Biopsy Diagnosis of Peripheral Neuropathy Gyl Midroni, Juan M. Bilbao, M.D., F.R.C.P.(C) M.D., F.R.C.P-(C) Neurologist Neuropathologist Associate Professor, Dept. of Pathology, University of Toronto with the technical assistance of Sandra M. Cohen, r.t. St. Michaels Hospital University of Toronto Butterworth-Heinemann Boston · Oxford · Melbourne · Singapore · Toronto · Munich · New Delhi · Tokyo Copyright © 1995 by Butterworth-Heinemann -β^ A member of the Reed Elsevier group All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of the publisher. Every effort has been made to ensure that the drug dosage schedules within this text are accurate and conform to standards accepted at time of publication. However, as treatment recommendations vary in the light of continuing research and clinical experience, the reader is advised to verify drug dosage schedules herein with information found on product information sheets. This is especially true in cases of new or infrequently used drugs. ^ Recognizing the importance of preserving what has been written, Butterworth-Heinemann prints its books on acid-free ^ paper whenever possible. Library of Congress Cataloging-in-Publication Data Midroni, Gyl, 1964- Biopsy diagnosis of peripheral neuropathy / Gyl Midroni, Juan M. Bilbao : with the technical assistance of Sandra M. Cohen, p. cm. Includes bibliographical references and index. ISBN 0-7506-9552-8 (hard cover : alk. paper) 1. Nerves, Peripheral—Biopsy. 2. Nerves, Peripheral—Diseases— Diagnosis. I. Bilbao, Juan M., 1938- . II. Title. [DNLM: 1. Peripheral Nervous System Diseases—diagnosis. 2. Biopsy 3. Sural Nerve—anatomy & histology 4. Sural Nerve— pathology 5. Peripheral Nervous System Diseases—pathology. WL 500 M629b 1995] RC409.M47 1995 616.87—dc20 DNLM/DLC for Library of Congress 95-8618 CIP British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library The publisher offers discounts on bulk orders of this book. For information, please write: Manager of Special Sales Butterworth-Heinemann 313 Washington Street Newton, MA 02158-1626 10 9 8 7 6 5 4 3 21 Printed in the United States of America Conventions The following abbreviations are used frequently throughout the monograph: CIDP chronic inflammatory demyelinating polyradiculoneuropathy EM electron microscopy GBS Guόlain-Barrι Syndrome H&E hematoxylin and eosin HMSN hereditary motor and sensory neuropathy LCA leukocyte common antigen LM light microscopy MF myelinated fiber NMSC non-myelinating Schwann cell OB onion bulb PAS periodic acid-Schiff PNS peripheral nervous system SC Schwann cell ScSu Schwann cell subunit UF unmyelinated fiber The following conventions apply to all figures: • Paraffin-embedded tissue is histostained with H&E unless stated otherwise. Semi-thin (plastic-embedded) tissue is always stained with Toluidine Blue. • Magnifications are provided for electron micrographs only. • Where appropriate, axons are labeled with "a." • ΑΠ iUustrations are from sural nerve biopsies unless otherwise stated. Preface This monograph is directed at neurologists and pa­ Bruni, T.C. Chen, Sam Cheung, Joseph Chu, Sharon thologists who require a work that combines a Cohen, Michael Cusimano, Al-Noor Dhanani, Farouk practical approach to nerve biopsy interpretation in Dindar, Eric Duncan, Sherali Esmail, Ignatio Fong, peripheral neuropathy with an overview of the prog­ Victor Fornasier, Richard Gladstone, Warren Gold­ ress that has accumulated over the past two decades in stein, Allan Gordon, Trevor Gray, Arthur Gryffe, Mark the understanding of peripheral nerve pathobiology Guttman, Gaspar Israelian, Dennis Izukawa, Ralph Also covered are clinical features of polyneuropathy Kern, Edward Keystone, Peter Kopplin, Colin Lam­ and the microscopic anatomy of nerves, both essential bert, Arnold Lang, Richard Magder, James Mahoney, to the interpretation of pathological change. In this Zbieg Manowski, Ronald McDonald, Donald McGil­ book attention is on the usefulness and limitations of livray, Stephen McKenzie, Arline McLean, Gary Mod- nerve biopsy as a diagnostic tool, and for a broad del, David Morganthau, Ayoob Mossanen, Richard perspective the literature has been extensively re­ Moulton, Paul Muller, John Norris, Paul O'Connor, viewed. We hope that our efforts to provide a rich Richard Perrin, Ah Qizilbash, Paul Ranalli, Gordon pictorial exposition of peripheral nerve histology and Sawa, Carol Sawka, Jacob Schneiderman, Daniel pathology will be helpful to the initiate. Selchen, Raold Serebrin, David Sutton, Bryan Temple, Our material consists of nearly 700 consecutive William Tucker, Jean Turley, Felix Tyndel, C. Peter nerve biopsies collected over a 22-year period at St. Watson, John Wherrett, Ronald Wilson, Robert Yufe, Michael's Hospital in Toronto. In the study of this and Catherine Zahn. tissue we employ techniques of classical histology, Dr. Peter Ashby, Dr. Kaiman Kovacs, and Dr. James resin histology, immunohistochemistry, electron mi­ Perry gave valuable suggestions for the monograph. croscopy, and—to a much lesser degree—morphome­ We thank Dr. Rob Macaulay for his many contributions try and fiber teasing. to this work. The skilled technical assistance of Wayne We acknowledge that biopsy of a subcutaneous Ozanne (ART) is gratefully acknowledged. Daniel A. nerve provides only a window into the morphological Hunter, RT, of Washington University School of Medi­ alterations in peripheral neuropathies; thus through­ cine, St. Louis, Missouri, generously performed mor­ out the text we refer to other works for a more com­ phometry. The librarial services at St. Michael's plete view of the pathology of the human peripheral Hospital were critical to the collection of reference nervous system. material. The authors wish to thank Leica Canada and This work was made possible through the support the Mount Sinai Hospital for permission to use their of the Sisters of St. Joseph, the St. Michael's Hospital equipment. Lianne Friesen and George Trogadis cre­ department of pathology, and Dr. Alan Hudson. Our ated the original artwork. Dr. Maja Steinlin and Dr. editor Susan Pioli, who supervised this work from Steffen Albrecht provided German translations. inception, was a constant source of encouragement. Sandra M. Cohen wishes to thank Steven M. Doyle We are grateful to the following pathologists who of the department of microbiology at the University of sent us specimens from conditions not available in our Toronto for his invaluable technical assistance and collection: Dimitris Agamanohs (Ohio), George David­ generosity in the loan of a diamond knife. son (Toronto), John Deck (Toronto), Venita Jay (To­ One of us is a clinical neurologist with a special ronto), Edward Johnson (Edmonton), Jacques interest in peripheral neuropathy; the other is a neuro­ Lamarche (Sherbrooke, Quebec), John Maguire (Ham­ pathologist with remote training in clinical neurology ilton), Jean Michaud (Montreal), and David Munoz and a special interest in the pathology of nerves. Juan (London, Ontario). M. Bilbao is especially indebted to the late Dr. Morri­ To the many physicians who shared their cases with son Finlayson. Gyl Midroni would like to acknowl­ us goes our appreciation: Peter Ashby, Sheldon edge his parents' influence and to dedicate his work to Baryshnick, Neville Bayer, Catherine Bergeron, Henry his wife. Dr. Phuong Nguyen, for her continuing pa­ Berry, Alexander Birnbaum, Donald Borrett, Joseph tience and support. xiii Foreword Of all the many pathology subspecialties, perhaps and the comprehensive coverage of the bibhography none appears more arcane to the general surgical obviously contribute a great deal to this achievement. pathologist than that dealing with the diseases of pe­ Other welcome features are the systematic approach, ripheral nerves. I have often wondered how much of the emphasis on a sound knowledge of the normal this inscrutability is due to the intrinsic difficulty of the structure of nerve, and the incorporation of recently subject, and how much to the absence of a work which acquired cytogenetic and molecular genetic data. One gathers the widely scattered body of hterature pertinent is also impressed by their objective and dispassionate to this field and presents it in a user-friendly format. consideration of the diagnostic utility of nerve biopsy Drs. Midroni and Bilbao have given us the answer. in many clinical situations. The book is punctuated by To be sure, the field is a complex one. Yet, they have illustrative case reports, including some "rare birds," shown us in this lucidly written treatise that adherence such as lymphomatoid granulomatosis involving to a simple and logical method can lead to apprecia­ nerve, and the neuropathy of adult onset Niemann- tion of the pathogenesis of the disease even by the Pick disease. uninitiated. Their approach is a practical one, well Drs. Midroni and Bilbao have made a significant suited for pathologists on the front lines of tissue and much needed contribution to the field, for which diagnosis, as well as for neurologists and neuropathol­ the community of anatomical pathologists and periph­ ogists who take a special interest in peripheral nerve eral nerve specialists should be most grateful. morphology. The uniformly superb quality of their illustrations Juan Rosai, M.D. XV C H A P T ER 1 Peripheral Neuropathy and the Role of Nerve Biopsy I. EPIDEMIOLOGY OF PERIPHERAL most frequent inflammatory and infectious neuropa­ NEUROPATHY thies are Guillain-Barré Syndrome (GBS), Chronic In­ flammatory Demyelinating Polyradiculoneuropathy Diseases of the peripheral nerve present an impor­ (CIDP), vasculitic neuropathies, and leprosy. Malig­ tant neurologic problem: the incidence of polyneurop­ nancy is associated with neuropathy through a variety athy in the United States is estimated at 40 per 100,000, of mechanisms, including autoimmunity, metabolic comparable to that of epilepsy or parkinsonism.^^ Since derangement, and infiltration of nerve. The most com­ not all patients are investigated, data regarding the mon familial neuropathies are HMSN types 1 and 2. In etiologic composition of polyneuropathies are hard to the modern literature, 10% to 20% of neuropathies obtain. For example, Dyck et al. estimate that only 10% remain cryptogenic (vide infra). Table 8-4 provides of affected patients with hypertrophic hereditary mo­ a detailed etiologic tabulation of peripheral nerve tor and sensory neuropathy (HMSN-1) present as a diseases. direct consequence of symptoms produced by this disease.^^ Reported series based on consecutive patients from well-defined geographic areas give some insight into Furthermore, many individuals with mild or sub­ the relative frequency of these neuropathies. Prineas^^ clinical neuropathy, whether genetically determined or reviewed 278 consecutive English patients hospital­ acquired, do not require investigation or treatment. ized with a diagnosis of polyneuropathy between 1957 The largest and best documented peripheral neuropa­ and 1966. The study eliminated 49 of these patients, thy series originate from specialized centers or are whose diagnoses were questionable or whose neurop­ selected from biopsy cases and so are not representa­ athies were inadequately investigated, and concen­ tive of neuropathy in the general population. trated on the remaining 229 patients. Approximately 30% to 35% were associated with well-known toxic or The etiologies of peripheral neuropathy metabolic causes, with nutritional polyneuropathy It is helpful to classify the major etiologies of pe­ and alcoholism forming the largest group. Vitamin B12 ripheral neuropathy into acquired toxic or metabolic, deficiency and drug-induced neuropathies each ac­ inflammatory or infectious, neoplasia and parapro­ counted for 16 of 229 patients, while diabetes occurred teinemia associated, and genetically determined. Ac­ in only 5% (12 of 229). Although not classified as such, quired toxic or metabolic factors probably account for examination of the data suggests that GBS accounted the majority of neuropathies. Common causes of such for 20% to 25% of the entire group of 229 patients and neuropathies include diabetes, alcohoHsm, nutritional CIDP another 7% to 15%. Ten famihes had inherited deficiencies, and pharmaceutical neurotoxins. The neuropathies. However, the chronic idiopathic group 1 Biopsy Diagnosis of Peripheral Neuropathy may also have included some genetically determined Cryptogenic peripheral neuropathy neuropathies for which a family history was not ob- Despite widely variable selection criteria the pro­ tained.^3 Fifteen of 229 patients (7%) had neuropathy portion of neuropathies that remain cryptogenic is associated with malignancy. remarkably constant in modern series, ranging from In a report on 120 consecutive patients with poly­ 10% to 15%.2'9'i^'28'49 Series from the 1950s and 1960s neuropathy studied over a three-year period in an reported 56% to 70% of polyneuropathies as crypto- EMG clinic in Israel, diabetic neuropathy accounted genic.2^'^^ However, investigators in that era did not for 22%. Twenty percent had GBS, CIDP, or vascuUtis. include several currently accepted "etiologies" of pe­ Thirteen percent were genetically determined, 13% ripheral neuropathy, such as GBS, CIDP, paraprotein metabolic, 6% drug induced, 4% associated with associated neuropathy, paraneoplastic neuropathy, malignancy, 8% miscellaneous, and 12% remained un- and genetically determined neuropathies without a diagnosed.2 A 380-patient prospective peripheral neu­ family history.^^ When carefully assessed patients with ropathy series from France revealed alcohol or drug a chronic idiopathic neuropathy are re-studied, an etio­ toxicity in 23% of patients, diabetic neuropathies in logic diagnosis often can be made. McLeod and co­ 16%, GBS and CIDP in 16%, hereditary neuropathies workers reported that approximately one-third of their in 10%, and neuropathies associated with collagen chronic idiopathic neuropathies subsequently could be diseases in 10%."^^ Diabetes is consistently underesti­ diagnosed at a one-year interval,^^ and Grahmann and mated because many diabetics do not consult neurol­ colleagues reported similar numbers.^^ In a review by ogists and are rarely admitted to hospital because of Dyck and coworkers of 205 previously unclassified neuropathy. patients, an etiologic diagnosis was made in 76%.^^ Thus, despite the absence of information about the However, this group was probably not as well studied types and frequency of polyneuropathy in the general prior to referral as the cryptogenic patients in the series population, a first estimate suggests that in the devel­ of McLeod et al.^^ These studies emphasize the impor­ oped world metabolic, toxic, and nutritional causes tance of careful history taking in diagnosis, especially account for 50% or more of neuropathies; inflamma­ for evidence of a familial neuropathy or of toxic expo- tory neuropathies (mainly GBS, CIDP, and vasculitis) sure.^^'^^ Examination of family members, even if not for 10% to 20%; familial neuropathy for 10% to 20%; known to be symptomatic, can prove extremely help- neoplasia associated neuropathy for 5% to 10%; and ful.^^ In other patients the etiology becomes clear with approximately 10% to 20% are idiopathic. A large vari­ time, as malignancy or other systemic diseases emerge. ety of rare neuropathies make up the rest of the group. The above figures represent the experience with adult patients in the developed world. In pediatric II. USEFULNESS OF NERVE BIOPSY IN neuropathy the etiologic spectrum differs, with fa­ EVALUATION OF PERIPHERAL NEUROPATHY milial neuropathies predominating.^^ Moreover, geo­ For several classes of patients, biopsy is rarely neces­ graphic factors are pertinent in determining the sary: patients with an established cause of toxic or relative frequency of neuropathies in a given popula­ metabolic neuropathy, patients with a familial neuropa­ tion. In a large Indian series 66% of neuropathies were thy, those with the Guillain-Barré Syndrome, or in pa­ diabetic, 14% leprous, and 11% GBS.^^ Leprous, nutri­ tients who are known to have a circulating paraprotein tional, and familial neuropathies were under- or systemic malignancy. In addition, patients with sus­ represented and GBS over-represented in this hospital pected CIDP may not require biopsy (vide infra). Nev­ inpatient series.^^ In a 1980 review, Osuntokun esti­ ertheless, biopsy continues to have a significant role in mated that in Africa 25% to 40% of neuropathies seen the diagnosis of peripheral nerve disease (Table 1-1). by neurologists were nutritional, with 5% to 10% each due to leprosy, alcoholism, and the Guillain-Barré Syn- drome.^^ Diabetes was accorded equal ranking with Review of experience in our laboratory porphyria at about 3% each! However, this author Histological material and clinical data noted that when all patients are considered, including those not coming to neurologists' attention, leprosy Since 1972 the peripheral nerve pathology labora­ was by far the most important cause of neuropathy, tory at St. Michael's Hospital has processed nearly 700 and diabetic neuropathies became more numerous.^^ nerve biopsies. These consist almost exclusively of Chapter 1 Peripheral Neuropathy and the Role of Nerve Biopsy Table 1-1 Specific diagnoses that can be made by segmental demyeUnation, mixed axonal/demyelinat­ nerve biopsy ing, and inflammatory or non-inflammatory) or as Inflammatory/infectious changes that permit a specific diagnosis in the absence of clinical information (e.g., vascuhtis, amyloid). The • Neuropathy with macrophage mediated demyelination (CIDP, GBS) clinician most involved in the case used all available • Vasculitic neuropathy information, including biopsy findings and follow-up, to make the final diagnosis. Based on the available • Leprous neuropathy, especially primary neuritic leprosy clinical material, the nerve biopsy was classified by • Sarcoidosis or granulomatous neuropathy ourselves as being of no value, helpful, or essential for • CMV neuritis in immune suppressed patients optimal patient management. Neoplasm/paraprotein associated A biopsy of no value did not modify management • Non-amyloid paraprotein associated neuropathies other than providing an impression of the severity and —IgM anti-MAG paraprotein (widely spaced myelin)** activity of the disease. Helpful biopsies a) supported —The POEMS syndrome (uncompacted myelin)** the diagnosis of a suspected etiology for which treat­ —Immunoglobulin deposition disease ment or genetic counseling is available (e.g., CIDP, • Primary amyloidosis HMSN); b) ruled out a working diagnosis (i.e., finding • Neoplastic infiltrative neuropathy of prominent active wallerian changes in suspected • Lymphomatoid granulomatosis HMSN-2); or c) distinguished between two alternative Metabolic/toxic diagnoses with therapeutic implications (e.g., leprous • Amiodarone neuropathy vs vasculitic neuropathy in a patient from a leprosy • Hexacarbon neuropathy endemic region). Biopsies essential to management revealed abnormalities that a) permitted definitive di­ Genetically determined agnosis of the cause of the neuropathy; or b) revealed • Hereditary neuropathy with liability to pressure palsies* suggestive but not completely diagnostic findings, re­ • Amyloid neuropathy (familial*) sulting in altered management (e.g., noncaseating • Giant axonal neuropathy granulomata suggesting sarcoidosis, or findings typi­ • Neuroaxonal dystrophy cal of CIDP in an electrophysiological^ "axonal" neu­ • Polyglucosan body disease ropathy). • Hereditary sensory and autonomic neuropathies Results • Storage diseases —Metachromatic leukodystrophy* In a review of nerve biopsies and clinical records for —Adrenoleukodystrophy* 267 patients with a mean age of 57.5 (Figure 1-1), the records of 234 patients had sufficient clinical material —Globoid cell leukodystrophy* to permit accurate clinical-pathological correlation. In —Niemann-Pick disease* these 234 patients, a final etiologic diagnosis was —Fabry's disease* reached in 76% (Table 1-2). Diagnostic findings were —Tangier disease* seen in 16% of all nerve biopsies (Table 1-3). —Neuronal ceroid lipofuscinosis We judged nerve biopsy essential for management in 48 patients (21%). In most of these cases tissue * Noninvasive biochemical or genetic tests may obviate need for tissue examination. examination revealed a specific diagnosis (e.g., vascu­ ** Strongly but not invariably associated. Utis, amyloidosis, leprosy). On three occasions biopsy revealed a picture typical of previously unsuspected sural nerves, with infrequent utilization of cutaneous CIDP. In one patient clinically diagnosed with HMSN- branches from the radial or ulnar nerves or the lateral 1, biopsy showed unequivocal inflammatory features peroneal nerve. During this period of time, the same suggestive of CIDP. In two patients with a clinical pathologist (JMB) has used a consistent protocol diagnosis of HMSN-2, biopsy revealed additional find­ (p. 119) to interpret these biopsies. The reports of 267 ings that led to disease-modifying treatment; in one consecutive nerve biopsies examined in the years 1986 patient a toxic medication was discontinued and in to 1993 were reviewed and the histological changes another steroids were given. classified as either nonspecific (axonal degeneration. We considered nerve biopsy helpful in 52 patients Biopsy Diagnosis of Peripheral Neuropathy 50 Analysis of our material suggests that no difference exists in the diagnostic value of nerve biopsy between neuropathies with axonal vs demyelinating or mixed electrophysiological features (Table 1-4). Indeed, nerve 40 biopsies in patients with axonal neuropathy reveal important and often unsuspected pathology such as vasculitis and amyloidosis. In addition, three patients with axonal electrophysiology showed a picture typi­ . 30 cal of CIDP on tissue examination. 0) Si ε Literature data on the usefulness of nerve biopsy ^20 Fifty-three of 120 consecutive patients presenting to an EMC clinic underwent nerve biopsy.^ Thirty-eight percent of the 53 biopsies yielded diagnostic or con­ tributory information that altered management in half 10 (19%) of the patients. In another series of 56 sural nerve biopsies, 27% provided information considered crucial for diagnosis. In an additional 37% the biopsy was Jl 1 nonspecific but supported the suspected cHnical diag­ 5 20 35 50 65 80 95 nosis and excluded elements of the differential diagno- years sis.^^ Reviewing personal experience with 385 biopsies, Oh^i was able to make a specific diagnosis in 24% of the cases. Fig. 1-1 Age distribution of patients for 267 consecutive nerve biopsies 1986-1993. III. INDICATIONS FOR NERVE BIOPSY (22%). In most of these cases biopsy supported a diag­ Indications for nerve biopsy have not been delin­ nosis of CIDP, HMSN-1 or HMSN-2. In other instances, eated. A 1967 position paper from the Research Group nerve biopsy enabled physicians to distinguish be­ in Neuromuscular Diseases of the World Federation of tween paraneoplastic and chemotherapy-related neu­ Neurology indicated that ". . . nerve biopsy is an in­ ropathy by showing prominent inflammation. Biopsy vestigation which should only be carried out in se­ also confirmed the diagnosis of GBS in a patient with lected centers where facilities for full and detailed atypical history and physical findings. In one patient a histologic studies are available, and the clinical indica­ "helpful" nerve biopsy revealed selective small my­ tions for the use of this method of investigation are elinated fiber loss without amyloid deposition. The few."'^^ Although we find this a restrictive statement, possibility of amyloidosis was raised by this pattern, we agree that interpreting pathologic change in a and this suspicion was confirmed by a persistent nerve specimen requires experience. Most problems search in muscle and rectal biopsy material. that beset the beginner relate to artifacts produced by The 267 consecutive patients studied were identified the biopsy procedure (crush, stretch), during process­ in our pathology laboratory and thus were biased to­ ing (rough handUng, delayed fixation, hyperosmolar wards diagnostically challenging neuropathies and to­ fixative), and during embedding (lack of orientation of wards etiologies for which a biopsy is used to confirm a fascicles, type of plastic). However, the techniques nec­ suspected treatable diagnosis. This circumstance pre­ essary for processing nerve tissue can easily be sumably accounts for the over-representation of CIDP adapted from those found in any established histology and vasculitis in the final diagnoses and also for the laboratory. These techniques include plastic embed­ relatively high rate of cryptogenic neuropathies (24%) ding; electron microscopy; and immunohistochemis- (Table 1-2). Approximately half the patients with cryp­ try in paraffin, frozen, and plastic sections. Diagnosis togenic neuropathy suffered from a minimally dis­ rarely requires the more difficult technique of nerve abling and non- or slowly progressive neuropathy, as teasing. With respect to the indications for nerve bi­ has been previously documented.^^'^^ opsy, the position quoted above was taken before a) Chapter 1 Peripheral Neuropathy and the Role of Nerve Biopsy Table 1-2 Clinicopathologic correlations in 267 nerve biopsies Nonspecific histology Specific Non-inflammatory Inflammatory Normal Etiology Ν histology axonal mixed/dem axonal mixed/dem Final diagnosis obtained (N=177) CiDP(GBS) 51(1) 5 12 7 27 1 Vasculitis 19 19 0 0 0 0 0 Paraprotein 12 2 3 3 1 2 1 NASID 11 0 3 1 4 2 1 Diabetes 10 0 5 2 2 1 0 Toxic/nutritional* 10 3 6 1 0 0 0 HMSN-1 5 0 0 5 0 0 0 HMSN-2 7 0 7 0 0 0 0 Amyloid 5 4 1 0 0 0 0 Uremia 4 0 3 1 0 0 0 Paraneoplastic 4 0 2 0 2 0 0 Other** 13 10 2 0 1 0 0 Not neuropathy 25 0 7 3 0 0 15 Final diagnosis not obtained (N=90) Cryptogenic 52 0 29 10 7 2 4 Insufficient data 33 1 19 2 6 2 3 Random 5 0 2 1 0 0 2 Total 267 39 94 41 30 36 27 15% 35% 15% 11% 13% 10% * Alcohol (3), amiodarone (3), INH/ethambutal (1), dilantin (1), propafenone (1), severe malnutrition (1). " Hereditary neuropathy with pressure palsies (3), granulomatous neuropathy (2), leprosy (2), infiltrating lymphoma (2), Fabry's disease (1), idiopathic sensory ganglionitis (1), spinocerebellar degeneration (1), hypothyroid (1). "* Four biopsies demonstrated macrophage mediated demyelination on EM. See footnote for Table 1-3. full appreciation of CIDP as a treatable neuropathy; b) the establishment of immunostaining techniques (for Table 1-3 Specific findings in 43 biopsies example, identification of amyloid major proteins); and c) the emergence of such entities as nonsystemic Necrotizing vasculitis 20 peripheral nerve vasculitis,^^ giant axonal neuropathy,^ Amyloidosis 4 (2 familial-TTR , 2 primary) and hereditary neuropathy with pressure palsies/ Fur­ CIDP (MMD)* 4 thermore, using plastic embedded material for light Tomaculous neuropathy 3 and electron microscopy studies permits a far more Amiodarone 3 informative examination of axons and myeUn than do Leprosy 2 the standard paraffin-based histological techniques Paraprotein** 2 employed in the 1960s and earlier. Granulomatous (sarcoid) 2 More recently, Asbury and Gilliatt have emphasized Lymphoma (infiltrative) 2 the usefulness of nerve biopsy in asymmetric and mul­ tifocal neuropathies and indicated that . . distal Fabry's disease 1 symmetrical polyneuropathies, whether subacute or * MMD: macrophage mediated demyelination. This finding is specific to CIDP chronic, axonal or demyelinating, are not further clari­ and GBS, and much more common than 4 of 52 cases. However, electron microscopic search for MMD is unnecessary when the clinical data and light fied by nerve biopsy. . . ."^ However, in our experience microscopic pictures are congruent. ** Paraproteinemic neuropathy does not usually have characteristic findings. a significantly asymmetric neuropathy occurred in However, one of these biopsies demonstrated widely spaced myelin strongly associated with circulating IgM paraprotein having activity against myelin- only a third of the instances noted above in which we associated glycoprotein. The other demonstrated endoneurial effacement with massive deposits of kappa light chain. rated biopsy essential for management. Significantly