attempts must be made to restore afferent impulses to the spinal cord.
12.1.2 Events After Nerve Injury A tide of changes sweeps centripetally after wounding of a nerve which can be described in the wound itself, in the proximal axon and dorsal root ganglion, in the dorsal horn and within the central nervous system. Wall (1985) put it thus: “with a latency of milliseconds combinations of afferent signals and of descending controls operate a rapid and powerful gate control. With a latency of minutes, impulses in C-fibres change the excitability of peripheral endings and of spinal cord circuits. With a latency of days, chemical transport in C-fibres from areas of damage further modifies cord connectivity with a disappearance of inhibition and an expansion of receptive fields. With a latency of weeks and months, anatomical degeneration produces secondary changes in deafferented cells with atrophy, sprouting, and abnormal firing patterns.” These events lead to a number of changes in the behaviour of the nervous system which underly the phenomena displayed by patients with neuropathic pain, some of which are so bizarre that they may lead the clinician to the conclusion that the patient is not of sound mind. This conclusion is almost always incorrect. 1. T here is spontaneous firing in nociceptor neurones, and abnormally increased firing in response to noxious stimulation. 2. There is sensitisation of the axons, their cell bodies in the dorsal root ganglion and in the dorsal horn of the spinal cord so that pain and sensory disturbance is increasingly experienced in new areas outside the distribution of the injured nerve. 3. There is sensitisation of mechano-sensor neurones or there is a change in their behaviour so that nonpainful stimuli are perceived as pain. 4. Afferent and efferent fibres in the sympathetic system, which are conveyed with the nerve trunks, are involved in the injury, affected by it, and may to some extent provoke or maintain symptoms and signs arising from abnormalities at the level of lesion or more centrally.
Pain
It is important to remember that the changes induced in the neurones of the nervous system by a peripheral lesion may be reversed by correction of that lesion. Refinements in methods of neurophysiological investigation have contributed much to the understanding of changes in the membrane of neurones in neuropathic pain. More understanding has been provided by the application of methods which enable the study of molecular changes. Much of this has been carried out on small mammals. The work that has been done in humans is likely to prove more rewarding in the long term because the findings can be interpreted against the symptoms expressed by the patient, and against the findings of single fibre function studies, quantitative sensory testing and skin biopsies. Some of these methods have been described in Chap. 5. Some of the studies carried out by Anand and his colleagues have already been described in Chaps. 2, 3 and 4. The Proximal Axon and Dorsal Root Ganglion Many chemical agents are released by damaged tissue, blood vessels and by cut nerve fibres. Michaelis et al. (1998) showed that fibres within the sural nerve of the rat which had been cut responded to the application of an “inflammatory soup” of inflammatory mediators by lowering the threshold to mechanical stimulation in most C-fibres and in about one half of the myelinated fibres. A few fibres, normally non responsive to touch, became sensitised. Wall and Gutnick (1974a and b) identified three major changes in neuromas made in the rat sciatic nerve. First, the nerve fibres became spontaneously active, whereas in the normal state they are silent. This is the basis of spontaneous sensory symptoms, of paraesthesiae and dysaesthesiae. Next, the fibres become sensitive to mechanical stimulation, an explanation for Tinel’s sign. The spontaneous activity and the mechanical sensitivity were suppressed by stimulation of the intact nerve proximally, by antidromic stimulation mimicking the action of transcutaneous nerve stimulation. This inhibition lasted for minutes or hours. Finally, the nerve sprouts become sensitive to noradrenalin. Coward et al. (2000) studied the sodium channels SNS/PN3 and NaN/SNS3, which are preferentially expressed in nociceptors in injured human neurones. SNS/PN3 accumulated at the site of nerve injury and also in nerve fibres in skin from patients with mechanical allodynia and hyperalgesia. Specific potassium channels, which normally inhibit conduction, decreased in the dorsal root ganglion and nerve. This may contribute to dysaesthesiae and hypersensitivity (Boettger et al. 2002). Spontaneous Impulse Generation was revealed by Michaelis et al. (2000) in a series of experiments in the rat in which the nerve to the gastrocnemius and soleus and the sural nerve were cut. After an interval of 6 days irregular, bursting, spontaneous discharges were recorded from the cell bodies of afferent fibres passing through the motor nerve, with a conduction velocity ranging from 5 to 30 m/s. Some, at least, of these were nociceptors. In contrast, the cell bodies of fibres
531
within the sural nerve remained quiet. These workers comment: “these findings provide the first evidence that after peripheral nerve injury both axotomised as well as intact afferent neurones supplying skeletal muscle but not skin afferents generate ongoing activity within the DRG.” This may explain the deep, poorly localised and cramping pain experienced by many patients with pain following nerve injury. Ephapses One more change, a structural one, is the development of new synapses between sprouting nerve fibres, ephapses, which permit the stimulation of one new fibre by an adjacent one (Seltzer and Devor 1979). These were described in the rat; they are rare in monkeys, and of doubtful relevance in man. However the concept of functional coupling may provide an explanation for the severity of pain so often seen after partial injuries of nerves. The ragged wound caused by a bullet or a spike of bone or a coarse suture or tip of a needle, will divide some axons while leaving others intact. There is focal demyelination of some of those nerve fibres which have not been transected. The threshold of the intact axons is diminished, they become sensitised by local chemical agents, noradrenalin and by the spontaneous activity in regenerating sprouts springing from the divided axons. Peter Misra (2008) offers the interesting analogy of L’Hermittés sign in early multiple sclerosis, in which painful spasms are induced by ephapses between nerve fibres in areas of demyelination. Changes in the Dorsal Horn There are profound changes within the dorsal horn after injury to a peripheral nerve. Their nature is similar to those which occur peripherally, chiefly sensitisation, but to this is added loss of inhibition both from peripheral afferent and central descending pathways. There is disturbance and diminution of the actions of endogenous enkephalin and opiates which activate post synaptic potassium ion channels and inhibit presynaptic calcium ion channels in the membranes of the DRG cell bodies (Lawson 2005). Woolf (1995) describes three mechanisms which bring about central sensitisation. First, the dorsal horn cells respond to an increase in C-fibre activity. There is a maintained state of central sensitisation from ectopic activity in the neuroma and dorsal root ganglion. Next, there is decreased inhibition because of loss of activity in A-beta fibres whose input is diminished (Devor 1991). Third, there are new synapses within the spinal cord. Woolf et al. (1992) described the highly ordered segregation of primary afferents in the dorsal horn of the spinal cord of rats. Low threshold mechanosensors terminate in laminae III and IV whereas high threshold nociceptors terminate in laminae I, II and V. After ligation of the sural nerve, low threshold A-beta afferents were shown to sprout into lamina II. These workers comment: “if the low threshold mechanoreceptive afferent terminals that sprout into lamina II establish functional contacts with cells that normally would have a monosynaptic C-nociceptor input, this could lead to inappropriate responses to innocuous peripheral stimulae.” McLachlan and colleagues
532
(1993) studied changes in the dorsal root ganglion of rats after axonopathy. New sympathetic fibres formed arborisations or baskets around large cells. This arborisation was reduced by blocking the action of nerve growth factor (Ramer et al. 1999), who suggest that this is more than an epiphenomenon but that it represents a functional coupling between somatic afferent and sympathetic fibres. The rapid and extensive changes in the receptive fields of some neurones within the dorsal horn were studied by Wall et al. (1999). A number of mechanisms which increase the size of the receptive field were identified: (1) Temporal summation by peripheral stimulation induced a very large increase in the receptive field of relevant cells within the dorsal horn. (2) Deafferentation, either by cutting the dorsal root or by cutting the peripheral nerve was followed by enlargement of the receptive field within days, several days before sprouting of the afferent terminal arborisations. (3) Repetitive stimulation of unmyelinated afferents led to prolonged hyperexcitability of neurones in the dorsal horn, a phenomenon described as “wind up” (Mendell 1966). Magerl et al. (1998) studied the effects of intradermal injection of capsaicin, an agent which first stimulates and then blocks C-fibre activity, in humans. They suggested that both “windup” and hyperalgesia reflect central sensitisation. The hyperalgesia induced by pinprick is an example of “windup,” brought about by homosynaptic facilitation within the nociceptor pathway. The perception of light touch as pain suggests heterosynaptic facilitation, with the involvement of afferent pathways which do not normally signal pain. Mendell (1999) raises the possibility of the involvement of “inactive synapses” as one contribution towards central sensitisation. Petersen-Zeitz and Basbaum (1999) point out that the n methyl D Asparate (NMDA) receptors, which are synthesised in the cell body and transported to the central terminals of the primary afferents, are activated by pain but this activation is insufficient to sustain the long term changes following persistent pain. A second messenger pathway involving calcium ion channels becomes involved. Cell bodies in the laminae III and IV of the dorsal horn receive a diverse input and have complex receptor fields including nociceptor afferents, low threshold afferents and some cells that respond to both skin and visceral afferents. Roberts (1986) characterised these as wide dynamic range (WDR) or multireceptive neurones. Some cells of laminaV also receive impulses from both A-beta and A-delta and some C-fibre afferents, and they respond to mechanical or thermal stimulae, increasing their discharge frequency as the stimulus intensifies (Price and Mayer 1974). These changes provide an explanation not only for the development of allodynia, but also the extension of allodynia into adjacent skin. Lamotte and colleagues (1992) recorded and stimulated single A-beta fibres in human volunteers. The receptor field, which was usually a small area of skin, was defined by stimulation. Stimulation of the fibres evoked the usual innocuous response. Capsaicin was injected. There followed a period
Surgical Disorders of the Peripheral Nerves
when normally painless stimuli were perceived as painful ones. Wasner et al. (1999) also studied the effect of capsaicin and suggested that the dynamic mechanical A-beta fibre mediated allodynia is not mediated by central presynaptic interaction between A-beta mechanosensors and nociceptive C-afferents, but by “sensitisation of second order multi receptive (wide dynamic range) neurones in the dorsal horn, as supported by ample experimental evidence.” The underlying neurophysiological change is the summation of slow synaptic potentials from small diameter axons which depolarise neurones in the dorsal horn. There is “windup” of action potential discharge in these spinal neurones but there is also a prolonged increase in the excitability of their membranes. Evidently urgent steps should be taken to reverse the cause of these peripheral and central states of excitation and disinhibition, but some changes are irreversible. Chief amongst these is the death of neurones following axonotomy (Fig. 12.1).
Fig. 12.1 Central sensitisation in post traumatic neuralgia. Injury to a branch of the lateral cutaneous nerve of forearm in a 34 year old man. He developed severe mechanical allodynia and allodynia initially in the distribution of the nerve and then over the course of the next 6 months into the territories of the median and radial nerves.
Pain
The Role of Nerve Growth Factor (NGF) The first case of absence of NGF was described in 1991 (Anand and colleagues 1991). The patient presented with severe postural hypotension and investigation confirmed autonomic failure. The sensory neuropeptides NGF and substance P were undetectable. There was no sensory axon reflex; Capsaicin injection was not painful. The threshold to heat pain was elevated but the patient did feel pin prick and sharp pain. Other studies in human neuropathies show a possible role for NGF in the maintenance of chronic neuropathic pain. The concentration of NGF is low in the skin of leprosy (Anand and colleagues 1994). It is increased in sunburnt skin which is notoriously hyperalgesic. NGF is diminished in diabetic neuropathy, which is usually painless (Anand and colleagues 1996) NGF is still produced by the target organs in skin denervated by nerve injury, unlike the situation in diabetic neuropathy and it may be that there is a relative excess of the neurotrophin which is encountered by fewer than normal, regenerating, nerve fibres. It is possible that the hyperalgesia and exquisite capsaicin sensitivity of the skin of patients in the early stages of recovery after nerve suture is explicable on this basis (Anand and colleagues 1997). The concept of moving a painful neuroma away from skin and subcutaneous tissues, which are rich in nerve growth factor, to areas which are poor in the neurotrophin, such as bone and muscle, is explored by Atherton and his colleagues (2006).
12.1.3 Central Pain: The Preganglionic Injury of the Brachial Plexus The onset, distribution, and qualities of this pain are set out in Tables 9.4, 9.5 and 9.6. Typically it has two components. First there is the constant pain, usually felt in the insensate hand, which is crushing or boring, bursting or burning. Superimposed on this are lightning like shooting pains, which course down the whole of the limb. These last a few seconds, they may occur up to 30 or 40 times in every hour, and are exceptionally severe. It is usually the case that patients will report that the distribution of these pains is within the territory of avulsed nerves. A shooting pain is not experienced in those parts of the limb that have not been denervated. Two thirds of conscious patients experienced constant pain and did so from the day of injury while more than one half of conscious patients experiencing the shooting pain did so on the day of injury while in others it became apparent at intervals ranging from 14 h to more than 4 weeks after injury. Any explanation of the mechanisms underlying this pain must acknowledge the early onset. It has been said that the clinician cannot rely on patients’ memory, a view which can be safely discarded because of the urgency with which these patients have been seen in our Unit. Of the 198 patients described in Chap. 9, 98 were referred within 7 days of
533
injury, 27 more were referred during the second week after injury. However, Taggart (1998) showed, in a prospective study of 85 patients, that pain reached a maximum intensity at a mean of 3 months after injury (Kato et al. 2006). Bonney (1973) thought of the pain as arising in the spinal cord as a result of damage and later gliosis in the substantia gelatinosa, and this has been confirmed in the lumbar spinal cord of the cat after dorsal rhizotomy or avulsion (OvelmenLevitt et al. 1984). Consistent pathological changes were noted after avulsion, and included damage to the medial portion of Lissauer’s tract, and gliosis within the substantia gelatinosa. Extra cellular microelectrical recordings from neurones within the deep laminae of the dorsal horn showed differences between rhizotomy and avulsion models. The receptive fields extended more widely after rhizotomy. Spontaneous activity from recorded neurones was seen in both preparations, but the pattern is different, with more regular firing being seen in the chronically avulsed cells. In a subsequent paper, Ovelmen-Levitt (1988) found that continuous high frequency activity was characteristic of avulsion. The neurophysiological sequelae of deafferentation were described by Loeser and Ward (1967) and by Anderson et al. (1977) who showed that experimental deafferentation produced spontaneous firing in laminae I and V of the posterior horn- the sites of relay of the nociceptor fibres. This spontaneous firing began a few days after the lesion, and gradually increased to reach a maximum at about 1 month. Two varieties of activity were observed: a continuous barrage and superimposed paroxysmal burst of high frequency firings. Microelectrode recordings made from the deafferented cords of human subjects show paroxysmal discharges that coincided with the recurrence of painful flexor spasms. AlbeFessard and Lombard (1983) showed experimentally that spontaneous firing found in the posterior horn after experimental section of the roots was gradually transmitted more rostrally. Some months after injury impulses could be detected in the thalamus. If transcutaneous nerve stimulation was applied before and immediately after the lesion was inflicted, spontaneous firing could be reduced considerably. Rinaldi et al. (1991) recorded electrical activity from single cells of the thalamus in ten patients with chronic pain associated with deafferentation. One of their patients suffered a traction injury to the brachial plexus; lesions of peripheral nerves accounted for seven others, in two cases pain followed cerebral infarctional spinal cord injury. They confirmed spontaneous activity in neurones in the nucleus ventralis postero lateralis (VPL), but also showed, for the first time in humans, hyperactivity within the medial thalamus. They commented: “the findings of a dual representation of neuronal hyperactivity in the lateral and medial thalamus is perhaps not surprising in view of the fact that these are the two primary thalamic targets for the pain related relays from the spinal cord, carried by the neospino-thalamic and paleospino-thalamic systems.” Banati and colleagues (2001) and Banati (2002)
534
detected activation of microglia in the thalamus by chronic pain and proposed that this may relate to synaptic change or to inter relation between glia and neurones.
12.1.4 The Sympathetic Nervous System and Pain The sympathetic nervous system has been implicated in the onset and maintenance of many neuropathic pain states notably causalgia which is discussed later in this chapter. Nathan (1983) concluded that:“are we then to conclude that in any human case in which there is pain associated with extreme sensitivity of the tissues and in which sympathetic blocks remove the pain and abnormal sensitivity – that in such cases we have outgrowing small nerve fibres sensitive to sympathetic transmitters? I think the answer to this question must be no. One reason is that in many of these cases there is no peripheral nerve lesion. If the sympathetic algodystrophy starts with a Colles fracture or a thrombophlebitis, there is no peripheral nerve lesion. A further reason is that one sees the same state with lesions of the central nervous system.” Loh and Natahan (1978) proposed that these pain states which appeared to be related to sympathetic activity were brought about by abnormal firing of peripheral nerve fibres – in particular the mechano-sensors - which alter the behaviour of neurones in the substantia gelatinosa and lamina 5 either by diminishing their inhibition or facilitating their activity or by both. “Thus, stopping the emission of the normal sympathetic transmitter stops the firing of peripheral nerve fibres and that stops the abnormal function within the central nervous system. The abnormal state is kept going by mechano-sensors and the large afferent fibres. The sympathetic outflow is acting on these fibres or receptors to cause the input to the spinal cord which causes a state of disinhibition or facilitation spreading from the original site of input.” Until quite recently the concept of sympathetically main tained versus sympathetically independent pain (Roberts 1986) was used extensively. The idea was closely scrutinised by Schott in a number of important papers. Indeed, he (Schott 1988) advised that interrupting the sympathetic outflow was “a futile procedure for many patients.” He (1994) pointed out that the vasomotor and sweating changes are often seen in painless injuries and they may persist after relief of pain. An effective sympathetic block, one which causes vaso and sudomotor paralysis may not relieve pain and sympathetic blocks may in fact worsen pain. “I suggest, as did Kuntz and Farnsworth (1931)that the effect may be due to blocking transmission in afferent fibres rather than sympathetic efferents.” In considering the central terminations of the visceral afferents (Sugiura et al. 1989)(see chapter 2), Schott (1994) comments: “these diffuse and extensive central terminations,
Surgical Disorders of the Peripheral Nerves
and interactions possible with other systems, particularly at spinal level, would also account for the various autonomic and motor phenomena that can occur in the affected area. Whilst tremor seen in reflex sympathetic dystrophy may respond to sympathetic blockade, dystonia and more complex movement disorders unfortunately rarely improve.” Sympathectomy is by no means a benign procedure, and sympathectomy for sweating can induce pain and allodynia at the border zone which is sometimes associated with pronounced increase in sweating in that area. Anand (1985) refuted the suggestion that this might be due to a delayed rise in sensory neuropeptides in somatic nerves. Thomas and Ochoa (1993) pointed out that the disorders labelled as causalgia-reflex sympathetic dystrophy might be no more than a group of symptoms provoked by a wide range of causes each demanding accurate diagnosis and treatments. “This classification apparently carries pathophysiological connotation in that it endorses or rejects a role for the sympathetic system in maintaining the pain. The term has been understandably welcomed by anesthesiologists who are handed the privilege and responsibility of issuing a pathophysiological diagnosis that may condemn neurological patients to multiple diagnostic therapeutic blocks or to sympathectomy. While sympathetic events might well influence the abnormal sensory physiology in some of these patients, it is not scientifically rigorous to assume that the sympathetic nervous system has a pathogenic role in determining pain, on the weaknesses of the subjective response of a patient to a ritualistic medical intervention.” To this severe caution is added the suggestion of Verdugo and Ochoa (1994) that pain relief in the absence of successful sympathetic block is in fact a placebo effect. Bonney (1973) suggested this some years ago adding that, unfortunately, relief after successful block might also be a placebo effect! Perhaps inevitably, dissatisfaction with the terms reflex sympathetic dystrophy and sympathetically maintained and sympathetically independent pain has led to the classification proposed by Stanton Hicks and colleagues (1995) and adopted by the International Association for the Study of Pain. These writers state that: “SMP can be associated with a number of diseases and a positive response to a sympathetic block should neither be a factor in the nomenclature of the disease nor define the disease.” New terms are proposed to replace the older ones of causalgia and reflex sympathetic dystrophy. They are:
12.2 Complex Regional Pain Syndrome (CRPS) Type 1 and (CRPS) Type 2 The diagnostic criteria which must be satisfied include (1) an initiating noxious event; (2) spontaneous pain or allodynia, not necessarily limited to the territory of a single peripheral
Pain
nerve and disproportionate to the inciting event; (3) evidence at some time of vasomotor or sudomotor abnormality in the region of the pain.
12.2.1 CRPS Type 2 (Causalgia) The diagnostic criteria include: (1) the syndrome develops after a nerve injury; (2) Spontaneous pain or allodynia occurs not necessarily limited to the territory of the injured nerve; (3) there has at some time been a vaso or sudomotor abnormality. The presence of other pathological states which would account for the pain and loss of function excludes the diagnosis. This classification casts a very wide net. It is characteristic of the pain caused by injury to a nerve that it is disproportionate, that it extends beyond the distribution of the nerve and that there is vaso and sudomotor disturbance. Although suggestions have been made to refine the IASP classification (Bruel et al. 1999) it must be said that any classification which relies on symptoms and signs which evolve and change without considering the underlying cause is unsound. Spontaneous resolution of some of the symptoms and signs of causalgia is not rare, leaving the patient with the residual symptoms and signs of the partial injury to the main nerve. The neurostenalgia of the sciatic nerve subjected to the action of expanding haematoma in the buttock after arthroplasty of the hip is characterised by severe burning or bursting pain and by intense allodynia throughout much of the lower limb. There is, for the first hours at least, increased sweating and obvious abnormality of vasomotor tone. If the cause is not treated the lesion of the nerve becomes a degenerative one, allodynia diminishes and vaso and sudomotor instability is replaced by sympathetic paralysis. An increase in the area of allodynia, usually accompanied by obvious vaso and sudomotor instability is common after injuries, especially incomplete ones, of main nerves and nerves of cutaneous sensation. These are not examples of complex regional pain syndrome Type I or type II, they are examples of post traumatic neuralgia. We have seen many hundreds of patients in whom an earlier diagnosis of CRPS Type I or Type II led to lengthy, fruitless, and at times, hazardous treatment by drugs and by various forms of nerve block (Fig.12.2). We think that the concept of CRPS Type 1 is flawed by the assumption that there is no injury to a nerve. Oaklander and her colleagues (2006) used QST and the measurement of epidermal nerve fibre density in skin biopsies in a case of CRPS Type I and found a 29% reduction in the density of the nerve terminals in the skin where there was mechanical allodynia and heat pain, hyperalgesia. They suggested that CRPS Type I is, in fact, specifically associated with focal distal nerve injury and degeneration of small diameter axons. Albrecht and colleagues (2006) examined tissues from the limbs
535
amputated for complex regional pain syndrome in two patients and they observed that “in these CRPS Type I patients the myriad of clinical symptoms observed had detectable neuropathologic correlates.” The pathological changes demonstrated included numerous abnormal fine calibre neurofilament positive axons, myelin basic protein negative axons innervating hair follicles, a measurable decrease in the innervation of the sweat glands and vessels of the skin, an inappropriate expression of neuropeptide Y in axons passing to the superficial arterioles and sweat glands, a loss of vascular endothelial integrity and an extraordinary vascular hypertrophy. These workers responded to an editorial in the publishing journal by writing: “we found extensive peripheral neuropathologies involving virtually all components of the cutaneous innervation, which included not just small fibres but also A-beta fibre innervation such as that to the hair follicles.” We have confirmed these observations in two patients who came to below knee amputation having been treated, for a number of years, as examples of CRPS
Fig. 12.2 Extreme central sensitisation in post traumatic neuralgia. Intraoperative incomplete section of the medial plantar nerve in a 51 year old woman. She was treated for many months by drugs and blocks for CRPS 1. By 3 years she could not walk because of intense allodynia in the leg and the foot, where there was profuse sweating and discolouration.
536
Fig. 12.3 The posture of the foot in a case of CRPS Type I. There was dystonia.
Type I. Tissues from the amputated limbs were studied by Professor Anand and some of his findings are illustrated in Figs. 4.26a–d, 4.27a–d. It is important to emphasise that these two patients came to amputation only after all other reasonable methods of treatment available had been exhausted and because the foot and ankle had become so severely deformed that the patients could not function at all and pleaded their case very effectively. The outcome for the first patient is illustrated in Fig. 13.70. The second patient’s course was very much more difficult. Case Report A 26 year old woman developed back pain radiating to her left foot and ankle, and sprained that ankle while marching with a weight on her back. She developed severe pain in her leg and foot and this was followed by progressive equino varus deformity (Fig. 12.3). There was dystonia. She made a good recovery after left below knee amputation and, by 3 months was able to walk for up to 2 miles with her prosthesis. She was highly motivated and set herself a rigorous set of targets. She wrenched her left knee 5 months after operation, her pain returned, and she developed flexion contracture of the knee which required tenotomy. She became wheel chair bound again and extremely depressed. It was at the initiative of Dr Lacoux, Dr Swingler and Mr Eljamel (Dundee) that a spinal cord stimulator was implanted which greatly improved her situation. Evidently operation must be undertaken with considerable apprehension and performed with very great care in any patient with true complex regional pain syndrome, that is to say in a patient who does not fall into the main groups of causalgia, neurostenalgia, and post traumatic neuralgia. Indeed, ill advised operations are a common cause of CRPS Type I. Case Report A 48 year old woman fell at work onto her left elbow. There was extensive bruising around the medial aspect of the joint and she developed symptoms consistent with a diagnosis of a lesion of the ulnar nerve. However nerve conduction studies were normal. The ulnar nerve was decompressed at the elbow and at the same operation the median nerve was decompressed at the wrist. Her pain was
Surgical Disorders of the Peripheral Nerves
made very much worse and proved resistant to repeated sympathetic blocks and also to prolonged infusion of local anaesthetic into the posterior triangle of the neck. Further neurophysiological investigations and findings from quantitative sensory testing (QST) were normal. Woolf and his colleagues (1998) put forward the important concept of classifying pain on the basis of underlying mechanisms. It is based on four questions: (1) what contribution is coming from the sensitisation of primary afferents? (2) What is the contribution from ectopic discharge of primary afferent neurones? (3) What is the extent of the involvement of the sympathetic nerve fibres? (4) What is the extent of sensitisation or disinhibition within the central nervous system? This proposal has the advantage of challenging the clinician to think carefully about each individual patient. However it must be admitted that different mechanisms vary in their extent and they evolve during the time in which any individual patient experiences neuropathic pain. This is true of the classification which we have used for more than 20 years and which we have found useful. 1. C ausalgia. 2. Pain due to persistent compression, distortion or ischaemia. We have called this pain neurostenalgia, from õTevos (stenos) a strait as in stenosis added to veurou (neurone) a nerve or tendon. It is agreeable that the verb otevzN means to moan or groan, so that there is a dual reference to the idea of pain. This group undoubtedly contains many examples of Seddon’s “irritative” lesions. 3. Post traumatic neuralgia, which can be divided between that following major injury to a major nerve trunk and that following injury to a nerve of cutaneous sensation. 4. Central pain. This too can be divided between that following rupture of spinal nerves at the level of the transitional zone leading to pure de-afferentation and that following avulsion of roots of a limb plexus from the spinal cord itself, a lesion which is truly an injury of the central nervous system. 5. Complex regional pain syndrome Type 1 (reflex sympathetic dystrophy). 6. Pain maintained either deliberately or subconsciously by the patient in response to a challenge such as the wish to obtain compensation, or resentment against a public body such as an insurance company or other provocation.
12.2.2 The Painful, Stiff, Swollen Part After Fracture or Soft Tissue Injury (CRPS Type I, Reflex Sympathetic Dystrophy) This strange disorder has, at other times, been called sympathetic algodystrophy, shoulder hand syndrome, Sudek’s atrophy amongst other terms. There is usually an injury in which
Pain
no major nerve trunk is damaged. This is followed by spontaneous pain, which spreads, and there is allodynia. There is vasomotor and sudomotor disturbance: excess sweating; edema; discolouration of skin and, later, changes in the nails, the hair with stiffness of the joints. The most common presentation to orthopaedic surgeons is after fractures or other injuries to the hand or foot but the syndrome also presents after injuries or operations at the shoulder, the elbow and at the knee. The initial stage is characteristic. The pain is diffuse and disturbs sleep. The hand is swollen and stiff. It is often wet and warm; it may be cold and blue. If untreated, fibrosis follows and the hand ends as a useless stiff appendage. We have seen cases where the pain and allodynia resolved spontaneously and we were faced with a very difficult problem of restoring function. Bonica (1990) described the progress of severe cases in three stages. First, the acute stage, is characterised by pain, edema and warmth. The next stage, the dystrophic, by cold skin with trophic changes. Lastly there is the atrophic stage when the part is wasted and the joints fixed in deformity. We have seen patients coming to us in the last stage and it is usual for them to describe a progression through the others. The atrophic stage represents the unrecognised or untreated case. Trophic changes are interesting and significant but indicate that earlier and more easily treated stages have passed. Nails become brittle and deformed, there is increased hair growth and the skin becomes atrophied and glossy. It is often dry and scaly. It is not unusual to see thickening of the palmar or plantar fascia reminiscent of Dupuytren’s contracture. The term complex regional pain syndrome is preferable to reflex sympathetic dystrophy which implies an unproven reflex, a sympathetic dependence which may be no more than an epiphenomenon and a dystrophy which is seen only in late cases. But it remains deeply unsatisfactory because of the suggestion that there is no nerve injury. In fact, there almost always is, either from compression or ischaemia of the trunk nerves such as the median or tibial from fracture haematoma, or direct injury to those nerves or to their terminal branches by displaced bone fragments, foreign bodies or wounds. The great majority of cases in orthopaedic practice can be prevented by early treatment and this calls for alertness in detection of the early stages of the syndrome. It is an error to label every painful, stiff and swollen hand presenting in a fracture clinic as RSD or CRPS type 1. More often these are examples of inadequate primary treatment. Some publications suggest Colles fracture is complicated by RSD in as many as 30% of cases. One might reasonably question the method of treatment used in these studies and contrast that with the rarity of RSD as a complication of Colles fractures treated by the regime proposed by John Crawford Adams which was followed for a number of years in the fracture clinics at St Mary’s Hospital. Adams emphasised that the fracture of the wrist must be well reduced and that it should be immobilised in a splint or slab, not a complete plaster, for the first week. The patient attended daily for the first 7 days
537
after injury for rebandaging of the splint. This ensured that the splint was still holding the reduction as the swelling diminished. At every attendance the patient joined a class and was taken through an exercise regime to maintain full range of movement of the shoulder, the elbow and fingers and thumb with some vigour. In those patients with increasing pain or swelling more vigorous steps were taken which entailed individual treatment or even admission for a few days when the limb was elevated and exercised with adequate analgesia. The key elements of Adams’ policy include insistence on accurate reduction; holding that reduction; and, above all, early functional use which prevented swelling. Zyluk and his colleagues provide much sound advice in three valuable papers. Zyluk (2001) studied 94 patients with reflex sympathetic dystrophy 1 year after completion of treatment finding that just under one third (29%) were free of symptoms. More than half of the patients continued to experience pain which was worsened by changes in the weather, indeed there was cold intolerance in 44% and there were persisting sensory changes in one third. The second paper (Zyluk 2005) reported a prospective study of patients with fractures of the distal radius and found no evidence that psychological factors played a part in the development of CRPS Type I. Zyluk and Puchalski (2008) set out a method of treatment for the established condition. The two cardinal features in diagnosis were pain and stiffness. The treatment involved admission for 1 week for a period of elevation and active movement with the object of restoring full finger flexion by the end of the admission, and the use of mannitol and dexamethasone, which scavenge inflammatory toxic oxygen and hydroxyl free radicals. Sympatholytic blocks were sometimes used. Zyluk and Puchalski comment: “this need for Hand and Orthopaedic surgeons and the therapy departments to be looking for incipient CRPS Type I and also to start treatment without delay cannot ... be emphasised too greatly.” Furthermore, delay in treatment is avoided: “if treatment can be carried out in the Hand or in the Orthopaedic Department and not referred elsewhere.” Perhaps this motto should be placed on the walls of all fracture and hand clinics. Hord and Oaklander (2003) offer a balanced review of different methods of treatment for the established condition. The use of physical methods and of TNS is supported and these writers emphasise that there must be inevitable consequences of chronic pain. Early treatment is more effective than late treatment and cautions are offered about the significant side effects of such regularly used agents as tricyclic antidepressants. We are forced to the conclusion that too many cases of this disorder are a reflection of lax medical care. In most early cases a cause will be found by those who look. The splint or bandage which is too tight; a limb improperly supported in a sling; a patient who is too frightened to keep the part elevated and in function. The surgeon should be alert for an underlying lesion of compression or irritation of a trunk nerve either from swelling from a displaced bone fragment. The mainstay of treatment is use of the part.
538
12.3 Diagnosis The history is essential and the patient must be given ample time to tell their story. In late cases there may be diffidence about expressing symptoms which seem so bizarre and which may already have been dismissed by others. It may prove necessary to put leading questions which indicate that the clinician is indeed listening to the patient, believes what they are saying and has an understanding of what is being said. Certain features are particularly important. (1) Onset. The immediate onset of pain after a wound or on awakening from an operation implies that the lesion has already been inflicted whereas delayed onset suggests a later event such as haematoma. (2) Distribution. The patient is asked where the pain started and where it went to. Did the pain spread beyond an earlier well defined area and if so, over what period of time? (3) Qualities. Was the pain there all the time, was it episodic, was it constant or intermittent? Many different terms are used, burning, bursting, crushing, compressing, “the hand in a vice,” “the bones of the foot are coming out of my skin,” “a hot needle or a hot file rasping on the skin” are common descriptions. Was the pain on the surface or was it deep? Episodic or convulsive pain is often described as lightning like, electrical, shooting or lancinating. (4) Aggravating and relieving factors. Many patients give a clear description of the phenomenon of allodynia. One which is commonly related is the increasing intolerance of a sheet on the leg and the foot after compression of the sciatic nerve by haematoma in the thigh or buttock. The effect of changes in temperature or the weather or of associated illness are important and frequently described features. (5) The effect of the pain upon life, upon work or study, on social activities and on sleep provide an insight into the severity of the pain. Examination must be done with gentleness, and in some patients no more than inspection is possible. Important features include trophic changes, vaso and sudomotor abnormality, the posture of the part, and the presence of spontaneous movements. After this allodynia, in its various forms, must be sought before deep palpation of the muscle compartments. Tinel’s sign is sought. There are three characteristics of pain caused by injury to a nerve which are extremely important for the clinician. These are dysaesthesiae, allodynia, and Tinel’s sign. With the evidence provided so far the clinician ought to be able to arrive at an accurate diagnosis about what nerve has been injured, where it was injured, and have a view about the cause of that injury and of the underlying mechanisms. The analysis of neuropathic pain in the absence of injury is more difficult. The clinician should never lose sight of the fact that pain is a symptom of an underlying disorder which may be a generalised neuropathy, a systemic disease, or indeed a focal lesion of the nerve from entrapment or tumour. It seems necessary to restate that although some clinicians may prefer to see themselves as specialists in one particular
Surgical Disorders of the Peripheral Nerves
field or in one particular joint patients do not come with a label around their neck indicating “hip,” or “knee” or other disorder. Dr Robert Spinner, of the Mayo Clinic (Spinner et al. 2007) has kindly provided details about a patient whose story is by no means unusual. A previously fit man experienced pain in his left ankle and foot which was not related to any injury and which increased in intensity over the course of some months. An operation was done to decompress the tibial nerve. This did not help his pain. He then went through a prolonged course of treatment involving drugs and blocks during which time his pain steadily worsened. He was then referred to Dr Spinner who made a diagnosis of schwannoma of the tibial nerve at his first meeting with this patient, whose pain was abolished by successful operation (Fig. 12.4). Willner and Low (1993) set out some principles governing the treatment of neuropathic pain. They include: (1) Removal of the cause. (2) Promotion of healing or regeneration. (3) Correction of the microenvironment of the nerve. (4) Restoration of afferent pathways. (5) Modulation of central inhibitory pathways. (6) Reduction of sympathetic over activity, and (7) changing pain thresholds by modification of emotional or behavioural components of pain interpretation. These are excellent principles. Neuropathic pain following a focal injury of a nerve will require operation in 1, 2, 3, 4, and 6 whereas drugs and other measures short of operation, have a part to play in the last three.
12.3.1 Drugs and Other Measures Short of Operation Drugs are disappointingly ineffective in reducing the severity of neuropathic pain. About 60% of patients do not experience even moderate improvement (Rowbotham 2005, Anand and Birch 2009, O’Connor and Dworkin 2009). Rowbotham (2005) outlines some mechanisms of action. Drugs may work by (1) reducing transmitter release, (2) by inhibiting post synaptic excitatory receptors, (3)by potentiating inhibitory transmitters, and (4) by the blockade of specific sodium and other ion channels. O’Connor and Dworkin (2009) provide a consensus of evidence gathered and analyzed under the auspices of the Neuropathic Pain Special Interest Group of the International Association for the study of Pain. They say: “compared with patients with non neuropathic chronic pain, patients with neuropathic pain seem to have higher average pains scores and lower health related quality of life (even after adjusting the pain scores); to require more medications; and to report less pain relief with treatment.” They also point out that: “patients with neuropathic pain are usually not prescribed medications with demonstrated efficacy for their condition and when they do receive appropriate treatment they receive dosages that are,
Pain
539
Fig. 12.4 Schwannoma at the ankle. Above left: the swelling and scar from previous operative incision. Above right: the MR scan. Below: the tumour exposed. By courtesy of Dr Robert Spinner, Mayo Clinic.
on average, far below the dosages with demonstrated efficacy in randomised controlled trials.” Recommended first choice drugs include certain antidepressants, both tricyclic and dual reuptake inhibitors of serotonin and nor epinephrine, calcium channel alpha2- delta ligands, such as gabapentin and pregabalin, and topical lidocaine; opioids and tramadol were recommended as second line treatments that might be considered for first line use in selected clinical circumstances.
Transcutaneous Nerve Stimulation The basis of this is to increase input to the posterior columns, and thereby to promote inhibitory impulses inhibiting rostral
transmission of impulses from the dorsal horn. It will not do simply to give the patient a stimulator and tell him or her to get on with it. Various positions of the stimulator and variations of frequencies, duration and amplitude of pulse are tried. Patients are encouraged to wear the stimulator for many hours a day and for at least 2 weeks. The effect of use on daily activity, drug intake and subjective feelings are all recorded in a “pain diary.” There seems to be no relation between the duration of symptoms and the effectiveness of the stimulator (Fig.12.5). An extreme case is that of the patient who had been in pain for some 50 years, whose pain was almost completely relieved by 2 days of treatment. There is no relation between the type of lesion and effectiveness. The particular advantage of transcutaneous stimulation is, of course, that it
540
Fig. 12.5 Transcutaneous nerve stimulation. An 8 year old boy sustained a complete preganglionic injury of the brachial plexus and he experienced terrible pain. This was abolished by transcutaneous nerve stimulation. His pain returned at the age of 18 years.
is non invasive, simple, has no side effects and is considerably cheaper than long term treatment with drugs. Sindou and Keravel (1980) commented that nerve stimulation cannot work if the fibres of the dorsal columns are wholly degenerate; the technique can only work in post ganglionic lesions or in cases where some of the fibres are preserved. Miss P. Barsby has provided the following very useful information about the use of transcutaneous nerve stimulation in neuropathic pain. She, with her colleagues, in the physiotherapy department of the Royal National Orthopaedic Hospital studied 281 in patient episodes from January 1995 to January 1998. High intensity, low frequency stimulation was found to be preferable, in “bursts” or continuously for “stretching” pain; low intensity, high frequency stimulation, again in “bursts” or continuously, was preferable for “neuralgic” pain. Transcutaneous nerve stimulation was not useful in the treatment of hyperaesthesia. About 80% of patients gained relief, but the results deteriorated after 3 to 6 months in more than half of the patients. Acupuncture has been used over the last 14 years; it has been helpful in a few cases. Continuous Peripheral Nerve Blocks The prevention of pain after operation is an often neglected duty of the surgeon. There are severe risks associated with percutaneous introduction of catheters or needles for the purpose of interscalene block, spinal nerve block, or main nerve block (see Chap. 3). It is safer to place the catheter adjacent to the nerve to be blocked under direct vision (Fig 12.6). However, skillful practitioners using ultrasonography have effectively prevented or reduced the pain in war wounds by continuous peripheral nerve blocks (CPNB). Buckenmaier and his colleagues have a very great experience with the method Stojadinovic et al. (2006) described 361 continuous
Surgical Disorders of the Peripheral Nerves
Fig. 12.6 Insertion of a catheter for prolonged supraclavicular brachial plexus block.
peripheral nerve blocks with a mean catheter infusion time of 9 days in 287 battle casualties who required multiple operations. Thirty five per cent of these patients had amputation. Complications relating to the catheter occurred in 11.9% of patients and these were technical or minor in nature. The infection rate related to catheter use was 1.9%. An important decrease in the pain score over 7 days was recorded. In a subsequent paper Buckenmaier et al. (2009) studied 110 wounded soldiers treated between July 2007 and February 2008 finding an improvement in pain in those treated by CPNB catheters over those who were not so treated which was statistically highly significant. No persisting causalgia has been seen in the Nerve Injury Clinic at Headley Court from more than 120 battle casualties with nerve injuries. Neuropathic pain is a problem for some but it is chiefly the syndrome of neurostenalgia from strangulation by scar tissue. In these cases secondary operations, which usually involved replacement of scar by full thickness skin, are necessary. The approach to the control of pain which is used for the civilians and the soldiers alike injured in Afghanistan has been summarised by Aldington and his colleagues in a series of papers. Connor et al. (2009) describe the efficacy of CPNB catheters, which often obviate the need for opiates. Edwards et al. (2009) describe the approach adopted at the Selly Oak Hospital (Birmingham) and outline some of the misconceptions so commonly encountered in the treatment of patients with neuropathic pain. These include: “pain is expected after all surgery and illness; patients requesting more analgesia are weak or seeking morphine; opiates are dangerous and cause addiction; analgesia will mask symptoms and delay diagnosis, and if he has gone for a fag it cannot be that bad’’. The use of opioids during rehabilitation after war injuries is discussed by Jagdish and Aldington (2009)
Pain
who point out that opioids continue to occupy a central role in the pain management of trauma patients from the battle field and that there is certainly a place for their carefully controlled use in chronic non cancer pain. It is emphasised that the dose should not be escalated in the absence of any improvement in pain or function and that there should be a ceiling dose. Linson et al. (1983) used continuous sympathetic block for the treatment of patients with causalgia or “reflex sympathetic dystrophy,” having been impressed by the results of continuous marcaine blocks of the prevertebral sympathetic chain in 100 cases by Betcher et al. (1953). Carr and Todd (1991) introduced a catheter percutaneously by the anterior approach and successfully relieved pain in 22 of 29 patients diagnosed as suffering from reflex sympathetic dystrophy following fractures or tunnel release, crush injury and other causes. We have used CPNB for over 20 years in patients in whom it was difficult to disentangle the contribution from the focal injury of that nerve from more general consequences. In some patients these blocks enabled successful rehabilitation without further intervention. It has proved helpful to many patients who have become understandably very wary about the idea of operations to see for themselves that the injury to the nerve is indeed the explanation for their pain and that correction of that lesion offers them a reasonable chance of improvement or even of cure. Control of Pain in Lesions of the Brachial Plexus The particular difficulties of treating pain caused by injuries to the brachial plexus are considered by Anand and Birch (2009). Tricyclic antidepressants probably work through descending inhibitory aminergic systems and also by blocking sodium channels and amitriptylline, at a starting dose of 10mg, increased thereafter in 10mg steps is recommended. Should side effects prove intolerable desipramine might be considered. Anticonvulsants, which reduce the transport of sodium and potassium ions across cell membranes, potentiate pre and postsynaptic inhibition, reduce post synaptic potentiation and the amplitude of evoked potentials, are useful and usually combined with an antidepressant. Capsaicin may be helpful in the hypersensitivity of reinnervation by depleting substance P and may act on nerve terminals so that there is initially a worsening of the pain. Topical anaesthetic agents may be useful and in some intractable states very high doses of capsaicin (5–10%) may be combined with regional anaesthesia. Tramadol, a centrally acting weak opiate, which modulates central serotoninergic and noradrenergic inhibition of pain has a low risk of addiction and, probably, less side effects than narcotics. Dr. Jonathan Berman (Royal National Orthopaedic Hospital) has developed a set of recommendations for the early control of pain following brachial plexus injury which have proved helpful to admitting clinicians (Berman 2007). (1) The pain of avulsion is scarcely ever relieved by a single analgesic agent. (2) When a new drug is introduced, it should be titrated as rapidly as is compatible with safety. (3) When a drug is not
541
effective at the upper end of its therapeutic range but side effects are minimal, a second agent should then be added and titrated in the same way. (4) If a combination is helpful but side effects are unacceptable, each agent should then be separately titrated downward until an acceptable compromise between side effects and analgesia is reached. The recommended first line agent is pregabalin. In a normal adult the initial dose is 75 mg bd which is increased to 150 mg bd after 1 week and then to 300 mg bd a week after that. In cases where pain remains significant it is reasonable to prescribe outside its license, titrating more rapidly, increasing doses every 3 or 4 days. If pregabalin is not available, gabapentin may be used as an alternative, starting at 300 mg bd and, if necessary, titrating rapidly to 600 mg tds over 3 to 4 days. This is the maximum licensed dose but, if necessary, the dose may be increased beyond this to a maximum of 900 mg qds over a further 1 to 2 weeks. Berman recommends opioids as the preferred second line agents for there is good evidence of their efficacy in some patients with central pain; “Whilst there must be concerns about physical dependence, tolerance, addiction and problems with respiratory failure when introduced too rapidly none of these are reasons for the late introduction of opioids where no specific risk factor has been determined. It is of course necessary to assess the risk of addiction by direct enquiry before starting opioids and this should include a history of alcohol and recreational drug use:.......in the opioid naive patients, oramorph 10–20 mg 6 hourly, or oxynorm 10 mg 6 hourly can be used for initial titration. Short acting opioids however should not be used in the long term. Sustained release opioids should be introduced early on, initially in the form, where tolerated, of buprenorphine patches. In the young adult an initial dose of 10 mg per hour, Butrans (one weekly Buprenorphine patch) should be commenced with escalation to 20mg per hour if necessary after a further 24 hours.” A maximum ceiling dose must be agreed and adhered to. Amitriptylline is now used as a third line agent, and it may be appropriate particularly for pain at night as it is sedating. This can be introduced early on in a young adult with an initial dose of 25 mg at night increasing to 50 mg after 3 or 4 days. Three to four days beyond this the dose can be increased to 75 mg. Anticonvulsants and cannabinoids are used as fourth line agents and these, together with any alterations of the schedule outlined above are prescribed and supervised by Dr. Berman and his colleagues of the Pain Management Team. The response to various methods of treatment short of operation in 198 patients with adequate follow up during the years 2000–2004 is set out in Table 12.1. All had significant or severe pain in the early months after their injury, with a PNI score of 3 or 4 during this time. The response was considered good when the pain score dropped by two levels and fair when it dropped by one. A good or fair response was seen in more than 50% of patients who were treated with major opioids (including
542
Surgical Disorders of the Peripheral Nerves
Table 12.1 Response to drugs and other methods of treatment in 198 patients, 2000–2004. Good Fair Poor TOTAL Amitriptylline
5
28
23
56
Other antidepressants
1
2
2
5
Pregabalin
7
1
2
10
Gabapentin
13
31
15
59
3
7
11
21
Anticonvulsants
3
33
12
48
Minor opioids
Anti inflammatory
11
33
11
55
Major opioids
24
39
6
69
Diazepam and other Anxiolytics
0
0
4
4
Cannabis
4
4
3
11
27
2
3
32
1
3
Physiotherapy, TNS
2
Social activity/study
33
1
2
36
Work
30
2
1
33
tramadol) and in those using gabapentin or pregabalin. However nearly all of these patients were taking two or more drugs, 20 declined major opioids, and 48 more stopped taking drugs because of side effects. Amitriptylline, gabapentin, and major opioids were particularly poorly tolerated and the findings relate to the 130 patients who continued with medication for a minimum of 6 months. Severe chronic neuropathic pain imposes great difficulties on patients and on those treating them. Scrupulous and prolonged follow up is essential, so is careful communication with the family practitioner. It is important to make sure that the patient should not, finding that he or she is being given an antidepressant, come to feel that the clinician feels that his or her pain is imaginary. The rationale for giving an antidepressant should, whenever possible, be explained. It is rare, in our experience, to find patients becoming addicted to morphine and the synthetic opiates. One of the tragic and paradoxical consequences of addiction before the injury is the uselessness of those agents in controlling pain. Wall (1997) urged a less timid approach in the prescription of opioids, so that patients are advised that such a prescription is not “an early route to the grave.” On the other hand, Nashold and Ostdahl (1979) commenting on 17 patients with pain after avulsion of their plexus, noted that addiction to narcotics played “a major role” in nearly half their cases and similar problems were encountered by Jebara and Bechir (1987) in their patients with causalgia.
12.3.2 Indications for Operation Operation should be considered when an injury to a peripheral nerve provokes continuing pain. Worsening pain with
deepening of the lesion of the nerve shows that the agent responsible is still at work and this is a strong indication for urgent operation. Prompt recognition and early correction prevents or reduces, lasting pain and disability (Ellis 1958). The association between relief of pain and reinnervation of muscle is a particularly strong indication for urgent operation in severe injuries to the brachial plexus should the patients condition permit. There may be occasions when neuropathic pain suggests a focal lesion of a nerve, tumour or entrapment, and it is in these cases that operation should be seen as an important method of investigation and as a means of establishing diagnosis. The place for operation in the three syndromes, causalgia, neurostenalgia, and post traumatic neuralgia are now considered, the place for operation in the relief of pain caused by injuries to the brachial and lumbosacral plexus has been outlined in Chaps. 8 and 9.
12.3.3 Causalgia The severity of this terrible pain is matched only by that caused by malignant infiltration of the lumbo sacral or brachial plexus or by the worst examples of central pain following avulsion of the spinal nerves forming either plexus. The best clinical definition of the syndrome was provided by Barnes (1954): “(1) it is severe, spontaneous and persistent; (2) It usually has a burning quality; (3) it may spread beyond the territory of the injured nerve or nerves and (4) it is invariably aggravated by both physical and emotional stimuli”. It is this peculiar aggravation by examination or any form of distraction which distinguishes causalgia from all other neuropathic pain syndromes. The following description is typical. Case Report one of our patients, a 23 year old man, an innocent bystander, was shot in the arm by a bullet from a hand gun during a street robbery. He experienced immediate and intense searing and a bursting pain in the arm, forearm, thumb, and index, middle and ring fingers. Nothing relieved this. We first saw him at 3 weeks. He was quiet, even withdrawn, answering questions in monosyllables. He had lost weight and was exhausted from lack of sleep. The affected arm was held cradled against his body; he could not tolerate any light touch or the pressure of contact from a sleeve. Light draught was intolerably painful. The hand itself was swollen and mottled with alternating red and blue discolouration. There was profuse sweating. The nails were coarse, the hair on the back of the hand had overgrown. This man found some relief from keeping the hand and forearm wrapped in a moist towel, onto which he frequently poured cool, but not cold, water. An injection of local anaesthetic into the cervical sympathetic chain brought relief within minutes, the relief coming seconds after his showing Bernard-Horner sign, and a short while before vaso-and sudomotor paralysis appeared in the hand. Sympathectomy abolished the searing bursting pain
Pain
Fig. 12.7 The nerve lesion in causalgia. The bullet partially severed the median nerve: the ulnar nerve was unscathed.
and also the intense allodynia: this unmasked the deficit from the injury to the median nerve itself. The nerve was repaired by partial resection and grafting: two thirds had been transected by the bullet (Fig. 12.7). Pain relief was complete. At 3 years the patient noted only imperfect sensation in the thumb and index finger, but little else abnormal. The diagnosis of causalgia is one of the most straightforward exercises in clinical medicine and it can be made from the history alone. There has been a wound to the proximal part of the limb involving a major nerve trunk. That damage is usually shown, at later operation, to be partial. There is intense spontaneous persistent pain, often but not always of a burning nature and that pain is localised principally in the distribution of the nerve or nerves affected but spreading beyond that distribution. The pain is aggravated by physical and emotional stimuli and it is associated with intense mechanical allodynia and hyperpathia. The pain is also associated with disturbance of circulation and of sweating. Barnes accorded to Mitchell et al. (1864) and Paget (1864) the credit for the first description of this condition to which Mitchell gave the name causalgia, burning pain. However Guthrie (1827) described the case of a patient who was clearly suffering from this disorder: “a soldier received, at the Battle of Waterloo, a wound in the back part of the thigh, which stunned at the moment, but was shortly followed by considerable pain, not in the part actually injured, but on the outside of the leg below the knee, in the sole of the foot, and in the toes. This pain gradually increased rather than diminished, became at times intolerable, and rendered the sufferers life miserable. It seemed to increase by paroxysms, during which the man was in agony; the pain not only being permanently intense in the foot, but darting down to it, and accompanied by spasms of the whole extremity. No kind of medicine had any effect upon it; and at these periods the poor fellow found relief only by putting his foot on the cold stone, or by enveloping it in cloths wet with water or the liq. plumbi subacet. dilutes. This man was discharged, without having obtained any permanent
543
benefit, although from habit he became less sensible of the pain, and more accustomed to the necessity of jumping out of bed, and placing his foot on a stone or keeping it constantly wet and cold.” Mitchell (1872) thought that causalgia was “the most terrible of all the tortures which a nerve wound may inflict” and he described the particular characteristics of the pain, that it is burning “mustard red hot,” or like a “red hot fire rasping the skin” …. “Exposure to the air was avoided by the patient with a care which seems absurd.” Some of his patients obtained relief by keeping the affected skin wet, indeed two men found some ease by pouring water into their boots. Pain invariably preceded the trophic changes in the skin. Tinel (1917) confirmed Mitchell’s observations: “what patients most dread is contact with the air and dryness of the hand; tepid water often relieves them, and we see them wrapping round their hands moist cloths which they constantly renew. It is also to be noted that profuse perspiration of the hand frequently takes place.” Tinel drew attention to the role of external stimulation: “a strong emotion, an unexpected sound, or a brilliant light might bring on pain.” Incidence and Response to Treatment Barnes (1953) described 48 cases of causalgia and he analyzed the response of pain to sympathetic block (ganglion block) and sympathectomy. In the upper limb there was always a partial lesion of the lower trunk or medial cord of the brachial plexus or of the median nerve. In the lower limb there was always a partial lesion of the tibial division of the sciatic nerve. In all but two cases the lesion lay proximal to the knee or elbow and the cause was bomb or shell splinter or high velocity missile. Twenty seven of 31 complete sympathetic blocks brought complete relief, in three there was partial relief, there was one failure. The results of sympathectomy were better in the lower limb than in the upper limb. Preganglionic sympathectomy in which the stellate ganglion and second thoracic ganglia with intervening trunk were removed proved more effective than post ganglionic sympathectomy. Barnes performed an important investigation in two patients with causalgia in the lower limb: “in two patients a low spinal anaesthetic was administered with the intention of blocking the pain fibres from the lower limb and at the same time preserving the sympathetic innervation of the leg. This is possible because the lowest white ramus for the sympathetic chain is given off by the second lumbar nerve root. One of the patients had severe causalgia after a sciatic nerve lesion at the level of the buttock. With the patient sitting, 3.5 cucm of 5% procaine was injected into the spinal theca. There was immediate relief of pain long before the preganglionic sympathetic fibres were paralysed, for there was no rise in the skin temperature of the big toe as measured by a thermocouple until eight minutes after administration of the anaesthetic. The other patient had severe causalgia following a lesion of the medial and lateral popliteal nerves in the popliteal fossa. One cubic centimetre of procaine was injected into the spinal theca and complete anaesthesia was
544
obtained up to the level of the third lumbar dermatome. The pain in the foot was abolished but the sympathetic outflow to the lower limb was unaffected as there was no rise in the skin temperature of the first and fifth toes.” The impulses interpreted as pain arose from the lesion of the nerve and they were conveyed centrally through afferent fibres in the affected trunk. The pain was relieved before the sympathetic efferent fibres had been blocked. Richards (1967) reviewed data drawn from the Second World War on the centenary of the original publication of Weir Mitchell, Morehouse and Keen. There were 377 cases of causalgia from 9,781 peripheral nerve injuries, an incidence of 3.9%. In 83% the median or tibial or trunks of origin were damaged and in most cases the damage was partial. In 88% the injury lay proximal to the knee or the elbow. Multiple nerve injuries were common. Vasomotor disturbance was usually seen, the part often being warmer. Most patients found relief by keeping the part wet. The onset was within 24 hour in over 50%, and within 7 days in over 80%. He made a significant observation: “It was a common experience of those working in nerve injury centres during World War Two to meet patients who gave a good history of causalgic pain which had subsided before the patient reached the centre.” Sympathectomy was followed by excellent or good results in 89% of cases and if the operation was successful even when performed years after injury in prisoners of war. Omer and Thomas (1972) described 77 cases of causalgia from the war in Vietnam. In 41 the nerves were injured in the lower limb. The lesion was proximal to the knee or elbow in 61 cases. Onset of pain was immediate in 38 cases and it developed in 30 more by 3 weeks from injury. The initial treatment in these cases was by infusion of local anaesthetic close to the damaged nerve. If this was unsuccessful then central sympathetic blockade was introduced. These methods secured pain relief in 47% of cases. Surgical sympathectomy was reserved for the 41 cases where pain persisted and it was successful in 34. In an extensive review of 110 papers up to the year 2000 Hassantash and colleagues (2003) analyzed aspects of 1528 cases of causalgia. The nerves most commonly damaged were the median and the sciatic and 92% of the lesions were incomplete. Missile wounds accounted for 77% of all cases. The onset of pain was less than 7 days in 72% and less than one month in 90%. About one half of patients experienced some relief by moistening the part with cool or warm water. Pain was improved or abolished by sympathetic block in 88% of cases, sympathectomy cured the pain in 94%. An important study comes from Jebara and Bechir (1987) described 20 patients whose causalgia was caused by high velocity missiles. The main vessel was wounded in 17 cases, all were successfully repaired. Pain started within 6 h in 15 patients. The diagnosis of causalgia was made on the basis of intensity and distribution of the pain, the resistance to morphine and a positive response to sympathetic block. All patients were cured by sympathectomy. Cervical sympathectomy involved excision
Surgical Disorders of the Peripheral Nerves
of the lower part of the stellate ganglion and the upper ganglia, including T2, T3, T4 and T5 which were approached by lateral transaxillary exposure. Lumbar sympathectomy was performed by an antero extraperitoneal approach, excising the ganglia L2 to 4 with the intervening chain. Pain relief was instant and lasting; there was one Bernard Horner syndrome. Symptoms from the injury to the nerve persisted in two patients. Jebara and Bechir say this:“many aspects of the causalgia suggested a form of psychoneurosis and this diagnosis has often and most unfairly been applied to these patients.” This was certainly the case in two of the three children that we have treated. Case Report A 12 year old boy fell at school. A pencil lodged in his blazer pocket penetrated through the outer side of his right buttock. He experienced immediate and intense pain and was unable to stand. At the first hospital clinical examination confirmed a near complete right sided sciatic palsy and plain radiographs showed a retained foreign body in the buttock. At first operation the wound was explored and a 6 cm. A piece of pencil was removed. There was no improvement in the nerve palsy. The pain remained severe, resistant to morphine and to gabapentin. Two weeks later a second operation was performed in which the sciatic nerve was exposed. No lesion of the nerve was identified. Treatment for the pain was undertaken by the hospital Pain Team. A child psychiatrist was invited to advise on the boy’s “behavioural disturbance.” At 28 days nerve conduction studies were performed. There was no peripheral conduction in either the tibial or the common peroneal divisions. Extensive axonopathy was demonstrated in muscles innervated by the common peroneal nerve. A computerised tomographic (CT) scan showed a retained foreign body. He was transferred to our Unit at day 30. The boy was clearly in very great pain, having lost weight and he was unable to sleep. The right lower limb was flexed at his knee and ankle. There was excessive sweating in sole of the foot. Vasomotor instability was evident, the leg and the foot were at times red and others blue. Examination of the leg and foot was impossible because of intense allodynia. He had causalgia caused by partial transection of the sciatic nerve, and retention of a foreign body. This was confirmed at the third operation when a 5cm length of the pencil was seen impaling the sciatic nerve at the level of the ischial tuberosity. About 50 bundles (three quarters of the nerve) had been transected. The pencil was removed and a 4cm defect in the nerve trunk was repaired using nine grafts from the ipsi-lateral sural nerve. A tissue catheter was placed to permit the infusion of local aesthetic about the sciatic nerve proximal to the level of lesion for 5 days. At 2 weeks the hip spica was replaced with plaster of Paris splint holding the knee in flexion. This was discarded at 6 weeks. Pain was abolished by the first post-operative day. The boy was re-admitted for 5 days of in-patient rehabilitation work at 3 months from the third operation. There was no allodynia. Vigorous stretching work to overcome the fixed flexion
Pain
deformity at hip, knee and ankle was now possible. To clinical examination, sympathetic function within the tibial division of the nerve was normal. Light touch sensation had returned to the sole of the foot. The medial hamstring muscles were graded at MRC 4, the heel flexors at MRC 3 and toe flexors at MRC 2. There was a strong Tinel sign for the tibial nerve in the middle of the leg and for the common peroneal division at the neck of the fibula. By 18 months from the third operation the fixed deformity had been overcome. The boy was now playing sport. The hamstring, heel flexor and toe flexor muscles were graded at MRC 5, the power of ankle dorsi flexion and eversion at MRC 4, the extensors of the toes at MRC 3. There was no over reaction to stimulation of the skin. The significance of this child’s pain was not grasped, nor was it in the case of the 13 year old girl with severe spastic quadriparesis whose case is summarised in Chap. 3. In her case, the bluish discolouration and excessive sweating, of the skin was clearly demarcated at the level of the knee and although the area of allodynia coincided with the distribution of vasomotor disturbance her mother related that, even as late as 18 months after the second operation, there was colour change in the foot, indicating persisting vasomotor instability long after relief of mechanical allodynia. It is refreshing to turn to an example where the syndrome was recognised in a fracture clinic so that urgent treatment was successful not only in relieving pain but in preventing the severe psychological disturbance and the severe secondary fixed deformities which marked the course of the other two children. Case report: A 9 year old girl fell from a swing sustaining a fracture of the right proximal humerus. Initial records recorded an appropriate level of pain from the fracture, a normal pulse at the wrist and no evidence of nerve injury. At about 48 h, she complained of numbness in the right little and ring fingers. This progressed over the next 7 days to involve the hand and the forearm as far as the elbow and was associated with progressive weakness of the right upper limb. Pain became severe at about day 5 and intolerable by day 14, when she was transferred to our Unit. The hand was discoloured; there was profuse sweating. She could not move wrist or fingers. Light touch of the hand and of the medial aspect of the forearm caused severe pain. A weak radial pulse was present. The fracture site was explored on the following morning. A spike of the distal fragment had displaced the brachial artery and had injured the trunk nerves, two of which, the ulnar and the radial, were acutely angled over it. No bundles had been disrupted. There was a short segment of contusion in both nerves. The fracture was reduced and stabilised. A local anaesthetic infusion was maintained for 48 h after operation. Pain was abolished by the first postoperative morning. There was rapid recovery of function for the median and for the radial nerves; recovery was slower for the ulnar nerve, but it was virtually complete by 6 months. There was considerable delay before diagnosis and appropriate treatment in two children yet all showed the cardinal
545
diagnostic features of causalgia: severe, intractable pain expressed throughout the limb; worsening of pain by examination, noise, disturbance, vaso and sudomotor instability and intense mechanical allodynia. Experience at St Mary’s and the Royal National Ortho paedic Hospitals Some details of 48 operated cases in whom a diagnosis of causalgia was made at the time of presentation are set out in Tables 12.2, 12.3. Spontaneous remission seems to have occurred in 17 more patients who sustained sciatic or femoral neuropathies during operations for hip arthroplasty, and in 18 others with wounds of the brachial plexus. These patients described severe burning pain, allodynia and vaso and sudomotor disturbance which lasted for several weeks but which had resolved by the time we saw them. We would probably now consider that the true diagnosis in patients 14 (Table 12.2) and 11 (Table 12.3) was neurostenalgia, in which the pain was provoked by the continuing action of a bullet upon nerves which were in continuity. The main artery had been damaged in 21 patients and in 15 the nerve trunks were subjected to traction and distortion by a false aneurysm or arterio venous fistula. Six of the 10 examples of causalgia identified in the 58 patients with penetrating missile injuries to the brachial plexus were associated with such arterial injury (Stewart and Birch 2001). In 14 patients a successful response to sympathetic block was followed by sympathectomy which abolished the pain in 11 patients. In 32 other patients causalgia was cured by correcting the lesion of the nerve and of the adjacent axial artery. The reasons why our practice has altered, so that sympathectomy was abandoned in 1998, are now described. 1. T he blocks of the cervical sympathetic chain invariably extended to the adjacent spinal nerves. 2. In earlier years sympathectomy was performed during the same operation in which the arterial and the nerve lesions were corrected so that it was not possible to say which component of the intervention had actually relieved pain. 3. In one case of arterio venous fistula between the subclavian artery and vein the lesion was successfully treated by embolisation and there was rapid resolution of pain without any intervention upon the sympathetic chain. There was extensive spontaneous recovery of the associated lesion of C5 and C6. 4. In one patient with massive arteriovenous fistula in the arm which led to right heart failure, the distension of the veins in the posterior triangle of the neck ruled out any form of sympathetic block or sympathectomy. Correction of the arteriovenous fistula and decompression of the nerves abolished this man’s pain. 5. Infusing local anaesthetic about the nerves proximal to the lesion for 48–72 h not only blocks the somatic afferent fibres but also the autonomic afferent and efferent fibres. It would be unwise to reject the extensive evidence that sympathetic blockade or sympathectomy can be extremely
546
Surgical Disorders of the Peripheral Nerves
Table 12.2 Causalgia: Response to sympathetic block or sympathectomy (16 cases). Case # Age Sex Nerve and level Cause Pain response to sympathetic block
Pain response to sympathectomy
Other treatment
Pain relief
Nerve repair
Pain relief
Neurolysis
1
23
M
Median
2
48
M
C7
Arm
Hand gun
3
24
M
Medial cord
Shot gun
Relief
Pain relief
Removal pellets, neurolysis
4
31
M
Sciatic
Thigh
Fragments
Relief
Pain relief
Removal fragments, neurolysis
5
34
M
Lateral cord.
A-V fistula
Shot gun
Relief
Pain relief
Correction of fistula
6
43
M
Median, axilla. Rupture axillary artery
Shot gun
Relief
Pain relief
Repair artery and nerve
7
24
M
Upper and middle trunks. Retroclavicular subclavian aneurysm
Shot gun
Relief
Pain relief
Repair artery and nerve
8
31
M
Median, ulnar Axilla
Fragments
No relief
Pain relief
Removal fragments, neurolysis
9
28
M
Sciatic
Hip
Rifle bullet
Relief
Pain relief
Neurolysis
10
34
F
Median Brachial aneurysm
Arm.
Iatrogenous
Relief
11
55
F
Sciatic
Hip
Iatrogenous Cement burn
Some relief
No relief
Some relief from graft of tibial divisions
12
64
F
Sciatic
Hip
Iatrogenous Retractor injury
Relief of allodynia
No relief
Neurolysis with tissue catheter. Relief of burning pain
13
68
F
Femoral
Hip
Crush injury by retractor
Relief for 7 days
No relief
Neurolysis with tissue catheter. Transient relief
14
21
M
C7
Bullet
Relief
Relief
Neurolysis
15
43
M
Median Axillary artery rupture
Bullet
Relief
Relief
Repair artery. Delayed graft nerve
16
24
M
Upper trunk. Aneurysm subclavian artery
Arm.
Hand gun
Bullet
effective in the relief of causalgia but clinicians faced with severe neuropathic pain ought to be alert to the possibility of a continuing active provocative agent at work upon that nerve. We should not forget the case described by Kline and Hudson (1995) in which causalgia following a gunshot wound to the infraclavicular part of the brachial plexus was relieved only temporarily by repair of a false aneurysm of the axillary artery and neurolysis of the lateral cord and only later sympathectomy cured the pain, nor should we forget the impressive evidence presented by Jebara and Bechir (1987). We think that causalgia arises from the lesion of the nerve and that its characteristics may be explained by events at that level. The wound is untidy and often contaminated and the lesion of the nerve is partial so that some fibres have been divided, some have undergone Wallerian degeneration, some have
Relief
Repair artery, neurolysis
Repair aneurysm. Graft upper trunk
become demyelinated while others remain intact. These are subjected to the actions of local chemical agents and also neurotransmitters which lower the threshold of those nerve fibres which have not been divided, including of course the sympathetic efferent and afferent fibres. These events may explain the observation made by Schott (1994): “despite the more eloquent effects of efferent sympathetic dysfunction which occur perhaps as epiphenomena it is the afferents which subserve those pains that appear ‘sympathetic dependant’”. The nerve is subjected to ischaemia, to continuing irritation by retained foreign bodies, or by the pulsatile compression of an expanding haematoma or fistula which heightens the effects upon individual nerve fibres which remained intact because the perineurium of some of the bundles has become disrupted. These different provocative agents continue at work unless
31
25
21
49
37
19
20
21
51
16
17
81
15
18
63
34
10
14
34
9
31
21
8
13
32
7
16
37
6
21
27
5
11
49
4
12
28
24
2
3
18
1
M
M
M
M
M
M
F
F
M
F
M
M
M
M
M
M
F
M
M
M
F
Posterior cord, lateral cord rupture. Arterio-venous fistula
Radial Axilla
C5,C6. Arterio-venous fistula
Aneurysm axillary artery
Ulnar Axilla
Median Axilla
Infraclavicular plexus. Rupture posterior circumflex artery. Enormous false aneurysm
Infraclavicular plexus. Rupture axillary artery
Infraclavicular plexus. Rupture axillary artery
Sciatic Thigh
Sciatic Thigh
Sciatic Buttock
Lateral, medial cords. Aneurysm axillary artery
Lateral, medial cords. Aneurysm axillary artery
C5. Arterio-venous fistula
Arterio-venous fistula
Median Axilla.
C5,C6,C7
Median, ulnar, musculocutaneous axilla
Radial Axilla
C5,C6,C7,C8
Medial cord
Tibial Leg
Table 12.3 Causalgia: response to direct treatment only (32 cases) Case # Age Sex Nerve and level
Shotgun
Fragment
Fragment
Rifle bullet
Repair fistula Nerve graft
Nerve graft
Repair fistula. Nerve graft
Repair artery. Nerve graft
Graft median
Repair artery. Neurolysis
Fracture dislocation shoulder
Hand gun
Repair artery. Neurolysis
Repair artery. Neurolysis
Removal pellets. Neurolysis
Removal fragments. Graft Good of partial lesion
Removal pellets. Neurolysis
Repair aneurysm. Neurolysis
Repair aneurysm. Neurolysis
Embolisation of fistula. Graft C5
Repair fistula. Graft median nerve
Graft C5,C6, C7
Graft all three nerves
Repair nerve
Neurolysis C5,C6. Graft C7, C8
Graft medial cord
Repair aneurysm
Treatment
Closed dislocation shoulder
Dislocation shoulder
Closed fracture
Shot gun
Fragments
Hand gun
Rifle bullet
Fragments
Hand gun
Shot gun
Fragments
Rifle bullet
Rifle bullet
Rifle bullet
False aneurysm tibial artery. Iatrogenous
Cause
Good
Good
Good
Good
Good
Good
Good
Good
Good
Good
Good
Good
Good
Good
Good
Good
Good
Good
Good
Good
Good
Pain response
Lateral cord good, posterior cord fair (continued)
Good
Good
Good
Good
Radial, musculocutaneous good, median ulnar fair
Good
Good
Good
Good
Good
Good
Good
C5 good
Median poor
C5,C6 Good, C7 poor
Musculocutaneous good, median fair, ulnar poor
Good
C5,C6 good. C7, C8 poor
Poor
Good
Recovery
Pain 547
24
35
25
17
22
28
36
11
10
12
23
25
26
27
28
29
30
31
32
M
F
M
M
F
M
F
M
M
M
M
30
22
24
Sex
Table 12.3 (continued) Case # Age
Bleeding from popliteal artery.
Tibial, common peroneal. Knee
Sciatic. Buttock
Posterior, medial cords. Axillary artery
Median. Arterio-venous fistula axillary vessels
C7,C8,T1 (hematoma)
C8,T1 lesion in continuity
Median false aneurysm axillary artery
C7 rupture. C5,C6 partial
C5,C6,C7,C8,T1
C5,C6,C7
Medial cord. Aneurysm subclavian artery
Nerve and level
Iatrogenous. Correction deformity
Retained foreign body
Entrapped in fracture humerus
Hand gun
Iatrogenous. (sympathectomy for hyperhydrosis
Iatrogenous. Injury subclavian artery (sympathectomy for hyperhydrosis)
Iatrogenous
Rifle bullet
Shot gun
Rifle bullet
Rifle bullet
Cause
Good
Neurolysis
Neurolysis
Removal foreign body. Graft of nerve
Release artery. Neurolysis
Tibial good. CPN poor
Good
Good
Good
Good
Neurolysis
Repair fistula. Neurolysis
Good
Good
Good
Good
Good
Pain response
Repair aneurysm. Nerve graft
Nerve graft
Neurolysis
Neurolysis
Repair aneurysm. Nerve graft
Treatment
Ultimately good
Good
Good
Good
Good
Good
C5 useful, C6 good, C7 useful Good
Useful
C5 good. C6,C7 useful
Useful
Recovery
548 Surgical Disorders of the Peripheral Nerves
Pain
549
and until the wound has been properly treated by adequate debridement, bleeding has been controlled, axial flow restored and the nerve repaired, actions which should be supplemented by prolonged local anaesthetic block of the nerve or nerves affected through an infusion proximal to the level of lesion.
12.3.4 Neurostenalgia It is with a sense of relief that we turn now to this group of pain states which respond so well to surgical intervention. These patients, together with those suffering from true causalgia, are the most rewarding to treat by operation. In most cases the nerve trunk is intact, the lesion is neurapraxia or prolonged conduction block or at worst axonotmesis. The nerve is in some way irritated, tethered, compressed or ischaemic and treatment of the cause relieves the pain. Tethering or Entrapment of the Nerve induces pain of a sharp, lancinating, quality which is expressed in the distribution of the nerve. Case Report A 43 year old woman experienced intense pain and foot drop immediately after operation for ligation of the short saphenous vein. She could find comfort only by lying immobile with the knee and ankle flexed. Any attempt at stretching of the leg brought on agonising pain. The common peroneal nerve was re-explored 7 days later. It had been encircled by a suture which had reduced its diameter to about one half of normal. The epineurial vessels in the distal trunk were empty although the epineurium had not been breached. The distal epineurial vessels rapidly filled up after removal of the suture. She was free of pain on awakening and her foot drop recovered by 12 months (axonotmesis). Case Report An 11 year old boy fell through a plate glass window. There was torrential bleeding from both upper limbs. His life and limbs were saved by a master of vascular surgery who repaired both common brachial arteries. After operation a right sided ulnar palsy was noted but in addition there was excruciating pain on attempted abduction of the shoulder and extension of the elbow. This alert surgeon recognised the cause and asked one of us to deal with it. As predicted, at reexploration 5 days later, a suture which had been used to stem the hemorrhage had passed through the sheath of the ulnar nerve. It was removed. There was immediate relief of pain and rapid recovery of the nerve (Figs. 12.8, 12.9). This pattern of neurostenalgia was recognised at least two centuries ago. The first Lord Nelson, who suffered constant stump pain after the amputation of his right upper limb through the arm was relieved of his pain, quite suddenly, when the ligature around the median nerve came loose and was discharged through the wound (Crumplin and Harrison 2005). Guthrie (1827) described a number of cases: “when a ligature happens to include a nerve, it is a long time before it comes away, and is frequently broken off close to the knot,
Fig. 12.8 Neurostenalgia. This 11 year old boy could not tolerate attempted extension at the elbow. Pain abolished by removal of a suture which had passed through the sheath of the ulnar nerve.
Fig. 12.9 Neurostenalgia. Removal of exostosis from proximal fibula caused severe pain in a 14 year old boy, which was worsened by attempted extension of the knee and dorsiflexion of the ankle. Pain was abolished by neurolysis 4 months later but many weeks passed before the secondary deformities were overcome.
leaving the noose behind, which causes a great deal of misery to the patient; and if the wound closes, never, perhaps, does come away.” The case described by Denmark (1813)
550
has been cited as an example of causalgia but it is more likely that Denmark was in fact describing a case of neurostenalgia. The patient had been injured by a musket ball in the arm at Badajoz in 1812: “I always found him with the forearm bent and in the supine posture supported by the firm grasp of the other hand;… a small tumour could be felt in the site of the wound on the anterior part of the arm, which he could not bear to be touched without evincing additional torture.” Attempted extension of the wrist or elbow worsened the pain which had a burning quality, and was so violent as to cause continual perspiration from his face. The pain began at the extremities of the thumb and all of the fingers except the little one, and extended up to the arm to the part wounded. Although Denmark recognised that the radial nerve had been wounded it was still working and he proposed to the patient (Henry Croft) that the limb might be saved by removing part of the nerve, above the wound; “which he willingly consented to; but observed that he would rather have the arm amputated at once, than run the risk of a second operation.” Amputation was done and Denmark dissected the arm and found a small portion of the musket ball embedded within the nerve which was “blended with and intimately attached” to the adjacent structures. Ischaemia pain, in the conscious patient, is of course the cardinal sign of ischaemia of limbs or compartment syndromes. Some of this pain arises from the receptors in the dying muscle but much of it from compression and ischaemia of the nerve trunks. Relief of pain is the indication that surgical decompression has been done in time and has been adequately performed. The pain is constant, deep seated and rather poorly localised, and usually described as burning, crushing or bursting. A haematoma, which falls short of causing ischaemia of a limb, will resolve into scar, which can be a potent cause of tethering of the nerve and of pain. Case Report A 71 year old man with a history of cerebrovascular disease, for which he was taking aspirin, fell in the street dislocating his right shoulder. The joint was reduced under general anaesthetic and when he awoke he described intense pain. This proved resistant to treatment and when we saw him, 6 months later, his medication included: carbamazepine 200 mg tds, cyclizine 50 mg tds, gabapentin 1,800 mg a day (divided), coproxamol, tramadol 50 mg 8 times a day, amitriptylline 75 mg bd and diclophenac 75 mg bd. He had a degenerative lesion of the median nerve. There was intense mechanical allodynia in the distribution of that trunk and sweating was increased in the thumb. The median sensory action potential was absent. Median motor conduction for the median was reduced in amplitude and velocity to less than 50%. Electromyography revealed spontaneous activity in the abductor pollicis brevis and in pronator teres but there were some low amplitude units. The lesion was one of axonopathy with elements of demyelination. At operation the median nerve was found tightly bound down within a
Surgical Disorders of the Peripheral Nerves
thickened axillary sheath. It was decompressed, an indwelling catheter was inserted and a local anaesthetic infusion was maintained about the nerves, proximal to the level of lesion, for 48 h. His pain was abolished and over the next 6 months there was considerable recovery for the nerve.
12.3.4.1 Duration of Symptoms One remarkable feature of neurostenalgia is the relief of pain by operation carried out years after onset. One such example was described by Henry (1957) of a naval officer whose long standing neck pain was abolished by section of the greater occipital nerve which had become embedded in torn muscle. Another example is that of case 2 described by Birch and St Clair Strange (1990). The patient suffered severe pain with peroneal palsy after successful treatment of a fracture of the pelvis. Nearly 3 years after the first operation the sciatic nerve was exposed at the notch. Its lateral part was narrowed to a translucent web by callus and the latter was removed. Pain was abolished; the nerve made a wholly unexpected recovery over the next 3 weeks. Montgomery and his colleagues (2005) described the case of a woman who endured severe pain for 5 years from a lesion of the sciatic nerve incurred during arthroplasty of the hip. She recollected that there had been extensive bruising in the buttock in the early days after operation. She presented with marked weakness of muscles innervated by the common peroneal division but NPI confirmed that the lesion was predominantly one of conduction block. The nerve was found tethered by adhesions between it and adjacent tissue at the level of the neck of the femur. The pain was abolished by the following day and she had full recovery of power shortly after that. The case of neurostenalgia of the femoral nerve reported by Phang and his colleagues (2009) has been described in Chap. 3. In that case pain continued for 10 years after pelvic osteotomy and the patient underwent arthroplasty of the hip in the mistaken belief it might relieve her pain. In fact, her pain continued and was relieved by operation upon the nerve. Even more remarkable is the case described by Camp et al. (2008) of intractable neurostenalgia of the ulnar nerve abolished by neurolysis 18 years after the injury. The patient was a farmer who had sustained severe injuries to the left arm from a combined harvester in 1987 which included rupture of the left brachial artery. This was successfully repaired using a reversed vein graft and the operating surgeon recorded that the median and ulnar nerves had “the appearance of having been stretched.” The patient became aware of pain 7 days later. This was burning, initially in the left hand, later radiating to the medial aspect of the forearm and it increased in intensity over the years. By 13 years he was no longer able to work regularly and sleep was constantly disturbed. Pain was worsened by cold but eased by physical and mental activity.
Pain
By 18 years, when we first saw him, pain disturbed every activity of his daily life and sleep was possible only with high doses of morphine and diazepam. He requested amputation, and indicated that he had become suicidal. Pain was confined to the distribution of the ulnar and medial cutaneous nerves of forearm. There was wasting of the small muscles of the hand. Sweating was reduced in the little and ring fingers but their colour was similar to that of the rest of the hand. Touch and warm thresholds were elevated. There was no mechanical or cold allodynia. Heat was felt as pain at 46.5°C, he experienced hyperalgesia to pin prick in the skin of the little and ring fingers. At operation the vein graft was intact but dilated and the ulnar nerve had become densely adherent to it over a distance of 6 cm. Pulsation from the vein graft could be seen transmitted to the ulnar nerve. The epineurium appeared to be intact but the epineurial blood vessels were scanty compared with those in the nerve trunk proximal and distal to the lesion. The ulnar and medial cutaneous nerves of the forearm were removed from scar and separated from the artery. A local anaesthetic infusion was maintained for 48 h after operation. Gentle active movements at the shoulder and elbow were commenced immediately after operation to maintain gliding of the nerves. Relief of pain was dramatic and lasting, by 1 year he was taking no medication and rated his PNI Pain score at 1 from an earlier maximum of 4 and the VAS score at 2 from an earlier level of 10. There was improvement in skin sensation, but the small muscles of the hand did not recover.
12.3.5 Post Traumatic Neuralgia Even after taking out cases of causalgia, central pain from traction injury to a limb plexus and those of continuing irritation there remain many patients with persisting and sometimes severe pain caused by injury to a nerve. Most clean transections of main nerves do not provoke severe pain, with the important exceptions of the spinal accessory nerve and the nerve to serratus anterior. The problem is particularly severe after injuries to the terminal branches of nerves of cutaneous sensation (Fig. 12.10). It is a remarkable and unexplained fact that the clean removal of these nerve trunks for the purpose of repairing a major nerve is very rarely attended by severe pain, whereas significant pain is really quite common after accidental damage to the terminal branches of those same cutaneous nerves. There are certain aspects of this phenomenon which might repay further investigation. First of these is warning, the patient has been told that a nerve will be taken, and that there will be an area of sensory loss which they did not have before. Next is the fact that the section of the nerve is complete when it is to be used for a graft, it is not a partial wound, it is not a wound of a
551
Fig. 12.10 A 34 year old man suffered a military rifle bullet wound through both arms damaging the superficial radial nerves. The nerve on the right was completely severed. There was no pain. The nerve on the left was incompletely divided and this caused severe pain
terminal twig. Third is the fact that during elevation of graft, when it is done properly, the donor nerve is sectioned deep to the deep fascia. Many of the worst cases of post traumatic neuralgia followed damage to a terminal branch where it lies in the superficial fascia. The only method of treatment with any hope of lasting success is the restoration of a normal input of non painful impulses; this normally involves successful repair of the affected nerve or nerves. It is these cases that produce the most difficulty in treatment and it is a salutary reflection that many of these injuries were caused by surgeons and doctors in the course of treatment. Dyck, Dyck and Engelstad (2005) described the experience of one of them during and after fascicular biopsy of the sural nerve of the ankle at the ankle without injection of local anaesthetic directly into the nerve. “Immediately with nerve transection, a sharp stinging burning discomfort occurred in a region just below the lateral malleolus and spreading to the Achilles tendon. Although severe, the pain lasted only a few seconds. This pain could have been avoided by local anaesthetic injection of the proximal nerve. In the first three or four post operative days there was soreness at the site of the incision, especially when the tissues were stretched. On stretching the nerve, as in bending, sharp electric shock like pain was experienced in the lateral heel …. When the sural nerve was palpated at the level of the proximal transection site, no local discomfort was experienced. In the cutaneous distribution of
552
innervation a mild prickling, unpleasant burning sensation was produced. When the affected dermatome was touched, some sensation was elicited. Although the threshold for touch and pin prick was greatly heightened, stronger stimulae produced a diffuse unpleasant prickling burning in the affected region. These symptoms gradually faded away, returning intermittently with less intensity over several years. At 5 years the symptoms had disappeared, although touching the affected dermatome still elicited a slightly unpleasant and altered sensation.” The initial characteristics of the pain reflect qualities of the nerve fibres involved. Injury to a nerve of cutaneous sensation causes spontaneous pain which is often intermittent or even convulsive against a background of constant dysaesthesiae. Mechanical allodynia is common and warm and cool allodynia are frequent. By contrast the pain after section of the spinal accessory nerve or nerve to serratus anterior, neither of which innervates skin, is characterised by a boring, deep seated, dull, aching pain which is poorly localised around the shoulder girdle. The pain caused by injury to these two nerves is quite characteristic yet its significance is rarely recognised. The propensity of the pain caused by injury to cutaneous nerves to evolve so that it spreads to adjacent skin with extension of the area of allodynia, the development of excessive sweating and of vasomotor instability often leads to diagnosis of “complex regional pain syndrome,” an error which consigned several hundreds of patients whom we have seen to fruitless and even to harmful treatment by infusions, blocks, powerful drugs and even to a suggestion that the patient is unbalanced and has some form or other of psychological disturbance.
Fig. 12.11 Post traumatic neuralgia misdiagnosed as CPRS Type I in a 12 year old girl. Left: a branch of the dorsal division of the ulnar nerve has been partially transected by the arthroscope. Right: the nerves were bathed in fibrin clot glue after neurolysis: a catheter inserted for prolonged nerve block.
Surgical Disorders of the Peripheral Nerves
Case Report A 12 year old girl developed persisting pain at the wrist after a fracture of the distal ulna. An MR scan suggested a tear of the fibrocartilage which was repaired using an arthroscope. She awoke in severe pain, with numbness affecting the whole hand. She was referred to a Pain Clinic and given a prolonged trial of drugs which impaired her intellect and so interfered with school work that an educational psychologist was asked to see her, and recognised deterioration in non verbal reasoning, spatial skills, information processing and defects in auditory and working memory. A measurable decline in reading and spelling was recorded. That educational psychologist requested further investigations, and an MR scan of the brain was performed which was normal. The child was discharged from hospital. Her family practitioner requested a further opinion and when we saw the child at 12 months after her injury she was withdrawn, indeed wary, and protected her hand from examination. However, she gave a clear description of severe shooting pain in the ulnar territory of the hand when she awoke after operation and it seemed inescapable that the ulnar nerve had in fact been damaged by the arthroscope (Fig. 12.11). A branch of the dorsal division of the ulnar nerve had been partially transected while the main trunk had clearly been distorted without any breach of the epineurium. The nerves were decompressed, and an infusion of local anaesthetic was maintained for 48 h after operation through an indwelling catheter. There was rapid improvement and she dispensed with medication 6 weeks later. She still had the mechanical pain in her wrist joint, which was helped by an orthosis and she was earnestly advised to keep well away from surgeons.
Pain
The credit for this outcome rests with the child’s mother, the educational psychologist and the family practitioner who were not content with the diagnosis of CRPS type 1 nor with the child being discharged from hospital care. Case Report A 68 year old, previously healthy woman sustained a fracture of the distal one third of her left ulnar which was treated by open reduction and internal fixation with a plate and screw. She awoke in agonising pain, which was worst in the little finger and extended to the whole hand and forearm. She was referred to a pain clinic where an unavailing course of regional blocks using local anaesthetic and sympatholytic agents was given with no lasting relief. Eighteen months later a further opinion was sought by her family practitioner who had been provided with a diagnosis of intractable CRPS type 1. The practitioner raised the possibility of a focal injury to the ulnar nerve and enquired whether there might be a case for sympathectomy. The diagnosis of injury to the dorsal branch of the ulnar nerve was clear from the history and from physical examination. She had a strong, painful Tinel sign on percussion over the nerve at the proximal end of the scar. Operation revealed that most of the dorsal branch had been crushed by the plate. The damaged segment was resected and replaced by a freeze thawed muscle graft. A local anaesthetic infusion was maintained for 72 h after her operation. Her neuropathic pain disappeared but it was then necessary to embark on a prolonged course of treatment to overcome the secondary fixed deformities of thumb web space. This case is but one example from many in which the operating surgeon did not consider the possibility of direct damage to a nerve, choosing instead to follow the broad and easy road of passing responsibility to another colleague without having made reasonable efforts to establish cause. Many patients with post traumatic neuralgia remain with pain, dysaethesiae, and even hyperpathia and allodynia resistant to all forms of treatment. Wounding of part of a nerve by a needle or scalpel is particularly dangerous. Barton and Smith (2008) describe the work of Joseph Swan, a pioneer of research on peripheral nerves who was active in the first half of the nineteenth century. They relate one of Swan’s cases, that of a woman who had sustained damage to the median nerve at the elbow after venesection. Her pain was so severe that she developed convulsions and went into coma. Swan suspecting a partial injury of the nerve divided it above the level of the wound seeking, as he said:“to destroy any connection with the sensorium.” Pain relief was immediate, dramatic, and lasting. The femoral nerve is particularly vulnerable to damage during nerve block or femoral arteriography and it seems difficult to justify resection and repair of the damaged segment of the nerve when it is still working, as it usually is (Fig. 12.12). We have seen patients whose lives have been ruined by an apparently trivial injury to a peripheral nerve. Prevention is the cure.
553
Fig. 12.12 Post traumatic neuralgia. Partial laceration of the femoral nerve by the needle during femoral nerve block induced intractable pain which was not relieved by neurolysis of the nerve. Spinal cord stimulation was necessary.
12.3.5.1 Post Traumatic Neuralgia After Injury to a Main Nerve The Amputation Neuroma Methods of prevention of pain have been described in Chap. 7. Bach et al. (1988) used epidural infusion before and for up to 72 h, after lower limb amputation. Tomaino and his colleagues (1998) emphasised three elements drawn from their experience with 300 amputations of the lower limbs. Patient confidence must be established by their involvement in making decisions from the outset. Next, nerves are blocked with local anaesthetic before they are divided, coagulation diathermy is used as little as possible and the nerve stumps are not implanted into muscle. Thirdly, a regime including diazepam, imipramine and amitriptylline is started before and maintained for 6 weeks after operation. Pain of Regeneration It is usual for patients to experience dysaesthesiae in the months or years after repair of a nerve and some never progress beyond this (see Chap. 4). Most describe hyperalgesia, particularly to cold. However, crippling pain is rare after properly executed repair and it is uncommon even after replantation of amputated hand or forearm. Our own experience with such cases is poor, resection and regrafting of the previous repairs when pain has been longstanding for a year or more, as it usually has, fails with dismal regularity. A common feature of these cases is that the first nerve repair was not well performed and in particular there had been inadequate treatment of the associated soft tissues. It is unusual for such cases to respond to transcutaneous nerve stimulation or sympathetic or local anaesthetic blockade. A decisive course of sensory retraining and desensitisation sometimes helps perhaps by a form of counter irritation. The Spinal Accessory Nerve and the Nerve to Serratus Anterior Division of either of these nerves usually causes severe pain and the response of that pain to repair of the nerve is quite dramatic. Camp (2010) studied pain in 93 patients with injuries of the spinal accessory nerve nearly all of which were iatrogenous. After repair, the severity of the
554
pain was reduced by two levels on the PNI system, from an average of 3 to an average of 1 in about three quarters of them, irrespective of the duration of symptoms. Relief generally occurred long before there was any evidence at all of reinnervation of the paralyzed trapezius. The pain relief was not followed by muscle recovery in three patients. The immediate onset of pain and the early relief of that pain by repair means that it cannot be explained by traction upon the nerves in the neck from dependence of the upper limb. It is more likely that the proximal stump of the divided nerve is subjected to the action of inflammatory and other noxious agents. There is, often, an element of traction upon the proximal stump because of tethering by scar to the floor of the posterior triangle (Fig. 12.13).
Fig. 12.13 A 16 year old girl suffered severe pain after iatrogenous division of the spinal accessory nerve. The nerve was repaired after an interval of 4 years. Her pain had disappeared by the day after operation. By 9 months there was good recovery into the trapezius muscle.
Surgical Disorders of the Peripheral Nerves
12.3.5.2 PTN After Injury to Minor Nerves Occasionally a cutaneous neuroma goes unrecognised for years. The value of accurate diagnosis, and of simple operation is exemplified by a number of our patients whose treatment course included psychiatric referral. In these, the psychiatrist insisted that there must be some injury to a peripheral nerve! Case Report A 34 year old woman with severe pain in the left side of her face, neck, arm and chest for 8 years, after internal fixation of fracture of the clavicle. There was allodynia, but there was also a well localised area of exquisite tenderness just above the scar of earlier incision. Movements of the neck brought on pain: the patient adopted a position of torticollis. At operation, a neuroma of supraclavicular nerve, 8mm in diameter, was found tethered to the clavicle. The neuroma was excised and the stump was placed deep to the fat pad. Pain was abolished. The treatment of painful neuroma is deeply unsatisfactory. Herndon and Hess (1991) wrote: “when operating on or near nerves, prevention of injury and neuroma formation should be foremost in the surgeon’s mind … great care is taken when working around nerves to prevent the iatrogenic injury during exploration.” A painful neuroma may not be the sole explanation for patients’ loss of function and, of itself, is not an indication for operation. As Dobyns (1991) emphasises the clinician should take time to establish the diagnosis and elicit important background information about the cause of injury as well as contributing factors including litigation and compensation. Occupational therapists can help by showing how the part can be maintained within the activities of daily function and simple padded finger stalls or custom made padded straps can protect a neuroma from mechanical irritation or protect a hypersensitive, tender digit. Many patients develop their own methods to cope with post traumatic neuralgia and two examples illustrate an approach which is, fortunately, common. Case Report A 58 year old man, working as a charge nurse, sustained injury to his right sural nerve during an operation for varicose vein surgery. He experienced severe pain and although the diagnosis was confirmed by nerve conduction studies no action was taken. This patient purchased his own support stocking which improved his symptoms to some extent and he used a hair dryer to relieve the allodynia of the skin of his calf and heel. By 6 months he had regained control of his pain but he had, by this time, lost his position. Case Report A 26 year old woman, who worked as a ski instructor, sustained injuries to both sural nerves after operation for bilateral calcaneal exostoses. She experienced severe pain and intense allodynia in the skin of the heels. She was prescribed a number of drugs, including gabapentin and amitriptylline and abandoned these because they made her too drowsy. She treated her pain herself by techniques of
Pain
desensitisation, and by applying silicone gel dressings. She modified her shoes and realised that ski boots were the most comfortable to wear and so was able to return to her job. The partial neuroma is a common and difficult problem. Neurolysis, followed by a 48 h infusion of local anaesthetic during which time the patient is encouraged to move the part, is sometimes effective for nerves tethered by a focus of scar. Gould (1991) has set out a systematic approach. The nerve is blocked with local anaesthetic and this is followed by a course of anti inflammatory and other agents, combined with desensitisation and other therapy techniques. If these fail his preferred operation involves neurolysis and wrapping the nerve by a vein or vascularised fascia. It is best to reconnect, where possible, the central and the distal stumps of the damaged nerve so that regenerating axons can grow to their proper targets. Repair offers the chance of restoring normal afferent volleys of impulses and so restore the normal inhibitory effects of these impulses on central neurones as predicted by the gate theory. If reconnection is impossible then the damaged proximal stump must be moved to the less hostile and less exposed place. Translocation of the proximal stump into muscle may permit a new connection between regenerating axons and host tissues. As Atherton and his colleagues (2006) suggest, the success of these methods may rest on moving the nerve away from the skin and subcutaneous tissues which are rich in nerve growth factor (NGF) to areas which are poor in NGF, such as bone and muscle. Many operations have been described. They fall into one of four groups. It is important that the bed upon which the nerve is repaired or the bed into which it has been transferred should be healthy. Suppression of pain after operation by infusion of local anaesthetic for at least 24 h after operation is essential. It is essential also that the patient be encouraged to use the injured limb from the outset. Post operative splinting must do no more than protect the repair or the transferred nerve. 1. The prevention of a neuroma. Although injection of sclerosants (Smith and Gomez 1970) have been shown to be successful, it is of course, impossible to prevent nerve regeneration unless the parent neurone is dead. Langley and Anderson (1904) may have approached a state of inhibition of regeneration in their proposal to suture one nerve trunk to another. Kon and Bloem (1987) applied this to palmar digital nerves by suturing one to another, or, when the injury was unilateral, suturing the dorsal to palmar branches. 2. Containment of a neuroma. The nerve end is secluded by epineurial flaps (Martini and Fromm 1989, Tupper and Booth 1976, Kline and Hudson 1995) or by silicone rubber caps (Swanson et al. 1977). Dahlin and Lundborg (2001) inserted the proximal end into silicone tubes at least 20mm in length and observed that the proximal stump extended only a few mm into the tube to form a cone shaped structure. Axonal regeneration is limited by
555
the length of the tube and the absence of a distal stump, and the axons do not show spontaneous activity. 3. Translocation. Translocation of the nerve without excision of the neuroma was practised by Littler (Herndon et al. 1976, Herndon 1999). Every effort is made to keep the neuroma intact with its mature encapsulating scar while transposing it en bloc to another area free of scar so that it is not subjected to repeated trauma. The neuroma with its fibrous capsule, is carefully isolated and it is transferred to lie deep to a muscle, in a web space, or between the shafts of adjacent metacarpals. Transposition into bone (Mass et al. 1984, Goldstein and Sturim 1985) aims to contain the nerve stump within a compartment and to protect it from direct injury. Hazari and Elliot (2004) emphasise that the nerve must be free of tension after transfer and that it should not be angulated at the point of entry into bone. 4. Translocation of nerve into muscle. Dellon and MacKinnon (1986) found that relocating the superficial radial nerve into the under surface of the brachioradialis was effective. Evans and Dellon (1994) and Sood and Elliot (1998) successfully treated neuromas of the wrist and hand by translocation into the pronator quadratus. It seems that the method is not as reliable when the nerve is transferred into the small muscles of the hand. We have used the technique in more than 50 cases. One patient was made worse, most experienced useful improvement. 5. Repair. This seems the most logical. When pain spreads beyond the territory of the nerve, with allodynia and even hyperpathia in adjacent skin, then the only possible way to rectify the situation is to restore volleys of afferent impulses from re-innervated skin, and so hope to reverse the changes described earlier in the dorsal horn; events predicted by the gate theory. The idea of using the patient’s own muscle as a graft is a significant advance in the treatment of these cases (Thomas et al. 1994) and the technique has been described in Chap. 7.
12.4 Interventions upon the Central Nervous System There is now unassailable evidence that the pain caused by preganglionic injuries to the brachial plexus is usually improved by distraction, by return to work and by reinnervation of muscle. Our findings are related in Chap. 9 but there are, sadly, a number of patients in whom it is impossible to re-innervate the limb either by graft or by transfer. These are patients with complete avulsion in whom there is associated damage to the spinal cord, and who show the signs of partial Brown Séquard syndrome. Their pain is truly central, it arises from damage to the central nervous system and it is both worse than and more intractable than de-afferentation pain.
556
We should now review the case for central interventions, either by destructive lesions in the spinal cord or implantation of central stimulators in the treatment of these patients. Case Report A 24 year old right handed apprentice printer sustained a complete avulsion injury of the brachial plexus on the dominant side in a motor cycle accident, a diagnosis which was confirmed by hemilaminectomy in 1964. He experienced severe pain from the day of injury which proved to be resistant to all forms of treatment then in use but in spite of this he returned to full time work as a printer, in a post which he maintained for the following 30 years. A stimulator was implanted into the thalamus 24 years after his injury, which failed. An intercostal nerve transfer was performed 31 years after his injury with the object of improving his pain, and this also failed. The arm was amputated 37 years after his accident, by which time he had retired from work, because he considered it a useless impediment. After retirement he explained that pain was becoming more and more dominant in his life. Dorsal root entry zone lesion was performed 40 years after his accident, after appropriate psychological and psychiatric assessment, which he thought did somewhat improve matters. There is no question about this patient’s motivation and courage and it is salutary to see how little benefit he gained from a wide range of treatments. Another example of a patient who showed great stoicism throughout and who came to terms with his pain by his own efforts and initiative is described Case Report An 18 year old right handed builder sustained a severe injury to the right brachial plexus when scaffolding collapsed onto the tip of this right shoulder. The injury occurred in 1962. He experienced immediate pain and complete paralysis of the upper limb. The pain was initially constant, of a crushing burning and bursting quality, felt in the hand and in the forearm. At about 1 week he experienced new shooting pains, radiating from the neck to the radial aspect of the hand. The plexus was never explored, and a trial of treatment included transcutaneous nerve stimulation and various drugs was unavailing. He returned to work at 12 months, and moved on to running his own business and employing a number of people. He explained how he had found that the choice for him lay between distraction by work and activity, and destruction. He abandoned all drugs because they made him too drowsy. He refused to consider taking opiates. A trial of cannabis actually made things worse. He continued in full time work until the age of retirement after which he moved to warmer climes as he had learnt that pain was very much worse in cold or wet weather and worsened when there was any change in the climate. So much depends upon the attitude of the patient. One man suffered a complete lesion of the brachial plexus, one in which four spinal nerves were avulsed from the spinal cord and which was complicated by rupture of the subclavian artery. Pain was severe. He described his experience in the
Surgical Disorders of the Peripheral Nerves
Rehabilitation Unit, and pointed out that although medication was certainly helpful in the early months after injury the control of pain rested on his determination to overcome his disability and to reintegrate in daily life. He (James 1988) makes this important observation which should be remembered by all engaged in this difficult field: “it is important for me not to ‘think disabled’ and I must act as if my splint is quite normal.” This patient returned to complete his nursing training and subsequently to full time work.
12.4.1 Interventions on the Central Nervous System: Stimulation With the increase of knowledge of inhibitor mechanisms, Shealy et al. (1967) introduced posterior column stimulators in some cases. At that time, the advantages of this non destructive method were widely canvassed. Technical advances (North and colleagues 2007) have led to the increasingly common use of spinal cord stimulation (SCS) in a growing number of pain problems. Stuart and Winfree (2009) describe the indications for SCS and describe spinal nerve root stimulation which is proving effective in foot pain caused by peripheral neuropathy and in the treatment of inguinal neuralgia after operations for hernia. Stuart and Winfree point out that stimulation of the dorsal root and of the dorsal root ganglion affects fibres destined for the spino thalamic tracts. Canavero and Bonicalzi (2003) provide an extensive review and indicate that SCS is of little value in brain lesions and unpredictable in the central pain caused by cord lesion. Meyerson (2005) considers motor cortex stimulation, and indicates that several mechanisms may be relevant, including presynaptic inhibition, and depression of activity in spino thalamic tract neurones. Stimulation of the prefrontal cortex attenuates the behavioural responses to peripheral noxious stimulation. He comments: “hopefully we can avoid repeating the somewhat sad story of spinal cord stimulation, where despite its extensive use for over three decades evidence for its efficacy is still limited.” Proposals for stimulating the natural production of enkephalins were canvassed by Richardson and Akill (1977): using paraventricular electrodes they were able to locate the sites at which stimulation produced maximal secretion of these natural analgesics. Our earlier experience with deep brain stimulation was not very promising but the successful application of the method is illustrated by a more recent case. Case Report A 26 year old right handed welder sustained a complete lesion of the brachial plexus in which C5 was ruptured and C6 to T1 were avulsed. He experienced severe pain in his right upper limb, which commenced on the day of injury and which included both a constant crushing and burning or bursting element involving the hand, with
Pain
superimposed lightning like shoots of convulsive pain running from the arm into the hand. Morphine gave him some relief but the side effects of this with other agents were such that he stopped taking all drugs. Intercostal transfer was performed 22 months later which seemed to improve his pain for about 3 months but then it returned with unabated vigour. Dorsal root entry zone lesion was performed 37 months later and this relieved pain for about 1 year. Pain then returned and put paid to his tentative efforts to get back to work. By now he admitted to drinking heavily. Sixty two months after the injury a deep brain stimulator was implanted, with electrodes placed into the periventricular gray and into the sensory thalamus (Professor Aziz, Oxford). Stimulation of the former electrode caused a feeling of warmth from his face to his chest and he developed hypertension and anxiety, “all indicative of correct electrode placement.” Stimulation of the sensory thalamus at low frequency improved the excruciating pain in his hand, which was worst in the thumb. Relief of pain was dramatic and he was fully implanted 1 week later. His pain relief is maintained at the time of writing.
12.4.2 Interventions upon the Central Nervous System: Ablation The first serious attempts to relieve brachial plexus pain by operation arose from the identification of the cord as the site of origin of symptoms. The refinements of cordotomy introduced in this country by Lipton (1968) made that procedure far safer than it had formerly been. “Percutaneous” cordotomy became a standard treatment in cases of intractable pain in the terminal stages of malignant disease. It was tried in a few cases of plexus pain: the early results were good, but it soon became apparent that a procedure applicable in cases of terminal disease was inapplicable in the case of otherwise healthy young persons. There was recurrence of pain within a year. This disadvantage was additional to the theoretical one of intervening on the healthy side of a cord already damaged unilaterally. Cordotomy rests on the principle of rather strict segregation of nerve fibres by their function. That assumption must be modified by the findings of many workers such as those of Wall (1970), and of Nathan et al. (1986). Nathan and his colleagues studied the sensory effects in man of lesions of the posterior columns and of some other afferent pathways and showed that lesions of the posterior columns caused extensive loss of function and in fact increased the perception of pain, tickle, warm and cold. However, the description of a new pathway for visceral pain in the posterior columns by Hirshberg and colleagues (1996) enabled Nauta and his colleagues (1997) to achieve an important triumph in the case of the young woman with overwhelming pain arising from the bowel, the bladder and the urethra after she was treated by
557
radiotherapy for carcinoma of the cervix: “she was literally dying of lower abdominal pain.” The operation involved: “a strictly midline lesion of dorsal T8 segment by needle, designed to interrupt the newly described visceral pain pathway which passes from neurones at the base of the dorsal horn and passes ipsilaterally in the dorsal midline to synapse in the nucleus gracilis.” The patient’s pain was abolished. Sindou (1972) perfected an operation of selective posterior rhizotomy at the dorsal root entry zone (DREZ), by microsurgical technique, which rested on the topographical distribution of afferent fibres in the dorsal root injury zone (Tarlov 1937). Small myelinated and unmyelinated fibres were found laterally, passing to the dorsal horn neurones: larger fibres, responsible for discriminative touch and proprioception were concentrated medially, passing to the dorsal columns. Microsurgical lesion in the lateral part of DREZ achieves analgesia, and selective microcoagulation in the dorsal horn itself destroys hyperactive neurones responsible for de-afferentation pain. Sindou et al. (2001) reviewed the technique in cases of pain from injuries to the spinal cord or cauda equina, in 44 patients. They found that the best results were obtained when pain was paroxysmal, and when it was “segmental,” that is when the expression of pain related closely to the lesion. They point out that spinal cord stimulation is effective only in deafferentation pain syndromes if the dorsal column – lemniscal systems remain at least partially functional. The target for the selective dorsal root entry zone lesion is defined as: the ventro lateral part of the central portion of the dorsal rootlets where there is a lateral regrouping of fine fibres; the medial part of the Lissauer’s tract, and the dorsal most layers of the dorsal horn “where the afferent fibres establish synaptic contacts with spino-reticulo-thalamic tract cells”. A new proposal was made in 1979 by Nashold and Ostdahl, that of coagulation of the “DREZ,” on the basis that this procedure destroyed the part of the gray matter in which spontaneous firing was occurring - namely, the substantia gelatinosa. Radio frequency coagulation was used, lesions being made at intervals along the intermediolateral sulcus of the posterior aspect of the cord. Of 21 patients, 17 had suffered avulsion of the brachial plexus; one had suffered avulsion of the upper limb. Good results were obtained in 13 cases but Nashold and Ostdahl note that in 11 cases there was some degree of residual weakness of the ipsilateral or occasionally both lower limbs. Since that time, Nashold (1981) has modified duration of coagulation, amplitude of current and depth of penetration, with favourable effect on the incidence of long tract affection. The principal objection to coagulation of the entry zone is that it inflicts another lesion on the side of a cord already damaged. Thomas and Jones (1984) found evidence of subclinical affection of the posterior columns in 50% of their patients with avulsion of the brachial plexus coming to operation for relief of pain. It is not uncommon in the acute case to find clinical evidence of affection of the ipsilateral
558
corticospinal tract. There are serious potential complications: affection of the ipsilateral corticospinal tracts; damage to the ipsi lateral posterior column; impotence; unpleasant paraesthesiae. There is an incidence of failure of around 40%, and an unquantified evidence of late recurrence. Sometimes the recurrent pain is worse than that which went before. Thomas and Kitchen (1994) supply a long term review of this operation. In 44 patients, at a minimum interval of 63 months from DREZ lesion for brachial plexus pain, 35 stated that there had been significant and lasting pain relief. Eight cases (`18%) had persisting neurological deficit - usually mild. It is plain that central operation of any sort, and entry zone coagulation in particular, should be reserved for patients who (1) have very severe an intractable pain persistently marring life: (2) have failed to respond to a full range of treatment short of operation; (3) have failed to show spontaneous improvement after 3 years or whose pain actually increases during this period; (4) have not experienced any relief after an otherwise successful neurotisation; (5) specifically require further treatment for persistent pain and understand well the drawbacks and hazards of central operation - particularly those of entry zone coagulation. Patients addicted to controlled drugs are not suitable subjects for central operation. We have seen two cases of severe pain after partial lesions of the spinal cord associated with traction lesions of the brachial plexus, which resolved with recovery of motor function (Flannery and Birch 1990) and the history of one of these is related in Chap. 9. It is important to remember that the late onset of pain may signify a new lesion within the spinal cord and examples of this event are also described in Chap. 9. We have also seen cases of particularly severe, but painless injuries where it seems that the patient performed his or her own DREZ lesion. Case Report An 18 year old student sustained an extremely severe lesion of the left brachial plexus when she was knocked down by a car. A CT myelogram confirmed that there was avulsion of the roots of C5, C6 and C7 central to the transitional zone and hemi atrophy of the ipsilateral side of the cord. The sensory action potential for the median nerve was preserved but it was reduced for the ulnar, the radial and the medial cutaneous nerves of forearm. Quantitative sensory testing confirmed that the lesion was virtually complete, but the flare response was preserved. At operation an undisplaced avulsion lesion (type 5) was confirmed. This patient never, at any time, experienced pain, and it is possible that the lesion induced ischaemic damage to the dorsal root entry zone.
12.5 Summary Wall (1985) commented that the dualistic separation of physical from mental, which is one of the foundations of western
Surgical Disorders of the Peripheral Nerves
scientific thought has been one of the factors responsible for confusion and error in analysis of pain. He sees this in the doctors’ question about pain “is it real or is it in the mind” and in some ways he re-states Aristotle “pain is an agony of the soul.” We hope that doctors treating patients with neuropathic pain will find our classification helpful and that they will recognise that appropriate surgical intervention is usually successful in true causalgia, and in neurostenalgia. It is increasingly clear that re-innervation of the limb is important in relieving brachial plexus pain but by how much remains uncertain because of the favourable natural history in some untreated patients. Success in these types of neuropathic pain works by modifying abnormal afferent and efferent somatic and sympathetic activity; by removal of an irritative lesion; and by restoring afferent impulses to the spinal cord. With the outstanding exception of the spinal accessory nerve, operations on nerves have much less to offer in post traumatic neuralgia, where our success, of about 50%, may be no more than placebo. Such operations have scarcely anything at all to offer in the treatment of reflex sympathetic dystrophy (CRPS Type 1). For these patients a programme designed to modify their response to pain may help, and we shall discuss this in the chapter Rehabilitation.
References Albe-Fessard D, Lombard MC (1983) Use of an animal model to evaluate the origin of and protection against deafferentation pain. In: Bonica JJ, Lindblom U, Iggo A (eds) Advances in pain research and therapy, vol 5. Raven, New York, pp 691–700 Albrecht PJ, Hines S, Eisenberg E, Pud D, Dr F, Connolly MK, Pare M, Davar G, Rice FL (2006) Pathologic alteration of cutaneous innervation and vasculature in affected limbs from patients with complex regional pain syndrome. Pain 120:244–266 Anand P (1985) Post-sympathetic pain and sensory neuropeptides. Lancet 1986: 512 Anand P, Birch R (2009) Pain following brachial plexus injury and its management. In: Venkataswami R (ed) Surgery of the injured hand. Jaypee Press, Delhi, pp 359–363 Anand P, Rudge P, Mathias CJ, Dr S, Ghatei MA, Naher-Noe M, Sharief M, Misra VP, Polak JM, Bloom SR, Thomas PK (1991) A new autonomic and sensory neuropathy with loss of sympathetic adrenergic sympathetic function and sensory neuropeptides. Lancet 337:1253–1254 Anand P, Pandya S, Ladiwake U, Singhal B, Sinicropi D, WilliamsChestnut RE (1994) Depletion of nerve growth factor in leprosy. Lancet 344:129–130 Anand P, Terenghi G, Warner G, Kopelman P, Williams-Chestnut RE, Sinicropi DV (1996) The role of endogenous nerve growth factor in human diabetic neuropathy. Nat Med 2:703–707 Anand P, Birch R, Terenghi G, Wellmer A, Cedarbaum J, Lindsay RM, Williams-Chestnut RE, Sinicropi DV (1997) Endogenous nerve growth factor and ciliary neurotrophic factor levels in human peripheral nerve injury. NeuroReport 8:1935–1938 Anderson SD, Basbaum AI, Fields HI (1977) Response of medullary raphe neurones to peripheral stimulation and to systemic opiates. Brain Res 123:363–368
Pain Atherton DD, Taherzadeh O, Facer P, Elliot D, Anand P (2006) The potential role of nerve growth factor (NGF) in painful neuromas and the mechanism of pain relief by their relocation-to-muscle. J Hand Surg 31B:652–656 Bach S, Noreng MF, Tjellden NU (1988) Phantom limb pains in amputees during the first twelve months following limb amputation, after peri-operative epidural blockade. Pain 33:297–301 Banati RB, Cagnin A, Brooks DJ, Gunn RN, Myers R, Jones T, Birch R, Anand P (2001) Long term transynaptic glial responses in the human thalamus after peripheral nerve injury. Neuro Report 12:3439–3442 Banati RB (2002) Brain plasticity and microglia: transynaptic glial activation in the thalamus after limb denervation linked to cortical plasticity and central sensitisation? J Physiol 96:289–299 Barnes R (1953) The role of sympathectomy in the treatment of causalgia. J Bone Joint Surg 35B:172–180 Barnes R (1954) Causalgia: a review of 48 cases. In: Seddon HJ (ed) Peripheral nerve injuries by the nerve injuries committee of the medical research council, Medical Research Council Special Report Series 282. HMSO, London, pp 156–185 Barton NJ, Smith BD (2008) Joseph Swan (1791–8174): pioneer of research on peripheral nerves. J Hand Surg 33E:252–259 Beecher HK (1946) Pain of men wounded in battle. Ann Surg 123:96–105 Berman J (2007) Guidelines on the early control of pain following brachial plexus injury. Internal Publication Pain Control Team Royal National Orthopaedic Hospital, Stanmore, UK Betcher AM, Bean G, Casten DF (1953) Continuous procaine blocks of paravertebral sympathetic ganglions: observations in 100 patients. JAMA 151:288–292 Birch R, St Clair Strange FG (1990) J Bone Joint Surg 72B:312–313 Boettger MK, Till S, Chen MX, Anad U, Otto WR, Plumpton C, Trevise DJ, Tate SN, Bountra C, Coward K, Birch R, Anand P (2002) 2002 calcium-activated potassium channel SK1 – and IK1-like immunoreactivity in injured human sensory neurones and its regulation of neurotrophic factors. Brain 125:252–263 Bonica JJ (1990) Causalgia and other reflex sympathetic dystrophies. In: Bonica JJ (ed) The management of pain, Vol. 2. Lea and Febiger, Philadelphia, pp 220–243 Bonney G (1973) Causalgia. Brit J Hosp Med 9:593–596 Bruel S, Harden RN, Galer BS, Saltz S, Bertram M, Balkonja M, Gayles R, Rudin N, Bhugra MK, Stanton-Hicks M (1999) External validation of IASP diagnostic criteria for complex regional pain syndrome and proposed research diagnostic criteria. Pain 81:147–154 Buckenmaier CC, Rupprecht C, McKnight G, McMillan B, White RL, Gallagher RM, Polomano R (2009) Pain following battle field injury and evacuation: a survey of 110 casualties from the wars in Iraq and Afghanistan. Pain Med 10:1487–1496 Camp SJ (2010) Lesions of the spinal accessory nerve. J. Bone Jt Surg B in press Camp SJ, Milano R, Sinisi M (2008) Intractable neurostenalgia of the ulnar nerve abolished by neurolysis 18 years after injury. J Hand Surg 33E:45–46 Canavero S, Bonicalzi V (2003) Spinal cord stimulation for central pain. Pain 103:225–226 Carr DB, Todd DP (1991) Therapeutic blocks and continuous infusion in the treatment of chronic pain. In: Gelberman RH (ed) Operative nerve repair and reconstruction. J B Lippincott, Philadelphia, pp 1563–1569 Connor CJ, Ralph JK, Aldington DJ (2009) Field hospital analgesia. J Army Med Corps 155:49–56 Coward K, Plumpton C, Facer P, Birch R, Carlstedt T, Tate S, Bountra C, Anand P (2000) Immunolocalisation of SNS/PN3 and NaN/ SNS2 sodium channels in human pain states. Pain 85:41–50 Crumplin MKH, Harrison A (2005) A fatal shot: the death of Admiral Horatio Viscount Lord Nelson. Ann Roy Coll Surg Eng 87:302–308 Dahlin LB, Lundborg G (2001) Use of tubes in peripheral nerve repair. In: Kliot M (ed) Neurosurg Clin, N Am 2001 12(2):341–352
559 Dav R, Ariely D, Frenk M (1995) Effect of past injury on pain threshold and tolerance. Pain 60:189–193 Déjerine PJ (1901) Sémiologie du système nerveux. In: Traité Pathologie Générale. Paris, Masson, p 559 Dellon AL, MacKinnon SE (1986) Treatment of the painful neuroma by neuroma resection and muscle implantation. Plast Reconstr Surg 77:427–438 Denmark A (1813) An example of symptoms resembling tic douloureux produced by a wound in the radial nerve. Med Chir Trans 4:48 Devor M (1991) Neuropathic pain and injured nerve: peripheral mechanisms. Brit Med Bull 47:619–630 Dobyns JH (1991) Pain dysfunction syndrome. In: Gelberman (ed) Operative nerve repair and reconstruction. JB Lippincott, Philadelphia, pp 1489–1495 Dyck PJ, Dyck PJB, Engelstad J (2005) Pathological alterations of nerves. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy Elsevier Saunders, Philadelphia, 4th edition, Chapter 32:733–829 Edwards D, Bowden M, Aldington DJ (2009) Pain management at role IV. J Army Med Corps 155:58–61 Ellis H (1958) Disability after tibial shaft fractures. J Bone Joint Surg 40B:190–197 Evans GRD, Dellon AL (1994) Implantation of the palmar cutaneous branch the median nerve into the pronator quadratus for treatment of painful neuroma. J Hand Surg 19 A:203–206 Flannery MC, Birch R (1990) Acute compression of the cervical spinal cord: a complication of preganglionic injury to the brachial plexus. Injury 24:247–148 Foerster O (1927) Die Leitungsbahnen des Schmerzgefuehls und die chirurgische Behandlung der Schmerz-zuständ. Urban Schwarzenberg, Berlin/Vienna Goldstein SA, Sturim HS (1985) Intraosseous nerve transposition for treatment of painful neuromas. J Hand Surg: 270–274 Gould JS (1991) Treatment of the painful injured nerve in continuity. In: Gelberman RH (ed) Operative nerve repair and reconstruction. Lippincott, Philadelphia, pp 1541–1550 Gowers WR (1886) A manual of diseases of the nervous system. In: Diseases of the spinal cord and nerves, vol 1. JA Churchill, London Guthrie GJ (1827) A treatise on gunshot wounds, on inflammation, erysipelas and mortification, on injuries of nerves and wounds of the extremities, 3rd edn. Burgess and Hill, London Hassantash SA, Afrakteh M, Maier RV (2003) Causalgia. A meta analysis of the literature. Arc Surg 138:1226–1231 Hazari A, Elliot D (2004) Treatment Of end-neuromas, neuromas-incontinuity and scarred nerves of the digits by proximal relocation. J Hand Surg 29B(4):338–350 Henry AK (1957) Extensile exposure, 2nd ed. Upper Limb. E and S Livingstone, Edinburgh, p116 Herndon JH, Eaton RG, Littler JW (1976) Management of painful neuromas in the hand. J Bone Joint Surg Am 58:369–373 Herndon JH, Hess AV (1991) Neuromas. In: Gelberman RH (ed) Operative nerve repair and reconstruction. JB Lippincott, Philadelphia, pp 1525–1540 Herndon JH (1999) Neuromas. In: Green DP, Hotchkiss RN, Pederson WC (eds) Green’s operative hand surgery, 4th edn. Churchill Livingstone, New York, pp 1469–1480 Hibbert C (1998) Wellington. A personal history. Harper Collins, London, pp 182–183 Hirschberg RM, Al-Chaer ED, Lawan NB et al (1996) Is there a pathway in the posterior funiculus that signals visceral pain? Pain 67:291–305 Hord ED, Oaklander AL (2003) Complex regional pain syndrome: a review of evidence supported treatment. Curr Pain Headache Rep 7:1531–3433 Htut M, Misra P, Anand P, Birch R, Carlstedt T (2006) Pain phenomena and sensory recovery following brachial plexus avulsion injury and surgical repair. J Hand Surg 31B:596–605
560 Jebara VA, Bechir S (1987) Causalgia: a war time experience – report of 20 treated cases. J Trauma 27:519–523 Jagdish S, Aldington D (2009) The use of opioids during rehabilitation after combat-related trauma. J Army Med Corps 155(1):64–67 James D (1988) Brachial plexus injury. Nurs Times 84(10):38–39 Kato N, Htut M, Taggart M, Carlstedt T, Birch R (2006) The effects of operative delay on the relief of neuropathic pain after injury to the brachial plexus. J Bone Joint Surg 88B:756–759 Kirkup J (2007) A history of limb amputation. Springer-Verlag, London Klein CJ, Dyck PJ (2005) Hereditary sensory and autonomic neuropathies. In: Dyck PJ, Thomas PK (eds) Peripheral Neuropathy, 4th edn. Elsevier Saunders, Philadelphia, Chapter 78, pp 1809–1844 Kline DG, Hudson AR (1995) Pain of nerve origin. In: Nerve Injuries. WB Saunders, Philadelphia, Chapter 20, pp 513–524 Bloem KM (1987) The treatment of amputation neuromas in fingers with a centrocentral nerve union. Ann Plast Surg 18:506–508 Kuntz A, Farnsworth DI (1931) Distribution of afferent fibres via the sympathetic trunks and grey communicating rami to the brachial and lumbosacral plexus. J Comp Neurol 53:389–399 Lamotte RH, Lundberg LER, Torebjork HE (1992) Pain, hyperalgesia and activity in nociceptive C units in humans after intradermal injection of capsaicin. J Physiol 448:749–764 Langley JN, Anderson HK (1904) The union of different kinds of nerve fibres. J Physiol 31:365–391 Lawson SN (2005) The peripheral sensory nervous system: dorsal root ganglion neurones. In: Dyck PJ, Thomas PK (eds) Peripheral Neuropathy, 4th ed., Elsevier Saunders, Philadelphia, Chapter 8, pp 163–202 Lewis T (1936) Cutaneous hyperalgesia. Clin Sci 2:373–4231 Linson MA, Leffert R, Todd DP (1983) The treatment of upper extremity reflex sympathetic dystrophy with prolonged continuous stellate ganglion blockade. J Hand Surg 8:153–159 Lipton S (1968) Percutaneous electrical cordotomy in relief of intractable pain. Brit Med J 2:210–212 Loeser JD, Ward AA (1967) Some effects of deafferentation on neurones of the cat spinal cord. Arch Neurol 17:629–636 Loh L, Natahan PW (1978) Painful peripheral nerve states and sympathetic block. J Neurol Neurosur Ps 41:664–671 Magerl W, Wilk SH, Treede RD (1998) Secondary hyperalgesia and peripheral wind-up following intradermal injection of capsaicin in humans. Pain 74:257–268 Martini A, Fromm B (1989) A new operation for the prevention and treatment of amputation neuromas. J Bone Jt Surg 71B: 379–383 Mass DP, Ciano MC, Tortosa R, Newmeyer WL, Kilgore ES (1984) Treatment of painful hand neuromas by their transfer into bone. Plast Reconstr Surg 74:182–185 McLachlan EM, Janig W, Devor M, Michaelis M (1993) Peripheral nerve injury triggers noradrenergic sprouting within dorsal root ganglia. Nature 363:543–545 Melzack R, Wall PD (1965) Pain mechanisms: A new theory. Science 150:971–979 Mendell LM (1966) Physiological properties of unmyelinated fibre projection to the spinal cord. Exp Neurol 16:316–332 Mendell LM (1999) Neurotrophin action on sensory neurones in adults: and extension of the neurotrophic hypothesis. Pain (Suppl 6): S127–S132 Merskey HM (1986) Classification of chronic pain: descriptions of chronic pain syndromes and definitions of pain terms prepared by the International Association for the Study of Pain Subcommittee on Taxonomy. Pain (Suppl 3):1–226 Meyerson B (2005) Motor cortex stimulation, Editorial. Pain 118:6–7 Michaelis M, Vögel C, Blenk K-H, Arnarson A, Jänig W (1998) Inflammatory mediation sensitize acutely axotomised nerve fibres to mechanical stimulation in the rat. J Neurosci 18:7581–7587 Michaelis M, Lui XG, Janig W (2000) Axotomised and intact muscle afferents but not skin afferents develop ongoing discharges of dorsal
Surgical Disorders of the Peripheral Nerves root ganglion origin after peripheral nerve lesion. J Neurosci 20(7):2724–2748 Misra P (2008) L’Hermittés sign in early multiple sclerosis. Personal communication Mitchell SW (1872) Injuries of nerves and their consequences. JB Lippincott, Philadelphia Mitchell SW, Morehouse GR, Keen WW (1864) Gunshot wounds and other injuries of nerves. J B Lippincott, Philadelphia, p 100 Montgomery AS, Birch R, Malone A (2005) Sciatic neurostenalgia: caused by total hip arthroplasty, cured by later neurolysis. J Bone Joint Surg 87-B:410–411 Nagasako EM, Oaklander AL, Dworkin RH (2003) Congenital insensitivity to pain. Topical review. Pain 101:21–219 Nashold BS (1981) Modification of lesion of DREZ technique. J Neurosurg 55:1012 Nashold BS, Ostdahl RH (1979) Dorsal root entry zone lesions for pain relief. J Neurosurg 51:59–69 Nathan PW (1977) Pain. Brit Med Bull 33:149–158 Nathan PW (1983) Pain and the sympathetic system. J Auton Nerv Syst 7:363–370 Nathan PW, Smith MS, Cook AW (1986) Sensory effects in man of lesions of the posterior columns and some other afferent pathways. Brain 109:1003–1041 Nauta HJW, Hewitt E, Westlund KN, Willis WD (1997) Surgical interruption of a midline dorsal column visceral pathway. J Neurosurg 86:538–542 Noordenbos W (1959) Pain. In: Problems pertaining to the transmission of nerve impulses which give rise to pain. Elsevier, Amsterdam North R, Shipley J, Prager J et al (2007) American Academy of Nerve medicine practise parameters for the use of spinal cord stimulation in the treatment of chronic neuropathic pain. Pain Med 8:S200–S275 O’Connor AB, Dworkin RH (2009) Treatment of neuropathic pain: an overview of present guidelines. Am J Med 122(10A):S22–S32 Oaklander AL (2001) The density of remaining nerve endings in human skin with and without postherpetic neuralgia after shingles. Pain 92:139–145 Oaklander AL, Rismiller JG, Gelman LB, Zheng L, Chang Y, Gott R (2006) Evidence of focal small-fibre axonal degeneration in complex regional pain syndrome-1 (reflex sympathetic dystrophy). Pain 120:235–243 Omer G, Thomas S (1972) Treatment of causalgia: a review of cases at the Brooke General Hospital. Texas Med J 67:93–96 Ovelmen-Levitt J, Johnson B, Bedenbaugh P, Nashold BS (1984) Dorsal root rhizotomy and avulsion in the cat: a comparison of long term effects on dorsal horn neuronal activity. J Neurosurg 15(6):921–927 Ovelmen-Levitt J (1988) Physiology of the dorsal horn as related to the deafferentation pain. Appl Neurophysiol 51:104–116 Paget J (1864) Clinical lecture on some cases of local paralysis. Med Times Lond 1:331–332 Petersen-Zeitz KR, Basbaum AI (1999) Second messengers, the substantia gelatinosa and injury induced persistent pain. Pain (Suppl 6):S5–S12 (Tribute to P Wall) Phang IS,Biant LC, Jones TS (2009) Neurostenalgia of the femoral nerve. A treatable cause of intractable hip pain in a young adult. J Arthroplasty PMID:19577882 Ploner M, Freund H-J, Schnitzler A (1999) Pain affect without pain sensation in a patient with post central lesion. Pain 81:211–214 Price DD, Greenspan JD, Dubner R (2003) Neurones involved in the exteroceptive function of pain. Pain 106:215–219 Price DD, Mayer DJ (1974) Physiological laminar organization of the dorsal horn of M. Mulatta. Brain Res 79:321–325 Rainville P, Carrier B, Hofbauer RK, Bushnell MC, Duncan GH (1999) Dissocation of sensory and affective dimensions of pain using hypnotic modulation. Pain 82:159–172 Ramer MS, Thompson SWN, McMahon SB (1999) Causes and consequences of sympathetic basket formation in dorsal root ganglia. Pain Suppl 6:111–120
Pain Richards RL (1967) Causalgia. A centennial review. Arch Neurol 16: 339–350 Richardson DE, Akill H (1977) Pain reduction by electrical brain stimulation in man. J Neurosurg 47:178–183 Rinaldi PC, Young RF, Albe-Fessard D, Chodakiewitz J (1991) Spontaneous neuronal hyperactivity in the medial intralaminar thalamic nuclei of patients with deafferentation pain. J Neurosurg 74:415–421 Roberts WJ (1986) A hypothesis on physiological basis for causalgia and related pains. Pain 24:297–311 Rowbotham MC (2005) Mechanisms and management of neuropathic pain. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 4th edn. Elsevier Saunders, Philadelphia, Chapter 118, pp 2637–2652 Schott GD (1988) Interrupting the sympathetic outflow in causalgia and reflex sympathetic dystrophy. Brit Med J 316:792–793 Schott GD (1994) Visceral afferents: their contribution to “sympatheticdependent” pain. Brain 117:397–413 Schott GD (2004) Communicating the experience of pain: the role of analogy. Pain 108:209–212 Shealy CN, Mortimer JT, Reswick JB (1967) Electrical inhibition of pain by stimulation of the dorsal columns. Anaesth Analg 46:489–491 Sherrington CS (1903) Selectivity for a certain quality of stimulus. J. Physiol 30:39 Seltzer Z, Devor M (1979) Ephaptic transmission in chronically damaged peripheral nerves. Neurol 29:1061–1064 Sindou M (1972) Étude de la jonction radiculo-médullaire postérieure. La radicellotomie postérieure sélective dans la chirurgie de la douleur. Thése Med. Lyons Sindou M, Keravel Y (1980) Analgésie par la méthode d’électrostimulation transcutanée: résultats dans les douleurs neurologiques à propos de 180cas. Neurochirurgie 26:153–157 Sindou M, Mertens P, Wael M (2001) Microsurgical DREZotomy for pain due to spinal cord and/or cauda equina injuries. Long term results in a series of 44 patients. Pain 92:159–171 Smith JR, Gomez NH (1970) Local injection therapy of neuromata of the hand with triamcinolone acetamide. A preliminary study of twenty two cases. J Bone Joint Surg 52A:71–83 Sood MK, Elliot D (1998) Treatment of painful neuromas of the hand and wrist by relocation into the pronator quadratus muscle. J Hand Surg 23B(2):214–219 Spinner RJ, Scheithauer BW, Amrami KK (2007) Medial plantar nerve compression by a tibial artery schwannoma. J Neurosurg 106: 921–923L Stanton-Hicks M, Janig W, Hassenbusch S et al (1995) Reflex sympathetic dystrophy: changing concepts and taxonomy. Pain 63: 127–133 Stewart M, Birch R (2001) Penetrating missile injuries. J Bone Joint Surg 83B:517–524 Stuart RM, Winfree CJ (2009) Neurostimulation techniques for painful peripheral nerve disorders In: eds. RJ Spinner., CJ Winfree. Neurosurg Clin 20:111–120 Stojadinovic A, Auton A, Peoples GE, McKnight GM, Shields C, Croll SM, Bleckmer LL, Winkley J, Maniscalco-Theberge ME, Buckenmaier CC (2006) Responding to challenges in modern combat casualty care: innovative use of advanced regional anesthesia. Pain Med 7:330–338 Sugiura Y, Terui N, Hosoya Y (1989) Difference in distribution of central terminals between visceral and somatic unmyelinated (C) primary afferent fibres. J Neurophysiol 62:834–840 Swanson AB, Boeve NR, Lumsden RM (1977) The prevention and treatment of amputation neuromata by silicone capping. J. Hand Surg 2:70–78 Sweet WH (1984) Deafferentation pain after posterior rhizotomy, trauma to a limb, and herpes zoster. Neurosurgery 15:928–932 Taggart M (1998) A prospective long term study of pain and rehabilitation In: Birch R, Bonney, G, Wynn Parry CB (eds) Surgical disorders of the peripheral nerves. Churchill Livingstone Edinburgh, London, Philadelphia, pp. 461–463
561 Tarlov I (1937) Structure of nerve rot. Nature of the junction between the central and peripheral nervous systems. Arc Neurol Psychiat 37:555–583 Thomas DGT, Jones SJ (1984) Dorsal root entry zone lesions (Nashold’s Procedure) in brachial plexus avulsion. Neurosurgery 15:966–968 Thomas DGT, Kitchen ND (1994) Long term follow up of dorsal root entry zone lesions in brachial plexus avulsion. J Neurol Neurosur Ps 57:737–738 Thomas M, Stirratt A, Birch R, Glasby M (1994) Freeze thawed muscle grafting for painful cutaneous neuromas. J Bone Joint Surg 76B: 474–476 Thomas PK, Ochoa J (1993) Diseases of the peripheral nervous system – clinical features and differential diagnosis. In: Dyck PJ, Thomas PK (eds) Peripheral neuropathy, 3rd edn. WB Saunders, London, pp 760–761 Tinel J (1917) Nerve wounds. Ballière Tindall and Cox. London. Authorised translation Rothwell F. Revised and edited by Joll CA, pp 187–192 Tomaino B, Anract P, Ouaknione M (1998) Psychological management prevention and treatment of phantom pain after amputation for tumour. Int Orthop 22:205–208 Tupper JW, Booth DM (1976) Treatment of painful neuromas of sensory nerves in the hand: a comparison of traditional and newer methods. J Hand Surg 1:144–151 Verdugo R, Ochoa J (1994) Sympathetically maintained pain. I. Phentolamine block questions the concept. Neurol 44:1003–1010 Wall PD (1970) The sensory and motor role of impulses travelling in the dorsal columns towards the cerebral cortex. Brain 93:505–524 Wall PD (1978) The gate control theory of pain mechanisms. A reexamination and re-statement. Brain 101:1–18 Wall PD (1985) Future trends in pain research. PhilosTrans R Soc (Lond) B 308:393–405 Wall PD (1997) The generation of yet another myth in the use of narcotics. Pain 73:121–122 Wall PD, Gutnick M (1974a) The properties of afferent nerve impulses originating from a neuroma. Nature 248:740–742 Wall PD, Gutnick M (1974b) Ongoing activity in peripheral nerves: the physiology and pharmacology of impulses arising from a neuroma. Exp Neurol 43:580–593 Wall P, Devor M (1981) The effect of peripheral nerve injury on dorsal root potentials and on transmission of afferent signals into the spinal cord. Brain Res 209:95–111 Wall PD, Lidierth M, Hillman P (1999) Brief and prolonged effects of Lissauer tract stimulation on dorsal horn cells. Pain 83:579–589 Wasner G, Baron R, Jänig W (1999) Dynamic mechanical allodynia in humans is not mediated by a central presynaptic interaction of AB mechanoreceptors and nociceptive C afferents. Pain 79:113–119 Willner C, Low PA (1993) Pharmacologic approached to neuropathic pain. In: Dyck PJ, Thomas PK, Griffin JW, Low PA, Poduslo JF (eds) Peripheral europathy, 3rd edn. Saunders, Philadelphia, Chapter 94, pp 1709–1720 Woolf CJ (1995) Somatic pain – pathogenesis and prevention. Brit J Anaesth 75:169–176 Woolf CJ, Shortland P, Coggleshall RE (1992) Peripheral nerve injury triggers central sprouting of myelinated afferents. Nature 355: 75–78 Woolf CJ, Bennett GJ, Doherty M, Dubner R, Kidd B, Koltzenurg M, Lipton R, Loeser JD, Payne R, Torebjörk E (1998) Toward a mechanism based classification of pain. Editorial. Pain 77:227–229 Zyluk A (2001) Sequelae of reflex sympathetic dystrophy. J Hand Surg 26B:151–154 Zyluk A (2005) CRPS1 after distal radius fracture: a prospective study of the role of psychological factors. J Hand Surg 30B:574–580 Zyluk A, Puchalski P (2008) Treatment of early complex regional pain syndrome Type I by a combination of mannitol and dexamethasone. J. Hand Surg 33E: 130–136
13
Reconstruction
Am I god to kill or make alive, that this man doth send unto me to recover a man of his leprosy? 2 Kings 5:7 Reconstruction: Definition; Limitations; Principles; Causes of fixed deformity. Correction of fixed deformity; Active reconstruction by restoration of active movement: upper limb (scapulo-humeral, glenohumeral, elbow and wrist joints: the hand): lower limb (hip, knee, ankle and foot); transfer of vascularized bone and muscle. Amputation.
We are indebted to Peter Smith (2006), librarian to the Royal National Orthopaedic Hospital, for advising us about the life and work of Stromeyer who by subcutaneous tenotomy cured William Little of his equinovarus deformity (Fig. 13.1). Little was the founder of the Orthopaedic Institution of London which became, later, the Royal National Orthopaedic Hospital. Stromeyer acquired great experience in control of infection, in war injuries, and also in the treatment of injuries inflicted during the construction of the canal between the Danube and the Rhine. It is to Jones (1916, 1921) that credit is due for establishing hospitals dedicated to rehabilitation; to him also goes the accolade for defining the most important principles that should be followed in reconstruction. The joints must be congruent and supple, the transferred muscle must be strong enough; the course of the transferred muscle and tendons should be direct, and the transplant should be attached with slight tension. The work of Jones and his colleagues at the great Alder Hay Military Hospital was reviewed by McMurray (1919). McMurray’s essay describes techniques still in use in anterior transfer of the hamstring muscles, the use of extensor carpi radialis longus in high median palsy and in his account of transfers for radial palsy he emphasises restoration of gliding planes and of avoidance of splinting in extreme positions. McMurray condemns trapezius transfer in the flail shoulder recommending arthrodesis instead. The failure of nerve grafting and of entubation in repair of the radial nerve is described. We use the term reconstruction to signify operations, other than by nerve repair, designed to restore function. These include: the release or correction of fixed deformities; transfer of musculo-tendinous units to restore balance across joints, similar transfer to restore lost active movement, osteotomy and, on occasion, arthrodesis or amputation. These are operations of palliation. Fixed deformity signals failure of primary treatment; arthrodesis acknowledges failure to regain active movement. No
musculo- tendinous transfer matches the outcome from good nerve regeneration. No muscle transfer reliably restores flexion and abduction at the hip or abduction at the
Fig. 13.1 Louis Stromeyer (1804–1876) (By kind permission of Dr. Mechler, of the Historisches Museum, Hannover).
R. Birch, Surgical Disorders of the Peripheral Nerves, DOI: 10.1007/978-1-84882-108-8_13, © Springer-Verlag London Limited 2011
563
564
Surgical Disorders of the Peripheral Nerves
shoulder. Those available for paralysis of the trapezius or serratus anterior muscles fall far short of normal. The principles underlying this work became clear from the treatment of deformities in the lower limb, and in particular, of deformity provoked by poliomyelitis, an upper motor neurone lesion and leprosy. Preserving or restoring sensation with early correction of fixed deformity is essential in the insensate foot or hand (Brand 1958; Antia et al. 1992; Srinivasan and Palande 1997; Srinivasan 2004). Anderson (2006) provides an admirable review of the treatment of the hand in leprosy. Sharrard (1955) analysed the relationship between peripheral paralysis and the loss of neurones in the anterior horn. Such studies help clinicians to predict deformity provoked by muscle imbalance and Sharrard’s (1971) later work provides extensive information about the early correction of such imbalance as well as the methods of correction of the established deformity. Anderson (2000) provides helpful guidance about the treatment of the upper limb severely affected by poliomyelitis. Persisting muscular imbalance deepens deformity up to, and even beyond, skeletal maturity. Continuity of care is essential. Too many Paediatric Units fail to ensure continuity and discharge patients at an arbitrary age. Our own experience is drawn largely from work with loss or disturbance of function from injury to the lower motor neurones (Table 13.1). Our experience in the treatment of cerebral palsy or brain injury is limited. Some general lessons are important. We are grateful to Mr Mark Paterson (of the Royal London Hospital) for providing us with the report prepared by SCOPE (2008). Young people with cerebral palsy in transition from paediatric to adult health services. Best practice recommendations, which was prepared by a panel chaired by Magid Bakheit, Professor of Neurological Rehabilitation to the Mount Gould Hospital in Plymouth. A number of recommendations are made: transition to the “adult service” should not be governed by
age, but by need and by the readiness of the patient; continuous service provision must be arranged. The SCOPE Report found that support services between the ages of 16–18 years are often inadequate, and that there are serious defects in enabling these patients to get up to the hospital, and in securing continuing provision for orthoses, communication aids, and seating and standing systems (Figs. 13.2 and 13.3). Whilst there is nothing so special or difficult about the standard musculo-tendinous transfers for lesions of the radial, the median or the common peroneal nerves that “referring on” is necessary, the situation is quite different after brain injury or after lesions of the spinal cord. The patterns of deformity after brain injury are complex and it is possible to make things worse by well intentioned interventions. “The worst mistake is to perform soft tissue procedures.... on patients with athetosis” (Zancolli 2005). It is a very serious mistake to focus on the limb, and it is usually the upper limb, rather than on the patient as a whole. These children do not present with such neat diagnostic labels as “shoulder and elbow,” hand,” or “foot and ankle.” A severe strain is imposed upon the family by unnecessarily frequent and poorly coordinated visits to the hospital. LeClerq (2003) insists on scrupulous diagnosis by repeated examinations and functional assessments, supplemented by video recordings. Temporary nerve or muscle block helps to distinguish between spasticity and fixed deformity and it can also be used to reinforce some muscles, to mimic the outcome of proposed operation and as an adjunct to that procedure. As much as possible should be done at one operation taking care, always, to avoid over correction. Close attention to the prevention of pain is essential, permitting early active movement. LeClerq (2003) outlines some absolute contraindications to operation which include: severe impairment of the deep and cutaneous sensory systems, the presence of dystonia or athetosis, and also
Table 13.1 3,368 operations for reconstruction: by cause and region. 1977–2006. Tetraplegia and other cord lesions
42
Spasticity from brain injury including cerebral palsy
85
Poliomyelitis
11
Lesions of the Brachial Plexus in the Adult
Peripheral Nerve Injuries Upper Limb
Peripheral Nerve Injuries PNI Lower Limb
Shoulder/shoulder girdle
149
Shoulder/ shoulder girdle
Hip
Elbow
402
Elbow
Wrist
466
Wrist
214
Ankle and foot
361
Hand
430
Flexion or extension of MCP and IP joints of fingers
240
Thumb
239
42 6
Knee
Operations for correction of scar, non union, mal-union and fixed deformity 143
305
5
210
Data about operations of reconstruction in birth lesions of the brachial plexus (BLBP) are set out in Chapter 10.
18
Reconstruction
565
unrealistic expectations by the patient or the family. As LeClerq (2003) says “the clinical picture always varies from one child to another, depending on the amount and the extent of the spasticity and the paralysis. There are no two identical cases and often they do not fit accurately into any of the described categories.”
13.1 Requirements
Fig. 13.2 A young adult with cerebral palsy developed severe pain from dysplasia of both hips which were successfully replaced (By courtesy of Mr. Aresh Hashemi Nejad (RNOH)).
Fig. 13.3 A young adult with cerebral palsy incapacitated by dislocation of the hip. Total hip arthroplasty was successful. (By courtesy of Mr. Mark Paterson (Royal London Hospital)).
There are two essential requirements. First, fixed deformity must be overcome. Next, in dynamic deformity, the causal force must be realigned and or so modified that the affected joints are rebalanced. This is of the utmost importance in the growing child or in the spastic deformities after head injury or cerebral palsy. Some of the pre-requisites for success are as follows. (1) There must be a clearly defined loss of function which can be remedied only by transfer, the prognosis for neural function being known. (2) Reconstruction in the adult after peripheral nerve injury requires a well motivated patient. Ideally, the patient defines what he or she needs from experience of work or from daily life. It is not for the surgeon to argue the case for operation, rather, it is for the patient to persuade the surgeon that there is indeed a justification for intervention. Operations for reconstruction in adult patients in legal process for compensation usually fail. (3) The treatment of continuing neuropathic pain takes precedence. (4) Poor sensation in the part is by no means a contraindication. Indeed, urgent correction of fixed deformity in the anaesthetic foot or hand is necessary to avoid ulceration and osteomyelitis (Srinivasan
566
2004). Citron and Taylor (1987) showed, from their study of transfers to improve hand function, that sensation improves after the operation. Dahlin et al. (1998) recorded improvement in stereognosis in 36 cases of cerebral palsy in the upper limb after operation and Eliasson et al. (1998) improved sensory function by rebalancing the spastic hands of 32 children. The muscle to be transferred should be dispensable: as our colleague Donal Brooks said “all transfers rob Peter to pay Paul.” The price should not be too high. It is too easy to replace one deformity with another. The forces acting across a joint must be rebalanced to avoid causing new problems, such as a lateral rotation deformity at the shoulder after transfer of the medial rotator muscles (Fig. 13.4), pronation deformity in the forearm after re-routing of biceps, hyperextension of the wrist after flexor to extensor transfer, hyperextension of the knee
Fig. 13.4 A 6 year old child with group IV birth lesion of the brachial plexus (BLBP) in whom transfer of latissimus dorsi and teres major to infraspinatus induced severe lateral rotation deformity.
Surgical Disorders of the Peripheral Nerves
after anterior transfer of the hamstring muscles and varus or valgus deformity of the heel after muscle transfer to the evertor or invertor muscles. At least one adequate motor must always be retained to balance the joint. Some muscles are indispensible: trapezius, serratus anterior, extensor carpi radialis brevis (ECRB) and abductor pollicis longus (APL) are examples. The muscle should be of sufficient power. When the indication for transfer is to restore extension of the knee or ankle, nothing short of full power will do. Re-innervated muscles are predictably unreliable. They may be weak; there is often a degree of co-contraction after repairs of the brachial plexus; there is, in particular, deficiency of afferent function of proprioception and muscle spindle control so that retraining is difficult or impossible. Beevor (1904) showed that the sensorimotor cortex controls movements rather than individual muscles and this permits the conversion of antagonists to agonists. It is common to see patients actively extending the knee or the ankle and toes as soon as the post operative splint is removed after hamstring to quadriceps transfer or anterior transfer of tibialis posterior. This conversion cannot occur if the muscles are “blind,” if there is a defect in the deep afferent pathways from the muscle spindles and the tendon receptors. The skeleton must be stable and joints must be congruent (Fig. 13.5). A split skin graft, or the deep scars following sepsis present an insuperable barrier to the transferred muscles. Only healthy full thickness skin will serve. The surgeon does need to know how to do the operation. It should be done cleanly and with kindness. Gentle handling of the tissues and meticulous haemostasis are as important in this as in any other operation. Pain must be reduced or prevented by infusion of local anaesthetic along the line of incision, and through an indwelling catheter. The preferred plane of passage is subcutaneous, in fatty areolar tissue, through unscarred tissue, and preliminary operations to improve skin may be necessary (Fig. 13.6). Brand (1987) used the term “drag” to sum up all those factors offering soft tissue resistance and friction to the transfer. He argued persuasively for preserving paratenon where possible, avoiding contact with naked bone or scarred fascia and using gentle tunnelling technique to provide an investing layer of living fat or loose areolar tissue. Compromise may be necessary; some muscles such as flexor carpi ulnaris, pronator teres or brachioradialis have lengthy attachments to periosteum or fascia so that considerable dissection is necessary to mobilise them. Others, such as the tendons of flexor digitorum superficialis or extensor indicis, can be detached from their insertion and re-routed through small incisions with a minimum of dissection. Tendon suture is preferred over other methods of insertion. The weak muscle may recover; the union of tendon to tendon is stronger than that of tendon to bone. When a tendon is passed around another to create a pulley neither
Reconstruction
Fig. 13.5 Infected non-union of humerus was treated by vascularized fibular graft before proceeding to muscle transfers for the associated high median palsy.
should be scarified. The paratenon of both should always be respected. The most important part of the operation is the splinting and post operative care, and this is the responsibility of the surgeon (Fig. 13.7). Swelling must be avoided and only those joints which require immobilisation should be splinted. Thereafter a balance is sought between risking the new attachment by premature or over vigorous force and too prolonged immobilisation leading to extensive scar formation. Protection of the transfer needs to be prolonged following such transfers as tibialis posterior for dorsiflexion of the ankle or hamstring transfer for knee extension, when some support is necessary for, at least, 3 months. We use dynamic or static functional splints wherever possible before operation and re-apply these as soon as possible after (Fig. 13.8). Some patients with foot drop, radial palsy, weakness of elbow flexion or weakness of the thoraco-scapular muscles prefer orthoses to operation.
567
Fig. 13.6 Flexor carpi ulnaris tunnelled subcutaneously to the dorsum of the wrist. Note the relation to the dorsal branch of the ulnar nerve.
Fig. 13.7 The first post operative splint for flexor to extensor transfer. The distal interphalangeal joints are free.
In the early stages of retraining it is often useful to introduce the muscle to its new function by mimicking the old: pectoralis major to serratus transfer is retrained by the patient strongly adducting the arm. This is a useful manoeuvre after
568
Surgical Disorders of the Peripheral Nerves
Fig. 13.8 Simple functional splinting. Above illustrates the use of a combined “knuckle duster” and opposition splint for a patient with a lesion of the median and ulnar nerves. The “lively” splint confers flexion of the metacarpophalangeal joints. At rest the joints are kept flexed; active extension is, however, possible. (Courtesy of Miss Lydia Dean Dip COT) Below the use of a combined “lively” and opposition splint in a patient with progressive neuropathy. (Courtesy of Mrs. Kathryn Johnston DipCOT).
elbow flexorplasty by transfer of pectoralis minor or latissimus dorsi. Transfers for wrist extension are first encouraged by clenching the hand.
13.1.1 The Tension of Transfer Standring (2005) offers a helpful description of muscle contractility. A muscle fibre consists of a number of myofibrils, about 1mM in diameter, which are regularly divided into
linear series of contractile units, the sarcomeres. These are defined by the Z-discs which anchor the thin (actin) filaments. The human sarcomere is 2.2 mM in length and will shorten, at the most, by 1 mM. The tension developed by the sarcomere reduces as it is stretched and also as it become shorter: “The force developed by an active muscle is the summation of the tractive forces exerted by millions of crossed bridges as they work asynchronously in repeated cycles of attachment and detachment. This force depends on the amount of contractile machinery that is assembled in parallel, and therefore on the cross sectional area of the
Reconstruction
muscle. The range of contraction generated by an active muscle depends on the relative motion that can take place between the overlapping arrays of thick and thin filaments in each sarcomere…. Since the sarcomeres are arranged in series, the muscle fibres shorten by the same percentage. The actual movement that takes place at the end of the fibres will depend on the number of sarcomeres in series, i.e., it will be proportional to fibre length” (Standring 2005). Frídén and Lieber (1998) measured sarcomere lengths of transferred muscles by projecting helium neon laser beams through muscle fibres onto a prism. The average sarcomere length was 3.78 mM in most of the muscles studied, implying that too much tension had been applied during suture. Some muscles had been so lengthened that no force could be generated by the transferred muscle. Lieber et al. (2005) showed, in an intra-operative study of brachioradialis transfer for high cervical cord lesion, that some muscles are transferred under too much tension: “the muscle generates near optimal tension at a slack sarcomere length where passive tension is zero.” Three dimensional computerised models have been used to study different muscle transfers. Saul et al. (2003) showed that zealous proximal advancement of the flexor origin during elbow flexorplasty interfered with elbow extension, wrist extension and with supination. Undue tension must be avoided in operations designed to improve the posture of the thumb and it is advisable to splint the thumb web space open after transfers for adduction, flexion or opposition. One of the advantages of performing these operations under local anaesthetic is that the excursion of the transferred tendon can be demonstrated to the surgeon by the patient. We use the following method for transfer of flexor carpi ulnaris to the dorsum of the wrist. The distance between the pisiform and the medial epicondyle is measured with the wrist in neutral. The tendon is detached from the pisiform, the muscle mobilised, and passed to the dorsal compartment of the forearm. The transferred tendon is sutured so that the muscle is at the same length as it was before detachment.
569
in birth lesions of the brachial plexus. (2) Post ischaemic fibrosis of muscle. The importance of this is frequently underestimated; it is seen particularly involving the compartments of the leg the flexor muscles of forearm and the small muscles of hand. (3) Untreated pain may cause severe flexion deformity of the wrist with extension of the metacarpophalangeal joints and flexion of the proximal interphalangeal (PIP) joints. Fixed flexion deformity at the hip, the knee and the ankle is a common complication of persisting pain. (4) Neglect. It must be said that many cases of fixed deformity are a reflection of elementary principles in the treatment of paralysed limbs. Two of these are particularly common. They are the fixed extension deformity of the metacarpo-phalangeal (MCP) joints of the hand and flexion deformity in the lower limb marring the function of hip, knee and ankle. We shall elaborate on two of these causes.
13.1.2.1 Persisting Imbalance During Growth Our own experience from peripheral nerve lesions is illustrated by some cases (Figs. 13.9–13.11). These illustrate the importance of progression of deformity in the growing hand and foot. Forward planning is needed, based on a clear understanding of the prognosis of the nerve lesion. This is particularly important when the object of operation is the restoration of active movement rather than the rebalancing of forces acting across a joint with the object of preventing progressive deformity. It is all too easy to transfer muscles prematurely and be faced with the problem of
13.1.2 Fixed Deformity The chief causes of fixed deformity of such severity that active treatment is necessary include: (1) The unopposed action of muscles during growth. Dislocation of the hip from unopposed action of the flexor and adductor muscles in cases of cerebral palsy or spina bifida, is a severe, often avoidable, complication. The effect on the posture of the foot after an irreparable tibial or common peroneal lesion in the growing child is severe. We have already seen the effects of muscle imbalance at the shoulder
Fig. 13.9 The foot of a 6 year old girl in whom a lesion of the common peroneal nerve went unrecognised and untreated for 3 years.
570
Surgical Disorders of the Peripheral Nerves
13.1.2.2 Post Ischaemic Fibrosis
Fig. 13.10 The foot of a 17 year old girl. Innominate osteotomy for hip dysplasia, performed at the age of 4 years, damaged the sciatic nerve.
“In the absence of peripheral arterial pulses take immediate steps to relieve possible pressure on a main artery by reducing any displaced fracture or dislocation and, if still necessary, exploring the artery. With failing nerve or muscle function in a tensely swollen limb, do not be deceived by the presence of a good peripheral pulse – consider immediate decompression.” (Parkes 1973a). It is extremely important to distinguish between the paralysis of the small muscles of the hand or foot caused by degenerative lesions of nerves, whether from ischaemia or compression or both, from ischaemic contracture of these muscles. The prognosis for recovery of function is determined by the fate of the main nerves (Figs. 13.12–13.14). Nerves are a good deal more resistant to ischaemia than muscles. It has long been our policy to define the prognosis for the nerves by exposure and decompression of the median, the ulnar, and the tibial nerves. Many patients experienced early relief of pain, later followed by improvement of sensation, of vaso and sudomotor function and recovery of the small muscles, notably of those innervated by the tibial and ulnar nerves. This may be a reflection of the fact
Fig. 13.11 The foot of an 8 year old girl with plexiform neurofibroma of the lumbo sacral plexus.
over correction. The long duration of myelination, so that conduction velocity approaches adult levels only after 5 years, shows that the deep afferent pathways are immature and possibly unstable. The prolonged evolution of hand control has been related in Chapter 10. For these reasons, muscle transfers which seek to restore active movements across the joint are best avoided until after the age of 5 years. These considerations do not apply if the object of operation is to prevent progression of deformity. This deformity can be predicted by sure knowledge of the prognosis for the injury to the nerve.
Fig. 13.12 Post ischaemic fibrosis of the deep flexor compartment in a 31 year old man. The tibial nerve and artery were entrapped in the fracture of the distal third of the tibia and were further damaged by the passage of an intramedullary nail.
Reconstruction
571 Table 13.2 372 operations for post ischaemic contracture: 1977–2006. Location Number Cause (number of limbs) of contracture of limbs
Fig. 13.13 Examples of post ischaemic fibrosis of the anterior compartment of the leg complicating intramedullary nailing. Above, showing contracture of extensor hallucis longus; below, the contracture involves extensor digitorum longus and extensor hallucis longus.
Shoulder
15
Crush or fracture Coma
8 7
Arm
14
Crush or fracture Coma
7 7
Forearm
134
Hand
55
Fracture or fracture dislocation at elbow Other fractures Drug causes Injection Proximal arterial Injury
53 30 18 10 78
Hip, thigh and knee
11
Crush Coma
6 5
Leg and foot
143
“Compartment syndrome” Arterial injury
73 70
that the ulnar neurovascular bundle runs in a discrete fascial compartment and is separate from the main body of the flexor muscles. The tibial nerve is similarly situated. We agree with Gülgönen (2001) who drew from his great experience in the aftermath of the Kocaeli earthquake: “I have not encountered a single Volkmann’s contracture case where the nerve tissue has not been affected to some extent” (Table 13. 2). We have seen an alarming increase in the numbers of patients referred with post ischaemic fibrosis of the muscles of the leg and the forearm. Evidently Parkes’ excellent advice is not being followed: Gülgönen (2001) reinforces that advice: “in short, the clinical picture, with pain, paraesthesia and paralysis, ought to be given precedence; and when a patient is examined in the light of this possibility, there is little chance that a compartment syndrome will be missed.” We believe that free muscle transfer is a better way forward than conventional tendon transfer in the most severe cases of post ischaemic fibrosis of the muscles of the forearm. Electromyography is useful in such cases. Insertion of the concentric needle into an irretrievably fibrosed muscle is met with characteristic gritty resistance. There is scanty or no electrical activity.
13.1.3 Correction of Fixed Deformity 13.1.3.1 Serial Splinting
Fig. 13.14 Volkmann’s ischaemic contracture of forearm and intrinsic muscles. The flexion deformity of the wrist has been over corrected by flexor muscle slide. The metacarpophalangeal joints are fixed in flexion; passive flexion for the proximal interphalangeal joints was impossible.
Sister Evelyn Hunter RGN, Miss J Laverty MCSP and Donal Brooks developed, over a number of years, an extremely useful technique for the treatment of fixed deformities throughout both upper and lower limbs. More than 1,000 such deformities involving the elbow, the wrist, the digits, the knee and the ankle have been corrected without operation.
572
Surgical Disorders of the Peripheral Nerves
were necessary, the time taken to achieve the result was 42 days. Adams (1966) described his method of correction of fixed deformity at the hip by “well leg” traction. The technique requires skill, and close attention to detail. All interested clinicians should read this outstanding monograph.
13.2 Principles of Operation
Fig. 13.15 Below: Fixed flexion deformity of the proximal interphalangeal joint following fracture. Above: A full range of movement was restored after three treatments with serial splinting.
Wynn Parry and his colleagues, first at R.A.F. Chessington and later at the Royal National Orthopaedic Hospital used similar techniques, achieving excellent results (Wynn Parry CB 1981) Carefully moulded plaster of Paris splints are applied and bandaged into position. After a few days the joints are stretched and new splints are fashioned. The technique demands a very high level of skill in the safe application of plaster of Paris splints, often onto insensitive and atrophic skin. Hunter et al. (1999) used the method for correction of the common and vexing deformity of flexion of the proximal interphalangeal joints of the fingers (Figs. 13.15 and 13.16). Seventy nine cases of flexion deformity of proximal interphalangeal joints were treated. The mean deformity before treatment was 85 degrees; after treatment this was reduced to 15 degrees. The duration of treatment ranged from one session to forty two; on average, six plaster changes
Tenotomy is indicated when the muscle is irretrievably damaged as a prelude to conventional transfer or as the first step in free functioning muscle transfer. Step elongation of the tendons is indicated when the muscles are not wholly destroyed. It is particularly valuable in the correction of equinovarus deformity, or fixed flexion or extension of the toes. It may be indicated in cases of post ischaemic fibrosis confined to the superficial or the deep flexor muscles of the forearm. Post ischaemic fibrosis of the muscles of the anterior compartment of the leg is more difficult. Occlusion of the anterior tibial artery causes necrosis of most of the tibialis anterior muscle but offsets from the peroneal artery ensure partial survival of the extensor muscles of the toes. In compartment syndrome, on the other hand, extensor hallucis longus (EHL) and extensor digitorum longus (EDL) are often worse affected than tibialis anterior. These muscles may prove insufficiently strong to work against gravity after tendon elongation and some patients, who have coped reasonably well, find themselves with the foot drop. Muscle slide is particularly useful when the shortened muscle remains powerful. Cavus deformity of the foot is regularly improved by release of the intrinsic muscles from the heel. Muscle slide is the operation of choice for the shortened adductor pollicis or pronator quadratus. Forearm flexor muscle slide is a considerable undertaking and there is a risk of over correction if the muscles are more seriously damaged than was appreciated. Osteotomy is useful to restore congruency of joints or to
Fig. 13.16 Closed fracture of shaft of femur led to post ischaemic fibrosis of the flexor muscles of the thigh in a 14 year old boy. This was fully corrected by two serial plaster of Paris splints. Note the position of protective padding.
Reconstruction
573
enhance their stability. The place for osteotomy in correction of posterior dislocation of the shoulder, or for supination deformity of the forearm, has been described in Chapter 10. If a deformity was provoked by muscular imbalance then it is likely to recur if that imbalance has not been corrected. It is important to remember that no arthrodesis should be performed in the growing limb unless the muscular imbalance provoking that deformity has been corrected. Indeed, deformity may recur even after skeletal maturity if muscular imbalance persists. We now describe four valuable operations in the treatment of fixed deformity. These have been indicated most often for post ischaemic fibrosis. They have also been used for severe spasticity from head injury or cerebral palsy, and for the treatment of joints contracted because of severe pain.
13.2.1 The Equino-varus Deformity of Ankle and Foot A 34 year old man presented with pain, a degenerative lesion of the tibial nerve and deformity after fracture of the proximal tibia. Below knee amputation had been recommended (Fig. 13.17). The contracture involved not only the flexor muscles of the heel but also the flexor muscles of the toes. The tibial neurovascular bundle was exposed through a postero-medial incision. The tibial artery was patent. The tendo-calcaneus, and the tendons of tibialis posterior, flexor hallucis longus and flexor digitorum profundus were all lengthened widely permitting the foot to come into a plantigrade posture. None of these muscles was completely infarcted. There was early relief of pain; functional plantar flexion of the heel and toes was evident at 1 year. Similar operations have been performed in 83 patients. A tourniquet is not used. The skin flaps must be handled tenderly; perforating vessels may have been damaged by the injury. The equinus deformity is corrected by elongation of the calcaneal tendon and inversion deformity by elongation of tibialis posterior. Then follows, as necessary, step lengthening of flexor digitorum longus and flexor hallucis longus. The ankle, the foot, and the toes are now held by an assistant in the desired plantigrade position and the tendons are repaired. The tibial nerve is widely exposed. It is always compressed; sometimes the epineurial vessels are obliterated. Decompression of the tibial nerve also releases constriction of the posterior tibial artery and improved flow through the vessel is usual. The contracted small muscles of the foot may require release from the heel. Infusion of local anaesthetic, through a catheter with its tip placed close to the proximal segment of the nerve, reduces postoperative pain. The correction is secured by a plaster of Paris slab which is completed at 2 weeks and the patient must be non weight
Fig. 13.17 Lateral view of left leg of a 34 year old man, 3 years after tibial fracture with ischaemic changes in the flexor muscles. Note the scar of the fasciotomy and the equino varus deformity of the foot.
bearing for 3 weeks. The splint can usually be discarded at 6 weeks. Patients are encouraged to move their toes as early as possible.
13.2.2 Flexor Muscle Slide, by the Technique of Brooks (1984) The operation is indicated where there is potential for function after partial recovery of long flexor muscles to at least MRC Grade 3. Exposure and decompression of the ulnar and median nerves is an essential component of the operation in cases of post ischaemic fibrosis. It is also useful in selected cases of spasticity following head injury or cerebral palsy. The brachial artery must be patent.
574
Surgical Disorders of the Peripheral Nerves
Fig. 13.18 Flexor muscle slide. Above: the incision. Below: the nerves and vessels exposed.
Ulnar n.
Brachial a.
Ulnar n.
Median n.
With a suprasystolic cuff in place an incision is made along the ulnar border of the forearm and lower arm from the wrist to above the elbow. The ulnar and median nerves are traced at the elbow, and the brachial artery is displayed (Fig. 13.18). The ulnar nerve is decompressed where it enters the cubital tunnel and dives deep to the flexor carpi ulnaris. The plane of dissection now moves distally, displaying the ulnar origin of the interosseous membrane. The flexor muscles are released from the ulna and radius, displaying and protecting the anterior interosseous nerve and vessels. At the proximal end of the incision the ulnar nerve is transposed forward, whilst detaching the superficial flexor origin from the medial epicondyle of the humerus so that the whole muscular mass displaces distally. It is important to release the radial origin of flexor pollicis longus and flexor digitorum superficialis and at times detachment of pronator quadratus from the ulna is also necessary. After release of the cuff and haemostasis the skin is closed and the forearm and hand are immobilised in a position of extension of wrist and digits. An indwelling catheter to permit the infusion of local anaesthetic, and placed adjacent to the median and ulnar nerves at the elbow helps to diminish post operative pain. The splint is removed at 6 weeks, when a resting or night splint is
Anterior interosseous a. and n.
re-applied in between periods of passive and active use of the wrist and of the digits. This operation displays and decompresses the arteries and the nerves whilst preserving muscular function. It has been performed on 42 occasions. Over correction occurred in four of these because the muscles were more damaged than expected. There is always some recurrence of deformity in the growing child (Fig. 13.19).
13.2.3 Fixed Extension at the Metacarpo-phalangeal Joints This operation was described by Campbell-Reid (1984). We have used it on 86 occasions and have found it consistently reliable in restoring a reasonable passive range of movement in the metacarpophalangeal joints of the hand. It is important not to intervene too soon. The inflammatory phase must have settled; there should be no swelling (Fig. 13.20). The MCP joints are approached through linear incisions centred between the metacarpal heads. The extensor tendons are lifted up from the capsule which is then incised transversely.
Reconstruction
575
Fig. 13.21 A useful static splint for the maintenance of correction in between periods of activity.
Fig. 13.19 A salutary tale. A 9 year old boy sustained a fracture of the distal humerus, which was treated by complete above elbow plaster of Paris splint. There was entrapment of the brachial artery and median nerve within the fracture. The ischaemic contracture of the forearm flexor muscles required a flexor muscle slide; the contracture of the small muscles of the hand was corrected by a series of operations, including Littler’s release, release of deep origin of the adductor pollicis slide of the first dorsal interosseous, and Z-plasty of the thumb web space. Neurological recovery was good. The hand shown at the age of 14 years.
Fig. 13.20 Release of the metacarpophalangeal joints.
This may be all that is required; in more severe cases division of the collateral ligaments as described by the originator is necessary. The joints are now flexed to 90 degrees or more. There is a tendency for the extensor tendons to displace from their proper position in which event a few fine absorbable sutures to the remnants of the capsule may prove necessary. The skin is closed and the position held with fine Kirschner wires
passed from the dorsum of the head of the metacarpal into the proximal phalanx. The hand is immobilised in two separate plaster of Paris splints. The wires are removed at between 2 and 3 weeks. The joints, however, are held in the position of flexion for a total of 6 weeks between spells of active and passive work (Fig. 13.21). It is a mistake to use a transverse incision to expose the joints, for it may be impossible to close the skin without tension. Close supervision of the maintenance of the position and throughout the period of mobilisation by the surgeon is essential.
13.2.4 Release of Contracted Small Muscles of the Hand Finochietto (1920) was probably the first surgeon to recognise the cause and effect on the hand of ischaemic contracture of the intrinsic muscles. Flexion at the proximal interphalangeal joint (PIPJ) is reduced when the metacarpophalangeal joint (MCPJ) is extended. However, when the MCPJ is flexed, flexion at the PIPJ is increased as illustrated in Fig. 13.22, which was drawn by, and kindly provided by Mr Donald Sammut (Bristol). Littler’s (1949) release is indicated in this situation. A dorsal linear mid line incision over the proximal phalanx permits exposure of the intrinsic lateral band and the interosseous triangular laminae. These are excised (Fig. 13.23). The hand is protected in a splint which holds the MCPJ in flexion; early active and passive flexion of the interphalangeal joints is encouraged. The operation is contraindicated when there is fixed flexion deformity of the metacarpophalangeal joints. In such cases it is usually necessary to perform extensive intrinsic release by cutting the interosseous tendons, the accessory collateral ligaments and releasing adhesions within the MCP joints (Gülgönen 2001). Adduction contracture of the thumb is common. Operations designed to restore active movements of the thumb will fail unless this is corrected. If adductor pollicis is still active it is approached through a palmar incision and detached from its origin from the middle metacarpal. In more severe cases we favour the operation of Brooks (1984) who recommended a
576
Surgical Disorders of the Peripheral Nerves
Z-plasty for the skin of the thumb web space combined with subperiosteal release, from their origin, of the first dorsal interosseous and adductor pollicis longus muscles. Rotation flaps or split skin grafts are sometimes needed (Gülgönen 2001; Srinivasan 2004). The thumb web space is splinted open for 3 weeks.
13.3 Methods of Reconstruction The priorities in reconstruction in the upper limb are, in order, restoration of: (1) stability of the scapula; (2) flexion and rotation at the shoulder; (3) extension and flexion at the elbow; (4) extension of the wrist; (5) flexion of the MCP joints of the fingers and adduction of the thumb and (6) opposition or prehensile grip between the thumb and tips of the fingers, that is, between the “eyes of the hand” (Moberg 1975). We think that, in the lower limb, the priorities are: (1) a stable hip; (2) extension of the knee; (3) dorsiflexion of the ankle; and (4) a balanced, supple, plantigrade foot.
13.3.1 The Shoulder Girdle 13.3.1.1 The Thoraco-scapular Joint Fig. 13.22 A test for contracture of the interosseous muscles. Drawing prepared and provided by Mr. Donald Sammut (Bristol).
Fig. 13.23 Excision of part of the interosseous hood. (Courtesy of Mr. Donald Sammut).
The spinal accessory nerve is particularly vulnerable to the attentions of surgeons and the consequences for the patient are crippling. The loss of movement at the thoraco-scapular and gleno-humeral joints is usually complicated by severe pain so that the disability is greater than that seen after transection of other main nerves in the upper limb. The consequences of lesions to the nerve to serratus anterior are nearly as severe. Results of repair of both of these nerves are generally rather good and the effects of delay are less severe than they are in most other nerve injuries. Muscle transfers to restore thoraco-scapular and gleno-humeral control are, at best, mediocre. Every reasonable effort must be made to reinnervate these muscles. Transfer of the Levator Scapulae and Rhomboids for Paralysis of Trapezius. Failure is guaranteed unless serratus anterior is strong and the gleno humeral joint is supple (Eden 1924; Lange 1951). The patient is prone. Through an incision one fingers breadth medial to the vertebral margin of the scapula, the paralysed trapezius is detached, to expose the levator scapulae and rhomboids. These are defined and mobilised, taking care for the dorsal scapular nerve (Fig. 13.24). The muscles are detached with an osteoperiosteal flap, respecting the insertion of serratus. The rhomboids are advanced by about 3cm deep to infra-spinatus and sutured to the scapula.
Reconstruction
577
The second incision exposes the insertion of trapezius in the spine of the scapula: levator scapulae is brought into this wound by a tunnel deep to trapezius and sutured to the lateral margin of the spine (Fig. 13.25). The supraspinatus muscle
and suprascapular nerve must be protected. The trapezius and infraspinatus are securely re-attached and the limb immobilised in a spica with the shoulder at 90 degrees of abduction for 6 weeks. Then, an abduction bolster is supplied. Passive and active movements of the gleno-humeral joint are encouraged. At 9 weeks a broad arm sling is worn and gentle active elevation and protraction of the scapula is introduced. Our results, from six cases, are at best moderate. These patients were in pain from the original injury to the accessory nerve and two had developed significant capsulitis of the shoulder. Bigliani et al. (1985) found good results with pain relief in six from seven cases. Romero and Gerber (2003) restored good function in nine from 16 cases. Transfer of the Sternal Portion of Pectoralis Major for Paralysis of Serratus Anterior. This is a useful operation in the treatment of the irretrievably paralysed serratus anterior and it does seem to ease the pain from the unstable scapula more reliably than the operation described above for trapezius palsy (Féry and Sommelet 1987). The patient is supine. The pectoralis major is exposed through a delto-pectoral incision (Fig. 13.26). The plane between the sternal and clavicular portion is defined and separated, respecting the medial pectoral neurovascular pedicle. The tendon of the sternal portion is the deepest of the three laminae and it is detached from the humerus. Now, through an incision in the mid axillary line, the scapula is exposed anterior to the thoraco-dorsal pedicle, and it is brought into the wound by a stout suture.
Fig. 13.25 Muscle transfer for trapezius paralysis: completed.
Fig. 13.26 Transfer for paralysis of serratus anterior. The incision.
Fig. 13.24 Muscle transfer for trapezius paralysis: the incisions.
578
A capacious tunnel in the axillary fascia is prepared so that the sternal pectoralis major glides easily over pectoralis minor and ribs . It is sutured to the lower pole of the scapula using an interposed tendon or fascia lata graft (Fig. 13.27). The arm is immobilised across the chest for 6 weeks. Passive and active abduction to about 40 degrees with lateral rotation to neutral is then begun and at 9 weeks from operation a more vigorous programme is followed with the aim of restoring a full range of abduction and lateral rotation by 12 weeks. At 12 weeks vigorous resisted work can begin. Narakas (1993) found some improvement in five cases after various muscle transfers for paralysis of serratus anterior. Nine from 13 of our patients reported improvement of function and some relief of pain at about 1 year from operation. In one patient the transferred tendon was torn from its new insertion following an injury; her symptoms recurred. Re-attachment of the detached muscle was possible; there was improvement.
Fig. 13.27 Paralysis of serratus anterior: completed.
Surgical Disorders of the Peripheral Nerves
Thoraco-scapular Arthrodesis. Fortunately, the indications for thoraco-scapular arthrodesis are rare, for this operation is a trial for surgeon and patient alike. The largest experience is that of Copeland and Howard (1978) who used it with great success in the treatment of patients with scapulohumeral dystrophy. Copeland (1995a) describes the operation fully; further description is superfluous. Diab et al. (2005) found good results following 11 operations in eight patients. We have used it in two cases for combined palsy of the eleventh cranial nerve and the nerve serratus anterior.
13.3.1.2 The Gleno-humeral Joint “There are no efficient musculo-tendinous transfers to replace the loss of function in the spinatii and deltoid.” (Narakas 1993). Gerber et al. (2006) studied 67 patients with ruptures of the rotator cuff who were treated by latissimus dorsi transfer. Detectable improvement was seen only when the subscapularis muscle remained intact. Ianotti et al. (2006) examined, by electromyography, latissimus dorsi muscle transferred for irreparable tears of the posterior superior rotator cuff. Magnetic resonance imaging (MRI) excluded atrophy of the transferred muscles but in only one of these was activity confirmed, by electromyography, in forward flexion. There was no electrical activity during lateral rotation at 90° of abduction in any patient. These studies indicate the importance of co-ordination of the numbers of muscles active across the gleno-humeral joint and the limitations of palliative muscle transfers. However, the situation may be different in cases where there is only partial denervation of muscle acting at the shoulder. Alex Lara y Muset (Raimondi et al. 2001) has shared his extensive experience derived from meticulous anatomical and clinical studies. He, with his colleagues, showed that transfer of the latissimus dorsi to the supraspinatus tendon significantly improved function in 43 cases of Birth Lesion of the Brachial Plexus (BLPP) where there was some shoulder function. However, in 27 children with flail shoulders treated by combined transfers of trapezius, latissimus dorsi and teres major the shoulder was so stabilised that it acted as a functional ankylosis. Function deteriorated as the children approached maturity. We have no experience of trapezius advancement (Yücetürk (2001)) to restore abduction. Narakas (1993) criticised the operation because of the disturbance of the insertion of the lateral part of deltoid and the risk of impingement upon the supraspinatus tendon. Jamieson (2003) found only little improvement in his series. Transfer of Latissimus Dorsi and Teres Major for Lateral Rotation of Shoulder. We follow the technique described by Roper (1972) (L’Episcopo 1939; Zachary 1947). This operation can be considered when the suprascapular nerve is irretrievably damaged. It is much less effective when deltoid is paralysed. It must not be used if
Reconstruction
579
Fig. 13.29 Transfer of latissimus dorsi and teres major: completed.
Fig. 13.28 Transfer of latissimus dorsi and teres major: the incision.
subscapularis is weak since the patient may lose forward flexion and medial rotation. The shoulder must be supple. It does no more than give the patient a sense of stability if the shoulder is flail. Pearle et al. (2007) provide an excellent description of the operation. The patient is placed on the side. The incision starts behind the acromion and follows the posterior border of the deltoid to curve forward into the axilla over the conjoined insertion of latissimus dorsi and teres major. It is extended inferiorly along the anterior margin of latissimus dorsi (Fig. 13.28). The tendon of that muscle is defined and followed to its insertion on the shaft of the humerus, where it is joined by teres major. The radial nerve, which crosses the anterior surface of the tendon, is seen and protected. The muscles are separated and both are released from the humerus. The posterior border of the deltoid is retracted to display the tendon of infraspinatus. The nerve stimulator is helpful in the identification of the circumflex nerve where it emerges from the quadrilateral tunnel. Lateral rotation reveals the supraspinatus tendon and the latissimus dorsi is passed deep to the deltoid and interwoven into it. Teres major is sutured to the tendon of infraspinatus (Fig. 13.29).
We no longer use a plaster of Paris spica for immobilisation but prefer an abduction bolster. This is removed at 6 weeks and gentle active lateral rotation is encouraged. By 9 weeks unrestricted active lateral rotation is permitted. Unrestricted activities follow from 12 weeks. Ross and Birch (1993) reviewed 22 of our cases. Two thirds of them found an improved sense of stability in the shoulder, one half regained active lateral rotation to a maximum of 20 degrees. None regained useful active abduction. The operation has been performed in 70 adults. Most recorded a sense of improved stability. Only a minority regained active lateral rotation (Fig. 13.30). Gleno-humeral Arthrodesis. There are occasions when a patient presents with such instability of the joint that arthrodesis is worthy of consideration as a method of transferring scapular function to the limb as a whole, and for the treatment of pain. The trapezius and serratus anterior muscles must be strong. We agree to three indications. (1) The flail shoulder causing pain because of subluxation; (2) the well motivated patient who uses a prosthesis after amputation of the upper limb for complete, untreatable and irreversible paralysis; (3) the patient with C5/6 or C5/6/7 palsy who has a useful hand, is in work, and needs a stable platform to use the hand. The lengthy debate about the position of arthrodesis is answered simply by considering the purpose of operation – it is to enhance the function of the limb. The patient should retain or regain the grip function between the arm and the chest and
580
Surgical Disorders of the Peripheral Nerves
We have performed 91 arthrodeses. There was no case of sepsis, there were two failures of fusion and one patient fractured the humerus in a subsequent fall (Fig. 13.31).
13.3.2 Extension of the Elbow Moberg (1975) emphasised the potential for valuable reconstruction in tetraplegics with sparing of C5 and C6. These patients have good control of the shoulder, flexion of elbow and extension of wrist and, perhaps most important of all, good sensation in the thumb, the index and the middle fingers. Moberg and Lamb (1980) clarified priorities in reconstruction, pointing out the benefit of active extension of the elbow and showing that this should take priority over tendon transfers within the hand. Lamb (1987) and Allieu et al. (1985) describe in detail the operation of transfer of the posterior one third of deltoid to permit active extension of the elbow. We have followed Lamb’s technique on six occasions but use the tendon of tibialis anterior to join the deltoid to triceps, as recommended by Landi et al. (1998). All patients regained full active extension of the elbow against gravity and three of them, who had a mild hyperextension posture of the elbow, were able to transfer from bed to chair. On two occasions we have used latissimus dorsi transfer for elbow extension in patients with injuries to the brachial plexus. The results were not impressive. Mulcahey et al. (2003) transferred biceps in eight tetraplegic patients in whom brachialis and supinator were strong. The results matched those from eight more patients treated by transfer of posterior deltoid with surprisingly little difference in flexion torque. Creasey and Keith (1996) offer a review of their approach to the upper limb in tetraplegia which is informed, extensive and stimulating.
Fig. 13.30 Some examples of latissimus dorsi and teres major transfers at the shoulder. Above, a 10 year old child, with BLBP Group 3. (Middle) A 17 year old girl with Group 3 BLBP. Both of these patients required relocation of posterior dislocation of the shoulder in earlier years. (Below) A 20 year old man with partial Brown Séquard lesion from knife wound to the cervical spinal cord. Pectoralis minor was transferred to biceps, latissimus dorsi to the supraspinatus and teres major to infraspinatus. Active flexion of elbow and some lateral rotation at the shoulder was regained but there was no active abduction at the shoulder.
should be able to bring the hand to the mouth. This gives us the desired position of between 20 and 30 degrees for abduction and more or less a similar angle for medial rotation and forward flexion. Copeland (1995b) uses a 10-hole pelvic reconstruction plate. Chammas et al. (1996) prefer combined external and internal fixation whilst Nagano et al. (1989) choose external fixation alone and theirs is a very large series.
13.3.3 Flexion of Elbow The number of operations described to restore the apparently simple movement of elbow flexion suggests that not one of them is particularly good. All fall short of successful reinnervation. Marshall et al. (1988) reviewed 50 of our cases, followed for up to 7 years. Their findings are disturbing, for in less than one half was there significant improvement in overall function. Five patients demonstrated a real increase in the active use of the damaged limb after the operation. Latissimus dorsi proved the best of all transfers for the range of elbow flexion, and for regaining supination. Triceps to biceps transfer achieved the greatest improvement in power but at the highest cost in function. Pectoralis major transfer was disappointing. Marshall and his colleagues thought that 27 patients
Reconstruction
581
Fig. 13.31 Before (above) and after (below) gleno-humeral arthrodesis in a 31 year old woman who suffered rupture of C5 with avulsion of C6, C7 and C8. Elbow flexion was restored by Oberlin’s operation, recovery through the graft of C5 and through accessory to suprascapular transfer was poor. The position of the fusion enables her to bring her hand to her mouth and maintains adduction.
had a good result: elbow flexion of MRC 4 or better. Fourteen more had a fair result; elbow flexion MRC 4 with no useful shoulder stabilisation, or MRC 3 with shoulder stability. The average power of elbow flexion was 1.5Kg. Those who achieved up to 8Kg or better had some biceps function, or had triceps to biceps transfer. The latissimus dorsi, pronator teres (PT) and flexor carpi radialis (FCR) are weakened if the 7th cervical nerve is damaged. The sternal head of pectoralis major and pectoralis minor are substantially denervated if the lesion includes C8 as well as C7. We have abandoned transfer of the sternal head of pectoralis major. The operation failed in more than
one half of 62 cases and function was marred in the remainder by the arm swinging into medial rotation. Anterior transfer of triceps is now performed very rarely. Thirty four of these operations have been done for patients with intractable co-contraction between triceps and biceps. Seven patients regretted going ahead with the operation which was reversed in two of them (Fig. 13.32). Pectoralis minor transfer (Le Couer (1967) is an elegant operation and it is certainly useful in improving power of elbow flexion by at least one MRC grade. It is appropriate when elbow flexors have regained power to MRC 2 or 3 and it also enhances active supination. Fusi (2003) studied 29 cases
582
Surgical Disorders of the Peripheral Nerves
Steindler (1918) proposed proximal advancement of the extensor origin of the forearm to the humerus, detaching a portion of bone and advancing this by up to 7 cm. Modifications of this operation have been widely practised. Brooks elevated the flexor muscles and interwove them into the medial intermuscular septum limiting proximal advancement to 3 cm. The operation achieves a limited range of flexion; it tends to increase the pronation deformity and we have had two cases of a significant lesion of the ulnar nerve from compression, in spite of exposing that nerve. The operation appears to achieve the same end as pectoralis minor transfer without the benefit of supination and it should be reserved for cases in which elbow flexors reach, at least, power MRC 2. However, Alnot and Abols (1984) and Egmond et al. (2001) reported generally good results in both adults and children. We suggest the following as guidances:
Fig. 13.32 (Above) 55 years after transfer of the whole of pectoralis major with lateral rotation osteotomy. The patent regretted the loss of adduction, the inability to bring his arm across his body and was aware of the alteration in appearance. (Below) Rupture of C5 and C6 was treated by transfer of pectoralis major and lateral rotation osteotomy 20 year previously. The patient is unable to bring the right hand to the mouth and he very much regretted the loss of medial rotation.
of pectoralis minor transfer in children with birth lesions of the brachial plexus. Elbow flexion power was increased on average, from MRC grade 2 to4 and the elbow function score from 2 to 4 (see Chap 10). In 12 patients the range of active flexion reached between 120° and 150° and in 11 more from 90° to 100° Extension was complete in 18 children, flexion deformity in excess of 30° was seen in nine. Nine children regained 90° of supination, nine more regained supination between 45° and 89°. We follow Fusi’s advice and interpose a tendon graft between the transferred pectoralis minor and the biceps tendon. Indeed, the first use of an interposed graft was prompted by finding complete rupture of the biceps in a child with birth lesion of the brachial plexus.
1. Some elbow flexion exists but it is weak; C7 and C8 and T1 have recovered. Pectoralis minor transfer is the first choice; proximal advancement of the flexor muscles comes second. 2. The elbow flexor muscles are completely paralysed; there is recovery through C7, C8 and T1. Bi-polar transfer of latissimus dorsi with a skin island is the first choice; after that, pectoralis minor transfer combined with proximal advancement of the flexor origin. 3. Free functioning muscle transfer is considered for other cases and in expert hands it may prove to be the first choice in cases of complete paralysis of the elbow flexors. Triceps to biceps transfer is reserved for cases of intractable co-contraction between triceps and biceps and it cannot be recommended in patients with good function at the shoulder nor for those who have defects in lower limb function. For many patients extension is more important than flexion at the elbow.
13.3.4 The Operations All of these are performed with the patient supine. Careful haemostasis and suction drainage are essential. In all the elbow is flexed at ninety degrees for 6 weeks after operation, in a plaster of Paris splint. At 6 weeks the arm is put into a triangular sling and gentle active and passive flexion work is commenced. Active extension work is not permitted until 9 weeks have elapsed, when the arm can be set free from restraint. 13.3.4.1 Pectoralis Minor Transfer The muscle is exposed through an incision which is made below the lower margin of pectoralis major. This should not be extended into the axilla and should end above the level of the nipple. The plane between pectoralis major and minor is
Reconstruction
583
Fig. 13.34 Pectoralis minor transfer: completed. Fig. 13.33 Pectoralis minor transfer: the incision.
opened. A nerve stimulator is advisable to enable detection of the medial pectoral pedicle and also of motor branches perforating pectoralis minor (Fig. 13.33). Now, the plane deep to pectoralis minor is identified and opened so that a finger can be passed deep to the muscle and hooked around the upper margin of the muscle. A skilful assistant will ease this exposure by elevating the arm so that the wound is opened out. The fascia attached to the upper margin of the muscle is incised taking great care for the medial pectoral pedicle which enters the lateral and superior margin of the muscle. The muscle is detached from the ribs, starting below and working from lateral to medial, taking a short flap of periosteum. Careful haemostasis of vessels passing between the periosteum and the muscle is required. Once the muscle has been detached from the ribs it can be gently mobilised superiorly and inferiorly. The detached end of the muscle is lightly sutured to form a tube. The biceps is now exposed through an anterior incision in the upper arm. The deep fascia is incised and a roomy tunnel is developed to join the two wounds. The pectoralis minor is drawn through this tunnel to lie on the surface of biceps. Next, the biceps tendon is exposed at the elbow through an incision on its lateral aspect, identifying and protecting the lateral cutaneous nerve of forearm. An interposed tendon graft is used, and this is usually taken from one of the extensor muscles of the toes (Fusi 2003). This is interwoven into the pectoralis minor and the distal end of the tendon is drawn into the elbow wound by a long pair of forceps which pass deep to the deep fascia of the arm. The tendon graft is interwoven into the biceps tendon with the elbow held at about 100 degrees of flexion and with the shoulder abducted to about 30 degrees (Fig. 13.34). Stimulation of the medial pectoral nerve provides some indication of tension. The wounds are now closed, that at the elbow is best done last with the arm in full adduction (Fig. 13.35).
13.3.4.2 Proximal Advancement of the Flexor Origin The flexor origin is exposed through a postero-medial incision centred over the olecranon, extended by some 10 cm curving forward so that the medial intermuscular septum can be displayed. The medial cutaneous nerve of the forearm must be seen and protected. The ulnar nerve is traced and decompressed, and the aponeurosis between the two heads of flexor carpi ulnaris is split. The median nerve is exposed anterior to the superficial flexor origin and the nerve(s) to the muscles identified with a nerve stimulator. The superficial flexor muscles are detached from the medial epicondyle preserving the capsule of the joint. The muscle mass is mobilised without provoking distortion or kinking of the ulnar nerve and its accompanying vessels. The ulnar origin of the flexor carpi ulnaris remains undisturbed. Now the anterior and posterior faces of the medial intermuscular septum are identified and the tendon of origin of the muscles is advanced proximally by, at the most, 3–4 cm, and interwoven into the medial intermuscular septum. On occasion the medial intermuscular septum is deficient and in the adult it may prove necessary to detach a piece of the medial epicondyle which is fixed into the humerus with a screw and a washer. This should never be done in a child. We have been asked to treat three cases of dislocation of the elbow following overzealous interference with the medial epicondyle. The elbow should not be flexed more than 90 degrees to diminish the risk of compression or of traction upon the ulnar nerve. Meticulous hemostasis is essential. A haematoma is a very serious complication. We avoid using the tourniquet (Fig. 13.36).
13.3.4.3 Latissimus Dorsi Transfer This operation cannot be performed if there has been rupture of the subclavian or axillary arteries. The thoraco-dorsal
584
Surgical Disorders of the Peripheral Nerves
Fig. 13.35 Outcome of pectoralis minor transfer by the method of Fusi in a 9 year old boy with BLBP Group 2.
identification of the thoraco-dorsal nerve and of the nerve to serratus anterior. The first incision runs along the anterior margin of the muscle, designing a skin paddle 12cms in length and five to six centimetres wide. The thoraco-dorsal pedicle is identified and mobilised proximally; the artery to serratus anterior requires ligation and division. The appropriate length of the muscle is measured to match the interval from the coracoid process to the elbow flexor crease and it is then detached below and posteriorly. Next, the tendon is detached from its insertion and re-inserted onto the coracoid. A lengthy incision, from the coracoid down the whole of the anterior aspect of the arm is prepared. With the elbow flexed at 90 degrees the flap is laid into the wound, and the muscle is sutured to the biceps tendon. Stimulation of the thoraco-dorsal nerve gives some indication of tension. The skin flap is now sutured (Fig. 13.37).
13.4 Paralysis of the Extensors to the Wrist and Fingers, and of the Abductors and Extensors to the Thumb Fig. 13.36 Poliomyelitis. The classical operation. Elbow flexion was restored by proximal advancement of the brachio-radialis and extensor muscles of the wrist, an operation performed by Mr. S. O’Connor, Cork in 1950.
pedicle must be handled tenderly, avoiding traction or torsion (Hovnanian 1956; Zancolli and Mitre 1973). Complete necrosis of the flap occurred in one of our cases because of twisting of this pedicle. A nerve stimulator is used in the
“It is entirely wrong for surgeons to believe that there is a standard procedure for tendon transfer for every single set of circumstances as is frequently described in articles and book”. (Lamb and Kuczynski 1981)
Boyes (1960) recognised no less than 58 different operations to restore extension of wrist and of the digits. To Jones (1916) goes the credit for the idea of using pronator teres for wrist extension and to Zachary (1946) that for recognising the
Reconstruction
585
Fig. 13.38 The importance of wrist extension. This 31 year old graphic designer sustained bilateral lesion of the brachial plexus. Avulsion of C7 on the right was treated by flexor to extensor transfer, achieving a power grip at 50% of estimated normal. On the left, C6, C7 and C8 were avulsed and only one FDS muscle was of available for transfer to EDC and EPL. Power grip was negligible. Fig. 13.37 Latissimus dorsi transfer in a 24 year old man with a C5, C6 lesion. The range of elbow flexion was 0–130 degrees; he was able to lift a weight of 2.5 kg.
importance of keeping at least one wrist flexor. We condemn arthrodesis of the wrist, for this abolishes the function gained from the “tenodesis effect.” The reader can demonstrate this quite simply by noting the posture of the fingers when their wrist is passively dorsi flexed and then permitted to drop into palmar flexion. It is an operation of last resort. Napier (1955) showed that in full palmar flexion the power of grip is 25% of the best achieved when the wrist is extended (Fig. 13.38). As Tubiana (2005a) pointed out, most wrists work in a position of semi pronation, and there is a synergy between the radial wrist extensors and the flexor carpi ulnaris (FCU). The extensor carpi ulnaris (ECU) is the partner of abductor pollicis longus (APL) and extensor pollicis longus (EPL). Furthermore, the excursion of APL and extensor pollicis
b revis (EPB) is only one half of the lateral extensor pollicis longus (EPL) so that one transfer will not do for all three. We have performed over 700 operations for this problem in adults, since 1975, more or less equally divided between paralysis from lesions of the spinal nerves and paralysis from lesions of the posterior cord and its terminal branches. The object of most of the 150 operations in BLBP was to rebalance the wrist, to improve abduction and extension of the thumb and to overcome the ulnar deviation which is so common after poor recovery of the seventh cervical nerve. Dunnett and Housden (1995) measured the range of passive and active movement, and power at the wrist, the metacarpophalangeal and interphalangeal joints in 49 cases of flexor to extensor transfers. Radial palsy accounted for 22 of these; 27 patients had lesions of the brachial plexus. Eight operations failed, all of these in the brachial plexus group and there had
586
Surgical Disorders of the Peripheral Nerves
been rupture of the subclavian or axillary artery in four of them. The power of wrist extension was, on average, 22%; of digital extension, 31%; and of grasp 40%. When FCU was transferred to EDC and EPL, power of grasp was close to 50%, whereas when FCR was transferred to EDC and EPL this fell to an average of 35%. There was no difference in the range of ulnar and radial deviation between these two groups. More than two thirds of the patients described impaired co-ordination and dexterity and over four fifths cited loss of endurance. We became dissatisfied with pronator teres transfer to the wrist extensors for this proved inadequate for those engaged in heavy manual work. Ropars et al. (2006) used the scoring system of Bincaz et al. (2002), and suggest that FCU should be preserved to avoid radial deviation deformity and cautioned against using flexor digitorum superficialis (FDS) because of impairment of finger dexterity. Valenti (2005) describes extending latissimus dorsi to ECRB to regain wrist extension in multiple nerve palsies. Bertelli and Ghizoni (2006) describe successful restoration of wrist extension and also of finger flexion using the brachialis muscle after lesions of the brachial plexus. Lim et al. (2004) has applied his work on the intramuscular distribution of motor nerves by splitting flexor carpi radialis along the nerve “watershed” using one half for EPL and the other half for EDC. Independent movement of fingers and of thumb was restored. The technique is selected according to the needs of the patient and by the availability of motors. Adequate extension of fingers and of thumb is seen in at least one third of patients with C5/C6/C7 lesions and in over 10% of cases where only the first thoracic nerve survives. Such work horses as PT, FCR and FCU are weak or paralysed in these lesions. We prefer to use three motors for straightforward radial palsy: one for wrist extension, one for extension of the digits and the third for abduction of the thumb ray, but such abundance is not always there. Gousheh and Arasteh 2006 achieved useful function in 180 cases, by transfer of FCU alone to EDC, EPL and extensor indicis proprius (EIP). Recovery was rapid: protected exercises were started on the first day after operation and splinting was abandoned at 28 days.
13.4.1 Pre and Post-operative Care A dynamic extension splint is applied some weeks before the proposed operation and the patient is encouraged to use the hand freely (Figs. 13.39 and 13.40). This informs about the potential benefit of the operation and may suggest modification of technique. After closure, a plaster of Paris splint is applied to the anterior aspect of the forearm and hand and secured by a non extensile bandage in a position of about 20–30 degrees of extension at the wrist, 20 degrees of flexion at the metacarpophalangeal joints of the fingers, with the
Fig. 13.39 A very elaborate splint for radial nerve palsy. Active extension for the wrist, fingers and thumb is provided. The appearance of impracticability is belied by the fact that the patient found this splint extremely useful, so useful that he declined operation. (Courtesy of Miss Lydia Dean DipCOT).
thumb in full abduction and extension. The terminal joints of the fingers are left free. At 2 weeks this plaster is removed and replaced by the dynamic splint made before operation. This is retained for a further 4 weeks. Supervised active flexion work is encouraged during this time. At 6 weeks, a simple static wrist support splint is fitted, which is worn for several hours a day for a further 6 weeks. Thumb and finger movement is
Reconstruction
587
EDC and EPL tendons; thirdly, a short, radial incision exposes the tendons of APL and EPB. Pronator teres (PT) is displayed through an incision in the middle of the anterior aspect of forearm, curving towards the flexor origin, protecting the radial artery, the lateral cutaneous and superficial radial nerves. PT must be adequately mobilised, and its tendon detached with a strip of periosteum. The tendon of FCU is sectioned close to the pisiform, having identified the ulnar nerve and artery (Fig. 13.41). This muscle is released from the ulna to just above the mid point of the forearm where a vascular pedicle enters the muscle accompanied by a branch to FCU. Mobilisation can be increased by incising the deep fascia. Palmaris longus is detached and mobilised. If it is
Fig. 13.40 Emergency repair of brachial artery and musculocutaneous nerve was successful in this 34 year old man: the graft for the radial nerve restored extension of the elbow. A low profile splint has been fitted as a prelude to flexor to extensor transfer.
unrestricted. The patient practices increasingly vigorous extension and flexion of the wrist against graduated resistance.
13.4.2 The Operations The Standard Transfer for Radial Palsy: pronator teres (PT) to extensor carpi radialis brevis (ECRB), flexor carpi ulnaris (FCU) to extensor digitorum communis (EDC) and extensor pollicis longus (EPL), palmaris longus (PL) to abductor pollicis longus (APL) and extensor pollicis brevis (EPB). The patient lies supine. A suprasystolic cuff is not essential and should not be used if the axial artery has been damaged or if there is any suggestion of polyneuropathy. The first incision curves along the anterior aspect of the forearm displaying FCU and PL; next, an incision is made on the dorsum, proximal to the extensor retinaculum to show the
Fig. 13.41 The flexor carpi ulnaris has been detached and mobilised. Note the position of the dorsal branch of ulnar nerve marked by the blue sling (above) and of the branches of the superficial radial nerve marked by the white sling (below). The tendon of ECRB is marked by a suture.
588
Surgical Disorders of the Peripheral Nerves
to ensure proper cadence for the fingers and for the thumb. It is important to avoid undue tension on EPL. FCU is passed through all five tendons and secured by a continuous interweaving suture. If a suprasystolic cuff has been used, this is now released and hemostasis obtained before closure (Fig. 13.42).
Fig. 13.42 A 51 year old man with radial palsy. FCU, PT and PL transfer. This unusually good result is, in part, explained by unexpected late recovery through repair of the nerve!
absent, FDS to the index finger is identified, protecting the median nerve, and then with the wrist and fingers in flexion, it is hooked up into the wound at the wrist and sectioned. There is no need for incisions in the palm of the hand. The transferred muscles are passed through separate tunnels deep to the skin, avoiding areas of scar. The best plane for FCU is around the ulnar side of the forearm and quite a large tunnel is necessary for this bulky muscle. On occasion, however, it must be passed around the radial border if the condition of the skin on the ulnar side is bad. The PL tendon is passed radially to EPB and APL whilst PT can be passed superficial or deep to brachio-radialis, whichever seems easier. PT is interwoven into the tendon of ECRB, in such tension that the wrist lies close to neutral. The two tendons, APL and EPB are identified at about the level of the radial styloid, taking care for the superficial radial nerve, which may still have some intact fibres within it. PL, or one of the FDS tendons is interwoven so that the CMC joint is abducted to at least 30 degrees and the MP joint to within 10 degrees of full extension. After this, the tendons of EDC and of EPL are now held by stay sutures and their tension is modified by the assistant
Fig. 13.43 FDS transfer. Note the display of median and ulnar nerves (above left), the subcutaneous passage of the muscles and the degree of tension.
13.4.2.1 Flexor Carpi Ulnaris Transfer to Extensor Carpi Radialis Brevis We prefer FCU for extension of the wrist in patients working in strenuous physical occupations. The power of grip after PT to ECRB transfer is about one third of normal, but usually exceeds 50% after FCU to ECRB transfer If this variation is followed pronator teres is available for EDC and EPL but a long strip of periosteum must be elevated from the radius and this tissue is, at times, tenuous. Flexor digitorum superficialis to the index finger is a reliable motor for EDC and EPL in such cases.
13.4.2.2 Transfer of Flexor Digitorum Superficialis This operation is used where wrist flexors and pronator teres are unavailable. Two muscles are required for adequate extension of the wrist. Two incisions are made. One runs from the wrist crease to the mid part of forearm, identifying and tracing the median nerve; a dorsal incision exposes EPL and EDC, and the tendon of ECRB. The FDS muscles to the index and middle are separated from those to the ring and to the little fingers as far proximal as the mid point of the forearm. FDS to index and middle are passed subcutaneously around the radial aspect of the forearm and then through part of the compartment of the extensor retinaculum containing the ECRB tendon (Fig. 13.43). A segment of this
Reconstruction
tendon is removed to make room but an adequate stump must be left for suture. The remaining two superficialis tendons are passed around the ulnar border of the forearm and interwoven into EDC and EPL. Passing FDS muscles through the interosseous membrane is needlessly complicated, and the muscles can become inextricably bound down (Fig. 13.44 and 13.45).
Fig. 13.44 A 21 year old student suffered rupture of C5 with avulsion of C6 and C7 and rupture of the subclavian artery. The artery was successfully repaired. The graft from C5 to C6 was unsuccessful. The spinal accessory nerve was transferred to the ventral root of C7 restoring a powerful triceps, pectoralis major and latissimus dorsi. FCU to ECRB, FDS (index) to APL and EPB, FDS (middle) to EPL and FDS (ring and little) to EDC improved hand function.
Fig. 13.45 Some examples of flexor to extensor transfer. (Above) Manual worker in full time work, in whom urgent repair of axillary artery and of circumflex and musculocutaneous nerves were successful. The graft of the radial nerve restored elbow extension. FCU to ECRB,
589
13.5 The High Median Palsy Whilst loss of sensation is the major defect in these lesions, we have seen cases where the territory of the superficial radial nerve and of the lateral cutaneous nerve of the forearm was so extensive that lateral grip between sensate skin of the radial border of the index and the ulnar border of the thumb was possible. A number of motors are available to restore flexion of the index and middle fingers, flexion of the interphalangeal joint of thumb and abduction of the thumb, but these operations will only work if powerful extension of the wrist is retained by a normal ECRB (Fig. 13.46). Sabapathy et al. (2005) reported good results after transfer of ECRL to FDP in cases of destruction of the forearm flexor muscles. ECRL and BR are the real work horses of hand reconstruction. The former is a beautiful muscle to work with. The excursion of the long tendon is greater than that of any other wrist muscle and it is easily mobilised to the mid forearm through small incisions. Brachio radialis is a little more surly and considerable dissection is needed to mobilise it. Identification and preservation of the superficial radial nerve and of terminal branches of the lateral cutaneous nerve is extremely important during elevation of these two muscles.
13.5.1 Abduction and Opposition of the Thumb The more distal and the more ulnar the pulley around which the transfer tendon passes, the greater is the adduction effect: the more radial and the more proximal the pulley is situated, the greater is the effect of abduction (ante position) (Moutet
PT to EDC and EPL, FDS (index) to APL and EPB. (Below) Urgent repair of axillary artery, and of musculocutaneous and circumflex nerves was successful in this 34 year old man. The radial nerve did not recover. FCU to ECRB, FDS to EDC, EPL and APL.
590
Surgical Disorders of the Peripheral Nerves
of the hypothenar eminence, where strong fibrous septa serve as a pulley (Fig. 13.47). Schwartz and MacDonald (2003) describe indications for, and assessment of, 156 operations of opponens plasty generally by transfer of the ring finger FDS. Results were good in 89% of cases.
The Operation
Fig. 13.46 “Tenodesis” effect of the wrist extensor muscles in a 13 year old child with high median and ulnar palsy. Active wrist extension is a pre-requisite for useful hand function.
2005, Bunnell 1938). If the object is to restore loss of abductor abductor pollicis brevis (APB) then the transferred tendon is looped around FCR. If all of the superficial thenar muscles are paralysed, then that tendon is better looped around the FCU tendon or through the subcutaneous tissues
Fig. 13.47 Two examples of opposition transfer. (Left) Undiagnosed non-union of clavicle provoked a partial lesion of C8 and T1 in a 21 year old woman, thumb function (above) was enhanced by transfer of FDS (ring). (Right) Opposition of the thumb improved by transfer of palmaris longus extended by a strip of the flexor retinaculum.
Extensor indicis (EI) to abductor pollicis brevis (APB); extensor carpi radialis longus ( ECRL) to flexor digitorum profundus (FDP); brachioradialis (BR) to flexor pollicis longus (FPL). Extensor indicis is exposed through three incisions. The first curves over the dorsum of the index MCP joint. EI tendon lies to the ulnar side of EDC tendon and it is detached. Traction on the tendon shows its course and it is mobilised through a second incision just distal to the extensor retinaculum. There may be inter-tendinous connections. The third incision is made proximal to the extensor retinaculum, the tendon is drawn into this and the muscle gently mobilised from the interosseous membrane (Fig. 13.48). Now, ECRL and BR are detached and mobilised through an incision along the radial aspect of the forearm. The tendons of FDP and FPL are exposed through an anterior incision and held with stay sutures. With the wrist in neutral such tension is applied to these that the tips of the fingers lie, in cadence, close to the palm and the IP joint of the thumb is flexed to 30 degrees and no more. EI is passed subcutaneously about the ulnar aspect of the wrist, and is looped around the tendon of FCR to be inserted into the tendon of APB. ECRL is interwoven into the
Reconstruction
591
Fig. 13.49 Extensor to flexor transfer in C8/T1 palsy following high median transfer in a 63 year old woman.
Fig. 13.50 BR to FPL in palsy of anterior interosseous nerve caused by neuralgic amyotrophy. The range of extension was full.
Fig. 13.48 Steps in the transfer of extensor indicis to abductor pollicis brevis.
tendons of FDP and BR into the tendon of FPL. Tension can now be checked: with the wrist passively extended the finger tips should meet the palm and the tip of the thumb come to the distal phalanx of the index finger. The digits extend when the wrist is dropped into palmar flexion. If a suprasystolic cuff has been used this is now removed and haemostasis obtained before closure. Two plaster of Paris splints are applied. The anterior splint extends to the mid palm supporting the wrist in 20 degrees of extension and the dorsal splint extends as a hood over the fingers and thumb to block excessive extension. Early gentle flexion and extension is encouraged from the first post operative day. The splints are replaced
by an orthosis which protects against forcible extension. This is removed for periods of supervised exercise and it is discarded at 6 weeks (Figs. 13.49 and 13.50).
13.6 The High Ulnar Palsy Whilst the insensate little finger is an encumbrance the most serious defect is loss of power. Power grip is reduced to about one half of normal, pinch grip between the thumb and index finger to no more than one quarter. Fixed flexion deformity of the little and ring PIPJ may occur rapidly with the recovery of the flexor muscles of the forearm after repair of high lesions and there is much to be said for early operation to restore MCPJ flexion in such cases. Parkes’ (1973b) essay on this subject credits Harold Styles with the first operation to replace lost intrinsic muscles and describes his own
592
Surgical Disorders of the Peripheral Nerves
operation of tenodesis which is valuable in some combined nerve palsies. Excellent reviews of the complex problem of loss of small muscles in the hand are provided by Zancolli (1979), Brand (1958), Tubiana and McCullough (1984), Smith (1987) and Tubiana (1993a, 2005b). Flexion of the little and ring fingers can be improved by side to side suture to adjacent flexor muscles or, if these are defective, by anterior transfer of brachio-radialis. We usually prefer the operation of Brand (1958) for regaining active flexion of the MCP joints of the fingers but follow Smith (1987) in using ECRL, never ECRB, and the tendon grafts are passed to the radial side of all fingers. This operation seems to be particularly suited to the supple hand whereas the dynamic lasso procedure (Zancolli 1957) is more suited to the strong, rather stiff, hand. In this operation FDS tendons are divided at the entrance to the flexor sheath, passed through a window in this, then sutured back onto themselves. The flexor sheath acts as a pulley. Persisting abduction of the little finger invites injury to that digit and it may persist after the operations of Brand or Zancolli (Fig. 13.51). A slip of extensor digiti minimi to the ring finger can be re-routed to draw the little finger towards the ring finger (Fig. 13.52).
Fig. 13.52 Transfer of part of extensor digiti minimi to the adjacent dorsal interosseous hood.
Adductor pollicis and the first dorsal interosseous muscles are consistently innervated by the ulnar nerve. The adductor is the most powerful of the small muscles of the hand and requires a strong motor to replace it. In uncomplicated ulnar palsy we favour FDS to the middle finger but whatever method is used it is important that the thumb web space is splinted in the open position. Molen and Evans (1996) describe an alternative method in which EI is detached from its insertion, passed deep to the index metacarpal, along the line of adductor pollicis and sutured into that tendon. One slip of APL tendon is extended by tendon graft to the insertion of the first dorsal interosseous muscle for, as Evans points out, the tendon of this essential muscle usually contains two or three separate slips (Fig. 13.53).
The Operation – ECRL for Flexion of MCPJ of Fingers
Fig. 13.51 High ulnar lesions. FDS (index and middle) were transferred as dynamic lassoos to the sheaths of the flexor tendons, and FDS (middle) was transferred to adductor pollicis. Although power grip was recorded at 75% and pinch grip at 60% the little finger lies in abduction.
Mild flexion deformity of the PIP joints can be overcome by stretching and by dynamic splinting. More severe deformity must be corrected before muscle transfer (based on Brand 1958). ECRL is detached and mobilised so that its tendon lies on the dorsum of the hand. It must be extended by a tendon
Reconstruction
593
Fig. 13.55 ECRL to interosseous muscles (Brand’s operation) and FDS (middle) to adductor pollicis in 25 year old man achieved a power grip of 80% and a pinch grip of 75%.
Fig. 13.53 Iatrogenous lesion of the ulnar nerve provoked causalgia in a 58 year old woman. Pain was relieved by repair of the nerve. FDS (ring) was transferred to adductor pollicis. The thumb web space was splinted open for 3 weeks after the operation.
side of the index finger. The tendon grafts are passed through the lumbrical canals, anterior to the deep transverse ligament to be inserted, by interweaving suture, into the interosseous hood. A fine gently curved pituitary rongeur is useful. Two plaster of Paris splints hold the wrist in extension and the metacarpophalangeal joints flexed to 90 degrees. Early active flexion and extension of the tips of the fingers is encouraged from day one. The dynamic splint, fitted and used before operation are reintroduced at 2 weeks and retained for four more (Fig. 13.55).
13.7 Combined Nerve Lesions and the Hand
Fig. 13.54 Transfer of ECRL which has been extended by tendon grafts.
graft, which is prepared as four tails (Fig. 13.54). One of the toe extensor tendons is best. The dorsal expansion of the interosseous muscles is exposed through incisions in the web space between the fingers and the tendon of the first dorsal interosseous muscle is exposed through an incision on the radial
These are particularly difficult cases. It is unusual to find a deficit of nerve uncomplicated by fibrosis and contracture from ischaemia or pain. No two cases are the same and each patient requires careful thought about the desired aim (Omer 1988). In particular, careful assessment of the power of the available muscles is necessary. Neurophysiological investigation (NPI) may reveal denervation not detected by clinical examination. The extent of remaining sensation in the hand is extremely important. Some elements of recovery of cutaneous sensibility are usual after repair of proximal lesions and if a patient can localise with accuracy to the digit and if they are able to discriminate between sharp and blunt and hot from cold stimulation, without over reacting to that stimulation, then they do have useful sensation. Sympathetic paralysis causes atrophy of the skin of the pulp; the hand may be cold from vasomotor paralysis and loss of sweating certainly diminishes the power of grip. A careful assessment of activities of daily living is
594
Fig. 13.56 Rupture of musculocutaneous, radial and median nerves were repaired by graft in a 47 year old man. Powerful flexion and extension of the elbow was restored. Flexor carpi radialis became strong enough to permit transfer of FCU to ECRB, so regaining extension of the wrist. An opposition splint has been applied as a prelude to opposition and flexion muscle transfer. Note the atrophy of the skin of the thumb and the index finger where there is persisting sympathetic paralysis.
always helpful. Orthoses help to control paralytic deformity and to correct fixed deformity and they can be particularly useful to patient and surgeon in designing the appropriate operation for restoration of control of the thumb (Fig. 13.56). It is important to maintain or restore stability of the MCP joint of the thumb which may have been weakened by the cause of the deformity, by the uncorrected deformity or by the operation to correct that deformity. It is wise always to use an orthosis or to stabilise, temporarily, the joints of the thumb by a fine Kirschner wire before proceeding to arthrodesis. A patient with an intact C5 and C6, or lateral cord, or median nerve, has an excellent platform for function whereas the patient in whom only C8 and T1 or the lower trunk, or the ulnar nerve are intact, is worse off. A good shoulder and a stable elbow permits positioning of the hand in space. The potential for the function of the hand itself depends a great deal on the quality of sensation, especially in the thumb, the index and the middle fingers (C7, lateral cord). We think that the most important objects of operation include restoration of extension of the wrist, flexion of the MCP joints of the fingers and grasp and release between the thumb and the fingers. Wherever possible we avoid arthrodesis of the wrist. Tenodesis of the wrist extensors by the method of Parkes (1973b) is often useful in these cases. The role of tenodesis in the forearm and hand is discussed by Revol (1993) (Figs. 13.57–13.60).
Surgical Disorders of the Peripheral Nerves
Fig. 13.57 A 21 year old man suffered tetraparesis after cervical laminectomy for spinal instability provoked by neurofibromatosis Type 1. Flexion of the MCP joints was regained by Brand’s operation and this was combined with opposition transfer.
numbers of orthopaedic surgeons has something to do with this, but it is surprising that whereas there are so many – perhaps too many – operations for reconstruction in the upper limb, there are certainly too few in the lower. We remain surprised to see the numbers of patients with paralytic deformities of the ankle and foot, the knee, and the hip for whom nothing has been proposed. Yet, nowhere else does skilful intervention performed appropriately bring about such benefit. The ability to cross a road or go upstairs is important. We might reflect that one of the disadvantages of the welcome strength of the discipline of hand surgery is that it has turned away from its roots at a time when there is undoubted increase in nerve injuries of bad prognosis in the lower limb from road traffic accidents, missile wounds and, as must be said, from intra-operative damage to trunk nerves. Many patients with complete sciatic palsy lead a very active life and can engage in such activities as mountaineering and skiing (Fig. 13.61). On the other hand palsy of the superior gluteal nerve is disabling, and a high lesion of the femoral nerve, abolishing flexion of the hip and extension of the knee is crippling.
13.8 The Lower Limb The extensive literature about reconstruction in the upper limb should not obscure the fact that the principles arose from early attempts to treat the deformed lower limb. It may be that the decline of poliomyelitis in countries with large
Fig. 13.58 A 55 year old man suffered avulsion of C7, C8 and T1. The hand is shown after high median transfer. This patient denies that he has any pain and continues in full time work.
Reconstruction
595
Fig. 13.59 Age is no barrier to useful regeneration. A 64 year old man suffered ruptures of C5 and of C7 and avulsions of C8 and T1. C6 was intact. At 6 days after injury C5 and C7 were grafted, the ventral root of C8 was reinnervated by the spinal accessory nerve and the ventral root of T1 by the proximal stump of C7. Function at the shoulder and the elbow was good, flexion and extension of the digits achieved power of grade 4 by the MRC System. There was paralysis of the small muscles of the hand (above). Later, function was improved by Brand’s operation combined with transfer of extensor indicis to APB (below).
Fig. 13.60 A 63 year old man suffered causalgia from false aneurysm of the brachial artery incurred during internal fixation of fracture of shaft of humerus. Pain was cured by correction of the aneurysm and decompression of the median nerve. Later, the rigid metacarpophalangeal joints were released and adductor pollicis detached from its origin. This was followed by opposition transfer. The hand shown 3 years after the injury.
Valuable lessons can be drawn from earlier, extensive studies with large numbers of patients with poliomyelitis. Close and Todd (1959) showed, by electromyographic studies which were done before and after operation, that the tibialis anterior and quadriceps muscles act during the swing phase whereas the heel flexors and the hamstrings are active
Fig. 13.61 The leg, ankle and foot in a 31 year old service man 4 years after fracture of pelvis which was complicated by a deep lesion of the lumbo sacral plexus. Gradual spontaneous recovery of cutaneous sensation and sympathetic function improved trophic state of the foot and healing of the heel sore. Paralysis of tibialis posterior and of the small muscles of the foot led to collapse of the medial arch. He returned to full active duties 3 months after his injury and continued his sporting activities.
instance phase. Peroneus brevis (PB) is much more likely to convert to a dorsi flexor of the ankle than peroneus longus (PL). Kuhlmann and Bell (1952) warned that deformity could be provoked by muscle transfer but also by arthrodesis. It is very important to remember that no arthrodesis should ever be performed in the growing limb unless the muscular imbalance provoking that deformity has been corrected. Reidy et al. (1952) confirmed that the results of muscle transfer were generally worse in younger children and that transfers for heel extension were more effective than transfers for heel
596
flexion. Schwartzmann and Crego (1948) studied anterior transfer of hamstrings in 134 cases of poliomyelitis. They confirmed that one strong flexor of the knee must be retained and that the transferred muscles must balance one another on either side of the patella, to prevent hyperextension of the knee or dislocation of the patella (McMurray 1919). The variety of progressive deformities seen in the growing foot is considerable and the surgeon should be alert to the need for a re-balancing operation to prevent fixed deformity of the skeleton. If this is permitted, later correction by osteotomy or arthrodesis is unavoidable. Each case must be taken on its merits. It is as necessary to be on the watch for progressive deformity arising from transfer as it is to be on the look out for that produced by the initial lesion. Care to maintain a balance between eversion and inversion of the heel is essential. The calcaneus deformity developing after lesions to the tibial nerve in children is corrected by posterior transfer of tibialis anterior (Herndon et al. 1956). However, we believe that it is important first to balance the forces of inversion and eversion before directing attention to the position of the heel.
13.8.1 The Insensate Foot Loss of sensation in the plantar skin is a severe defect which, for some patients, is more dangerous than denervation of the skin of the hand. The remarkable atrophy of the skin adds to the risks of unnoted injury and when denervation is compounded by fixed deformity then the risks of ulceration, of osteomyelitis and even of spreading infection are very real. One patient described the conventional ankle foot orthosis as a contraption of the Devil. We have seen far too many instances of ulceration and sepsis caused by the unthinking prescription of this device, often by clinicians who do not review their patients after ordering the appliance. It is most often prescribed for older patients with partial lesions of the sciatic nerve inflicted during arthroplasty of the hip. Many find that it is too painful when applied to the partially denervated skin of the calf. We have seen 18 cases of deep ulceration caused by this appliance, six patients required admission and treatment by intravenous antibiotics for cellulitis and septicaemia, and we were forced to below knee amputation in two cases. Other, and much better, orthoses are available but all too often it is left to the patient to find out about them and to buy them (Fig. 13.62). Srinivasan and Palande (1997) and Srinivasan (2004) emphasise that the aim of treatment in the foot in leprosy includes prevention of plantar ulcers or the healing of ulcers already established. They describe the development of deformities of the toes and recommend appropriate interventions. The toes progress from supple to fixed deformity culminating in rigid extension deformity at the metatarso-phalangeal joints (MTPJ). This leads to sores in the plantar skin. The long flexor tendon should be transferred to the extensors of the toes whilst
Surgical Disorders of the Peripheral Nerves
Fig. 13.62 A conventional ankle/foot orthosis was prescribed for a 63 year old woman who suffered a deep lesion of the sciatic nerve during arthroplasty of the hip. She developed such a severe sore in her insensate sole of foot that below knee amputation was necessary.
the deformity is still supple. Rigid proximal interphalangeal joints (PIPJ) are excised if the MTPJ are still supple. In the last stages of the deformity the dislocated MCPJ must be reduced, with excision of the condyle and transfer of extensor digitorum longus into the metatarsal head. In these cases, skin grafts for the dorsum of the foot are required. Schwartz et al. (2006) carried out arthrodesis in 115 neuropathic feet. The deformity was corrected in about three quarters of cases, and the fusion was successful in 88%. Ulceration was improved in most patients, so that 40% of them were able to wear normal shoes, another 50% used specially made insoles. Amputation became necessary in 10 limbs.
13.8.2 Abduction and Flexion at the Hip Loss of either of these functions is a most serious matter. Nowadays, it has to be said that both of these problems are usually iatrogenous. Re-innervation, wherever possible, must be attempted. We know of no muscle transfer which restores loss of flexion at the hip. Sharrard (1971) described transfer of the ilio-psoas to enhance abduction at the hip joint in children with spina bifida. The operation, which is a considerable one, is described at length in Sharrard’s work. The elements of the operation include detachment of the ilio psoas tendon with a piece of the lesser trochanter; mobilisation of this muscle which is passed deep and lateral to the femoral nerve; transfer of the muscle mass through a large hole in the ilium and then suture of the ilio psoas tendon through a tunnel fashioned in the greater trochanter of the hip. The hip must be protected in a spica for 12 weeks. After that, active weight bearing is encouraged, and in particular strengthening of hip flexion and of hip abduction is taught.
Reconstruction
We have had occasion to use it in two young adults who had suffered iatropathic injury to the sciatic and gluteal nerves in the course of operations for hip dysplasia. Both experienced relief of pain and an improved sense of stability, but neither patient could abduct the hip against gravity.
597
only when the patients fell and damaged themselves. In one case, fracture of the cervical spine afflicted by ankylosing spondylitis caused a deep lesion to the spinal cord.
13.8.3.1 Anterior Transfer of Hamstrings for Paralysis of Knee Extensors
13.8.3 The Knee It was a common experience, in times when poliomyelitis was common to see children and young adults walking quite well, even thought their quadriceps muscles were paralysed. Of course, they did this by a form of adaptation, in which the tensor fascia lata was responsible for the stabilisation of the knee. In many cases there was the added factor of hyperextension deformity at the knee and medial rotation posture at the hip. It is, however, quite wrong to assume that an adult with a deep femoral nerve lesion can walk comfortably without risk. In six cases of femoral nerve palsy caused during hip arthroplasty, the lesion was brought to the attention of those concerned
There must be no flexion deformity of the knee and the hamstring muscles must be of full power (Herndon 1961; Clark 1956). A tourniquet should not be used. The patient lies supine. Two separate incisions are made over the biceps and semitendinosis tendons, protecting the common peroneal nerve which lies in close relation to biceps muscle. The tendons are detached from their insertion and mobilised to the middle of the thigh, preserving the neurovascular pedicles. A large window is made in the lateral intermuscular septum for the passage of the biceps tendon; on the medial side, the passage for semitendinosus is subcutaneous. The tendons are interwoven through the quadriceps tendon through the retinacula of the patella and then sutured, end to end beneath its lower pole (Fig. 13.63).
Fig. 13.63 Hamstring to quadriceps transfers in a 24 year old woman. The anterior muscles of the thigh were infarcted by excessive exercise and they were excised by Mr. Simon Owen Johnston (Royal London Hospital). Six months later biceps femoris and semitendinosis were
detached and mobilised (above) noting and protecting the common peroneal nerve 1, the long saphenous vein and the saphenous nerve 2. 3 marks the fascia lata. The range of extension of the knee at 10 weeks after operation is illustrated below.
598
The limb is immobilised in a plaster of Paris cylinder for 2 weeks. The patient is encouraged to mimic quadriceps action during this time but weight bearing is not permitted. At 2 weeks the plaster is replaced by a splint which is hinged at the knee. Active extension, and controlled increasing active flexion is encouraged over the subsequent 10 weeks. Anti-gravity extension is introduced at 4 weeks. Seven from 12 of our patients regained active extension to neutral and were able to walk unencumbered and climb stairs without difficulty. One failure occurred because the medial hamstrings were damaged so that only the biceps femoris was available for transfer; another followed posterior subluxation of the transferred tendons. One patient required manipulation of the knee at 6 months to improve the range of flexion. In two patients knee control was restored but walking was greatly hampered by weakness of the flexor muscles of the hip.
13.8.4 Foot and Ankle: The Drop Foot from Common Peroneal Palsy This has been treated by arthrodesis, triple arthrodesis, or the more elegant operation of Lambrinudi (1927). This operation is technically difficult. When the operation was done in the young it was often followed by degenerative changes in the ankle. Seddon did this operation with mathematical precision: there was a significant failure rate. Those who did it less precisely, and smashed the bone up more had a better success rate. Angus and Cowell (1986) followed a number of triple arthodeses over an average period of 13 years. There were a large number of unsatisfactory results. Residual deformity persisted in over one half, pseudarthrosis was found in 22% and 25% of patients remained in pain. However, Smith and Wood (2007) show that severe valgus or varus deformity can be treated successfully by arthrodesis of the ankle. The ankle is approached from the front. A segment of fibula is excised in a varus deformity and dorsal wedge osteotomy of the foot metatarsals is done when there is fixed plantar flexion; 24 of 25 operations succeeded. Ober (1933) recommended forward transfer of the TP. Watkins et al. (1954) reviewed 25 cases where Putti’s idea of transferring the muscle through the interosseous membrane had been used finding good results in 17 of 25 cases. Turner and Cooper (1972) and Lipscombe and Sanchez (1961) reported with guarded optimism. Soares (1996) preferred transfer of the TP through the interosseous membrane into TA and the peroneal tendons. Srinavasan et al. (1968) described the two tail transfer of TP for correction of drop foot in leprosy and Richards (1989) described the results of 39 operations by this method. One tail of the transferred tendon was sutured to EHL, the other to EDL and peroneus tertius. Active dorsi flexion to at least 10 degrees was restored in 17 cases.
Surgical Disorders of the Peripheral Nerves
We have performed over 200 of these operations and the technique has changed over the years. Full passive dorsi flexion at the ankle is essential. In early years the tendon of tibialis posterior was passed through the tarsus, and sutured to a bar incorporated in the plaster. There were risks from pressure on the skin and also of detachment of the transferred tendon. Secure attachment of the tendon to the tarsus proved difficult. In one case where the tendon was passed superficial to the extensor retinaculum bow-stringing was so severe that the overlying skin broke down. Later on, tibialis posterior tendon was split into two tails, one interwoven in TA, the other into the PB which was divided above the ankle, and the distal tendon re-routed to the dorsum of the foot. Yeap et al. (2001a) studied 12 of these patients at an average of 90 months after operation. The power of dorsiflexion was measured using a Cybex 11 dynamometer. It was, on average, 30% of the normal side. In a further study Yeap et al. (2001b) used a scoring system for outcome which included seven sections: pain; requirement for an orthosis; the ability to wear normal shoes; level of activity; power of dorsiflexion; range of dorsiflexion, and posture of the foot. Eighteen patients were reviewed at, on average, 64.6 months after operation. No patient developed painful flat foot. Results were graded by the system as excellent in four patients; good in seven more. The results were graded as fair or poor in seven patients, who complained of persisting toe drop, persisting pain, or inadequate dorsiflexion. Whilst this operation gives balanced dorsiflexion at the ankle it was certainly no better than Adam’s (1980) method of transfer to peroneus tertius (Fig. 13.64). We now prefer the method of Srinivasan (2004).
13.8.4.1 Tibialis Posterior (TP) Transfer to the Extensors of the Toes The patient lies supine. The tourniquet is avoided. If, in spite of reasonable efforts, the tendo Achillis is still tight, it is elongated through three small incisions. That on the lateral side is long enough to ensure that only the tendon is divided and not the sural nerve. Two stab incisions are made on the medial side with a fine blade and about one third of the tendon is divided through each of the three incisions. Firm dorsi flexion corrects the deformity Now the tendon of the TP is exposed by a short incision over the medial instep where it is sectioned. The muscle belly is exposed through a postero-medial incision, extending from the base of the medial malleolus to the mid calf. The tibial nerve and posterior tibial artery are identified. The tendon is drawn into this wound and the muscle freed from the interosseous membrane (Fig. 13.65). Care is taken to avoid damage to the large deep veins. Then an incision in the anterior aspect of the leg is made, lateral to the subcutaneous surface of the tibia. Tibialis anterior (TA) is retracted from the interosseous membrane identifying and protecting the anterior tibial nerve and vessels (Fig. 13.66).
Reconstruction Fig. 13.64 Muscle transfers for foot drop. On the left, tibialis posterior has been transferred to peroneus tertius. On the right, tibialis posterior was transferred to EDL and EHL.
Fig. 13.65 Elevation of tibialis posterior. The tibial nerve and posterior tibial artery and accompanying veins are displayed.
599
600
Surgical Disorders of the Peripheral Nerves
A wide window is made in the membrane and the muscle and tendon of TP, are drawn through this. The tendon is split in two. Now, an incision is made over the dorsum of the foot, mid way between the tendons of TA and PT. The transferred tendon is passed deep to the retinaculum. One portion of it is sutured into the EHL tendon; the other into the tendons of EDL (Fig. 13.67). The tension of the transfer is such that the foot lies in about 10 degrees of extension, neutral between inversion and eversion. The transfer is protected with a plaster of Paris splint for 6 weeks during which time weight bearing is not allowed. At 6 weeks the splint is removed and the patient is fitted with an orthosis. Active exercises are started but the transfer is protected for a total of 3 months. Srinivasan (2004) passes the tendon of the tibialis posterior subcutaneously and relates that no problems with bowstringing or skin erosion have been encountered. 13.8.4.2 Paralysis of the Dorsi Flexor Muscles of Ankle and Toes, with Intact Evertors of the Ankle
Fig. 13.66 A large window is prepared in the interosseous membrane for transfer of tibialis posterior, displaying and protecting the anterior neurovascular bundle 1.
Fig. 13.67 Two tails of the transferred tibialis posterior, 3, are interwoven into EHL, 1, and EDL, 2.
This is very difficult. Tibialis posterior is in balance with peroneus longus brevis. Transfer of tibialis posterior is out of the question – transfer of peroneus brevis risks disturbing that balance. Transfer of either muscle in the growing skeleton virtually guarantees later deformity at the heel. We were not impressed by our results of anterior transfer of the medial half of gastrocnemius.
Reconstruction
13.9 Vascularized Bone and Muscle Transfer The free vascularized fibular graft was described by Taylor et al. (1975). Other bone grafts were described later. We touched on our experience with 18 cases in 1987 (Scott 1985; Birch 1987). Weiland and Russell Moore (1988) reported 91 cases, 19 of these for severe injuries to the upper limb: Pho (1991) provides an excellent description of the technique (Fig. 13.68). A precept of reconstructive work is stability of the skeleton and what has become, in skilful hands, a standard and reliable operation should be remembered for the difficult case. Although improvements in understanding of fracture healing have led to increased use of bone transport to such an extent that the technique is now rarely indicated in the lower limb, there is no doubt of its continuing validity in the upper limb. The experimental foundation of free functioning muscle transfer was laid by Tamai et al. (1970). Harii et al. (1976) performed the first free gracilis transfer for facial palsy. Douglas Harrison (Mount Vernon Hospital) has reviewed his immense experience in the treatment of facial palsy by free muscle transfer (Harrison 2002, 2005). Harrison prefers the pectoralis minor muscle, which is reinnervated through a 20 cm long sural graft, usually apposed to the buccal branch
601
of the normal facial nerve. The pectoralis minor is transferred at a second stage operation, 6 months later. Harrison makes two important observations: the first is that regeneration through the nerve graft is monitored by an advancing Tinel sign; next, the biopsies of the distal stumps of the grafts reveal only non myelinated fibres, for myelination will not occur until the axons have reached their muscle target. Pectoralis minor is a valuable muscle and its blood supply has been carefully described by MacQuillan et al. (2004). Surgeons in Shanghai performed the first free muscle transfer for reconstruction of flexor muscles of forearm (Shanghai, 6th Peoples Hospital 1976), and Ikuta, Kubo and Tsuge (1976) described a number of cases in the treatment of Volkmann’s contracture. Manktelow and McKee (1978) described their first two operations to restore active finger flexion. Both patients achieved good power grip, substantially better than that provided by conventional tendon transfer. The indications for and technique of operation in a series of 98 cases was discussed by Manktelow (1993). A good review comes from Morris and Zuker 2000, from Kay (2000) and Carlsen et al. (2009) provide an exceptionally good description of the indications and technique of the gracilis transfer. We have used the latissimus dorsi muscle for forearm flexor reconstruction in three cases and the medial gastrocnemius in a further patient. The latissimus dorsi was chosen because of our familiarity with it, but gracilis or gastrocnemius provide a better range of movement. Useful, powerful grip was restored in these patients, but the range of digital flexion was limited (Fig. 13.69a,b). We believe that free muscle transfer has a definite and valuable role in the treatment of the severely injured upper limb in which all the flexor muscles have been damaged by ischaemia or direct injury. We think that the operation is technically somewhat easier than the free vascularized bone graft. The key to success lies in the selection of a suitable donor nerve, and a muscle of appropriate length and quality, adequate for the purpose.
13.10 Amputation
Fig. 13.68 Excision of the radius, ulnar and extensor compartments for extensive aponeurotic fibromatosis in a 9 year old boy treated by free vascularized fibular graft and flexor to extensor transfer. The result at age 12 years.
In earlier years, many patients with severe injuries to the brachial plexus were treated by amputation and arthrodesis of the shoulder. Fletcher (1981) showed that some of these patients were greatly helped by this procedure. Elective amputation should certainly not be seen as a failure of treatment by either surgeon or patient, but rather as a positive step in rehabilitation (Wilkinson et al. 1993). The decision is made by the patient who has rehabilitated themselves, and appreciates that the limb is but a useless appendage, an impediment at work, in daily life, and a source of sepsis. Seventy three above elbow amputations have been performed. All, save one, of these patients felt a good deal
602
Surgical Disorders of the Peripheral Nerves
Fig. 13.69 (a, b) Free functioning muscle transfer. (a, above) A 31 year old woman lay in a coma for 18 h infarcting the whole of the flexor muscles of the forearm. The infarct was excised and replaced by free latissimus transfer. Useful power grip was gained , but the finger tips did not reach the palm. (b, below) A medial gastrocnemius free muscle transfer was used to restore flexor function in this 12 year old boy with severe Volkmann’s ischaemic contracture. There is full digital flexion even with the wrist flexed. (Courtesy of Mr David Evans FRCS).
better, describing improved posture of the spine and of the shoulder girdle. More than half related a significant improvement in pain, in spite of having been advised that the operation would do nothing to alleviate this. Clearly some of the pain that they were experiencing came from traction of the paralysed limb on the gleno-humeral joint and on the suspensory muscles of the scapula. No patient experienced new phantom pain. The one unhappy patient was the one case of arthrodesis which left the patient dissatisfied with the position of the stump. The level of section is through the middle of the upper arm, just distal to the deltoid insertion. Arthrodesis is not done at the first operation; it is reserved for those patients wholly committed to the idea of using a prosthesis. The idea needs to be introduced with tact. Some patients are reluctant to express their thoughts on the matter because they felt that they were “letting us down” for all the earlier work that had been done. The interval from injury to amputation lay between 30 months and 21 years. All patients were at work, or were retraining. The question of a prosthesis was discussed before going on to operation and those interested saw demonstrations of prostheses in our limb fitting centre. Amputation of upper or lower limb should be seen as a positive step towards rehabilitation when it is performed for the right reasons (Fig. 13.70).
Fig. 13.70 A 28 year old woman suffered an open fracture/dislocation at the ankle and the treating surgeon considered amputation at that time. The foot survived but she developed severe pain and significant deformity from ischaemic damage to the nerves of the small muscles of the foot. Subsequent treatment for a presumptive diagnosis of “complex regional pain syndrome Type 1” was unavailing. Three months after below knee amputation she became an upland sheep farmer.
13.11 Conclusion No musculo-tendinous transfer matches the outcome seen after good nerve healing. We have become disenchanted with muscle transfers for the thoraco-scapular and gleno- humeral joints, and
Reconstruction
are not particularly impressed by results of elbow flexorplasty in cases where there was complete paralysis of elbow flexors. No muscle transfer adequately restores flexion or abduction at the hip. Clinicians must do more to diagnose, urgently, and to treat, adequately, nerve lesions in such cases. Too many are neglected: we can only hope that the central purpose of this work bears some fruit. Transected nerves will not get better unless they are repaired: the sooner that is done, the better.
References Adams JC (1966) Ischio femoral arthrodesis. E&S Livingstone, Edinburgh Adams JC (1980) Standard orthopaedic operations, 2nd edn. Churchill Livingstone, Edinburgh Allieu JY, Teissier J, Triki F et al (1985) Réanimation de l’extension du coude chez la tétraplégique par transplantation du deltoide postérieur. Etude de 21 cas. Rev Chir Orthop 71:195–200 Alnot JY, Abols Y (1984) Réanimation de la flexion du coude par transferts tendineux dams la paralysies traumatique de plexus brachial de l’adulte. Rev Chir Orthop 70:313–323 Anderson GA (2000) The child’s hand in the developing world. In: Gupta A, Kay SPJ, Scheker LR (eds) The growing hand, 1st edn. Mosby, St Louis, MO New York, pp 1097–1114, ISBN OT 234 21331 Anderson GA (2006) The surgical management of deformities of the hand in leprosy. J Bone Joint Surg 88B:290–4 Angus PD, Cowell HR (1986) Triple arthrodesis. A critical long term review. J Bone Joint Surg 68B:260–265 Antia NH, Enna CD, Daver BM (1992) The surgical management of deformities in leprosy. Oxford University Press, Bombay Beevor CE (1904) The Croonian lectures on muscular movements and their represenation in the central nervous system. Adlard and Son, London, 1904 Bertelli JA, Ghizoni MF (2006) Brachialis muscle transfer to reconstruct finger flexion or wrist extension in brachial plexus palsy. J Hand Surg 31A:190–196 Bigliani LU, Perez-Sanz JR, Wolfe IN (1985) Treatment of Trapezius Paralysis. J Bone Joint Surg 67A:871–877 Bincaz LE, Cherifi H, Alnot JY (2002) Les transfer palliatifs de réanimation de l’extension du poignet et des doigts. A propos de 14 transferts pour paralysie radiale et dix transferts pour lésion plexique. Chir Main 21:13–22 Birch R (1987) The place of microsurgery in orthopaedics. In: Catterall A (ed) Recent Adv Orthopaed 5:165–186 Boyes JH (1960) Tendon transfers for radial palsy. Bull Hosp Joint Dis 21:97–105 Brand PW (1958) Paralytic claw hand. J Bone Joint Surg 40B:618–632 Brand PW (1987) Biomechanics of tendon transfer. In: Lamb DW (ed) The paralysed hand. Churchill Livingstone, Edinburgh, London, New York pp 190–214 Brooks DM (1984) Flexor muscle slide. In: Brooks DM, Birch R (eds) Rob and Smith’s operative surgery, 4th edn. Butterworths, London Bunnell S (1938) Opposition of the thumb. J Bone Joint Surg 20:269–284 Campbell-Reid DA (1984) Capsulectomy of the metacarpophalangeal joint and proximal interphalangeal joints. In: Brooks DM, Birch R (eds) Rob and Smith’s operative surgery, 4th edn. Butterworths, London, pp 226–234 Carlsen B, Bishop AT, Shin AY (2009) Late reconstruction for brachial plexus injury. In: Spinner RJ, Winfree CJ (eds) Neurosurgery clinics of North America. Peripheral nerves: injuries., pp 51–64
603 Chammas M, Meyer ZU, Beckendorf G, Allieu Y (1996) L’arthrodèse d’épaule pour paralysie post traumatique du plexus brachial. Rev Chir Orthop 82:386–395 Citron N, Taylor J (1987) Tendon transfer in partially anaesthetic hands. J Hand Surg 12B:14–18 Clark JMP (1956) Muscle and tendon transposition in poliomyelitis. In: Sir Harry Platt (ed) Modern trends in orthopaedics (second series). Butterworths, London, pp 116–143 Close JA, Todd FN (1959) The phasic activity of the muscles of the lower extremity and the effect of tendon transfer. J Bone Joint Surg 41A:189–208 Copeland SA (1995a) Glenohumeral Arthrodesis. In: Copeland SA (ed) Operative shoulder surgery. Churchill Livingstone, New York, pp 267–278 Copeland SA (1995b) Thoroscapular arthrodesis. In: Copeland SA (ed) Operative shoulder surgery. Churchill Livingstone, New York, pp 267–278 Copeland SA, Howard RC (1978) Thoracoscapular fusion for facioscapulo-humeral dystrophy. J Bone Joint Surg 60B:547–551 Creasey G, Keith MW (1996) Principles of upper extremity surgery in tetraplegia. In: Piemer CA (ed) Surgery of the hand and upper extremity. McGraw Hill, New York, pp 1483–1497, Chapter 64 Dahlin LB, Komoto-Tufvesson Y, Salgeback S (1998) Surgery of the spastic hand in cerebral palsy. Improvement in stereognosis and hand function after surgery. J Hand Surg 223B:334–339 Diab M, Darras BT, Shapiro F (2005) Scapulothoracic fusion for facio scapulo humeral muscular dystrophy. J Bone Joint Surg 87A: 2267–2275 Dunnet W, Housden P, Birch R (1995) Results of flexor to extensor tendon transfer. J Hand Surg 20B(1):26–28 Eden R (1924) Zur Behandlung der Trapezius Lähmung mittels Muskelplastik. Deutsche Zeitschrift Chir 184:387–397 Egmond V, Tonino AJ, Kortleve JW (2001) Steindler flexorplasty of the elbow in obstetrical brachial plexus injuries. J Paed Orth 21:169–173 Eliasson AC, Ekholm C, Carlstedt T (1998) Hand function in children with cerebral palsy after upper-limb tendon transfer and muscle release. Dev Med Child Neurol 40:612–621 Féry A, Sommelet J (1987) La paralysie du grande dentelé. Resultat de traitement de 12 cas dont 9 operes et revue generale de literature. Rev Chir Orthop 23:277 Finochietto R (1920) Retracción de Volkmann de lo Músculos Intrinsecos de las Manos. Bol Trab Soc Cir Buenos Aires 4:31 Fletcher I (1981) Amputations of the upper limb. In: Wynn Parry CB (ed) Rehabilitation of the hand, 4th edn. Butterworths, London, pp 331–354 Fridén J, Lieber RL (1998) Evidence for muscle attachment at relatively long lengths in tendon transfer surgery. J Hand Surg 23A:105–110 Fusi P (2003) Valutazione clinica dell’intervento di traspocione di piccoli pettorale pro bicipite nei deficit di flexxione di gomito consequenti a paralisi ostetrica di plesso brachiale. Tesi di Specializzazione 701728, Università degli studi di Brescia Gerber C, Maquieria G, Epinosa N (2006) Latissimus dorsi transfer for the treatment of irreparable rotator cuff tears. J Bone Joint Surg 88A:113–120 Gousheh J, Arasteh E (2006) Transfer of a single flexor carpi ulnaris for treatment of radial nerve palsy. J Hand Surg 31B:542–546 Gülgönen A (2001) Surgery for Volkmanns ischaemic contracture. Review article. J Hand Surg 26B:283–296 Harii K, Ohmari T, Torii S (1976) Free gracilis muscle transplantation with microneurovascular anastomoses for the treatment of facial paralysis. Plast Reconstr Surg 57:133–143 Harrison DH (2002) The treatment of unilateral and bilateral facial palsy using free muscle transfers. Clin Plast Surg 29:539–549 Harrison DH (2005) Surgical correction of unilateral and bilateral facial palsy. Postgrad Med J 81:562–567 Herndon GH (1961) Tendon transplantation at the knee and foot. In: American academy of orthopaedic surgeons, Instructional course lectures, vol 18. CV Mosby, St Louis, pp 145–168
604 Herndon CH, Strong JM, Heyman CH (1956) Transposition of the tibialis anterior in the treatment of paralytic talipes Calcaneus. J Bone Joint Surg 38A:751–760 Hovnanian P (1956) Latissimus dorsi transplantation for loss of elbow flexion or extension at the elbow. Ann Surg 143(4):493–499 Hunter E, Laverty J, Pollock R, Birch R (1999) Non operative treatment of fixed flexion deformity and the proximal inter phalangeal joint. J Hand Surg 24B:281–3 Ianotti JP, Hennigan S, Herzog R, Kella S, Kelley M, Leggin B, Williams GR (2006) Latissimus dorsi tendon transfer for irreparable posterosuperior rotator cuff tear. J Bone Joint Surg 88A:342–348 Ikuta Y, Kubo T, Tsuge K (1976) Free muscle transplantation by microsurgical technique to treat severe Volkmann’s contracture. Plast Reconstr Surg 58:407–411 Jamieson A (2003) (pers. com) Trapezius transfer for the flail shoulder in adult lesions of the brachial plexus. Paper read to the British Brachial Plexus Club. Glasgow 2003, organised by T. Hems Jones R (1916) On suture of nerves, and alternative methods of treatment by transplantation of tendons. Brit Med J 1:641–643 Jones R (1921) Orthopaedic surgery of injuries. Oxford University Press, Hodder, London Kay SPJ (2000) Free tissue transfer in children. In: Gupta A, Kay SPJ, Scheker LR (eds) The growing hand, 1st edn. Mosby, St Louis, MO New York, pp 969–986, ISBN OT 234 21331 Kuhlmann RF, Bell JF (1952) Clinical evaluation of tendon transplanting for poliomyelitis affecting the lower extremities. J Bone Surg 34A:915–926 L’Episcopo JB (1939) Restoration of muscle balance in the treatment of obstetrical paralysis. NY State J Med 39:357–363 Lamb DW (1987) The upper limb and hand in traumatic tetraplegia. In: Lamb DW (ed) The paralysed hand. Churchill Livingstone, Edinburgh, pp 136–152 Lamb DW, Kuczynski K (1981) The practice of hand surgery. Blackwell, Oxford Lambrinudi C (1927) New operation on drop foot. Brit J Surg 15:193–200 Landi A, Caserta G, Della Rosa N (1998) Patologia neurologia: il gomito nella tetraplegia. In: Monografia SICM. La Patologia non traumatico del Gomito Mattioli: Fidenza, pp 91–100 Lange M (1951) Die Behandlung der Irreperablem Trapezius Lähmung. Lagenbecks Arch Klin Chir 270:437–439 Le Couer P (1967) Procédés de Restoration de la flexion du coude paralytique pars transplantation du petit pectoral. Rev Chir Orthop 53:357–372 Leclerq C (2003) General assessment of the upper limb. Hand Clin 19:557–564 Lieber RL, Murray WM, Clark DL, Hentz VR, Friden J (2005) Biomechanical properties of the brachio-radialis muscle: implications for surgical tendon transfer. J Hand Surg 30A:273–282 Lim AYT, Lahiri A, Pereira BP, Kumar VP, Tan LL (2004) Independent function in a split flexor carpi radialis transfer. J Hand Surg 29A:28–31 Lipscomb PR, Sanchez JJ (1961) Anterior transplantation of the posterior tibial tendon for persistent palsy of the common peroneal nerve. J Bone Joint Surg 43A:60–66 Littler E (1949) Tendon transfer and arthrodeses in combined median and ulnar nerve paralysis. J Bone Joint Surg 31A:225–234 MacQuillan A, Horlock N, Grobelaar A, Harrison D (2004) Arterial and venous anatomical features in the pectoralis minor muscle flap pedicle. Plast Reconstr Surg 113:872–876 Manktelow RT (1993) Functional microsurgical muscle transplantation. In: Tubiana R (ed) The hand, vol 4. WB Saunders, Philadelphia, pp 313–318 Manktelow RT, McKee NH (1978) Free muscle transplantation to provide active finger flexion. J Hand Surg 3:416–426
Surgical Disorders of the Peripheral Nerves Marshall RW, Williams DH, Birch R, Bonney G (1988) Operations to restore elbow flexion after brachial plexus injuries. J Bone Joint Surg 70B:577–582 McMurray TP (1919) Discussion of the indications, technique and results of transplantation in gunshot injuries of nerve. J Orthop Surg 1(3):125–135 (old series xvii no. 3) Moberg E (1975) Surgical treatment for absent single hand grip and elbow extension in quadriplegics. J Bone Joint Surg 57A:196–206 Moberg E, Lamb DW (1980) Surgical rehabilitation of the upper limb in tetraplegia. Proceedings of the international conference in edinburgh. Hand 12:209–213 Molen Van Der A, Evans DM (1996) Results of transfer of the extensor indicis for restoration of adduction of the thumb. Presented to the combined meeting of Hellenic and British societies for surgery of the hand Morris SF, Zuker RM (2000) Functioning muscle transfers. In: Gupta A, Kay SPJ, Scheker LR (eds) The growing hand, 1st edn. Mosby, St Louis, pp 1021–1034, ISBN OT 234 21331 Moutet F (2005) Paralysis of the thumb. In: Tubiana R, Gilbert A. Taylor and Francis, Abingdon. United Kingdom. ISBN 1-85317494-7, Chapter 11, pp 167–186 Mulcahey MJ, Lutz C, Kozin SH, Betz RR (2003) Prospective evaluation of biceps to triceps, and deltoid to triceps for elbow extension in tetraplegia. J Hand Surg 28A:964–971 Nagano A, Obinaga S, Ochiai N, Kurokawa T (1989) Shoulder arthrodesis by external fixation. Clin Orthop 247:97–100 Napier JR (1955) Prehensile movements of the human hand. J Anat 89:564 Narakas AO (1993) Paralytic disorders of the shoulder girdle. In: Tubiana R (ed) The hand, vol 4. WB Saunders, Philadelphia, pp 112–125, Chapter 9 Ober FR (1933) Tendon transplantation in the lower extremity. New Eng J Med 209:52–59 Omer G (1988) Combined nerve palsies. In: Green DP (ed) Operative hand surgery, 2nd edn. Churchill Livingstone, Edinburgh, pp 1555–1568, Chapter 40 Parkes A (1973a) Ischaemic effects of external and internal pressure on the upper limb. Hand 5:105–112 Parkes A (1973b) Paralytic claw fingers – a graft tenodesis operation. Hand 5:192–199 Pearle AD, Voos JE, Kelly BT, Chehab EL, Warren RF (2007) Surgical technique and anatomic study of latissimus dorsi and teres major transfer. J Bone Jt Surg 89A(Suppl 2):284–96 Pho RWH (1991) Vascularised bone grafts in upper extremity reconstruction. In: Meyer VE, Black MJM (eds) Microsurgical pocedures. No. 7 in the hand and upper limb series. Churchill Livingstone, Edinburgh, pp 158–171 Raimondi P, Muset y Lara A, Saporits I (2001) Palliative surgery: shoulder paralysis. In: Gilbert A (ed) Brachial plexus injuries. Martin Dunitz, London, pp 225–238, Chapter 23 Reidy JA, Broderick TF, Barr JS (1952) Tendon transplantation in the lower extremity A review of end results in poliomyelitis. JB JS 34A:900–908 Revol MP (1993) Tenodesis. In: Tubiana R (ed) The hand, vol 4. WB Saunders, Philadelphia, pp 93–98, Chapter 7, English translation, ISBN 0–7216–8910–8 Richards BM (1989) Interosseous transfer of tibialis posterior for common peroneal palsy. J Bone Joint Surg 71B:834–837 Romero J, Gerber C (2003) Levator scapulae and rhomboid transfer for paralysis of trapezius. The Eden Lange procedure. J Bone Joint Surg 85B:1141–5 Ropars M, Dréano T, Siret P, Belot N, Langlais F (2006) Long term results of tendon transfers in radial and posterior interosseous nerve paralysis, J Hand Surg 31B:502–506 Roper BA (1972) External rotation palsy of the shoulder. Proc Roy Soc Med 65:375–386
Reconstruction Ross A, Birch R (1993) Reconstruction of the paralysed shoulder after brachial plexus injuries. In: Tubiana R (ed) The hand, vol 4. WB Saunders, Philadelphia, pp 126–133, English translation Sabapathy SR, Gowda DKL, Ranade AB, Venkatramani H, Sebastin SJ (2005) Functional outcome of extensor carpi radialis longus transfer for finger flexion in post traumatic flexor muscle loss. J Hand Surg 30A:267–272 Saul KR, Murray WM, Hentz VR, SL DELP (2003) Biomechanics of the Steindler flexorplasty surgery: A computer simulation study. J Hand Surg 28A:979–986 Schwartz RJ, MacDonald M (2003) Assessment and results of opponensplasty. J Hand Surg 28B:593–596 Schwartz RJ, MacDonald MRC, van der PAS M (2006) Results of arthrodesis in neuropathic feet. J Bone Joint Surg 88B:747–750 Schwartzmann JR, Crego CH (1948) Hamstring tendon transplantation for relief of quadriceps femoris paralysis in residual poliomyelitis. A follow up study of 134 cases. J Bone Joint Surg 30A:541–549 Scope Report (2008) Young people with cerebral palsy in transition from paediatric to adult health. Best practice recommendations. Scope. 6 Market Road, London. N7 9PW. HEMIHELP Camelford House, 89 Albert Embankment, London SE1 7TP Ref. Code: DYS06011 Scott WA (1985) Infected non-union of a femur treated with a vascularised fibular graft. Injury 16:455–456 Shanghai Sixth People’s Hospital (1976) Free muscle transplantation by microvascular neurovascular anastomoses. Chinese Med J 2:47–50 Sharrard WJW (1955) The distribution of the permanent paralysis in the lower limb in poliomyelitis. J Bone Joint Surg 37B:540–558 Sharrard WJW (1971) Paediatric orthopaedics and fractures, 2nd edn. Blackwell, Oxford Smith RJ (1987) Tendon transfers of the hand and forearm. In: Monographs in hand surgery. Little, Brown and Co, Boston, ASBN 0–136–801 747–7 C Smith PF (2006) Louis Stromeyer (1804–76): German orthopaedic and military surgeon and his links with Britain. J Med Bibliogr 14:65–74 Smith R, Wood PLR (2007) Arthrodesis of the ankle in the presence of a large deformity in the coronal plane. J Bone Joint Surg 89B:615–19 Soares D (1996) Tibialis posterior transfer for the correction of foot drop in leprosy. J Bone Joint Surg 78B:61–62 Srinavasan H, Mukherjee SM, Supramanian RA (1968) Two tail transfer of tibialis posterior for correction of dropped foot in leprosy. J Bone Joint Surg 50B:623–628 Srinivasan H (2004) Atlas of corrective surgical procedures commonly used in leprosy. Dr. H Srinivasan, Chennai, India Srinivasan H, Palande DD (1997) Essential surgery in leprosy, Techniques for District Hospitals. Action Programme for the elimination of leprosy, World Health Organisation Standring S (2005) Functional anatomy of the musculo-skeletal system. In: Standring S (ed) Gray’s anatomy, 39th edn. Elsevier, 135, pp 83–135 Steindler A (1918) A muscle plasty for the relief of flail elbow in infantile paralysis. Interstate Med J 35:235–241 Tamai S, Komatsu S, Sakamoto H et al (1970) Free muscle transplants in dogs with microsurgical neurovascular anastomoses. J Plas Recon Surg 46:219–225
605 Taylor GI, Miller GDH, Ham FJ (1975) The free vascularised bone graft. A clinical extension of microvascular techniques. Plas Reconstr Surg 55:533–544 Tubiana R (1993) Paralysis of the thumb. In: Tubiana R (ed) The hand, vol 4. WB Saunders, Philadelphia, pp 182–253, Chapter 13 Tubiana R (2005a) Paralysis of the radial nerve. In: Tubiana R, Gilbert A (eds) Tendon nerve and other disorders. Taylor and Francis, Abingdon, pp 147–165, ISBN 1-85317-494-7, Chapter 10 Tubiana R (2005b) Paralysis of the ulnar nerve. In: Tubiana R, Gilbert A (eds) Tendon nerve and other disorders. Taylor and Francis, Abingdon, pp 187–214, ISBN 1-85317-494-7, Chapter 12 Tubiana R, McCullough CJ (1984) Paralysis of the intrinsic muscles of the hand. In: Brooks DM, Birch R (eds) Rob and Smiths operative surgery. The hand, 4th edn. Butterworths, London, pp 318–343 Turner JW, Cooper RR (1972) Anterior transfer of the tibialis posterior through the interosseous membrane. Clin Orthop Rel Res 83:241 Valenti P (2005) Associated paralysis. In: Tubiana R, Gilbert A (eds) Taylor and Francis. Taylor and Francis, Abingdon, pp 215–222, ISBN 1-85317-494-7, Chapter 13 Watkins MB, Jones JB, Ryder CT, Brown TH (1954) Transplantation of the posterior tibial tendon. Journal Bone Joint Surg 36A:1181–1189 Weiland AJ, Russell Moore J (1988) Vascularised bone grafts. In: Green DP (ed) Operative hand surgery, vol 2, 2nd edn. Churchill Livingstone, Edinburgh, p 1245 Wilkinson MCP, Birch R, Bonney G (1993) Brachial plexus injuries: when to amputate. Injury 24(9)):603–605 Wynn Parry CB (1981) Rehabilitation of the hand, 4th edn. Butterworth, London Yeap JS, Birch R, Singh D (2001a) Long-term results of tibialis posterior tendon transfers for foot drop. Int Orthop 25(2):114–8 Yeap JS, Birch R, Singh D (2001b) A method of evaluating the results of tendon transfers for foot drop. Clin Orth Rel Res 383:208–213 Yücetürk A (2001) Palliative surgery: tendon transfers to shoulder in adults. In: Gilbert A (ed) Brachial plexus injuries federation of European societies for surgery of the hand (FESSH). Martin Dunitz, London, pp 115–122, ISBN 1–84184–015–7, Chapter 12 Zachary RB (1946) Tendon transplantation for radial paralysis. Br J Surg 33:358–364 Zachary RB (1947) Transplantation of teres major and latissimus dorsi. Lancet 2:757–758 Zancolli EA (1957) Claw hand caused by paralysis of the intrinsic muscles A simple surgical procedure for its correction. J Bone Joint Surg 39A:1076–1080 Zancolli EA (1979) Structural and dynamic bases of hand surgery, 2nd edn. JB Lippincott, Philadelphia Zancolli EA (2005) Surgery of the infantile spastic hand in cerebral palsy. In: Tubiana R, Gilbert A (eds). Taylor and Francis, Abingdon. ISBN 1-85317-494-7, Chapter 28, pp 409–426 Zancolli EA, Mitre H (1973) Latissimus dorsi transfer to restore elbow flexion. J Bone Joint Surg 55A:1265–1275
14
Rehabilitation
Effects of legislation; objects; origin, and the work of Robert Jones; the rehabilitation team at the Royal National Orthopaedic Hospital; methods of work and some changes in practice; motivation and examples of particular problems; persisting pain; the pain program, and the return to work. Paralysis not physically determined. Rehabilitation of the war wounded; the work of the Defence Medical Rehabilitation Centre (Headley Court) and of Col. John Etherington and his colleagues. The future. God help the man who falls into the hand of the state Dorothy Birch 1959
The state of rehabilitation in the National Health Service is, at best, variable and it is often deplorable. Exceptions are provided by the outstanding work of Special Units treating injuries to the spinal cord, the brain, or some progressive neurological disorders. In many other areas rehabilitation has become, yet again, the Cinderella speciality. Some reasons for this unacceptable state include:
ber of patients sustaining injuries. John Crawford Adams refused to countenance the establishment of a Sports Injury Clinic at St Mary’s Hospital for just this reason. 7. We fear that the chief responsibility for this state of affairs must rest with orthopaedic surgeons, far too many of whom seem to have forgotten about the chief reason for their speciality.
1. The abolition of continuity of care so that a fracture surgeon no longer sees their patient from start to finish. 2. The weakening or breakdown of team working and a fission of the components of that team so that departments of physiotherapy and occupational therapy often now run their own “waiting lists.” 3. The restricted availability of essential elements of a rehabilitation team such as speech therapy, clinical psychology and orthotist. The position of Hospital Employment Advisor has disappeared. 4. The farming out of rehabilitation work to private agencies, whose work is sometimes good, but whose referrals from insurance companies, solicitors and enlightened employers come far too late. 5. The confusion or ignorance of too many clinicians who seem unable to distinguish between convalescence and rehabilitation. A patient leaving the hospital after total hip arthroplasty ought to be much better off than when they came in whereas a patient losing a limb or suffering significant and permanent loss of function can never regain the level of function that they had enjoyed before their injury. This confusion has led to the loss of rehabilitation facilities, beds and wards, which have been converted into “step down” wards to enable a higher “through put.” 6. The glamor and reward of treating young healthy and highly motivated sportsmen and women has drawn attention, energy and resources away from the far larger num-
Legislation: It might be as well now to consider the implications of legislative changes. Formerly, employers of more than 20 persons were obliged to include a percentage of disabled people within their work force. That obligation was widely disregarded; it was in fact virtually unenforceable. All that was swept away by the Disability Discrimination Act of 1996. Disability was defined as a physical or mental impairment which has a substantial and long-term adverse effect on a person’s ability to carry out normal day to day activities. People with existing or earlier disability were covered by the Act. It became unlawful for employers with 20 or more staff to discriminate against current or prospective employees with disability, because of reasons relating to this disability, and employers were obliged to make reasonable adjustments if their employment arrangements or premises substantially disadvantaged a disabled employee or a disabled applicant. Unlawful discrimination was defined; circumstances were defined in which less favorable treatment of a disabled person is justified. Those covered by the Act were named; categories of exempt persons such as those serving in the Armed Forces, were defined. One may speculate that had these regulations been in force 200 years ago the victor of the Nile, Copenhagen and Trafalgar would not have come as he did to his triumph and his death. The balance was decisively altered: the employer was no longer obliged to include a proportion of disabled persons in the work force; rather, it was for the disabled person refused a
R. Birch, Surgical Disorders of the Peripheral Nerves, DOI: 10.1007/978-1-84882-108-8_14, © Springer-Verlag London Limited 2011
607
608
job to show that the refusal stemmed from discrimination because of his or her disability. Cox (1996) commented quite sharply that the guidance notes published to help interpretation of the provisions of the Act “signal a contentious time ahead for health professionals” He concluded: “…it is in the area of employment that doctors, particularly occupational physicians, will be most involved. Their expertize must ensure that many of the Act’s ambiguities and uncertainties are resolved with fairness to both disabled people and employers. It will require the wisdom of Solomon.” It remains to be seen whether the most recent Disability Discrimination Act (2005) will make a material difference. We think that it will prove an error to group together the problems facing those with disability under the umbrella of discrimination within the Equality and Human Rights Commission. Many patients do not consider themselves disabled simply because the injury impairs their ability to function, some, indeed, would reject the suggestion with disbelief or even anger. Many do not want the label of “disabled” attached, rather they seek for active and useful support to enable them to get on with their lives. The system is open to abuse from all sides. It is disheartening that parents of young children with defects of function at the shoulder from birth injuries of the brachial plexus ask for support in their applications for attendance allowance and disability car badges. The recent Act does not address the widespread failures of the National Health Service to assist people to overcome the aftermath of injuries to peripheral nerves and to their limbs and so enable them to lead independent lives within their own control; instead it widens the gulf between the hospital service and the Disability Employment Advisor at the Job Centre. It will bring about an adversarial state between employer and employee which will benefit only lawyers. It does not address the needs of those who, because of age or the severity of injury, will not be able to work. We are concerned here with the rehabilitation of patients who have suffered serious damage to the peripheral nervous system but it is a pleasure to acknowledge the great help provided by Dr. Frederick Middleton, Dr. Angela Gall and their team in the Spinal Injury Unit at the Royal National Orthopaedic Hospital from whom we learnt many important lessons which have been applied in the treatment of patients with peripheral nerve injuries. The objects of rehabilitation include: 1. The objective assessment of disability and the accurate measurement of the outcome of treatment 2. The reduction of the degree of disability by physical and other therapies 3. The return of the patient to his or her original work, to the original work modified or to suitable other work 4. The restoration of the patients’ ability to live in his or her own home, to enjoy recreation and social intercourse and to be independently mobile
Surgical Disorders of the Peripheral Nerves
The process starts with treatment of the injury and the consequences of that injury must be borne in mind from the outset so that plans are made to minimize disability. This is possible only with a clear and accurate diagnosis and prognosis. One of the most damaging consequences of adopting a policy of “wait and see” after injury to a main nerve is the effect that this has on the patient’s morale, on their ability to cope and on their ability to plan for the future. Appropriate and timely later operations, including those for correction of rigid deformity, pain, scar, revision of a nerve repair and palliative musculotendinous transfers are integral to the process of rehabilitation. Russell Davis (1949) performed an important study of 82 patients with peripheral nerve injuries, and concluded that: “Recovery had been better when the patient recognised an objective for treatment; the earlier in the treatment the objective had become explicit, the greater had been its value as an incentive. Neurotic secondary gain had been of no practical importance in delaying recovery.” We agree with London (1967): “any doctor looking after an injured or otherwise disabled person should accept responsibility for seeing that he or she is, far as possible, restored to their previous position in society.” Wynn Parry’s definition of rehabilitation as the “staged withdrawal of support” is the best there is.
14.1 Communication A great deal is made nowadays of “communication.” All too often this means using “hand outs” to patients which do not address their own specific needs and which lead to confusion and often to dismay. There is nothing better than a face to face conversation in which the problem is outlined based on a proper understanding of the extent of the injury, what has been done in the operation of repair and the prognosis. This clinical skill can be acquired only by practise. It is remarkable how so many patients, no matter what their educational background, grasp the essence of their case after such a conversation perhaps aided by a simple diagram of that patient’s own injury. Mr. Joe Connelly (of the Venture Trust) has had a very great deal of experience in working with young people who have gone off the rails and he relates this conversation that he had with a young man with whom he was trying to build a degree of confidence and self respect by using the simple analogy of making good lentil soup: “You take six to eight carrots, scrape them, chop them and then put them in a pot. There was a lengthy silence and then the young man looked at me and said ‘you’ll nae get 68 carrots into wan pot’.” It is important to be both precise and clear!
Rehabilitation
14.2 History In some respects Robert Jones’s greatest contribution was towards the rehabilitation of the wounded (Fig. 14.1). His principles for this included segregation and continuity of treatment. He wrote to Makins after the war as follows: “the treatment of those cases in the early part of the war was deplorable. I heard frequently “we cannot follow up our cases.” During the first 12 months of war no provision of any sort was made for cases crippled and/or deformed and this became a focus for seeming discontent” (Jones 1921). So was founded the Alder Hey Orthopaedic Hospital in early 1915 and then later the great military hospital at Shepherd’s Bush. Jones’ idea of a military rehabilitation hospital was that they would provide cohesion between the compartments of treatment. At its height there were 800 patients at Shepherd’s Bush, 500 of whom were employed at regular physical work and as he said: “the difference in atmosphere and morale in hospitals where patients have nothing to do but smoke, play cards or be entertained from that found in those where, for part of the day they have regular useful and productive work, is striking” (Jones 1921).
Fig. 14.1 Robert Jones 1857–1933.
609
By 1917 this principle of structured rehabilitation had been extended to 14 other centres and the fruits of this are described in the article by Goldthwait (1917) one of the orthopaedic surgeons from the United States of America who volunteered their services to Great Britain and the Western Allies long before their country’s formal engagement in conflict. Goldthwait relates how orthopaedic physicians and surgeons became responsible for so much of this work because they were trained to think of results in terms of function. Between 30% and 50% of all military hospital beds were occupied by orthopaedic conditions, including all cases of nerve injury but excluding the limbless patients who were treated in special hospitals. One of the earliest of these, Roehampton House, fitted nearly 8,000 limbs during the war under the direction of Muirhead Little. Goldthwait identified two particular benefits of the curative workshops. First was the psychological, from engagement in useful trades including carpentry, metal work, leather, cobbling, tailoring, net making, basket making and draughtmanship which enabled the injured men to enhance previous skills or to learn new ones and engage in the manufacture of splints, braces, shoes, clothing, and furniture for the hospital and other services. Second came the therapeutic advantage from appropriate modification of the daily work. A stiff wrist could be mobilized by using a carpenter’s plane, gradually increasing the length of the stroke as the wrist loosens; pronosupination could be improved by using the edge of the plane and rotating it. For stiff elbows a short bladed saw was used, increasing the length of the blade and of the cut as the elbow loosened. The jigsaw or other machine operated by the legs helped to improve the range of movement in the joints of the lower limbs. Stewart (2009) in his Robert Jones lecture “Wounds of mind and body” describes Osgood’s (1918) observations from the Old Mill Military Hospital in Aberdeen and the work of the patients on deep sea nets: “they can dispose of any number, and the exercise for stiff fingers in making the mesh, and for stiff arms and legs as they sway back on the stout cords is admirable.” The results were impressive, for of the first 1,350 patients admitted to Shepherd’s Bush 1,000 were fit enough to return to military duties and 350 returned to civil life but they were at least trained, and of use to themselves and the community. Goldthwait ends with this poignant comment: “plans are underway that involve the taking over of several of the largest of the orthopaedic centres after the war is over for the continuance and permanent development of this work.” This was not to be. Robert Jones’ biographer, Frederick Watson (1934) wrote of the Shepherd’s Bush Military Hospital that: “ with a little imagination this might have become a permanent national institution; instead of which it was ordained by some inscrutable fate that it should arise from the mediocrity of the poor house and into that forlorn destiny return.”
610
Fig. 14.2 Christopher Wynn Parry characteristically at work in the Rehabilitation Unit.
The Second World War saw the rapid reinstitution of rehabilitation units and rehabilitation hospitals, much of that work being undertaken in the five Special Hospitals designated for nerve injuries. One of the earliest was established at Torquay under the direction of Watson Jones and Osmond Clark and this was rapidly followed by the establishment of other hospitals, seven of which remained within the jurisdiction of the Royal Air Force. RAF Chessington, survived after the war and was amalgamated with Headley Court which had been granted a degree of autonomy away from the Ministry of Defence and the National Health Service. This is now the Defence Medical Rehabilitation Unit. Christopher Wynn Parry joined the staff of Chessington and of Headley Court in the l940s and in 1975 he took over the position of Director of Rehabilitation at the Royal National Orthopaedic Hospital (Fig. 14.2). One of us (RB) joined him as co-Director of rehabilitation in 1979. There were examples of enlightened employers continuing with the work of occupational rehabilitation, one such was the rehabilitation centre set up in Crewe by the London Midland and Scottish Railway Company. Other examples of strongly vocationally orientated rehabilitation centres were seen in both nationalized and private industries. Nicol, orthopaedic surgeon to Mansfield and Chesterfield, established a rehabilitation hospital for injured miners at Berry Hill; Plewes established a very effective link with the Vauxhall motor company in Luton. The continuing value of workshop practice is illustrated in Fig. 14.3a–d.
14.3 The Rehabilitation Team The conventional term “multidisciplinary team” should not obscure the essential purpose which is the long term benefit of the patient. The distinction between the disciplines should not be rigid; it is important to recognize the particular skills of individuals, whilst maintaining a clear sense of purpose.
Surgical Disorders of the Peripheral Nerves
The Consultant has specific responsibilities which include: (1) The explanation of the diagnosis and the prognosis to the patient based on the findings at operation. Ill informed comments about the prognosis from others should be avoided or firmly and clearly corrected. The approach should be one of reasonable optimism, with an indication of the interval before recovery may be seen. An outline of what further operations may prove useful should be provided. (2) The Coordination of treatment of the patient and adequate and timely communication with the family practitioner, the referring clinician and others involved in treatment at the first hospital and in the community. (3) The definition and the timing of later operation. (4) The support of staff against pressure or abuse from managers, patients, their families or other agencies. Nursing Staff should: (1) ensure that the policies of investigations and assessments and of consent have been followed in the outpatient department. There, nurses explain to the patient what is involved during an admission to hospital. Nursing staff are responsible for: (1) wound care; (2) the correction of fixed deformity in liaison with a physio or occupational therapist; (3) the in patient nursing staff have a particular responsibility for learning about the patients’ anxieties and concerns and reporting these. The Physiotherapists: (1) measure deformity and power and teach exercises which will prevent or correct deformities; (2) design a programme of exercises to improve power, coordination, endurance and to improve balance and gait; (3) are responsible for the use of the transcutaneous nerve stimulator; (4) the ward physiotherapists are responsible for the mobilization of a limb after it has been set free from splints or bandages and teaching the patient how to improve walking, balance, and confidence (Fig. 14.4a, b). The Occupational Therapists: (1) measure function by systems appropriately modified to the patient, by age, and by condition and which complement those made by nursing staff and physiotherapists; (2) manufacture static or dynamic splints to enhance function, to test whether musculo tendinous transfers will suit that patient, to apply such splints before operation and to modify them as necessary after operation; (3) treat the sensitive, painful or florid scar, in liaison with nursing staff; (4) show patients the possibilities of orthoses and liaise with the orthotist in their design, their fitting and training; (5) measure disadvantages affecting daily living and offer advice and treatment by special aids; (6) the inpatient occupational therapist has particular responsibilities for advice about returning to driving and, in the chronic pain programme, return to work. The clinical psychologist measures the cognitive disturbance provoked by head injury and the depression of mood caused by pain or injury and plans measures to help. The Orthotists are responsible for the provision of modified shoe wear, of splints to overcome foot drop or other deformities of the ankle and foot and the measuring, manufacture and fitting of orthoses for the paralyzed upper limb.
Rehabilitation
a
c
611
b
d
Fig. 14.3 (a–d)Workshop training in the Rehabilitation Unit. The patients are using an earlier model of the flail arm splint using various terminal attachments including the split hook and the pliers.
Other skills may prove necessary, such as speech therapy for some patients after head injury, or dieticians for those recovering from multiple injuries. The advice of consultant psychiatrists is sometimes necessary.
14.4 The Method of Work One significant change has been the move away from inpatient work to outpatient based work over the last 12 years. The number of patents admitted to the Rehabilitation Unit in the year 2001 was 156, an appreciable decline from earlier years. The current figure for inpatient admissions is now about 100. The number of patients whose treatment had been carried out in the outpatient occupational therapy department has nearly
doubled during those years and now stands at about 300 patients a year, each patient requiring, on average, 3.5 sessions of treatment. The changes in practice for patients undergoing operations of reconstruction are set out in Chap. 13. Physiotherapy and the birth lesion of the brachial plexus (BLPP): the recognition of late onset shoulder dislocation or other shoulder deformity and the growing appreciation of the problem of secondary contractures after successful relocation has stimulated the development of a more prolonged and vigorous programme of supervision, in which the parents and the child take the central part. Because of the far flung nature of the practice close liaison with local services is essential. Mr. Tony Betts (MCSP) advises that the annual number of new patients with nerve injuries treated as outpatients in the physiotherapy department is about 140, most of whom come from the London area, and he outlines the
612
a
b
Surgical Disorders of the Peripheral Nerves
need encouragement, for they are apprehensive about doing harm in a fretful child. They are advised about methods of engaging the upper limb in normal activities and play, and are shown how to encourage leading with and guiding the afflicted arm. This may go some way to overcome the problem of the apraxia which is so commonly seen in the first years, in which the child wholly ignores the afflicted limb (see Chap. 10). After the age of 5 years, and before adolescence, the children are encouraged to do much more themselves and specific exercises are taught involving both arms, using pulleys or a stick are taught, in which the joints of the shoulder girdle are worked through their maximum range. During these years such activities as dancing, gymnastics, swimming and ball games are certainly useful. The adolescent child may prefer not to focus on the arm, and often resent the idea of “disability.” For these, a session of massage and stretching which reveals a limb which is more supple and less painful encourages them to continue with their own efforts. Swimming, throwing, catching and racquet sports are particularly useful to encourage integration of the upper limb. Throughout treatment attention must be paid to posture and balance and particular attention is paid to the common problem of a secondary postural scoliosis. The fixed deformity of the elbow is treated by the method outlined in the previous chapter. A prolonged course of stretching with serial splinting is necessary. Outpatient work depends on the continuing engagement of local services, which is often good but not always so. There is a growing trend to “rationing” the number of attendances which should be resisted. All patients with severe injuries to the peripheral nerves require prolonged observation.
14.5 Some Technical Aspects
Fig. 14.4 (a, b) Patients are taught stretching exercises which they continue themselves.
current approach for about 60 new cases of BLPP in every year, about one half of whom require continuing supervision. Severe or rapidly progressing deformities are seen every week or every other week; after that, according to the response to treatment, attendances taper away to once in every 6 months. Particular attention is necessary during the periods of growth spurt, between the ages of 7 and 9 and again between 12 and 14 or thereabouts. In children aged 5 years or less the work is done by the parents who are taught techniques of massage, stretching, and particular stretches for the capsule and muscles of the glenohumeral joint. These exercises are carried out several times a day. Parents
Desensitization: a painful scar or hypersensitive part or limb is a serious bar to mobilization and strengthening. It inhibits active movement, walking, the use of splints, orthoses or prostheses. Ms. Jude Brook (Occupational Therapy Department, Royal National Orthopaedic Hospital) has devised a method which has proved remarkably effective. The degree of sensitivity is graded by the patient as (1) unbearable, (2) tolerable, and (3) minor or absent. The patient is then shown simple methods of desensitization by massage and is taught a method of desensitization in which the skin is exposed to a series of textures graded from one (very fine) to 20 (coarse). The skin is then exposed to a range of particles which are hard, such as lentils or rice, to very soft, such as tissue paper. These are graded from A to H. A suitable container can be used into which the hand or the foot can be immersed. The patient continues with this treatment at home, applying it 6–8 times in every day, and a record is kept of the response of symptoms to the different textures and particles. Their progress is reviewed at intervals of 2–3 weeks. Whilst the method is less effective in the
Rehabilitation
613
Fig. 14.6 A 68 year old man sustained a fracture of the proximal humeral shaft. This sling was provided, without any programme of passive and active work to maintain range of movement. Correction of the fixed deformities in the right upper limb required several weeks of intensive treatment.
conventional devices. Their true value cannot be determined until a system of financial audit is introduced within the National Health Service which actually measures the cost to Fig. 14.5 Pain and severe hypersensitivity following an operation upon the ulnar nerve at the elbow was improved by a course of desensitisation and the use of a padded silicone sleeve.
sensitive skin associated with a neuroma it has helped even some of these intractable situations (Fig. 14.5). Functional splinting: it is important to remember the limitations of splints which are generally available commercially (Fig. 14.6). One splint which is commonly used to stabilize the wrist, and which is secured in place by Velcro strips, usually extends beyond the palmar crease and immobilizes the metacarpophalangeal joints in extension and it often provokes swelling. The conventional standard ankle/foot orthosis has already been condemned. Mrs. Kathryn Johnson (Dip COT) emphasizes that splints must be made for the individual patient according to need and they should be seen as part of the exercise programme. Not only can such splints prevent deformity and enhance use by placing joints in a useful position, they also allow strengthening of weak muscles or of muscles after transfer by modifying the resistance of the spring or other restraining device. Well made splints enhance rebalancing, by controlling, stabilizing or modifying compensatory or “trick” movements. Their manufacture requires a high level of skill which can be acquired only by practice and becoming familiar with the very wide range of materials which are available. From these, materials are selected for their ease of working, their comfort and their strength (Fig. 14.7a–d). Orthoses: there are a wide range of orthoses fabricated from materials suitable for the hypersensitive, the insensate or the trophic skin but many of these are more costly than
Fig. 14.7 Examples of functional splinting By courtesy of Mrs Kathryn Johnston DipCOT. (a) Opposition splint for paralysis of thenar muscles. (b) Another type of opposition splint in a patient with a hereditary sensory motor neuropathy, (c) A useful splint for persisting weakness of the small muscles of the hand in a 28 year old man two years after primary suture of the median and ulnar nerves and of the common brachial artery in the axilla, (d) A simple ankle/foot orthoses helped this 82 year old woman with a foot drop complicating total hip arthroplasty. The previous conventional splint had provoked a sore.
614
Fig. 14.7 (continued)
Surgical Disorders of the Peripheral Nerves
Rehabilitation
Fig. 14.7 (continued)
615
616
the patient and to society of continuing dependence and which includes the costs of treatment of the infected trophic sore. Such a system does not exist, and it cannot exist with the present division between “purchaser” and “provider.” Mr. David Nursaw (orthotist, Royal National Orthopaedic Hospital) has provided details about two orthoses which we have used regularly for patient with foot drop. The silicone ankle foot orthosis is made by Dorset Orthopaedic, Ringwood, Dorset, UK (email: http://www.dorset-ortho.co.uk). A light weight orthosis, which is designed for use with or without a shoe, is manufactured by Technology in Motion, 132–134 Arthur Road, Wimbledon, London SW19 8AA (email: info@ technologyinmotion.co.uk (Fig. 14.8a–d). The flail arm splint: the concept and practice of an orthosis designed to restore some function to the complete paralyzed upper limb has existed in the Rehabilitation Unit at the Royal National Orthopaedic Hospital for over 40 years. It is agreeable to remember that one of those engaged in their fabrication was himself one armed because of an injury to the brachial plexus. However, earlier versions of the orthosis were cumbersome, heavy and difficult to don. Up to 1988 about one patient in three with a complete lesion found that the splint was of lasting benefit (Wynn Parry 1998). These defects were recognized by Dr. Robin Platt (Royal National Orthopaedic Hospital) and he was engaged in work to improve support of the subluxated shoulder, introduce lateral rotation and prono supination before his untimely death. It became clear that it was a mistake to offer the device to patients too soon after their injury, for many of these found the implication of lasting, of permanent disability, quite unacceptable. Instead the matter was discussed with patients at the subsequent meeting in the outpatient department, and those who expressed an interest were provided with further information before making any commitment to measuring, fitting and training. However, many patients gained great benefit from the earlier models (Fig. 14.9a, b). The last few years have seen important advances in the design of the orthosis. We are grateful to Dr. Martin Middleton (Royal National Orthopaedic Hospital) for providing information about current thinking and continuing development. A number of problems have been addressed. The correction of glenohumeral joint subluxation. The earlier splint achieved this by applying compression between the forearm and the shoulder girdle with the elbow flexed. That compression force was lost when the elbow extended. Some orthoses rely on a degree of circumferential compression of the arm with a supporting sling. One such device, the Omotrain (Bauerfriend UK Ltd. Telephone number 0121 698 8666) was introduced to us by Mr. Iain Henderson, Senior orthotist at the Raigmore Hospital (Inverness) and it has proved useful (Fig. 14.10). Dr. Middleton has developed an orthosis using a meshed, gently elastic, silicone tube which provides an effective suction
Surgical Disorders of the Peripheral Nerves
force upon the arm and is suspended from a simple shoulder harness (Fig. 14.11). The elbow locking component of the orthosis has been improved, by moulding the joint onto the medial side and by using a set of three point flexion forces which reduces the weight, the extent, and which eliminates additional strapping (Figs. 14.12a, b and 14.13a, b). Over 100 of these orthoses have been fitted. The rate of continuing use in daily life is high and there has been a good response from those using earlier models. Now, the device is offered to patients soon after operation as a temporary support whilst recovery is awaited. Interested patients are assessed in the occupational therapy department and shown examples and possibilities. After this that the patient is referred to the orthotist, for designing and for fitting. The training may be done as an outpatient for those patients who do not live too far away. Other patients are admitted for 1 week into the Rehabilitation Ward and this includes those who will use the splint in such trades as carpentry or paint spraying, and who will need to test the different terminal devices in the heavy workshop (Fig. 14.14). The rehabilitation Unit: The inpatient rehabilitation ward was developed by Wynn Parry as a 5 day facility (Monday to Friday). The patient is admitted with a clear plan of action and is assessed on the first day by nursing staff, physiotherapists and occupational therapists. These findings are analyzed at the rehabilitation ward round later that day. At that meeting the patient is advised of the findings at operation, and the plan of treatment. Most patients are admitted for 5 days, a decision may be made to extend that time towards the end of the first week. The results of treatment for return of work and pain relief in patients with lesions of the brachial plexus are described in Chap. 9. We shall now consider some examples of the process of rehabilitation which has been used with patients with nerve injuries or the consequences of nerve injuries. These examples illustrate the importance of motivation, the problems facing the older patient, the difficulties in progressive neurological disease, the choice of intervention, and chronic pain.
14.6 Motivation The motivation of the patient is the most important element in rehabilitation for without it, the process must fail. It may be necessary to take a firm line. Drunkenness, in patients in the Rehabilitation Ward, leads to final warning from nurse in charge; abuse of staff means instant discharge by the consultant. The usual outcome is a contrite patient at the outpatient clinic who begs to continue the course. The situation is more difficult for patients engaged in continuing legal process; it is not uncommon to receive letters from solicitors after every clinic attendance. It is for the consultant to offer the patient
Rehabilitation Fig. 14.8 (a–d) Some examples of useful ankle/foot orthoses. By courtesy of Mr. David Nursaw, Orthotist, Royal National Orthopaedic Hospital.
617
a
c
d
b
618
Surgical Disorders of the Peripheral Nerves
a
b
c
Fig. 14.9 Splints used for extensive paralysis of the upper limb (a) A 33year old woman sustained preganglionic injury of C7, C8 and T1 on the left side. She experienced severe pain. Pain was well controlled by transcutaneous nerve stimulation and she found the gauntlet component of the flail arm splint useful. (b) A 26 year old man with complete preganglionic injury of right brachial plexus used his flail arm splint
regularly whilst working full time as a gardener. (c) More than 40 patients with compete lesions of the brachial plexus had been able to retrain in such trades as paint spraying or welding using the flail arm splint. (d) A 48 year old man who had used his flail arm splint (the heavy duty variant shown here) whilst remaining in full time work.
Rehabilitation
619
Fig. 14.10 The Omotrain orthosis in a 32 year old man with irreparable injury to the suprascapular and circumflex nerves. Fig. 14.9 (continued)
the choice between working with the team or seeking compensation. Fortunately, it is more common to meet patients who rehabilitate themselves (Fig. 14.15a–d). Case report: A 48 year old right handed brick layer suffered a fracture of the right clavicle, of the contra lateral scaphoid, with rupture of the posterior divisions of the brachial plexus in a motor cycle accident. He was seen at 4 weeks by which time he was already back at work, He was supplied with heavy duty orthosis which overcame the painful subluxation at the gleno humeral joint and provided stability at the elbow and the wrist.. At operation, 6 weeks after injury, the ruptured posterior divisions were repaired and over the course of the next 18 months he regained power of abduction at the shoulder to MRC grade 4 and of extension at elbow to similar levels. A flexor to extensor transfer was performed 3 months after injury as the nerve repair could not be expected to restore extension of the wrist and of the digits. In this case, flexor carpi ulnaris was transferred to the extensor carpi radialis brevis for the sake of power. By 6 weeks after this operation his power of grip was 50% of the uninjured side and the
Fig. 14.11 Control of shoulder subluxation using a silicone sleeve with harness suspension. By courtesy of Dr. Martin Middleton.
620
a
Fig. 14.12 Modified light weight elbow splints. (a) A more recent pattern of elbow lock splint using carbon fibre (b) A modified (light weight) elbow splint in a 23 year old woman who sustained a rupture of C5 and avulsion of C6 and C7. Lateral rotation was restored by accessory to suprascapular transfer, repair of C5 was followed by useful
a
Surgical Disorders of the Peripheral Nerves
b
power in deltoid and in the clavicular head of pectoralis major, transfer of C5 to the ventral root of C7 restored elbow extension to grade MRC 4 and wrist extension to grade MRC 3 but elbow flexion did not recover. Courtesy of Dr. Martin Middleton.
b
Fig. 14.13 (a, b) The orthosis in current use from complete lesions of the brachial plexus.
Rehabilitation
Fig. 14.14 As a child aged 9 years this woman suffered complete lesion of the brachial plexus. Repair restored useful function at the shoulder and flexion at the elbow. She is now a skilled puppeteer.
range of active movement at the wrist and the digits was about 80% of normal. This patient had already done the work and it was the task of those treating him to provide, with a degree of urgency, correct advice about the appropriate orthosis, an accurate prognosis for recovery after repair of the nerves and an early palliative musculotendinous transfer. The splints were fitted and training given whilst he was an outpatient. Case report: A 28 year old, right handed unemployed man suffered a complete injury to his right brachial plexus in a motor cycle accident. He was subjected to a prolonged course of investigations, which were not contributory, over a period of 13 months during which time he became wholly demoralized. Operation was performed at 13 months which confirmed the clinical diagnosis of rupture of C5 and of C6 with avulsion of C7 and a recovering lesion of C8 and T1. The repair of the nerves probably contributed to an improvement in his pain and to stability at the gleno humeral joint but there was no recovery of elbow flexion nor of extension of
621
the wrist and the fingers. It was put to him that muscle transfers could improve function but these would not be undertaken unless and until he had demonstrated a clear commitment to finding and keeping work. He did just this and was admitted into the Rehabilitation Unit for a week of work to overcome fixed deformity and to improve muscle power. Two muscle transfers were performed at an interval of 9 weeks. Triceps was transferred to his biceps, restoring a power of elbow flexion at MRC Grade 4, with an active range of flexion of 110° He could lift a weight of five pounds. Then, flexor to extensor transfer was done, using flexor digitorum superficialis which considerably improved his hand posture. His power grip, which was 5% of normal before operation was improved to 20%. This patient was in full time work by 2 weeks after the final operation. Case report: An 18 year old right handed unemployed man sustained a severe wound to the right axilla when he was thrown out of the car in which he was a passenger. He was impaled upon a stake, and emergency repair of the axillary artery with cover of the defect by a myocutaneous flap was successful (Mr. Reid, Bristol). The musculocutaneous nerve and the biceps muscle had been destroyed, the median and ulnar nerves had been violently stretched. He then spent some months in prison for charges relating to the accident. We saw him at 8 months when he openly outlined this history and offered a firm commitment to “going straight.” The first question related to the prognosis for the nerves, for at first sight the outlook for the damaged upper limb seemed poor. However NPI confirmed extensive recovery for the circumflex and the radial nerves, and considerable reinnervation of muscles innervate by the median and the ulnar nerves. He was admitted to the Rehabilitation Unit for 2 weeks during which time work was directed towards overcoming deformity, relieving the hypersensitivity of the scar and strengthening those muscles which were active. Appropriate dynamic splints were made to enhance hand function after which the paralysis of the small muscles of the hand was mitigated by transfer of extensor carpi radialis longus to restore flexion at the metacarpophalangeal joints and of extensor indicis to enhance opposition of the thumb. He regained a pulp to pulp grip between the thumb and the index and middle fingers, his power grip improved from less than 10 to 25% of normal, and that of pinch grip from less than 5 to 20% of normal. He demonstrated assiduous dedication throughout his course of treatment and returned to full time work. He is now a happily married family man. The tasks of the rehabilitation team included the determination of an accurate prognosis, followed by appropriate work to improve the state of the limb leading to musculotendinous transfers designed to meet that individual patient’s needs. For some patients a severe injury may prove a catalyst for their taking their destiny into their own hands. In others, who have had a bad time, a subsequent injury may open out opportunities for rehabilitation which have been missed.
622
Case report: A 20 year old left handed student of accountancy sustained a severe head injury with multiple fractures of long bones in a motor cycle accident. No active rehabilitation was undertaken for 7 years. When he sought assistance for his memory defect and other problems of cognition he was told that he had come too late. Two years after this an operation as performed for synostosis of his right forearm during which the posterior interosseous nerve was severed. He was then seen in the Peripheral Nerve Injury clinic when it was put to him that a simple muscle transfer would improve function within his hand but that it was not even then too late to think about his future for he had been out of work since his initial injury. He then met with the clinical psychologist and our then Hospital Employment Advisor. It was clear that he could not
a
Surgical Disorders of the Peripheral Nerves
return to accountancy but he expressed an interest in working with wood. A flexor to extensor transfer was performed, and he was admitted into the Rehabilitation Unit for 2 weeks of vocational retraining after which he successfully found work as a carpenter and joiner. The experience of others may be particular helpful to patients who have become low in spirits. Many years ago one of us (RB) was treating a young man who had become very despondent because of the persisting pain from an infected non union of his tibia for which it seemed below knee amputation was the best way forward. The patient was horrified by the suggestion. Fortunately another patient then visiting the outpatient department kindly agreed to speak to him. He had been a merchant seaman in one of the
b
c
Fig. 14.15 Motivation (a) A 48 year old woman sustained infarction of the anterior spinal cord from interscalene block (see Chap. 3). She experienced lasting and severe pain. A prolonged course of treatment, which included a series of musculotendinous transfers for both of her hands which she bore with great fortitude. The range of movement at her wrist and hands is illustrated 10 years after the event. (b) A 63 year old engineer with complete paralysis of his previously dominant right upper limb made his own adaptations for activities of daily living. (c) At the age of 28 years one patient suffered a complete lesion of his right (dominant) brachial plexus. Recovery following repair was confined to the shoulder
and he designed a steering clamp to enable him to return to driving. The clamp was made by Hodge Mobility Ltd., Windsor, www.hodgemobility.co.uk. Other companies offering this type of service to our patients include Cowal Mobility Aids Ltd. 32, New Pond Road, Holmer Green. Bucks tel: 014949714400 and Reselco Engineering Ltd. Kew Bridge Pumping Station, Green Dragon Lane, Brentford, Middlesex TW8 0EN, tel: 0208 847 4509. (d) A 33 year old man sustained a complete and irreparable injury to his right brachial plexus in a motor cycle accident. He made an early return to work andcontinues to attend the outpatient department, driving to his appointments using a 650cc motor cycle.
Rehabilitation
Fig. 14.15 (continued)
Russian convoys when his ship was torpedoed in the Arctic Sea. He survived 3 days on a life raft in February but required an above knee amputation on one side and a below knee amputation on the other. He returned to live in Poplar, East London, he continued to work and he continued to walk with his prostheses without any aid. This man, by then in his early 70s, climbed the three staircases of St Andrew’s Hospital, Bow and marched the length of the ward to converse with the young patient who was, quite rightly, rather in awe of this seaman. He raised but one objection to the proposal of amputation and that related to the rolling gait shown by the older man to which objection came the reply: “son, that is the mark of a right seaman.” The older patient often has the burdens of underlying disease, social isolation and low income yet their needs are met irregularly and sometimes not at all. Some illustrative cases show what can be achieved by close coordination of surgical, nursing and therapy effort in liaison with local rehabilitation and social services. One very important element in the treatment of all of these patients was the close liaison initiated by the nursing and therapy staff with local services, a process
623
which began when the patients were first seen. Simple measures may tip the balance in the patient’s favour. In a personal communication Mannan and his colleagues (2006) have shown how a “hands free” device can assist those who cannot bear weight through one lower limb, by supporting that knee on a knee tray. Case report: An 82 year old widow who had sustained two strokes and was being treated for hypertension fell in the street fracturing her proximal humerus. This was treated by intramedullary nail. The operation was complicated by a complete radial palsy and severe pain. She was seen at 10 weeks with severe fixed deformities at the shoulder, the elbow, the wrist and the fingers. The pain clearly arose from the nerve and operation revealed a soft fusiform neuroma partly entrapped in the fracture. Neurolysis cured her pain and she remained as an in-patient in the Rehabilitation Unit for 2 weeks of treatment of the fixed deformity. She was provided with a useful and comfortable functional dynamic splint. After a successful trial of independent living at home with support from local services a flexor to extensor transfer enabled her to dispense with the splint. Case Report: A 70 year old man suffering from ischaemic heart disease and hypertension underwent excision of a sequestered disk at L3-4. This was complicated by severe pain and a foot drop. One year later a meningocoele was tied off which relieved his pain. He was admitted into the Rehabilitation Ward with a fixed equino varus deformity which was corrected by serial splinting and stretching over the course of 1 week. An orthosis was provided which he found comfortable. At operation 6 weeks later peroneus brevis was transferred to tibialis anterior, restoring dorsi flexion whilst retaining eversion through peroneus longus. After this he could dispense with the splint and return to driving. Case Report: An 84 year old widow experienced complete sciatic palsy following total hip replacement. There was later some recovery through the tibial nerve and this led to an increasingly severe and fixed equino varus deformity. At 6 years after her operation she was scarcely coping at home and was in considerable pain. She was admitted into the Rehabilitation Ward where a 2 week course of stretching and serial splinting, followed by capsulotomy, elongation of tendo Achilles and tendon transfer failed to correct the deformity. A modified boot was prepared by our orthotist, which allowed her to walk with one stick. Case Report: A 75 year old widow suffered a fracture dislocation of the ankle. An arthrodesis was attempted, which was complicated by division of the superficial division of the common peroneal nerve. She became wheel chair bound and remained in severe pain. Her family practitioner asked for advice, and raised the question of below knee amputation. She was admitted into the Rehabilitation Unit for a course of stretching, mobilization, gait improvement and pain relief. Three weeks later the arthrodesis was revised using a fibular
624
strut graft passed through the heel and the neuroma was implanted into muscle. Her pain was considerably improved and she managed to return to walking with an aid. Her treatment required 6 weeks of inpatient treatment. Case Report: A 78 year old retired teacher, who was living alone and wholly independent suffered a fracture of her proximal humerus which proceeded to non union. An operation was done at 4 months which was complicated by sepsis and by radial palsy. She was seen at 5 months when it seemed that the remedy was to hand of a flexor to extensor transfer. A dynamic splint proved useful, and a 5 day in patient course of work considerably improved the condition of the upper limb and enhanced her confidence. Flexor to extensor transfer was done which restored nearly full range of movement and a power grip of 25% of normal.
14.7 Rehabilitation in Progressive Neurological Disease The inexorable loss of function which is seen in some cases of hereditary sensory motor neuropathy, radiation neuropathy, certain myopathies or variants of motor neurone disease may induce a sense of hopelessness and of despair in those patients afflicted. Much can be done to maintain or restore independence by careful treatment of swelling, or lymphoedema and by proper care for the integument. Carefully made dynamic splints are invaluable. Occasionally muscle transfers are indicated but it is important to gauge the rapidity of progression of the disorder and to measure the power and the function of those muscles elected for transfer. Case report: An 82 year old widowed lady had been treated for carcinoma of the breast 40 years previously by mastectomy and radiotherapy. She developed a cervical myelopathy, which progressed from a right sided foot drop through hypothyroidism to increasingly severe sensory disturbance in her right upper limb with deepening weakness of the extensor muscles of the forearm. She had, in earlier years, been supplied with a conventional ankle/foot orthosis which induced pressure sores in her leg which took many months to heal. After functional assessments at her first attendance she was admitted for a 2 week course of treatment to improve her sense of balance, coordination, and confidence. A more effective ankle/foot orthosis was supplied. After a trial of functional splinting a flexor to extensor transfer was performed which considerably improved hand function on the right, her dominant, side.
14.8 The Choice of Intervention It is extremely important to advise operations according to the needs of an individual patient. It does not do to follow convention.
Surgical Disorders of the Peripheral Nerves
Case report: A 26 year old right handed motor cyclist sustained massive multiple injuries in a road traffic accident which were successfully treated by Mr. D Pring (Guernsey). A through knee amputation proved necessary on the right side. The injury to the right brachial plexus appeared, at initial examination, complete but there was early recovery through C8 and T1. The brachial plexus was explored 9 weeks after his injury, which confirmed avulsion of C7, a severe (type 4) rupture of C5 but a recovering lesion in C6, C8 and T1. In his case C7 innervated all of the extensor muscles of the arm and forearm. At operation one fascicle from the median nerve was transferred to the nerve to the long head of triceps and a flexor to extensor transfer was performed using flexor digitorum superficialis. He achieved a power grip of 70% of normal, with active wrist extension of 20° to MRC grade 4 and by 6 months his power of elbow extension was MRC Grade 4. This transformed his ability to cope with a wheel chair and subsequently enabled him to deal with the fitting and use of a lower limb prosthesis. Case report: A 43 year old right handed weigh bridge operator suffered severe multiple injuries in a motor cycle accident. These included an injury to the chest, a burst fracture of the tenth thoracic vertebra, a fracture of the left humerus and an open fracture of the left radius and ulna. These were successfully treated by Mr. Chakrabarti (Kings Lynn) and urgent NPI by Dr. Rosemary Eames (Kings Lynn) confirmed an extensive lesion of the lateral and posterior cords. In consultation with the referring surgeon, 2 days after his injury, it was agreed that further treatment for the left upper limb should follow stabilization of the spinal injury, recovery from other skeletal injuries and also the anticipated recovery of the ulnar and circumflex nerves. Four months later the brachial plexus and the nerves in the arm were explored. There was a severe traction injury of the suprascapular nerve, the median and musculocutaneous nerves had been ruptured, the radial nerve was irreparably damaged. The median and musculocutaneous nerves were repaired. He was admitted to the Rehabilitation Ward for 2 weeks of treatment to the scars and contractures and an orthosis was provided to stabilize gleno humeral, elbow and wrist joints. He was able to return to work at 9 months after his injury. By 3 years recovery through the ulnar nerve had been good and power of elbow flexion had reached MRC grade 2. There was considerable recovery through the graft of the median nerve, so that he could localize with accuracy to the thumb, the index and the middle fingers, without over reaction. There was useful recovery into the flexor muscles of the forearm innervated by the median nerve too. Although he was coping he preferred to dispense with the orthosis which, he said, “marked him out as a disabled man.” Elbow flexion was successfully enhanced by transfer of pectoralis minor, achieving a range of active flexion of 100° and of power of MRC grade3+. At the same operation the flexor carpi ulnaris was transferred to extensor carpi radialis brevis. No muscle
Rehabilitation
was transferred to the digits because of poor recovery into the intrinsic muscles and weakness of flexor pollicus longus. The patient was delighted with this result and continued to work full time without any orthosis. Case report: A 16 year old right handed nursing auxiliary sustained severe multiple injuries in a motor cycle accident. There was a deep lesion of C5, C6, C on the right side. He was treated by Mr. N Boughey (Lincoln) who discussed his case with us on the day of the injury. The brachial plexus was explored 3 weeks after his injury, finding a severe traction injury of C5, C6 and C7 with along segment of stretching involving the posterior cord. C5 did not recover, but he did regain elbow extension. This left him with a flail shoulder and lack of extension of the wrist and of the fingers. His rehabilitation was delayed because of difficulties with the open fracture of the left tibia and fibula but he was back at work, driving, by 2 years with the assistance of the Hospital Employment Advisor and a simple orthosis for stabilization of the gleno humeral and wrist joints. A flexor to extensor transfer at 8 months after injury restored power grip of 65% of normal, full range of extension of the wrist and digits. He came to see us again 2 years later by which time he was working as a nursing auxiliary and a farmer. His chief problem lay in the shoulder, for which we advised gleno humeral arthrodesis but he declined that suggestion pointing out that his work required the ability to laterally rotate his arm. Instead, latissimus dorsi and teres major were transferred to the rotator cuff. This improved the pain from the subluxation of the joint, and provided an active range of abduction and forward flexion of 20°and also 20° of active lateral rotation.
14.9 Patients with Neuropathic Pain The case notes of patients with significant or severe pain are reviewed every week by the patient’s consultant with consultants working in the Pain Management Team which is currently led by Dr. Jonathan Berman. The diagnosis is reviewed. Treatment plans are made which may involve a sequence of local regional or sympathetic blocks, spinal cord stimulation, dorsal root entry zone lesion (DREZ), or entry into the Chronic Pain Programme. All proposals are communicated to the family practitioner and referring clinician. The chronic pain programme: We emphasise again that no patient should be sent to a “pain” clinic or entered into a “pain” programme until an accurate diagnosis has been made. George Omer (1987) wrote “there are only two principles in the treatment of an established pain syndrome involving the upper extremity: to relieve the subjective pain experience and to institute active function of the involved extremity.” He further commented that “the best of functional activity occurs when the patient returns to daily work.
625
Patients who continue functional activities will ultimately “cure” themselves.” Some patients with pain retreat into themselves, cutting themselves off from family, friends and work. Fear of return to work, or of any painful activity leads to progressive dependence: learned abnormal behavioral patterns, including needless inactivity and display of suffering, become embedded. There may be an unconscious manipulation of the environment with the patient at the centre of things, with a family dominated by the patient’s pain behavior. For some of these, pain management programs do help: that is, carefully structured multi disciplinary programmes of rehabilitation which aim to modify behavior so as to increase function, increase activity and to relieve fear (Withrington and Wynn Parry 1984). Certain techniques are common to all structured pain management programmes. We shall outline some of those that we think are particularly important. 1. A full explanation of pain must be given by the clinician and time allowed for the patient to voice all fears and anxieties, many of which are unreasonable. Some patients have extraordinary ideas about the nature of their pain and its possible effects on life expectancy and they may be reluctant to open out these fears until they are actively encouraged to do so. Patients are usually much relieved when their pain is explained. Many patients with a paralyzed yet painful arm have believed that they must be taking leave of their senses, particularly when they have been ill advised that pain is evidence of recovery or that it is “all in the mind.” Understanding is the essential prelude to acceptance. We cannot over emphasize the value of a prolonged interview in which explanations are given and all questions answered. Empathy between the doctor and the patient is essential for the successful first faltering step of the long journey of rehabilitation. After that suitably qualified clinical psychologists or occupational therapists might assume responsibility for the overall programme. Once that has been established no further medical or surgical intervention is appropriate. The patient is encouraged to take personal responsibility for coping with the effect of his or her pain. Continuing review by physicians leads the patient to believe there is still some further remedy. 2. The patient is encouraged to examine how pain is affecting daily life so that the problem solving attitude is developed. 3. Factors which re-enforce pain behavior are identified and where possible modified. The spouse who treats the patient as an invalid protecting him or her from physical and emotional stress should gently be dissuaded from this attitude and encouraged rather to make the patient independent and supported in disregarding complaints. 4. The patient is encouraged to keep a pain diary. In this way, he or she recognizes that pain is not continuous, but that it may fluctuate, so permitting some activity.
626
5. Specific aims of a short, medium and long-term are set and scrupulously assessed. 6. The regular exercise programme is essential. 7. A planned reduction of medication is a sine qua non. 8. The patient is persuaded that he or she is not helpless and that there is “life with pain.” 9. Patients are taught relaxation and controlled breathing exercises which many find helpful in reducing spasms of pain. 10. Daily activities are increased regularly and monitored. 11. Some patients can learn self hypnosis to reduce severity of pain – the help of psychologists and suitably qualified hypnotherapists should be sought. 12. The constitution of the group is all important. Results may be significantly affected by one individual with an unyielding and morose sense of dependence or possessed of a determination to beat the system. On the other hand an enthusiastic individual can often “carry” some less well motivated members. The team leader has to be aware of these factors and be ready to take appropriate action. The work is demanding for all members of the multidisciplinary team and regular frank discussions with agreed objectives are vital. The failure of many chronic pain programmes to address the question of return to work is a major defect, one which has been addressed by the institution of a “return to work group” within the 3 week chronic pain programme at the Royal National Orthopaedic Hospital. The return to work group is currently led by Mrs Cara Lovell (BSc hons OT) and attendance for all patients aged less than 65 years is compulsory; it is optional for those who are older. There are diploma or degree courses in vocational rehabilitation at Canterbury Christchurch, Sheffield Hallam and Glasgow Caledonian Universities. These go some way to fill the gap left by the disappearance of the Hospital Employment Advisor. The question of return to work, or to holding on to a position at work, is addressed in the second week of the pain programme, and participants are invited to identify problems, after information has been provided about the implications of the Disability Discrimination Act (2005). The ethos is to encourage patients to move away from the idea that “I will never work again … all my friends and family agree that I will never work again” to a more positive view of the situation, to consideration and active engagement in retraining or seeking work or by supporting and maintaining their position at work. Lovell refers to the obstacles faced by the patient and by those helping them. Main and Burton (Main and Spanswick 2000) outline the concept of “flags” which group the different issues posing obstacles. Yellow flags identify the attitudes and beliefs that an individual may hold, for example the idea that work is harmful or impossible. Blue flags relate to such problems at work as lack of support, lack of job satisfaction or time pressure. Black flags are not
Surgical Disorders of the Peripheral Nerves
based on individual perception, they relate to the over arching impact of national and work place policies and Main and Burton suggest: “even the most sharply focussed intervention designed to remove the relevant yellow and blue flags may fail as a consequence of insuperable obstacles in the form of black flags.”
14.10 Paralysis not Physically Determined Most clinicians have seen or will see at least one patient with a paralysis of a limb for which no physical cause can be assigned. The most likely cause is a “conversion” disorder or “conversion hysteria,” in which resolution of some mental disorder is sought by converting it into a physical one. A few patients are actually malingering – consciously simulating a paralysis in order to claim compensation or to avoid some unpleasant task or experience. A very few are true exponents of the “Munchausen syndrome.”
14.10.1 Conversion Paralysis Abse (1987) makes this definition: “In conversion hysteria there are dramatic somatic symptoms into which the patient’s mental conflict is ‘converted’…” There may be gross paralytic, spasmodic or convulsive motor disturbance or perversion of sensation. Oscar Wilde (1890) may have been near an understanding with: “That is one of the great secrets of life – to cure the soul by means of the senses and the senses by means of the soul.” Tyrer (1989) noted that although the typical case of conversion disorder was in most instances easy to identify, the diagnosis provided a fertile ground for clinical error. Although early descriptions of “conversion hysteria” “emphasise the equanimity of the patient in the face of a major handicap (the belle indifférence of Charcot), that feature is not now considered of prime importance in diagnosis.” Indeed, patients with the very severe and doubtless physically determined pain of intradural injury of the brachial plexus seem to display a curious indifference to a pain described as constant and severe, with episodic exacerbations. Sunderland (1991) dealt with “conversion hysteria” and noted certain features: area of sensory loss not corresponding with anatomical distribution; sensory defect not associated with loss of function; absence of trophic changes; paralysis never an isolated nor predominant sign. None of these features is entirely constant, but that does not detract from the value of Sunderland’s criteria as an initial guide. Case Report: A man was sent at the age of 22 for fitting of a functional splint for a complete lesion of the brachial plexus of the right (formerly dominant) upper limb. He was then living alone, running his own business as a window
Rehabilitation
cleaner. He seemed perfectly well adjusted; apart from the upper limb, perfectly fit. The patient’s General Practitioner sent the medical record, which was enormously bulky and very informative. The patient, an only child, had been brought up by parents clearly obsessed with matters of health. Between the ages of 4 and 11 he had been seen intermittently by pediatricians for a variety of complaints, none of which was found to have a physical basis. At the age of 14 he developed a motor and sensory paralysis of the right upper limb, which was investigated but never consistently treated. The patient left home at the age of 18, and never returned. There was complete motor and sensory paralysis of the right upper limb, with marked wasting, brittle nails and other “trophic” changes. However, there was no history of unnoticed accidental injury of the limb. Electrophysiological examination (Dr. Nicholas Murray) showed, in effect, normal responses. The patient was admitted and kept in for 6 weeks, during which the basis of his condition was explained to him, and regular treatment was given to encourage him to resume the use of the limb. The response of the “paralysed” muscles to electrical stimulation was shown, and the attempt was made to overcome resistance to voluntary activity. A functional splint was fitted. All that was obtained was some abduction of the shoulder and some flexion of the elbow. Even that small gain was lost in the weeks after this young man left hospital; he did not wear the “functional” splint and the paralysis returned in its full extent. This case exemplifies the conversion paralysis in an extreme form. It is worth noting that there were undoubtedly trophic changes, though there was no history of unnoticed accidental injury. Over the years the patient had adjusted his life well to the disability: our intervention was in the event irrelevant. A number of other patients have been seen and we have found it necessary to conduct extensive investigations to exclude underlying progressing neuromuscular disease. It may then be helpful to involve a clinical psychologist and an occupational therapist after explaining to the patient that no serious underlying disease process has been detected. The whole approach to the patient must be supportive, and not hostile nor dismissive. It may be useful to explain to the patient that they are suffering from a form of conduction block, so that the instructions from the brain to the paralyzed part are just not getting through and that by their own efforts they may well be able to regain function. This is much easier when the patient is seen soon after the initiating injury or event. It is important to be aware of the individual who seeks to manipulate those attempting to help them by playing one off against the other or by transferring their problem onto others. Patients from a professional background can be difficult and expressions of contempt and disregard towards nursing staff, therapists and junior doctors are warning signs. Particular care is necessary with children presenting with this disorder. The parents must be involved from the outset and early involvement of child psychologists and psychiatrists is required.
627
14.10.2 The Munchausen Syndrome Richard Asher’s (1951) description of this syndrome has not been bettered, though the picture of hospital mores drawn in his article now seems archaic. Thus: “…it requires a bold casualty officer to refuse admission.” Nowadays, decisions about admission have largely been removed from the control of the “casualty officers”; indeed, from clinicians in general. Other features of a bygone age, such as the “experienced front gate porter” are found, but the article remains the classic account of an extraordinary condition by an altogether exceptional clinician and thinker. As Asher says, the person suffering from this condition resemble the famous Baron Munchausen in having or in affecting to have travelled widely; their stories, like those attributed to the Baron, are both dramatic and untruthful. The patient comes with apparently acute illness supported by a plausible, extensive and dramatic history. The last is largely made up of falsehoods; later, the patient is found to have attended a number of other hospitals and deceived many clinicians there. He or she nearly always discharges him- or herself against advice, after quarrelling violently with doctors and nurses. Asher goes on to list telltale signs, and names three well-known varieties of the syndrome: acute abdominal; haemorrhagic; neurological. Our particular recollection, dating from the 1960s, is of a sturdily built man in his early 1930s, who used to come with a complete motor and sensory paralysis of the right (dominant) upper limb, allegedly caused, in one version, by the fall of a heavy weight on the shoulder. There had been previous accidents: in particular, one sustained during service in the Royal Air Force, when the allegedly ruptured spleen had been removed. This man was admitted to St Mary’s Hospital Paddington, referred from another hospital for investigation of a recent lesion of the brachial plexus. At the time, there was a rather convincing motor and sensory paralysis of the right upper limb, with much complaint of pain. The abdomen showed the large scar of the splenectomy, with other scars. It was perhaps fortunate that at that time the prime method of investigating lesions of the brachial plexus was by myelography: soon after that course was suggested, the patient slipped away. Investigation revealed previous visits to several other hospitals in the London area where similar dramas had been enacted. We continued to hear of this patient from various hospitals for several years afterward. Evidently, his case combined the abdominal and neurological syndromes; as is so often the case, it was difficult to discover what advantage he sought from the deception. Asher suggested as possible motives: the desire to be the centre of interest and attention; a grudge against hospitals and doctors satisfied by deceiving them; a desire for drugs; a desire to escape the attentions of the police; a simple need for bread and lodging. Nowadays, of course, many more people than formerly harbor grudges against doctors and hospitals, just as
628
many more seek refuge from the police and many more seek a bed for the night. On the other hand, in the United Kingdom of today, prohibited drugs are readily available, at a price; attendance at hospital with symptoms and signs of serious disease no longer provides any guarantee of a bed for the night. Yet “Munchausen” patients are still seen and still cause unease and confusion (Kwan et al 1997). Asher commented on the remarkable tolerance of such patients to “the more brutish hospital measures”: he could hardly have forseen that 60 years later “brutish measures” would include keeping sick patients for up to 2 days on trolleys in accident departments. Even that prospect does not seem to have discouraged these patients. Diagnosis is hard even for the most experienced clinician. Those who have learnt from harsh experience to distrust on sight the open faced, frank man or woman who looks his or her interloculator straight in the eye and speaks openly, clearly and respectfully, will have less difficulty than will their more trustful colleagues. The “paralysis” can of course be shown by electrophysiological examination to be a fake; the sight of an abdomen looking like a battlefield is a sure giveaway. Nevertheless, the earnestness, clarity and frankness with which the patient initially expounds his or her condition constitute the first and most important warning sign. We are not aware that any clinician has successfully treated this condition.
14.10.3 Malingering We think that when the patient with the “hysterical” paralysis of the brachial plexus is alone at home, his or her paralysis persists. On the other hand, when the malingerer is home alone or with his or her intimates, the paralysis disappears or reverts to its true physical extent. The object of deception is the natural one of wishing to gain a large sum in compensation or to avoid some tiresome or dangerous task. The latter is, of course, the common cause of malingering by persons in the armed services in wartime. It is hard, even for the shrewdest clinician armed with the most advanced technical aids, to penetrate the cloud of deception and come to a true assessment of motor power, sensibility, joint movement, degree of pain and level of function. Modern advances have come to the aid of clinicians and insurers: the video camera permits recording of a claimant’s level of activity when he or she thinks him- or herself unnoticed. These are not agreeable pastures.
14.11 Rehabilitation in the War Wounded The present Medical Director of the Defence Medical Rehabilitation Centre at Headley Court, Colonel John Etherington OBE, has summarized the aims and the methods of rehabilitation of the war wounded (Etherington 2010)
Surgical Disorders of the Peripheral Nerves
The process starts at the time of injury, where early assessment is done by deployed medical rehabilitation teams based at the field hospital or even further forward so that a plan for rehabilitation is in place by the time the injured service men and women return to the base hospital. The expectation for the level of outcome is higher than in civilian practice. The whole patient must be considered, treatment cannot be compartmentalized into one area of therapy alone and the psychological effects of near death, of losing friends, of disfigurement, often heightened by “mild” brain injury have a major bearing on the ability to learn new activities or return to work. Patients are admitted to Headley Court once their general condition is stable and they are free from serious infection. During 2009, 170 new battle casualties were admitted to the Unit and because of planned readmissions the total of admissions into the complex trauma facility was 504. The total number of admissions during that year was close to 2000. Treatment is undertaken by a multidisciplinary team (MDT) which is led by physicians who have overall responsibility for coordinating work and case management. The objects of treatment in the individual case are defined a t MDT Meetings at which assessments are provided by medical, nursing, exercise therapy, physiotherapy, occupational therapy, social work and other disciplines, and from this realistic goals are set, for short term (3–4 weeks), medium term (2–3 months), and long term (6 months) which should be specific, measureable, achievable, realistic and timely. Almost inevitably this process receives the acronym of SMART. There is, throughout, a focus on the outcome with the aim of returning the patients to duty and patients must be involved throughout. Progress is reviewed at planned MDT meetings. Patients are admitted for between 3 and 4 weeks, and join a demanding process of physical training and exercises, in groups, for about 5 h a day. Peer support is extremely helpful, and classes in hydrotherapy, posture, gait, walking and running, recreation are used. Where necessary the programme is tailored to individual needs. This is a rigorous period of physical training and employment tasks during which formal assessments are done to ensure return to the best possible physical and vocational capability. The role of the remedial instructor is important, these are service physical training instructors who have undergone a further 6 months training programme in exercise based rehabilitation. The programme of work passes through several phases, the early is designed to meet preset outcomes by intensive physical activity, the intermediate phase moves towards improvement in function, confidence, and general physical condition whilst the late phase represents the limit of exercise based rehabilitation. Treatment of the war wounded requires particular skills in (1) wound care, (2) nutrition for patients admitted in a highly catabolic state, (3) orthotics and prosthetics. The social worker has important responsibilities offering support to the family and engaging resources from external agencies particularly for
Rehabilitation
those who require resettlement or retraining, rehousing and legal advice. Reintegration into society for those unable to remain within the armed forces is by no means easy and is made more difficult by the reluctance of some commissioning authorities to take on financial responsibility. An external charitable agency has become involved in this process. In any process of early and active rehabilitation, it is essential that specialist advice about problems of infection, skin cover, wound healing and bone union is available. It is for the clinicians in the nerve injury clinic to provide an accurate prognosis for the recovery of injured nerves, for these present a major obstacle to rehabilitation, a problem which is heightened until the prognosis is clarified. Arrangements have been made to ensure urgent NPI at the National Hospital (Queen Square) and quantitative sensory testing (Professor Praveen Anand, Imperial College). The nerve injury clinic works closely with the pain clinic directed by Colonel D. Aldington RAMC. Secondary operations for the relief of pain, the correction of deformity, the improvement of function by musculotendinous transfer, the resurfacing of scarred areas and the revision of nerve repairs are undertaken when they are needed. This close integration of the general process of rehabilitation with appropriate and timely intervention is one particular feature of the work of Headley Court.
14.11.1 Outcomes A number of measures are used to record improvement including standardized fitness measures, patient reported outcomes, hospital anxiety and depression scale amongst others. Of one group of admissions of 19 patients, 17 required assistance in activities of daily living on admission. Of these 16 were fully independent on discharge, of 13 who came in unable to drive, 9 were able to do so. Patients with brain injuries present a particularly difficult problem. Over a 12 month period 57% of such patients returned to military work whilst 3% were discharged to civilian life. Of a recent consecutive group of 58 lower limb amputees, 90% returned to employment with the military, 10% were discharged to civilian life. The mean interval before return to work for the transtibial amputees was 7.5 months. The lower limb amputee: It is inevitable that the circumstances of the battle field means that some amputation stumps may be not of optimal length or that skin cover is poor, but as Etherington says: “these are decisions for the cold light of day rather than on the battle field and an overlong or a badly scarred stump can be later revised.” It is important to remember that these are healthy patients and the experience of Headley Court cannot be applied unthinkingly to older patients with peripheral vascular disease. The ideal length for a below knee amputation is 8 cm of stump length for each meter of height, anything shorter than 7 cm is difficult to fit.
629
A transfemoral amputation requires a minimum of 15 cm above the medial tibial plateau to enable the fitting of a knee joint whilst retaining an adequate lever arm. The bulbous stump is difficult to fit, and this is more common when the classical long posterior flap is used as opposed to the skew flap which produces a more conical stump. The through knee amputation is very good, it is end bearing, and whilst fitting is demanding and the knee system must sit below the contra lateral knee, this is more than compensated by the stability and control from the long lever arm. The rehabilitation of amputees is much more rapid than those with fractures of the leg and foot and this means that some of the service men and women request unnecessary amputation. These requests require careful handling. Indeed the problem with most of the service men and women is that they need to be reined in rather than urged on Fig. 14.16. There have been very important advances in prosthetic design. The most important of these is the behind knee system in the above knee amputee, and microprocessor knees such as the CLeg® the Rheo® system has revolutionized knee control particularly where stability is a critical issue such as in the bilateral transfemoral amputee. These are expensive and may not be funded by Health Commissioners. Highperformance systems for the ankle and foot are now available, using a spring system such as the Carbon Fibre Flex-run® or Cheetah® designs. The example of the DMRC is outstanding and there are important lessons to be learnt for rehabilitation in the National Health Service as a whole. These include: the clarity in leadership, responsibility and purpose; the closely knit nature of the MDT, and the regular engagement with other medical and surgical disciplines in those cases requiring early intervention. It is also true that it is very rare to encounter problems with motivation or discipline!
Fig. 14.16 A 38 year old service man sustained amputation of left upper and lower limbs and massive injuries to the right lower limb whilst disarming an explosive device. He continues to work, training others in this field.
630
14.12 The Future There have been very great advances in the design of aids, orthoses, prostheses, and other artifacts so useful in rehabilitation. The work of the Cleveland Team (Peckham et al 2001) provide an outstanding example of the application of technological advance with their development of an implanted neuroprosthesis for restoring hand grasp in tetraplegic patients, work which has been extended to restoring elbow extension, pronation and grasp release (Peckham and colleagues 2001). The extensive work of Lundborg and his colleagues which is set out in the second edition of Lundborg’s work Nerve Injury and Repair (2004) opens out the possibility of restoring functional sensation in patients after amputation or nerve injury (Sebelius et al 2005). However these possibilities must be weighed against the generally disorganized, incomplete and unfocussed rehabilitation which has become so common throughout the National Health Service. We can only hope that work of Professor Keith Willett (currently National Clinical Director for Trauma Care, Department of Health) in this field will bear fruit.
References Abse DW (1987) Hysteria and related mental disorders. Wright, Bristol, p25 Asher R (1951) Munchausen’s syndrome. Lancet I:339–341 Cox RAF (1996) Avoiding discrimination against disabled people. Brit Med J 313:1346–1347 Etherington J (2011) Rehabilitation in the war wounded. In: Mahoney PF, Brooks AJ, Ryan JM, Schwab CW, Hodgetts TJ, Midwinter M. Ballistic Trauma 3 rd Edition, Springer (In Press) Goldthwait JE (1917) The place of orthopaedic surgery in war. J Orthop Surg XV(10):679–686
Surgical Disorders of the Peripheral Nerves Jones R (1921) Orthopaedic surgery of injuries. Oxford University Press, Hodder, London Joyce JL (1918) A study of a series of peripheral nerve injuries from a surgical aspect. Brit J Surg 23:418–457 Kwan P, Lynch S, Davy A (1997) Munchausen’s syndrome with concurrent neurological and psychiatric representations. J Royal Soc Med 90:83–85 Lundborg G (2004) Nerve injury and repair: regeneration, reconstruction and cortical remodelling, 2nd edn. Elsevier, Churchill Livingstone, Edinburgh Main C, Spanswick C (eds) (2000) Pain management: an interdisciplinary approach, 1st edn. Churchill Livingstone, London Omer GE (1987) The management of pain. In: Lamb DW (ed) The paralysed hand. Churchill Livingstone, Edinburgh, London, Melbourne, New York, pp 216–231 Osgood RB (1918) The orthopaedic centres of Great Britain and their American medical officers. J Bone Jt Surg(Am) s 2–16:132–140 Peckham RH, Keith MW, Kilgore KL, Grill JH, Wuolle KS, Thrope GB, Gorman P, Hobby J, Retz R, Carroll S, Hentz RV, Wiegner A (2001) Efficacy of implanted neuro-prosthesis for restoring hand grasp in tetraplegia. A multicentre trial. Arch Phys Med Rehabil 82:1380–1388 Russell Davis D (1949) Some factors affecting the results of treatment of peripheral nerve injuries. Lancet 253:877–880 Sebelius FCP, Rosen BN, Lundborg GN (2005) Refined myoelectric control in below-elbow amputees using artificial neural networks and a distal glove. J Hand Surg 30A:780–789 Stewart MPM (2009) Wounds of mind and body. Robert Jones Lecture given at The British Orthopaedic Association meeting, Manchester, Autumn Sunderland S (1991) Conversion hysteria. In: Nerve injuries and repair. A critical appraisal. Churchill Livingstone, Edinburgh, p 327 Tyrer P (1989) Conversion hysteria. In: Classification of neuroses. Wiley, New York, pp 106–107; 111–112 Watson F (1934) The life of Sir Robert Jones. Hodder and Stoughton, London Wilde O (1890) The picture of Dorian Gray. Lippincott’s Monthly Magazine, July 1990, pp 3–100 Withrington RH, Wynnparry CB (1984) Painful disorders of peripheral nerves. Post Grad Med J 60:869–875 Wynn Parry CB (1998) Rehabilitation. In: Birch R, Bonney G, Wynn Parry CB (eds) Surgical disorders of the peripheral nerves. Churchill Livingstone, London, Edinburgh, pp 451–466, Chapter 18
Index
A Abdomen and pelvis, 270–271 sympathetic system, 36 Ablation, central nervous system, 557–558 Acromegaly, 488 Action potential, ion channels, 56, 57 Adult cutaneous sensation, recovery of, 130 traumatic brachial plexopathy, 214 Afferent autonomic pathways, 37 Ageing, peripheral nerve fibres changes, 59–61 Alcohol, iatrogenous injuries, 487 Allodynia, 529 Allografts, large gaps repair, 125–126 Amputation limbs, 629 reconstruction, 601–602 revascularisation axillary artery, rupture, 337, 338 brachial plexus, 333–334 closed infraclavicular lesion, 334–339 incomplete avulsion injury, 336 median and ulnar nerves, 338 patterns, 335 suprascapular and circumflex nerves, 339 Anatomy autonomic nervous system, 2 (see also Autonomic nervous system) cervical nerves, from spinal cord, 1 cranial and spinal nerves (see also Cranial nerves; Spinal nerves) brainstem and cervical cord, laminectomy, 4 division of, 2 hemilaminectomy, 3 origin of roots, 2 spine and spinal cord, 3 musculo skeletal injury, nerves risk, 37–40 Ankle equino-varus deformity of, 573 eversion, 182 foot orthosis, 596 and heel, lumbar and sacral anterior primary rami, 30 schwannoma, 539 Anterior horn cell disorders, 224 Anterior interosseous nerve syndrome, 208 Anterior primary rami, of spinal nerves brachial plexus, 7, 10, 11 cervical plexus, 8, 9 circumflex (axillary) nerve, 13 dermatomes distribution, upper limb, 12 elbow, 15
forearm, superficial radial and lateral cutaneous nerves of, 8, 18 hand, median and ulnar nerves in, 17 median nerve, 16 musculocutaneous nerve, 14 phrenic nerve, course of, 9 radial nerve, 14 right axilla and arm, 12 suprascapular and circumflex nerves, 13 suprascapular nerves, 13 trunks and cords, 7 ulnar nerve, 16 Anterior tarsal tunnel syndrome, 220 Arterial thoracic outlet syndrome, 277–279 Articaine, 503 Atrophy, of foot, 110 Autonomic nerve fibres, sweat glands, 65 Autonomic nervous system afferent autonomic pathways, 37 efferent and afferent paths, 33 parasympathetic nervous system, 36 sympathetic system abdomen and pelvis, 36 cervico thoracic (stellate) ganglion, 34 in chest and abdomen, 35 Avulsed ventral root reinnervation, 399 repair, 410–411 Avulsion lesion, spinal nerves repair abdomen and pelvis, 270–271 anterolateral transvertebral approach, 262 brachial plexus, infraclavicular part of, 265 circumflex nerve, 266 endoscopic transforaminal approach, 262 lateral approach, 262–264 median nerves in, 266–269 plantar nerves and tibial nerve, 273, 275 posterior interosseous nerve, 267, 269 radial nerve, 266–267 sciatic nerve, 271–272 spinal accessory nerve, 264–265 suprascapular nerve, 265 tibial and common peroneal nerves, 272–274 ulnar nerves in, 266, 269–270 Axilla medial cord rupture, 153 reversed vein grafting in, 242 Axillary (circumflex) nerve, upper limb neuropathies, 213 Axon, 118, 119 Axonal transport, peripheral nerve fibres, 57
R. Birch, Surgical Disorders of the Peripheral Nerves, DOI: 10.1007/978-1-84882-108-8, © Springer-Verlag London Limited 2011
631
632 Axonotmesis and neurotmesis focal mechanical nerve injury, classification of, 82 hour glass constriction of, 82 peripheral nerve injuries, 204 Axoplasm, 43 B Bilateral lesion penetrating missile injuries, 332 repair, preganglionic injury, 407–409 Biopsy, 243–244 Birth lesions of brachial plexus (BLBP), 217–218, 611–612 deformities biceps pseudotumor, 463 C7 and C8, poor recovery for, 467 clavicle and scapula shortening, 464 elbow and forearm, 478–479 flexor origin, proximal advancement of, 467 injuries inflicted during, 463 limb, denervation of, 464 persisting muscular imbalance, 464–465 shoulder dislocation, closed manipulation of, 466 supination, 478–479 electrodiagnostic findings, clinical interpretation of, 449 epidemiology bilateral lesion, from breech delivery, 440 causation, 437 incidence, 436–437 Mallet score, 444 natural history, 441 outcome of, 441 ranges of active movement in, 445 risk factors, 437–438, 440–444 shoulder and of forearm rotation, ranges of movement, 443 untreated complex subluxation in, pain, 442, 444 gleno-humeral joint, posterior subluxation and dislocation (PS/PD) active range of movement, 473 AP and axial radiographs, after glenoplasty, 476 course to recovery, after relocation, 474–475 diagnosis and classification, 469–470 incidence of, 472 inferior scapulo-humeral angle, fixed contracture of, 478 lateral rotation osteotomy, shoulder dislocation, 471 medial rotation contracture, 470–471 radiological and operative findings in, 476 reduction, anterior approach, 472 secondary deformities, onset and progression of, 467–469 shoulder function, 477 group 4 lesion, spontaneous recovery in, 429 imaging, 453 nerve operations, 453 afferent function, method of repair, 459 breech delivery, 461 C5 and 6 rupture, C7, C8 and T1 avulsion, 462 case report, 459 C5, C6 and C7, repair of, 461 cocontraction, 463 C5 rupture, C6, C7, C8 and T1 avulsion, 462 findings, 460 grading of results of, 460 late reinnervation, 462 results of repair, 461 spinal nerves, 460, 461 ulnar nerve, 459
Index neurophysiological investigations (NPI) outcome, electrodiagnostic grade, 449 pre-operative NPI investigations, 450–452 operation incomplete lesion, 454 indications for, 453–454 intra operative conduction studies, 455 paralysis, evolution of, 454 post-operative care, 456–457 repair, methods of, 456 ruptured C5 repair, with avulsion, 457 recovery, in complete lesion elbow and hand, 445 evolution of, 446 group 4 lesion, 446 outcome, 447 spinal nerve in central affect, 431–432 types of, 430, 431 study methods C5, C6 and C7, 432 elbow, flexion and extension of, 434 green chart, 438–440 group 4 lesion, 434 hand function, analysis of, 434 inferior scapulo humeral angles, cocontraction and contracture, 436 Klumpke’s lesion, 434 limb growth, 434–435 Mallet chart, 435 Narakas classification, 432–433 records, 432 risk factor chart, 432, 433 scoring system for, 437 serratus anterior paralysis, breech delivery, 436 shoulder, function of, 433–434, 436 wrist and hand, 434 treatment, 447–448 Bleeding, control, 233–234 Blood supply, peripheral nerve fibres neck, resin cast of, 59 segmental medullary (radicular) arteries and anterior spinal artery, 60 spinal cord and cauda equina, dorsal vessels, 58 Bone and muscle transfer, reconstruction, 601 Brachial plexus amputation revascularisation, 333–334 anterior primary rami, 7, 10, 11 blood supply, 379–380 BLPP, 611–612 complete lesion, 621 cranial nerves cervical and, 9 excision of clavicles, 10 and great vessels, 11 left, 11 distal stump, 86–87 document used for, 186 infraclavicular part of, 265 lesions in, pain control, 540–541 lesions, upper limb neuropathies adult traumatic brachial plexopathy, 214 birth lesion of, 217–218 electrodiagnostic findings, in pre and post-ganglionic lesions, 214 elements, 215
633
Index neuralgic amyotrophy, 216–217 stinger injury, 215–216 true neurogenic thoracic outlet syndrome, 216 orthosis, 620 penetrating missile injuries, 326–327 preganglionic injury of, 533–534 Brachio radialis, elbow, 179 Brain derived nerve factor (BDNF), 44–45 Brainstem and cervical cord, laminectomy, 4 British Association of Otorhinolaryngologists (2002), 485 British Society for Clinical Neurophysiology, 206, 207 Buccal technique, third molar surgery, 509–510 C Carcinoma and lympho-proliferative disease, 489 Carpal tunnel syndrome, 99, 282 Cauda equina, dorsal vessels, 58 Causalgia case report, 542, 544, 545 definition of, 542 incidence and, 543–544 nerve lesion in, 543 response to direct treatment, 547–548 response to sympathetic block/sympathectomy, 546 Cell body and proximal stump, reactions to injury chromatolysis, 83 contralateral effects, 85 delay, 84 injection injury, 85 perineurium, wounds of, 84–85 TRPV1, 83 Central motor conduction time (CMCT), 194–196 Central nervous system definition, 1 pain ablation, 557–558 case report, 556 stimulation, 556–557 Cerebral palsy, reconstruction hip, dislocation of, 565 severe pain from dysplasia, 565 Cervical cord, laminectomy, 4 Cervical nerves, from spinal cord, 1 Cervical plexus, anterior primary rami, 8, 9 Cervico thoracic (stellate) ganglion, 34 Chest and abdomen, sympathetic system, 35 Circumflex nerve avulsion lesion, 266 reinnervation of, 252 rupture of, 154 Clavicular head, of pectoralis major, 178 Clinical psychologist, responsibilities, 610 Closed supraclavicular lesion. See Preganglionic injury Closed traction lesions, 384–385 Cocontraction, 463 nerve transfer, 132–133 Cognitive behavioural therapy, 515–516 Cold thermal nerve injury, 204 Common peroneal nerves, 272–273 lower limb neuropathies, 219–220 Complex dislocation (CD), 470 Complex regional pain syndrome (CRPS) causalgia, type 2 case report, 536 diagnostic criteria, 535 fracture/soft tissue injury, 536–537
Complex subluxation (CS), 470 Compound muscle action potential (CMAP), 193, 194, 210 Compound nerve injury amputation revascularisation brachial plexus, 333–334 closed infraclavicular lesion, 334–339 contraindications, 303 lower limb femoral nerve, 360–361 hip, 363–365 lumbo sacral plexus, 361–363 peroneal and tibial nerves, 365–367 Royal National Orthopaedic Hospital, 367–369 nerves and bone and joint injuries clavicle, 344–345 elbow, 356–359 forearm, 359–360 gleno-humeral joint dislocation, 345–351 iatrogenous injuries, 359 incidence, 351–355 musculocutaneous nerve, 355–356 nerve to serratus anterior, 344 pattern of fracture, 341–343 radial nerve and fractures of humerus, 351 spinal accessory nerve, 343–344 penetrating missile injuries ballistic peripheral nerve injuries, 331 bilateral lesions, 332 brachial plexus, 326–327 delayed primary suture, 323 grafts, 324–326 high energy transfer (HET), tissue damage, 327 incidence, cause and distribution, 330 low energy transfer (LET), tissue damage, 327 massive energy transfer (MET), tissue damage, 327 military rifle bullet wound, 323 MRCR evidence, 323–324 nerve lesion, 330 nerves, 331–332 pain, 328, 330–331 peripheral nerves, 328–330 recovery, 331 sacular aneurysm, 330 shot gun blast, 322 sciatic nerve, suture, 303 skin, 320–321 vascular lesion arterio-venous fistula, axilla, 316 catastrophic bleeding, 307 circumferential rupture, 311 elbow, 312 emergency arterial repair, 308 false aneurysm and arteriovenous fistulae, 313–316 femur and tibia, fracture, 314 iatrogenous false aneurysm, 317 iatrogenous ischaemic injury, 319–320 iatrogenous post ischaemic contracture, 319 ischaemia and nerve, 316–318 knee, 312–313 leaking aneurysm, 310 multiple fragment wounds, 307 Poiseuille’s law, 312 proximal humerus, fracture, 310 shoulder, 309–312 supraclavicular plexus and penetrating missile wounds, 309 tissue-compartment syndrome, 318
634 traumatic false aneurysms and arterio-venous fistulae, 309 ulnar artery repair, 307 war wounds contamination, 304 gas gangrene, 305 missile wounds, leg, 306 principles, 305 stages, 305–306 Compound sensory action potential (SNAP), 192, 193 Computerised tomographic scan, preganglionic injury, 396 Cone beam CT scanning (CBCT), of third molar, 511 Connective tissue disease, iatrogenous injuries carcinoma and lympho-proliferative disease, 489 case report, 488–489 immune brachial plexus neuropathy, 489–490 rheumatoid neuropathy, 488 Consultant, responsibilities, 610 Contact heat evoked potential stimulator (CHEPS), 165, 166 Conventional grafting, limitations, 124–125 Conventional nerve transfers, 410 Conversion paralysis, 626–627 Cord lesion haematoma and displacement, 380–381 level of injury, 380 Coronectomy, third molar surgery, 510–512 Cortical maps, 71 Cranial nerves brachial plexus cervical and, 9 excision of clavicles, 10 and great vessels, 11 left, 11 brainstem and cervical cord, laminectomy, 4 facial nerve, terminal branches of, 6 hypoglossal nerve, 7 spinal accessory nerve, 7 spinal cord and brain stem junction, 5 11th cranial nerve, formation of, 5 CRPS. See Complex regional pain syndrome Crush injuries, 97–98 Cutaneous sensation, recovery, 130 Cutaneous sensibility, somatic sensory system glabrous skin and hairy skin, 63 immuno reactivity in, 64 small diameter nociceptor cell bodies, 64 small fibre neuropathy, 64 Cutaneous sensory receptors, 63 low threshold mechanosensors, 65 nociceptors, 65–66 thermoreceptors, 65 D Deafferentation pain, 529–530 Deformities BLBP biceps pseudotumor, 463 C7 and C8, poor recovery for, 467 clavicle and scapula shortening, 464 elbow and forearm, 478–479 flexor origin, proximal advancement of, 467 injuries inflicted during, 463 limb, denervation of, 464 persisting muscular imbalance, 464–465 shoulder dislocation, closed manipulation of, 466 supination, 478–479 fixed, 569–572 Demyelination, electrophysiological consequences, 201
Index Denervation, peripheral effects, 109–110 Dental procedure, trigeminal nerve injuries, 501–502. See also Trigeminal nerve injury, iatrogenous injuries Dermatomes distribution, 24 upper limb, 12 Desensitization, rehabiltation, 612–613 Diabetes, iatrogenous injuries hypothyroidism and acromegaly, 488 neuropathy, 487 Dorsal horn, nocicipient, 531–532 Dorsal primary ramus, 385–386 Dorsal root ganglion (DRG) neurone, 43, 44 Dorsal root, regeneration, 136–137 Dorsi flexor muscles, paralysis, 600 Double crush syndrome, 223 E Elbow anterior primary rami, 15 collateral vessels, 38 deformities, 478–479 flexion, 179 and extension of, 434 locking component, orthosis, 616 reconstruction extension of, 580 flexion of, 580–582 ulnar nerve, 208–211, 282, 284 ulnar neuropathy, 284, 285 Electrical neural injury, 204 Electric shock, 104–106 Electrodiagnostic techniques. See Peripheral nerve injuries, clinical neurophysiology Electromyography (EMG), 195, 197 findings, in peripheral nerve injuries, 205 Electrophysiological examination, nerve injury, 185–187 Electrotonus, 194–195 Endodontic related nerve injury, 507–509 algorithm, 509 medicaments, 508 radiographs, 508 End-organs, regeneration, 131 Endoscopy, preganglionic injury, 398 Entrapment neuropathy first rib, 276 seventh cervical rib, 276–280 suprapleural membrane (Sibson’s fascia), 276 thoracic outlet syndromes, 275–276 Entrapment syndromes lower limb, 285 upper limb, 284–285 Entubation, 127 Envenomation, 108 Epineurium, 52 Epineurotomy, 243 Equino-varus deformity, of ankle and foot, 573 Extensor carpi radialis, 180 Extensor carpi radialis longus ( ECRL), 592–593 Extra osseous Ewing’s tumour, from sympathetic chain, 298 F Facial nerve, terminal branches, 6 Femoral and sacral plexuses, ganglionated sympathetic chain, 19 Femoral nerve, 20, 21 lower limb neuropathies, 221 sensibility loss, 158 and vessels, inguinal ligament, 39
635
Index Fiolle and delmas, operation, 256, 258 First rib, 276 Fixed deformity, 569–572 Flexor carpi ulnaris, 587, 588 Flexor digitorum superficialis (FDS), 180, 588–589 Foot and ankle, drop foot from common peroneal palsy dorsi flexor muscles, paralysis of, 600 tibialis posterior (TP) transfer, 598–600 atrophy, 110 equino-varus deformity of, 573 insensate, 596 orthosis, 596 posture of, CRPS type I, 536 small muscles, 182 Forearm deformities, 478–479 superficial radial and lateral cutaneous nerves, 18 transection of, 156 Free functioning muscle transfer (FFMT), 406 Freeze thawed muscle graft (FTMG), 126–127, 252 Functional splinting reconstruction, 568 rehabiltation, 613 G General Dental Council (GDC), 502 Generalised disorders, iatrogenous injuries, 486–487 Generalised polyneuropathy, 224 Glabrous skin and hairy skin, somatic sensory system, 63 Gleno-humeral arthrodesis, 579, 581 Gleno-humeral joint, 578–580 dislocation nerve lesions, 346 pain, 346–351 vascular injuries, 345–346 posterior subluxation and dislocation (PS/PD) active range of movement, 473 AP and axial radiographs, after glenoplasty, 476 course to recovery, after relocation, 474–475 diagnosis and classification, 469–470 incidence of, 472 inferior scapulo-humeral angle, fixed contracture of, 478 lateral rotation osteotomy, shoulder dislocation, 471 medial rotation contracture, 470–471 radiological and operative findings in, 476 reduction, anterior approach, 472 secondary deformities, onset and progression of, 467–469 shoulder function, 477 subluxation, 616 Glial cells, 43 Glial derived nerve factor (GDNF), 45, 46 Gluteus maximus, splitting, 272 Golgi tendon organs, 68–69 Grafts case report, 249 elevation and preparation of, 247 regeneration, 123–125 vascularised, 248–249 H Hand lesions, of ulnar nerve, 181 median and ulnar nerves in, 17 nerve lesions and, 593–595 Hemilaminectomy, 3 High resolution ultrasonography, nerve injury, 188
Hip abduction and flexion of, 596–597 dislocation, 565 Histamine induced flare response, 165 Holt’s method, suprascapular nerve, 411 Horse radish peroxidase (HRP), 136 Hyperpathia, 529 Hypoglossal nerve, 7 Hypothyroidism and acromegaly, 488 I Iatrogenous injuries case report, 484–485 causes alcohol, 487 connective tissue disease, 488–490 continuity of care, timing of operation, 493 diabetes, 487–488 generalised disorders, 486–487 specialisation, 493 teaching and training, 492–493 warning and consent, 490–492 incidence and audit, 485–486 prevention of audit and consent, 499 conduct of affairs, 499–500 recognition and action, 500 teaching and training, 499 radiation neuropathy malignancy, 498 operation, site of, 498–499 referrals, 483 total hip arthroplasty, nerve lesions in findings in, 494–496 incidence of, 494 urgent reexploration, indications for, 496–497 trigeminal nerve injury (see Trigeminal nerve injury, iatrogenous injuries) Immune brachial plexus neuropathy, 489–490 Implant related nerve injuries, 504–507 aetiology of, 506 algorithm, recommended management of, 507 bilateral IAN injury, 505 intra-operative, 505 post operative, 505 prevention of, 506–507 Impulse generation, 531 Infant, cutaneous sensation recovery, 130 Inferior alveolar nerve injury (IANI), 502, 518–520 Inferior scapula-humeral angle (ISHA), 170 Infraspinatus, 177 Injection injury, 85, 106–107 Injury. See Reactions to injury Insensate foot, 596 Insulin growth factor (Igf), 45 Intercostal nerves spinal accessory nerve, 250, 251 transfer, 410 ulnar to biceps transfer, 250–251 C6 and C7 avulsed ventral roots, reinnervation of, 252 C5 and of C6 avulsion, dorsal scapular palsy, 251 circumflex nerve, reinnervation of, 252 Interosseous muscles, contracture, 576 Intradural injury, regeneration dorsal root, 136–137 horse radish peroxidase (HRP), 136 ventral root, 136
636 Ion channels, peripheral nerve fibres mucosa and submucous plexus, in human rectum, 57 sodium channel staining, sural nerve, 56 Ischaemia acute, 87–88 and acute compression, neurovascular fascial compartments, 90, 92, 93 chronic, 97 and conduction, 232 neurostenalgia, 550 swollen muscle, 92–95 Marfan’s syndrome, 95 spinal accessory nerve, 95 staphylococcal septicaemia, 95 from tamponade case report, 90 interscalene block, permanent defects in, 91 traction, 95–96 K Knee, 597–598 collateral circulation, 40 extension of, 182 hamstring to quadriceps transfers in, 597 L Laminectomy, brainstem and cervical cord, 4 Late onset cord lesion, mechanical causes, 382 Latissimus dorsi transfer, 578–580, 583–584 Limb BLBP denervation of, 464 growth, 434–435 lengthening and traction, 497 lower (see Lower limb) upper (see Upper limb) Linear abrasions, 377 Lingual nerve injury (LNI), 502, 509, 516 Local analgesia (LA), trigeminal nerve injury, 503–504 prevention of, 504 Lower limb, 174–175, 177, 594–595 ankle, eversion, 182 compound nerve injury femoral nerve, 360–361 hip, 363–365 lumbo sacral plexus, 361–363 peroneal and tibial nerves, 365–367 Royal National Orthopaedic Hospital, 367–369 conventional ankle/foot orthosis, 596 entrapment syndromes, 285 foot and ankle, drop foot from common peroneal palsy, 598–600 foot, small muscles of, 182 hip, abduction and flexion of, 596–597 insensate foot, 596 knee, 597–598 extension of, 182 hamstring to quadriceps transfers in, 597 neuropathies common peroneal nerve, 219–220 femoral nerve, 221 lumbosacral plexopathy, 222 meralgia paraesthetica, 221 obstetric nerve injuries, 221–222 root lesions, electrodiagnosis in, 223 saphenous nerve lesions, 222
Index sciatic nerve, 219 tibial nerve and branches, 220–221 Lumbar and sacral anterior primary rami, of spinal nerves ankle and heel, 30 dermatomes, distribution of, 24 femoral and sacral plexuses, ganglionated sympathetic chain, 19 femoral nerve, 20, 21 left foot, sole of, 31 limb, posterior aspect of, 26 lower limb, anterior aspect of, 25 peroneal nerve, 29 popliteal fossa, 27, 28 sacral plexus, 21–22 sciatic nerve and components, 23, 24 Lumbosacral plexopathy, 222 Lympho-proliferative disease, 489 M Magnetic resonance imaging nerve injury, 187–188 preganglionic injury, 396–397 Malignancy, radiotherapy, 498 Malignant peripheral nerve sheath tumours (MPNST), 108 clinical features, 294 histopathological findings in, 295 operation, principles of extra osseous Ewing’s tumour, from sympathetic chain, 298 massive neuroblastoma, 298 neuroepithelioma, 297 NF1, 296 open biopsy, in thigh, 295 primitive neurectodermal tumours, 296 pathology, 294 Mallet chart, BLBP, 435 Marfan’s syndrome, 95 Massive neuroblastoma, 298 Mechanosensors, low threshold, 65 Medial rotation contracture (MRC), 470 Median nerves, 266–269 anterior primary rami, 16 in arm and axilla, 266 of distal stump, 86 injury, 156 primary suture, wrist, 246 proximal stump of, 118 sensibility loss, 157 thermal damage, 104 upper limb neuropathies anterior interosseous nerve syndrome, 208 case report, 208 forearm, post-operative appearance of, 208 high, 207–208 wrist, 206–207 Meralgia paraesthetica, 221, 285 Metacarpophalangeal joints, 574–575 Microscopic structure, of nervous system axoplasm, 43 cortical maps, 71 cultured human dorsal root ganglion neurone, 43 DRG neurone cultured and immunostained, 44 phase contrast image of, 43 glial cells of, 43 nerve growth factor BDNF, 44–45 GDNF, 45, 46
637
Index glabrous arm skin, basal epithelial cells in, 46 immuno reactivity in, 44 insulin growth factor (Igf), 45 neurotrophin 3 (NT3) immunostaining, 46 subepithelial fascicles of, 46 neuron theory, 44 peripheral nerve fibres (see Peripheral nerve fibres) somatic motor system, 61–62 somatic sensory system central connections, 69–70 cutaneous sensibility, 62, 64 cutaneous sensory receptors, 65–66 deep sensibility, 66, 67 golgi tendon organs, 68–69 skin, 64–65 visceral afferents, 70–71 synaptic activity, 72 Morton’s metatarsalgia, 220, 287 Motor evoked potentials (MEPs), 199 Motor neurone cell bodies, in ventral horn, 62 MPNST. See Malignant peripheral nerve sheath tumours Munchausen syndrome, 627–628 Muscle myopathic disorders, 224 spindles, 66, 68 somatic sensory system, 24, 26 transfer, reconstruction, 601 Muscle examination, nerve injury clinical, 168–171, 173–174 abduction, initiation of, 177 adductors and medial rotators, 173 brachio radialis, elbow, 179 clavicular head, pectoralis major, 178 deltoid, 174 elbow flexion, 179 extensor carpi radialis, activity in, 180 fingers, long flexors of, 174 flexor digitorum superficialis, 180 hand lesions, of ulnar nerve, 181 inferior scapula-humeral angle (ISHA) in, 170 infraspinatus, 177 intrinsic muscles, 174 rotator cuff, rupture of, 178 scapula movement, circumflex nerve rupture, 176 scapular winging, serratus anterior lesion, 173 serratus anterior, 170–171, 173 spinal accessory and serratus anterior, 171 spinal accessory palsy, 172 sternal head, pectoralis major, 178 suprascapular and circumflex nerves, 176 trapezius, 171 upper limb, elevation of, 175 late signs of, 177, 179–180 post ischaemic contracture, in hand, 183 post ischaemic fibrosis, 184 skin changes, 184 lower limb, 174–175, 177 ankle, eversion, 182 foot, small muscles of, 182 knee, extension of, 182 methods grip and pinch, measurement of, 169 shoulder, measuring power of, 170 pitfalls accessory insertion, 167 anatomical variations, 167
direct substitution, 167 motor recovery, 167 muscle power, measurement of, 167–168 rebound, 167 tenodesis action, 167 trick movements, 166–167 reinnervation, signs of, 180–182 Muscular neurotisation, 252 Musculocutaneous nerves anterior primary rami, 14 rupture, 154 sensory loss, 155 Musculo skeletal injury, nerves risk elbow, collateral vessels, 38 fascia, 37 femoral nerve and vessels, inguinal ligament, 39 knee, collateral circulation, 40 popliteal and anterior tibial arteries, 40 Myelinated nerve fibre, 50 Myelography, preganglionic injury, 394, 396 N National Health Service (NHS), 485 Neck, 254 close range shot gun blast, 146 course in, 378–379 superficial nerves in, 238 Nerve action potential (NAP), 192–193, 198 Nerve fibres neuronal cell bodies and, 136 reorganisation, 118–119 Nerve growth factor (NGF) BDNF, 44–45 GDNF, 45, 46 glabrous arm skin, basal epithelial cells in, 46 immunoreactive fibres in, 122 immuno reactivity in, 44 insulin growth factor (Igf), 45 neurotrophin 3 (NT3) immunostaining, 46 in pain, 533 subepithelial fascicles of, 46 Nerve injury, clinical aspects avulsion of C5, 151 C4, C5, C6, C7, 151 C4-T1, 150 C5-T1, sensory loss in, 150 axilla, medial cord rupture, 153 C5 and C6, rupture of, 152 C8 and T1, transection of, 152 circumflex nerve, rupture of, 154 diagnosis electrophysiological examination, 185–187 high resolution ultrasonography, 188 magnetic resonance imaging, 187–188 examination, 148 forearm, transection of, 156 history, 145 infant’s hand, palmar nerves, 159 level and depth of, 151–154 massive blast injury, 149 median nerve injury, 156 military rifle bullet, debridement, 148 muscles, examination of (see also Muscle examination, nerve injury) clinical, 168–171, 173–174
638 late signs of, 177, 179–180 lower limb, 174–175, 177 methods, 168 pitfalls, 166–168 reinnervation, signs of, 180–182 musculocutaneous nerves, rupture of, 154 neck, close range shot gun blast, 146 pain, 148 penetrating missile wound, 146, 147 Platt lesion, 146 quantitative sensory testing (QST) conduction, in somatic efferent pathways, 166 contact heat evoked potential studies, 166 histamine induced flair assessment, laser, 165 magnetic stimulation of, 167 monofilament perception, punctate touch, 164 sweat testing, 165 vibration perception thresholds, measurement of, 165 warm and cool thermal thresholds, testing of, 164 radial nerve, lesion of, 155 records, 182, 184–185 brachial plexus, document used for, 186 compound nerve injury, document used for, 185 red cross wound classification, 147 sacral plexus, cutaneous distribution of, 158 sciatic nerve, bullet wound, 146 sensibility, examination of, 161–163 chart recording improvement, measurement, 163 sensory recovery, 162 sensibility loss femoral nerve, 158 median nerve, 157 sensory abnormality, lateral cord in axilla, 152 sensory loss and clawing, ulnar nerve, 157 musculocutaneous nerve rupture, 155 peroneal nerve, 159 sciatic nerve transection, in thigh, 158 signs of, 154–156 sympathetic paralysis, transection of, 149 symptoms and signs, 148, 150–151 Tinel’s sign, 156, 157, 159 brachial plexus, traction lesions of, 161 in closed lesions, 159–161 degenerative lesions, prognosis in, 160 and recovery, 161 static and progressing, 160 wounds categories of, 148 system scores, features, 147 wrist, median and palmar cutaneous nerve, 156 Nerve repair, operation epineurium and perineurium, 249 free vascularised ulnar nerve graft, 249 grafting, 247–249 immobilisation, 252–254 nerve transfers, 249–252 primary suture of, 246, 247 principles, 245 stumps in traction injury, of radial nerve, 245 suture, methods of, 245–247 tidy wound, 244 wrist, median nerve, 246 Nerves and bone and joint injuries clavicle, 344–345 elbow
Index dislocation, 356 prevention, 358–359 Volkmann’s contracture and supracondylar fracture, 357–358 forearm, 359–360 gleno-humeral joint dislocation nerve lesions, 346 pain, 346–351 vascular injuries, 345–346 iatrogenous injuries, 359 incidence compound lesions, radial nerve, 353 decompression/extrication (neurolysis), 354 iatrogenous lesions, 354–355 lesions recovery, 352 open fracture, 352 Tinel sign, 353–354 musculocutaneous nerve, 355–356 pattern of fracture, 341–343 radial nerve and fractures of humerus, 351 serratus anterior, nerve to, 344 spinal accessory nerve, 343–344 Nerve transfer conventional, 410 elbow flexion, 411 repair technique, 398–399 Nervi nervorum, 59 Nervous system, definition, 1 Neuralgic amyotrophy, 216–217, 489–490 Neural outlet syndrome, without plain motor signs, 279–280 Neurapraxia, 79, 203–204 Neuroepithelioma, 297 Neurogenic thoracic outlet syndrome, 279 Neurolysis, 241–243 epineurotomy, 243 external, 242 futile, 243 Neuroma, 121–122, 518 containment of, 555 prevention of, 555 Neuronal cell bodies and nerve fibres, 136 Neuropathic pain, 528, 538, 625–626 Neurophysiological examination, 234–235 Neurophysiological investigations (NPI), BLBP outcome, electrodiagnostic grade, 449 pre-operative NPI investigations, 450–452 Neuroplasticity, 517 Neurostenalgia, pain case report, 549, 550 duration of symptoms, 550–551 ischaemia, 550 Neurotmesis and axonotmesis, 81–82 peripheral nerve injuries, 204 Neurotrophin 3 (NT3) immunostaining, 46 Neurotrophins, 44–46 Neurotropism, 115 Nociceptors, 65–66 Nocicipient. See Pain Nodes of Ranvier, 50, 52 Nursing staff, responsibilities, 610 O Obstetric nerve injuries, 221–222 Occupational therapist, responsibilities, 610 Omotrain orthosis, 619 Operations
639
Index anterior route, 280 apparatus and instruments hand, radial side of, 238 magnification, 235 neck, superficial nerves in, 238 neurophysiological examination, 234–235 normal and abnormal SSEP traces, 237 pain, prevention of, 236–237 records, 240, 241 severe pain and deep paralysis, gleno-humeral joint, 236 approaches neck, 254 postero-lateral route, 259–262 transclavicular exposure, 256–260 transverse supraclavicular approach, 254–258 avulsion lesion, spinal nerves repair (see Avulsion lesion, spinal nerves repair) birth lesions of brachial plexus (BLBP) incomplete lesion, 454 indications for, 453–454 intra operative conduction studies, 455 paralysis, evolution of, 454 post-operative care, 456–457 repair, methods of, 456 ruptured C5 repair, with avulsion, 457 bleeding, control of, 233–234 carpal tunnel syndrome, 282 complications and, 282 dangers of, 281 entrapment neuropathy first rib, 276 seventh cervical rib, 276–280 suprapleural membrane (Sibson’s fascia), 276 thoracic outlet syndromes, 275–276 entrapment syndromes lower limb, 285 upper limb, 284–285 indications and intervention diagnosis, 232 ischaemia and conduction, 232 persisting painful median palsy, 233 units, role of, 232–233 meralgia paraesthetica, 285 Morton’s metatarsalgia, 287 nerve repair epineurium and perineurium, 249 free vascularised ulnar nerve graft, 249 grafting, 247–249 immobilisation, 252–254 nerve transfers, 249–252 primary suture of, 246, 247 principles, 245 stumps in traction injury, of radial nerve, 245 suture, methods of, 245–247 tidy wound, 244 wrist, median nerve, 246 piriformis syndrome, 286 posterior approach, 282 for post ischaemic contracture, 571 pudendal nerve, entrapment of, 285–286 for reconstruction, 564 repair, methods of axillary artery, rupture of, 242 biopsy, 243–244 neurolysis, 241–243 vascular repair, 239, 241, 242
seventh cervical and first thoracic ribs, removal of, 281 tarsal tunnel syndrome, 286–287 technique of, open method, 282–284 transaxillary approach, 281–282 tumours of, pitfalls in intraneural ganglion, 289–290 MPNST, 293–298 solitary benign schwannoma, 287–292 solitary neurofibroma, 290–293 ulnar neuropathy, elbow anterior transposition, complications of, 285 decompression, 284 transposition, 284 Orthosis, 613, 616 ankle/foot, 596 Orthotists, responsibilities, 610–611 P Pain, 527, 528 causalgia, 542–549 central nervous system, interventions ablation, 557–558 case report, 556 stimulation, 556–557 complex regional pain syndrome (CRPS) causalgia, type 2, 535–536 fracture/soft tissue injury, 536–537 and deep paralysis, gleno-humeral joint, 236 nerve injury, 148 neurostenalgia, 549–551 nocicipient allodynia, 529 brachial plexus, preganglionic injury of, 533–534 deafferentation, 529–530 definitions of, 529 dorsal horn, changes in, 531–532 ephapses, 531 events, 530–533 gate theory, 530 hyperpathia, 529 nerve growth factor (NGF), role of, 533 post traumatic neuralgia, central sensitisation in, 532 proximal axon and dorsal root ganglion, 531 spontaneous impulse generation, 531 sympathetic nervous system, 534 operation drugs and measures of, 538–542 indications for, 542 during operation, prevention of, 236–237 post traumatic neuralgia amputation neuroma, 553 case report, 552–553 minor nerves, 554–555 regeneration, 553 spinal accessory nerve and serratus anterior, 553–554 regeneration and, 133–135 relief, preganglionic injury improvement in, 419–420 initial and final pain values, 420 left sided lesion, 417–418 peripheral nerve injury unit scale, 416–418 response, 420–421 Paralysis, 626–627 Parasympathetic nervous system, 36 Pelvis, 270–271 chondrosarcoma, 285
640 Penetrating missile injuries, compound nerves ballistic peripheral nerve injuries, 331 bilateral lesions, 332 brachial plexus, 326–327 delayed primary suture, 323 grafts, 324–326 high energy transfer (HET), tissue damage, 327 incidence, cause and distribution, 330 low energy transfer (LET), tissue damage, 327 massive energy transfer (MET), tissue damage, 327 military rifle bullet wound, 323 MRCR evidence, 323–324 nerve lesion, 330 nerves, 331–332 pain, 328, 330–331 peripheral nerves, 328–330 recovery, 331 sacular aneurysm, 330 shot gun blast, 322 Penetrating missile wound, 146, 147 Percussion injury, 106 Perineurium, 52 intraneurial ganglion, 100 and neoplasm/infiltration, 107–108 wounds of, 84–85 Peripheral nerve fibres afferent and efferent fibres paths of, 48 ventral root, 49 ageing, changes in nerves with, 59–61 avulsed dorsal root, transverse section of, 47 avulsed ventral root, intradural rupture, 47 axonal transport, 57 blood supply of neck, resin cast of, 59 segmental medullary (radicular) arteries and anterior spinal artery, 60 spinal cord and cauda equina, dorsal vessels, 58 conduction, 54, 56 DRG neurones and Schwann cells, in vitro cultures of, 50 epineurium, 52 fascicular arrangement of, 55 ion channels mucosa and submucous plexus, in human rectum, 57 sodium channel staining, sural nerve, 56 myelinated nerve fibre, 50 nervi nervorum, 59 nodes of Ranvier, 50, 52 perineurium, 52 Schmidt–Lanterman incisure, 51 Schwann cell development, 49 spinal cord, morphology of, 47 sural nerve double immunostaining of, 52 Wallerian degeneration, 54 transitional region/zone (TZ), 46, 47 ulnar nerve bundles and epineurial vessels of, 53 extrinsic epineurial vessels of, 53 unmyelinated axons, clusters of, 51 ventral root, rupture of, 50 Peripheral nerve injuries, clinical neurophysiology diffuse problems anterior horn cell disorders, 224 case report, 224–225 generalised polyneuropathy, 224 muscle/myopathic disorders, 224
Index electrodiagnostic techniques case report, 192 central motor conduction time (CMCT), 194–196 compound muscle action potential (CMAP), 193, 194 compound sensory action potential (SNAP), 192, 193 electromyography (EMG), 195, 197, 198 electrotonus, 194–195 F response, 194 H reflex, 193–194 limitations and pitfalls of, 199–201 and localisation, 199 mixed nerve action potential (NAP), 192–193 neuromuscular transmission, tests of, 194, 196 quantitative EMG (QEMG), 197 quantitative sensory testing, 195, 197 sensory evoked potentials (SSEPs), 195, 197 surface EMG, 197 intra-operative neurophysiological procedures free running and stimulated EMG, 198 motor evoked potentials (MEPs), 199 nerve action potentials, 198 somatosensory, dermatomal and root evoked potentials, 198–199 lower limb neuropathies, clinical applications common peroneal nerve, 219–220 disorders, 221–222 femoral nerve, 221 lumbosacral plexopathy, 222 root lesions, electrodiagnosis in, 223 sciatic nerve, 219 tibial nerve and branches, 220–221 pathophysiological response of axonotmesis, 204 classification and, 203 cold thermal nerve injury, 204 demyelination, electrophysiological consequences of, 201 electrical neural injury, 204 electromyography (EMG) findings in, 205 neurapraxia, 203–204 neurotmesis, 204 radiation injury, 204–205 regeneration and reinnervation, 205–206 temporal sequence of, 202 types of, electrophysiological consequences, 202–205 safety, 201 upper limb neuropathies, clinical applications axillary (circumflex) nerve, 213 brachial plexus lesions, 214–219 long thoracic nerve, 213–214 median nerve, 206–208 radial nerve, 211–212 spinal accessory nerve, 213 suprascapular nerve, 212 ulnar nerve, 208–211 Peroneal nerve, 29 sensory loss, 159 Phantom limb, reactions to injury, 110–111 Phrenic nerve, 9 Physiotherapists, responsibilities, 610 Physiotherapy and BLPP, 611–612 Piriformis syndrome, 219, 286 Plantar nerves and tibial nerve, 273, 275 Platt lesion, 146 Poliomyelitis, 584 Popliteal and anterior tibial arteries, 40 Popliteal fossa, 27, 28 tibial and common peroneal nerves, 272–274
641
Index Positron emission tomographic scan, 110 Posterior interosseous nerve, 267, 269 Posterior interosseus neuropathy, 212 Posterior primary rami, of spinal nerves, 32 Postganglionic lesion dorsal primary ramus, 385–386 somatosensory evoked potentials (SSEP), 386–388 Post ischaemic cord lesion, late onset, 381 Post ischaemic fibrosis, 570–571 Post operative infection related nerve injuries, 513 Post traumatic neuralgia (PTN), 551 amputation neuroma, 553 case report, 552–553 central sensitisation in, 532 extreme central sensitisation in, 535 iatrogenous division of, 554 minor nerves case report, 554–555 containment of neuroma, 555 prevention of neuroma, 555 repair, 555 translocation, 555 regeneration, 553 spinal accessory nerve and serratus anterior, 553–554 Preganglionic injury anatomy functional distribution, 379–380 micro anatomy, 379 neck, course in, 378–379 classification, 376 closed traction lesions, 384–385 diagnosis computerised tomographic scan, 396 endoscopy, 398 examination, 390 injury, cause, 389 inspection, 390 magnetic resonance imaging, 396–397 myelography, 394–396 operation, 398 pain, 389–390 plain radiographs, 392–394 rupture, 392 Tinel’s sign, 390–391 ultrasonography, 398 epidemiology, 388–389 function, patient recovery age, 415–416 results of, 412–415 vascularised ulnar nerve graft, 411–412 injury mechanisms, 376–378 linear abrasions, 377 pain relief improvement in, 419–420 initial and final pain values, 420 left sided lesion, 417–418 peripheral nerve injury unit scale, 416–418 response, 420–421 pre and postganglionic lesion dorsal primary ramus, 385–386 somatosensory evoked potentials (SSEP), 386–388 reimplantation, avulsed spinal nerves Bonney, George, 421–423 Carlstedt, Thomas, 423–425 repair strategies bilateral lesion, 407–409 C5 and C6, 400–402
C5, C6 and C7, 402 C5, C6, C7, C8, intact T1, 402–403 complete lesion, 404–406 free functioning muscle transfer (FFMT), 406 graft, 400 injury patterns, 400 left sided lesion., 406 lower lesion, 404 middle group, 404 quantitative sensory testing, 406, 407 right sided lesion, 407 ventral root (VR) repair, 403 repair technique direct repair, ruptured ventral roots, 399 nerve transfer, 398–399 reinnervation, avulsed ventral root, 399 vascularised ulnar nerve graft (VUNG), 399 results avulsed ventral roots repair, 410–411 conventional nerve transfers, 410 definitions, 409–410 methods, 409 spinal cord lesions cord lesion, 380 late onset cord lesion, 382–383 post ischaemic cord lesion, 381–382 unrelated cord injury, 380 total avulsion, 377 types, 375 Primitive neurectodermal tumours (PNET), 296 Progressive neurological disease, 624 PTN. See Post traumatic neuralgia Pudendal nerve, entrapment, 285–286 Q Quantitative EMG (QEMG), 197 R Radial nerve, 14, 98, 266–267 lesion, 155 upper limb neuropathies, 211–212 Radiation injury, 204–205 neuropathy, 109 neuropathy, iatrogenous injuries, 497 malignancy, 498 operation, site of, 498–499 and peripheral nerves, 108, 109 thrombosis, subclavian artery, 109 Reactions to injury axonotmesis and neurotmesis focal mechanical nerve injury, classification of, 82 hour glass constriction of, 82 case report, 78 cell body and proximal stump chromatolysis, 83 contralateral effects, 85 delay, 84 injection injury, 85 perineurium, wounds of, 84–85 TRPV1, 83 conduction block, 78, 79 deepening of lesion, 77 degenerative lesion, 79 denervation, peripheral effects of, 109–110 distal stump
642 brachial plexus, 86–87 dorsal root ganglion, 88 of median nerve, 86 nerve and roots detached from, 88 rupture and avulsion of, 89 sepsis, rupture, 87 ulnar nerve, 80 Wallerian degeneration in ventral root of, 89 envenomation, 108 ischaemia acute, 87–88 and acute compression, neurovascular fascial compartments, 90, 92 chronic, 97 swollen muscle, 92–95 from tamponade, 89–90 traction, 95–96 nerves damage to, 77 pressure effect on, 78 neurapraxia, 79, 80 non-degenerative lesion, 79 perineurium and neoplasm/infiltration, 107–108 phantom limb, 110–111 physical agents, lesion types acute compression, 98–99 chronic nerve compression, 99–100 crush, 97–98 electric shock, 104–106 injection injury, 106–107 percussion injury, 106 reperfusion injury, 96 thermal injury, 103–104 traction/stretch injury, 100–103 vibration injury, 107 radiation and peripheral nerves, 108, 109 Wallerian degeneration, distal stump of cervical nerve, 81 Reconstruction amputation, 601–602 cerebral palsy hip, dislocation of, 565 severe pain from dysplasia, 565 elbow extension of, 580 flexion of, 580–582 group IV BLBP, 566 high median palsy brachioradialis (BR) to flexor pollicis longus (FPL), 591 extensor to flexor transfer in C8/T1, 591 opposition transfer, 590 tenodesis effect of, wrist extensor muscles, 590 thumb, abduction and opposition of, 589–591 high ulnar palsy ECRL for, 592–593 extensor digiti minimi transfer, 592 iatrogenous lesion, 593 nerve lesions and hand, 593–595 operations for, 564 ankle and foot, equino-varus deformity of, 573 arthrodesis, 573 flexor muscle slide, 573–574 flexor origin, proximal advancement of, 583, 584 hand, contracted small muscle release of, 575–576 latissimus dorsi transfer, 583–584 metacarpo-phalangeal joints, fixed extension, 574–575 muscle slide, 572
Index osteotomy, 572–573 pectoralis minor transfer, 582–583 step elongation, 572 tenotomy, 572 paralysis of brachial artery and musculocutaneous nerve repair, 587 flexor carpi ulnaris, 587, 588 flexor digitorum superficialis, transfer of, 588–589 pre and post-operative care, 586–587 radial nerve palsy, splint for, 586 radial palsy, 588 wrist extension, 585 requirements fixed deformity, 569–570 flexor carpi ulnaris, 567 flexor, post operative splint for, 567 humerus, infected non-union of, 567 persisting imbalance during growth, 569–570 post ischaemic fibrosis, 570–571 pre-requisites for, 565 serial splinting, 571–572 simple functional splinting, 568 tension of transfer, 568–569 shoulder girdle gleno-humeral joint, 578–580 thoraco-scapular joint, 576–578 vascularized bone and muscle transfer, 601 Reflex sympathetic dystrophy, 536–537 Regeneration and recovery delay, 140 factors in prognosis age, 139 level of injury, 139–140 sensation, 139 function after nerve repair case report, 138 grading of results, 138–139 quality of, 137–138 recall, 138 intradural injury dorsal root, 136–137 horse radish peroxidase (HRP), 136 ventral root, 136 large gaps, repair of allografts, 125–126 conventional grafting, limitations of, 124–125 entubation, 127 failure of, vascularised sural nerve graft, 126 freeze thawed muscle graft (FTMG), 126–127 grafts, 123–125 methods of, 123 proximal suture line, 126 vascularised ulnar nerve graft, 125 nerve and axon response, to transection case report, 122–123 cellular response, 116 compartmentation/mini fascicle formation, 119 distal stump, 117 failure of, 120 graft, middle of, 120 guidance and selection, 119–120 maturation, 121 median nerve, proximal stump of, 118 nerve fibres, reorganisation of, 118–119 neuroma, 121–122 neurotrophic factor, 120
Index NGF immunoreactive fibres in, 122 proximal stump, 117 rate of, 122 ulnar nerve, proximal stump, 119 useless regeneration, 122, 123 Wallerian degeneration, in ulnar nerve, 117 nerve transfer case report, 129–130 cocontraction, 132–133 contra-lateral transfer, 129 cutaneous sensation, 130 deep afferent pathways, 130–132 end-to-side repair, 128–129 limitations of, 130 neuronal cell bodies and nerve fibres, 136 and pain, 133–135 skin staining, nerve markers, 135 neurotropism, 115 severity of injury, 140 Regeneration and reinnervation, peripheral nerve injuries, 205–206 Rehabiltation amputation, limbs, 629 anterior spinal cord infarction, 622 choice of intervention, 624–625 clinical psychologist, 610 communication, 608 consultant, 610 conversion paralysis, 626–627 desensitization, 612–613 elbow locking component, 616 flail arm splint, 616 functional splinting, 613 glenohumeral joint subluxation, 616 history, 609–610 inpatient unit, 616 legislation, 607 malingering, 628 modified light weight elbow splints, 620 motivation, 616–624 Munchausen syndrome, 627–628 neuropathic pain, 625–626 nursing staff, 610 objectives, 608 occupational therapists, 610 Omotrain orthosis, 619 orthoses, 613, 616 orthotists, 610–611 pain and hypersensitivity, elbow, 613 physiotherapists, 610 physiotherapy and BLPP, 611–612 progressive neurological disease, 624 proximal humeral shaft fracture, 613 restrictions, 607 shoulder subluxation, 619 stretching exercises, 612 war wounds outcomes, 629 treatment, 628–629 workshop training, 611, 612 Reimplantation, avulsed spinal nerves, 421–425 Bonney, George, 421–423 Carlstedt, Thomas, 423–425 Reperfusion injury, 96 Reversed vein grafting, in axilla, 242 Ruptured ventral roots, direct repair, 399
643 S Sacral anterior primary rami, of spinal nerves. See Lumbar and sacral anterior primary rami, of spinal nerves Sacral plexus, 21–22, 270–271 cutaneous distribution, 158 Saphenous nerve lesions, 222 Schmidt–Lanterman incisure, peripheral nerve fibres, 51 Schwann cells, 43, 46, 49–51 Schwannoma, ankle, 539 Sciatic nerve, 23, 24, 271–272 bullet wound, 146 lower limb neuropathies, 219 suture, 303 transection, in thigh, 158 Sensory action potential (SNAP), 192, 193 Sensory evoked potentials (SSEPs), 195, 197, 237 Shoulder dislocation, 466, 471 function, 436, 477 Shoulder girdle, reconstruction gleno-humeral joint arthrodesis, 579–581 latissimus dorsi and teres major, transfer of, 578–580 thoraco-scapular joint levator scapulae and rhomboids transfer, trapezius paralysis, 576–577 pectoralis major transfer, serratus anterior paralysis, 577–578 thoraco-scapular arthrodesis, 578 Sibson’s fascia. See Suprapleural membrane Simple posterior dislocation (SD), 470 Simple posterior subluxation (SS), 470 Skeletal muscle, afferent and efferent innervation, 67 Skin somatic sensory system, 64–65 staining, nerve markers, 135 Sodium channel staining, sural nerve, 56 Solitary benign schwannoma dumb-bell tumour, cervical region, 291 histopathological features in, 289 massive, in mediastinum, 292 symptoms and signs in, 288 treatment, 288–289 upper trunk of brachial plexus, 290 Solitary neurofibroma, 290–293 Somatic motor system descending tracts in, 61 motor neurone cell bodies in, 62 Somatic sensory system central connections, 69–70 cutaneous sensibility glabrous skin and hairy skin, 63 immuno reactivity in, 64 small diameter nociceptor cell bodies, 64 small fibre neuropathy, 64 cutaneous sensory receptors low threshold mechanosensors, 65 nociceptors, 65–66 thermoreceptors, 65 deep sensibility, 66, 67 golgi tendon organs, 68–69 muscle spindles, 66, 68 skin, 64–65 visceral afferents, 70–71 Somatosensory evoked potentials (SSEP), 386–388 Spinal accessory nerve, 7, 95
644 avulsion lesion, 264–265 upper limb neuropathies, 213 Spinal cord descending tracts in, 61 dorsal vessels to, 58 lesions, 380–383 Spinal nerves anterior primary rami of brachial plexus, 7, 10, 11 cervical plexus, 8, 9 circumflex (axillary) and suprascapular nerves, 13 circumflex (axillary) nerve, 13 dermatomes distribution, upper limb, 12 elbow, 15 forearm, superficial radial and lateral cutaneous nerves of, 18 hand, median and ulnar nerves in, 17 median nerve, 16 musculocutaneous nerve, 14 phrenic nerve, course of, 9 radial nerve, 14 right axilla and arm, nerves in, 12 suprascapular and circumflex nerves, 13 trunks and cords, 7 ulnar nerve, 16 in BLBP, 460, 461 central affect, 431–432 types of, 430, 431 lumbar and sacral anterior primary rami ankle and heel, 30 dermatomes, distribution of, 24 femoral and sacral plexuses, ganglionated sympathetic chain, 19 femoral nerve, 20, 21 left foot, sole of, 31 limb, posterior aspect of, 26 lower limb, anterior aspect of, 25 peroneal nerve, 29 popliteal fossa, 27, 28 sacral plexus, 21–22 sciatic nerve and components, 23, 24 posterior primary rami, 32 thoracic anterior primary rami, 18 Staphylococcal septicaemia, 95 Superficial radial nerve, 212 Supination deformity, 478–479 Suprapleural membrane, 276 Suprascapular nerve avulsion lesion, 265 transfer, 411 upper limb neuropathies, 212 Sural nerve double immunostaining of, 52 sodium channel staining, 56 Wallerian degeneration, 54 Sural neuropathy, 221 Sweat testing, nerve injury, 165 Sympathetic nervous system abdomen and pelvis, 36 cervico thoracic (stellate) ganglion, 34 in chest and abdomen, 35 and pain, 534 Sympathetic paralysis, transection, 149 Synaptic activity, 72 T Tamponade, ischaemia, 89–90 Tarsal tunnel syndrome, 220, 286–287 Thermal injury, 103–104
Index Thermoreceptors, 65 Third molar surgery (TMS), 502 buccal technique, 509–510 CBCT of, 511 coronectomy, 510–512 DPT radiographs, 510 juxta apical area, 510 lingual nerve injury, 509 Thoracic anterior primary rami, of spinal nerves, 18 Thoracic nerve, upper limb neuropathies, 213–214 Thoracic outlet syndrome (TOS), 216, 275–276 arterial, 277–279 neurogenic, 279 Thoraco-scapular arthrodesis, 578 Thoraco-scapular joint, 576–578 Thumb, abduction and opposition, 589–591 Tibialis posterior (TP) transfer, 598–600 Tibial nerve and common peroneal nerves, 272–273 lower limb neuropathies, 220–221 motor study, 194 Tinel’s sign compound nerve injury, 353–354 nerve injury, 156, 157, 159 brachial plexus, traction lesions of, 161 in closed lesions, 159–161 degenerative lesions, prognosis in, 160 and recovery, 161 static and progressing, 160 preganglionic injury, 390–391 Tissue-compartment syndrome, 318 TMS. See Third molar surgery TOS. See Thoracic outlet syndrome Total avulsion, 377 Total hip arthroplasty, nerve lesions findings in, 494–496 incidence of, 494 limb lengthening and traction, 497 nerves, abnormalities of, 495–496 outcome, 496 pain, 495, 496 urgent reexploration, indications for, 496–497 Traction/stretch injury, 100–103 Transclavicular exposure, 256–260 Transcranial electromagnetic evoked potentials (TCEMEP), 235 Transcutaneous nerve stimulation brachial plexus lesions, pain control, 540–541 catheter insertion, for supraclavicular brachial plexus block, 540 continuous peripheral nerve blocks, 540 response to drugs and methods of treatment, 542 Transverse supraclavicular approach, 254–258 Trapezius, 171 muscle wasting, in spinal accessory neuropathy, 213 paralysis, muscle transfer, 576, 577 Trigeminal nerve injury, iatrogenous injuries dental procedure, features of, 501–502 General Dental Council (GDC), 502 management of, 502 endodontic (ENOO), 507–509 evaluation of, 513–514 implant related nerve injuries, 504–507 improved consent, 521 inferior alveolar nerve injury (IANI), 502 interventions medical, 516 surgical, 516–517 and timing, 515
645
Index lingual nerve injury (LNI), 502 local analgesia (LA), 503–504 management of, 513 tools, 515 mandibular tooth extraction, 513 medico legal issues, 518–519 neuroplasticity, 517 post operative infection, 513 prevention of, 521 reassurance counselling/cognitive behavioural therapy, 515–516 signs and symptoms, 503 socket medications, 513 surgical technique decompression, 517 exploration, 517–518 IAN injuries, 518 scar tissue/neuroma, 518 third molar surgery buccal technique, 509–510 CBCT of, 511 coronectomy, 510–512 DPT radiographs, 510 juxta apical area, 510 lingual nerve injury, 509 U Ulnar nerves, 16, 269–270 in arm and axilla, 266 bundles and epineurial vessels of, 53 elbow, 282, 284 extrinsic epineurial vessels of, 53 hand lesions of, 181 iatrogenous lesion of, 593 proximal stump, 119 sensory loss, 157 upper limb neuropathies, 208–211 distal lesions, 210–211 elbow, 208–211 Wallerian degeneration, 117 Ulnar neuropathy, elbow anterior transposition, complications of, 285 decompression, 284 transposition, 284 Ultrasonography, preganglionic injury, 398 Upper limb, 175 entrapment syndromes, 284–285 neuropathies axillary (circumflex) nerve, 213 brachial plexus lesions, 214–219 long thoracic nerve, 213–214 median nerve, 206–208 radial nerve, 211–212 spinal accessory nerve, 213 suprascapular nerve, 212 ulnar nerve, 208–211 V Vascularised graft, 248–249 sural nerve, failure, 126 Vascularised ulnar nerve graft (VUNG), 125 recovery function, 411–412 repair technique, 399
Vascular lesion, compound nerve injury arterio-venous fistula, axilla, 316 catastrophic bleeding, 307 circumferential rupture, 311 elbow, 312 emergency arterial repair, 308 false aneurysm and arteriovenous fistulae, 313–316 femur and tibia, fracture, 314 iatrogenous false aneurysm, 317 iatrogenous ischaemic injury, 319–320 iatrogenous post ischaemic contracture, 319 ischaemia and nerve, 316–318 knee, 312–313 leaking aneurysm, 310 multiple fragment wounds, 307 Poiseuille’s law, 312 proximal humerus, fracture, 310 shoulder, 309–312 supraclavicular plexus and penetrating missile wounds, 309 tissue-compartment syndrome, 318 traumatic false aneurysms and arterio-venous fistulae, 309 ulnar artery repair, 307 Vascular repair artery/vein, 239 reversed vein graft, 239, 241, 242 Ventral root (VR) regeneration, 136 repair, 403, 411, 412 Vertebral artery thrombosis, preganglionic injury, 383 Vibration injury, 107 Visceral afferents, 70–71 Visual analogue scales (VAS), for pain, 528 Volkmann’s contracture ischaemic, 93 of forearm and intrinsic muscles, 571 and supracondylar fracture, 357–358 W Wallerian degeneration cervical nerve distal stump of, 81 ventral root of, 89 in ulnar nerve, 117 War wounds, compound nerve injury contamination, 304 gas gangrene, 305 missile wounds, leg, 306 principles, 305 stages, 305–306 Wounds nerve injury categories of, 148 system scores, features, 147 of perineurium, 84–85 Wrist median and palmar cutaneous nerve, 156 median neuropathy, 206 primary suture of median nerve, 246