Comprehensive Electromyography: With Clinical Correlations and Case Studies [1st ed.] 9781108578103

Table of contents :
Cover......Page 1
Half-title......Page 3
Title page......Page 5
Copyright information......Page 6
Dedication......Page 7
Table of contents......Page 9
Preface......Page 11
Acknowledgments......Page 14
Section 1 Introductory Chapters......Page 15
Introduction......Page 17
Charge......Page 18
Current......Page 22
Voltage......Page 24
Resistance......Page 25
Ohm’s Law......Page 26
Types of Circuits and Current......Page 27
Series DC Circuits......Page 28
Parallel DC Circuits......Page 32
Capacitors......Page 34
Alternating Current......Page 37
The Quantification of AC Signal......Page 38
Household Electricity......Page 40
Transformers......Page 44
Electromagnetic Radiation......Page 45
Recommended Reading......Page 46
Capacitance......Page 47
Filters......Page 48
Filter Arrangements......Page 54
Analog-to-Digital Converters and Digital-to-Analog Converters......Page 55
Stimulators......Page 57
Monopolar Amplifiers......Page 58
Differential Amplifiers......Page 59
The Common Mode Rejection Ratio......Page 61
Suggested Reading......Page 62
The Cell Membrane......Page 63
Motor Neurons......Page 64
Sensory Neurons......Page 65
Nerve Fiber Classification......Page 66
Ion Channels......Page 67
Transmembrane Potentials, Nernst Equilibrium Potentials, and Selective Permeability......Page 69
Depolarization and Repolarization......Page 71
The Absolute Refractory Period and the Relative Refractory Period......Page 72
Local Responses......Page 73
The Advantage of Myelin......Page 75
Action Potential Propagation Speed......Page 76
Connective Tissue Elements of Nerve Trunks......Page 79
References......Page 80
Introduction......Page 82
Synaptic Space......Page 83
Postsynaptic Region......Page 84
Postsynaptic Membrane Depolarizations......Page 85
References......Page 86
Structural Organization of Muscle......Page 87
The Sarcomere......Page 88
The CNS Influence......Page 90
Muscle Shortening......Page 91
Motor Unit and Muscle Fiber Physiology, Biochemistry, and Metabolism......Page 92
Motor Unit Size and Distribution......Page 93
References......Page 94
Section 2 Nerve Conduction Studies......Page 97
Introduction......Page 99
History......Page 100
Proper Placement of the Recording Electrodes......Page 101
The Basic Technique......Page 102
Volume Conduction......Page 104
Orthodromic versus Antidromic......Page 106
The Stimulating Electrodes and Their Proper Placement......Page 107
References......Page 108
Biphasic Morphology......Page 109
The E2 Electrode Is Not Inactive......Page 110
Positive Dip......Page 111
Nerve Stimulation......Page 112
Physiologic Temporal Dispersion......Page 113
What We Measure and What It Means......Page 115
Amplitude......Page 116
Negative AUC......Page 118
Distal Latency......Page 119
The Contributors to the Distal Latency Time......Page 120
Conduction Velocity......Page 121
References......Page 123
Technique......Page 125
Measurements......Page 126
Amplitude......Page 127
Orthodromic versus Antidromic Techniques......Page 128
Fixed Landmarks versus Fixed Distances......Page 130
Peak Latencies versus Onset Latencies......Page 131
Conduction Velocity......Page 132
Effects of Filtering......Page 133
Ideal Interelectrode Distance......Page 134
Mixed Nerve Conduction Studies......Page 135
References......Page 136
Effects of Focal Demyelination on Action Potentials......Page 137
Introduction......Page 138
The Pathology of Focal Demyelination......Page 139
Uniform Demyelinating Conduction Slowing......Page 140
Nonuniform Demyelinating Conduction Slowing......Page 141
Focal Demyelinating Conduction Block......Page 142
Conduction Failure......Page 145
The Major Advantages and Disadvantages of the Motor NCS......Page 146
The Sensory NCS Manifestations of Pathology and Pathophysiology......Page 147
The Timing of NCS Manifestations......Page 149
References......Page 150
The Effect of Focal Demyelination and Axon Disruption on the NCS......Page 151
Motor NCS......Page 152
References......Page 154
Introduction......Page 155
Technique......Page 156
Technical Errors......Page 157
Amplitude and Latency Measurements......Page 158
Utility of H Wave Testing......Page 159
Physiology and Technique......Page 160
Utility......Page 161
A Waves......Page 162
Blink Reflexes......Page 163
References......Page 164
Introduction......Page 166
Introduction......Page 167
Technique......Page 168
Postexercise Facilitation and Postexercise Exhaustion......Page 169
Technical Errors......Page 170
High-Frequency RNSS Technique......Page 171
References......Page 172
Section 3 The Needle EMG Examination......Page 173
Introduction......Page 175
Motor Unit Anatomy and Physiology......Page 176
Duration......Page 177
Motor Unit Recruitment (Force Generation)......Page 178
The Measurements We Make and Their Meanings......Page 183
Insertional Phase......Page 184
Miniature Endplate Potentials......Page 185
Endplate Spikes......Page 186
MUAP Amplitude (Peak-to-Peak)......Page 187
MUAP Duration......Page 189
The Number of Phases and Turns Composing the MUAP......Page 191
Needle Recording Electrode Types......Page 192
Concentric Needle Electrodes......Page 193
Monopolar Needle Electrodes......Page 194
References......Page 195
Chapter Organization......Page 197
Resting Phase......Page 198
Morphology and Duration......Page 199
The Auditory Characteristics and Firing Frequency of Fibrillation Potentials......Page 200
Specificity and Utility......Page 201
Fasciculation Potentials and Cramp Potentials......Page 202
Electrodiagnostic Features......Page 203
Myotonic Potentials......Page 204
Neuromyotonia......Page 205
Grouped Repetitive Discharges (GRDs) and Myokymia......Page 206
Activation Phase......Page 207
MUAP Duration......Page 208
Reinnervation via Collateral Sprouting......Page 209
MUAP Recruitment......Page 210
Grading Neurogenic Recruitment......Page 211
Decreased Spatial And Temporal Recruitment......Page 212
Non-Myopathic Motor Unit Disintegration Disorders......Page 213
MUAP Stability......Page 214
Motor Axon Loss Following Reinnervation via Collateral Sprouting......Page 215
References......Page 216
Single-Fiber Needle Electrodes......Page 218
Technique and Measurements Made......Page 219
Jitter Using a Concentric Needle Electrode......Page 220
Macro EMG Needle Electrodes......Page 221
References......Page 222
Section 4 Other Pertinent EDX Information......Page 225
The Importance of Proper Initial Management......Page 227
Correlations between Lesion Acuteness and the Underlying Pathophysiology......Page 228
Electrodiagnostic Grading......Page 229
Timing of the Study......Page 230
The Value of Motor Responses in the Assessment of Lesion Severity......Page 231
Collateral Sprouting......Page 232
Proximodistal Axon Regeneration......Page 233
Determining the Potential for Reinnervation......Page 234
Neurapraxia......Page 235
Sunderland Grade 2......Page 236
Sunderland Grade 5......Page 237
Surgical Intervention......Page 238
Approach to Axon Loss (Grades 2–5)......Page 239
Major Surgical Interventions......Page 240
Types of Nerve Injuries......Page 241
Compression Injuries......Page 242
Acute Compression......Page 243
Chronic Compression......Page 244
Ischemic Injury......Page 245
References......Page 246
Upper Motor Neuron Disorders......Page 249
Anterior Horn Cell Disorders......Page 250
Post-Poliomyelitis Syndrome......Page 251
Spinal Muscular Atrophy, Type 3......Page 252
Radiculopathies......Page 253
Plexopathies......Page 255
The Brachial Plexus......Page 256
The Lumbosacral Plexus......Page 259
Focal Neuropathies (Mononeuropathies)......Page 260
Acquired Axon Loss Polyneuropathies......Page 261
Acquired Demyelinating Polyneuropathies......Page 262
Chronic Inflammatory Demyelinating Polyradiculoneuropathy......Page 263
Hereditary Polyneuropathies......Page 264
Neuromuscular Junction Disorders......Page 265
Repetitive Nerve Stimulation Studies......Page 266
Needle EMG Study......Page 267
Myopathic Disorders......Page 268
References......Page 269
Studying Cool Limbs......Page 271
Nerve Conduction Studies......Page 272
Repetitive Nerve Stimulation Studies......Page 273
Nerve Conduction Studies......Page 274
Body Habitus–Related issues......Page 275
Introductory Comments......Page 276
MGA to the Hypothenar Eminence......Page 277
MGA to the Adductor Pollicis Muscle......Page 278
MGA with Concomitant Carpal Tunnel Syndrome......Page 279
MGA with a Concomitant Ulnar Neuropathy......Page 280
Atypically Proximal MGA......Page 281
Identifying an RCA......Page 282
Identifying an Accessory Deep Peroneal Nerve......Page 283
The Presence of a Positive Dip......Page 284
Stimulus (Shock) Artifact......Page 285
Stimulus Artifact Reduction......Page 286
Rotation of the Anode about the Cathode......Page 287
Submaximal Stimulation......Page 288
Excessive Stimulation......Page 289
60 Hz Power Artifact......Page 290
Measurement Errors......Page 291
Filtering Issues......Page 293
Needle EMG......Page 295
Changing the Sensitivity to Better Identify the Onset Latency......Page 296
Sweep Speed and Sensitivity......Page 297
Identifying Demyelination Based on the Latency of a Very Low Amplitude Response......Page 298
Measurement Errors......Page 299
References......Page 300
Electrical Injury......Page 304
Indwelling Medical Devices......Page 307
Bleeding......Page 309
Transmission of Infection......Page 310
Prevention......Page 312
References......Page 313
EDX Testing as an ‘‘Extension of the Neurological Examination’’......Page 316
The Timing of the EDX Examination......Page 317
The Limitations of Electrodiagnostic Testing......Page 318
The EDX Study Components Performed and Their Order of Performance......Page 319
Supervision......Page 320
Reviewing the Consult......Page 321
Explaining the EDX Study to the Patient (Verbal Informed Consent)......Page 322
Brief Discussion......Page 323
Tips to Lessen the Discomfort Associated with the Needle EMG Study......Page 324
Format of Report......Page 326
References......Page 327
Section 5 Case Studies in Electrodiagnostic Medicine......Page 329
Lesion Localization......Page 331
Identifying Nonorganic Lesions......Page 332
Initial Studies......Page 333
Abbreviations......Page 334
Needle EMG......Page 335
The Electrodiagnostic Exercises......Page 336
References......Page 493
Section 6 Appendices......Page 495
The Lumbosacral Plexus......Page 497
The Median Nerve......Page 498
The Ulnar Nerve......Page 499
The Radial Nerve......Page 500
The Obturator Nerve......Page 501
The Femoral Nerve......Page 502
The Sciatic Nerve Proper......Page 503
The Tibial Nerve......Page 504
The Peroneal Nerve......Page 505
Appendix 3: Myotome Tables for the Upper and Lower Extremities......Page 506
Appendix 4: The SNAP, CMAP, and Needle EMG Domains of the Brachial Plexus Elements^......Page 507
Reference......Page 509
The Median Sensory NCS, Recording Second Digit......Page 510
The Ulnar Sensory NCS, Recording Fifth Digit......Page 511
Superficial Radial Sensory NCS......Page 512
Median Motor NCS, Recording Abductor Pollicis Brevis (Thenar Eminence)......Page 513
Ulnar Motor NCS, Recording Abductor Digiti Minimi (Hypothenar Eminence)......Page 514
Radial Motor NCS, Recording Extensor Digitorum Communis......Page 515
Musculocutaneous Motor NCS, Recording Biceps......Page 516
Suprascapular Motor NCS, Recording Infraspinatus......Page 517
Superficial Peroneal Sensory NCS......Page 518
Peroneal Motor NCS, Recording Extensor Digitorum Brevis......Page 519
Tibial Motor NCS, Recording Abductor Hallucis......Page 520
Femoral Motor NCS, Recording Rectus Femoris......Page 521
Appendix 6: The Age-Related, Normal Control Values Used in Our EMG Laboratories......Page 523
Lower Extremity......Page 524
Needle EMG......Page 525
Myopathy......Page 526
Index......Page 527

Citation preview

Comprehensive Electromyography With Clinical Correlations and Case Studies

Comprehensive Electromyography With Clinical Correlations and Case Studies Mark A. Ferrante Professor, Department of Neurology Co-Director, Neurophysiology Fellowship Associate Director, Residency Training Program University of Tennessee Health Science Center Section Chief, Neurophysiology Memphis VA Medical Center, Memphis, TN, USA

University Printing House, Cambridge CB2 8BS, United Kingdom One Liberty Plaza, 20th Floor, New York, NY 10006, USA 477 Williamstown Road, Port Melbourne, VIC 3207, Australia 314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre, New Delhi – 110025, India 79 Anson Road, #06–04/06, Singapore 079906 Cambridge University Press is part of the University of Cambridge. It furthers the University’s mission by disseminating knowledge in the pursuit of education, learning and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/9781107562035 DOI: 10.1017/9781316417614 © Mark A. Ferrante 2018 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2018 Printed in the United Kingdom by Clays, St Ives plc A catalogue record for this publication is available from the British Library ISBN 978-1-107-56203-5 Paperback Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

.................................................................. Every effort has been made in preparing this book to provide accurate and up-to-date information which is in accord with accepted standards and practice at the time of publication. Although case histories are drawn from actual cases, every effort has been made to disguise the identities of the individuals involved. Nevertheless, the authors, editors and publishers can make no warranties that the information contained herein is totally free from error, not least because clinical standards are constantly changing through research and regulation. The authors, editors and publishers therefore disclaim all liability for direct or consequential damages resulting from the use of material contained in this book. Readers are strongly advised to pay careful attention to information provided by the manufacturer of any drugs or equipment that they plan to use.

I dedicate this textbook to the memory of my mentor, colleague, and good friend, Dr. Asa J. Wilbourn, whose work in the field has touched so many practitioners of electrodiagnostic medicine, whether they are aware of that contact or not.

Contents Preface ix Acknowledgments

xii

Section 1 Introductory Chapters 1

Basic Electricity, Electrical Concepts, and Circuits 3

2

Instrumentation

3

Anatomy and Physiology of Neurons

4

Anatomy and Physiology of the Neuromuscular Junction 68

5

Anatomy and Physiology of Muscle

33 49

73

Section 4 Other Pertinent EDX Information 16 Assessment and Initial Management of Peripheral Nerve Injuries 213 17 The Electrodiagnostic Manifestations of Disorders at Various Levels of the Neuraxis 235 18 Common Pitfalls and Their Resolution 19 Safety Issues in the EDX Laboratory

Section 2 Nerve Conduction Studies 6

Electrodes and Nerve Conduction Study Basics 85

7

Motor Nerve Conduction Studies

8

Sensory Nerve Conduction Studies

9

The NCS Manifestations of Various Pathologies 123

95 111

11 Late Responses and Blink Reflexes

141

12 Repetitive Nerve Stimulation Studies and Their Pathological Manifestations 152

Section 3 The Needle EMG Examination 13 The Needle EMG Examination

161

14 The Needle EMG Manifestations of Pathology 183 15 Single-Fiber EMG and Macro EMG

204

290

20 Nontechnical Issues, Pain Lessening Techniques, the Encounter, and the Report 302

Section 5 Case Studies in Electrodiagnostic Medicine Case 1 through Case 50

10 The Utility of NCS for Lesion Localization and Characterization 137

257

317

Section 6 Appendices Appendix 1 Plexus Anatomy

483

Appendix 2 Nerve Anatomy

484

Appendix 3 Myotome Tables for the Upper and Lower Extremities 492 Appendix 4 The SNAP, CMAP, and Needle EMG Domains of the Brachial Plexus Elements 493 Appendix 5 The Sensory and Motor NCS Techniques Used in Our EMG Laboratories 496

vii

Contents

Appendix 6 The Age-Related, Normal Control Values Used in Our EMG Laboratories 509 Appendix 7 Our Screening Sensory and Motor NCS and Needle EMG Muscles 510 Appendix 8 The Advantages and Disadvantages of the Individual EDX Studies 511

viii

Appendix 9 Needle EMG Findings with Lesions at Various Levels of the Neuraxis 512

Index

513

Preface

Although measurements of the muscle response were initially made in the late nineteenth century, motor nerve conduction studies (NCS) were not truly born until the 1940s. At that time, technology was quite limited. As technological advancements occurred over the next 30 years, the ingenuity and depth of thought of our electrodiagnostic (EDX) forefathers slowly advanced this field and addressed most of the technical issues (e.g., instrumentation, ideal electrode placement, the significance of the response parameters, etc.) surrounding EDX medicine. We have gone from analog signals (continuously varying signals), RC circuit filtering (filtering through the use of resistors and capacitors arranged in series), and oscilloscopes to digital signals (composed of a series of on–off pulses representing ones and zeroes), digital filtering, and monitors. As a result, modern-day EDX medicine is much easier to perform. However, as a result of this automaticity, many of the electrical, instrumental, and other technical issues surrounding EDX medicine are unfamiliar to modern-day EDX providers. An understanding of these areas remains mandatory for the proper elicitation and interpretation of the compound electrical potentials recorded during the EDX examination, as well as for the ability to troubleshoot the commonly occurring problem associated with the recording of these electrical signals. Thus, the technical side of EDX medicine must still be learned. Fortunately, the principles and concepts underlying EDX medicine are straightforward and readily mastered once a basic scientific foundation has been established. The electromyography textbook that had the most profound impact on my early years as an EDX provider was The Physiological and Technical Basis of Electromyography. It was the third EDX medicine textbook that I read as a resident. It contained a wealth of information, but, regrettably, my understanding was incomplete due to a lack of knowledge in certain aspects of electricity, electronics, physics,

and engineering, as well as unfamiliarity with some of the jargon. At the time, I would have benefited from a pre-EDX medicine textbook that reviewed the core knowledge necessary for a career in this field, including the basic and advanced principles and concepts of the basic sciences, electricity, electronics, and engineering, and that also defined the jargon and connected the dots. Such a textbook still does not exist. Although a number of excellent EDX textbooks preface their discussions with electrical principles and concepts, these discussions are often too superficial or, when of the proper depth, do not provide enough explanation. Consequently, this textbook provides a comprehensive review of the basic and advanced principles and concepts underlying EDX medicine at the beginning of the textbook, defines all jargon and terminology, and errs on the side of oversimplification. It begins with five introductory chapters that provide the core knowledge necessary for a complete understanding of the science pertinent to EDX medicine. All of the technical jargon and EDX terms used in this textbook are defined at their entry into the textbook, and for those interested, the mathematical calculations are solved in a step-by-step manner. The foundational material allows for a complete understanding of the basic and advanced principles and concepts necessary for the performance of quality EDX medicine. By the end of the textbook, the reader will have a full understanding of the measurements we make, how the measurements are generated, what the measurements mean, and how disease affects the measurements. Thus, this textbook addresses the physiological and technical basis of electromyography, as well as the interpretation of the study findings. Because we are providers of EDX medicine, and thus should possess the deepest possible understanding of EDX medicine, the opening two chapters on electricity and electronics are written at a level beyond that required for the performance of quality EDX medicine. The details

ix

Preface

included in these two chapters do not require memorization. They are provided to make the principles and concepts under discussion more understandable. To better understand how the material contained in this textbook is applied in the EMG laboratory, 50 case studies are included in the final section. These case studies not only reinforce the information contained in the textbook, they also introduce other material best taught using a case study format. Finally, a number of appendixes containing useful information for the day-to-day practice of EDX medicine are addended. Five sections provide the framework for 20 chapters. A review of the core disciplines underlying EDX medicine is provided in Section 1. The integration of this information with the nerve conduction studies (Section 2) and the needle EMG examination (Section 3) follows. Section 4 reviews other topics pertinent to the EDX medicine provider, including lesion characterization, lesion prognostication, common pitfalls and their resolution, safety issues, and details related to the patient encounter and EDX report writing. Following these 4 sections, 50 case studies are provided to reinforce the material presented throughout the text, demonstrate its application, and introduce additional teaching points better conveyed in a case study format. Rather than providing the data and then discussing it, the data is discussed as it is collected so that the reader obtains a better understanding of the orchestration of the various studies composition of the EDX encounter. Finally, although EDX studies are sensitive and reliable, the accurate localization and characterization of lesions involving the peripheral neuromuscular system require that the EDX procedures be performed properly. For this reason, the EDX techniques used in our EMG laboratories, along with their age-related normal control values, are provided in Appendix 5 and Appendix 6, respectively. A fuller understanding of this material, along with the material presented in the instrumentation and troubleshooting chapters and the appendix addressing EMG machine settings, will yield better EDX techniques that allow the EDX provider to collect more easily the various compound electrical potentials despite an environment of unfriendly electrical signal. Those topics with a greater impact on the practical aspects of EDX medicine, such as enhancing the signal-to-noise ratio and eliminating stimulus artifact, are given greater attention than the topics

x

with lesser clinical application. Chapter 1 functions as a stepping-stone to Chapter 2, which reviews instrumentation pertinent to an EMG laboratory. Once these electrical concepts are understood, membrane electrophysiology, including the resting membrane potential and the generation and propagation of action potentials, is straightforward. Section 2 of this textbook, which consists of 7 chapters, focuses on the NCS, including motor NCS, sensory NCS, mixed NCS, late responses, and repetitive nerve stimulation studies. How we collect these responses and the measurements we make from them are reviewed in detail. This is followed by a comprehensive discussion of the various pathologies and pathophysiologies affecting the peripheral nervous system (PNS) and the EDX manifestations associated with each. The final chapter of this section discusses our approach to lesion localization and characterization. Section 3, which consists of 3 chapters, discusses the needle EMG examination, the pathological manifestations observed during needle EMG, single-fiber EMG, and macro EMG. Section 4, composed of 5 chapters, reviews a number of topics important to the practitioner of EDX medicine, including a chapter on prognostication, a chapter overviewing the EDX manifestation of lesions at different levels of the neuromuscular axis, a chapter on EMG laboratory pitfalls and their resolution, a chapter on EMG laboratory safety, and a chapter discussing the patient encounter, tips to avoid patient discomfort, and report writing. Section 5 uses a case study approach to reinforce the information provided in the textbook and to introduce new material best illustrated through such an approach. Each EDX case is discussed in a step-bystep manner, with an interpretation of the study data as it is collected rather than all at once at the end. Each case study begins with the reason for referral, an abbreviated history and focused neurological examination, and a discussion of the NCS that should be employed initially. The focus of each case is lesion localization and lesion characterization, including temporal information regarding the chronicity of the lesion and its rate of progression. Each case study provides at least one unique teaching point and, when instructive, concludes with an impression written in the report format of the author. Although the discussions associated with some of these cases are partially repetitive, they should be considered akin to the scales

Preface

and arpeggios of musical practice. The basic concepts are introduced in the earlier cases and the more advanced ones in the subsequent ones. The final section of this textbook contains a number of appendixes. They are intended to provide immediately available data pertinent to the day-to-day practice of EDX medicine. These appendixes include: (1) anatomical information, such as the SNAP, CMAP, and muscle domains of the various nerve and plexus elements; (2) the routine and nonroutine NCS techniques used by the author, and their agedependent normal values; and (3) proper EMG machine settings. Other important information is also included here.

In some sections of this textbook, the information conveyed is necessarily redundant. This maintains the coherence of the current material without forcing the reader to return to a previous section of the text. In these instances, however, the material is abridged. The chapters are organized so that simple concepts beget more complicated concepts, thereby permitting selflearning to proceed as effortlessly as possible. Through the 20 didactic chapters and the reinforcing case studies, the reader will gain a mastery over the basic and more complicated aspects of EDX medicine and their clinical application. For explanatory purposes, the textbook includes a large number of illustrations.

xi

Acknowledgments

There are many individuals that I would like to thank for their contributions throughout my career. First and foremost, I would like to thank my children, Nicole, Kristen, and John, and my wife, Jung, for their sacrifices and patience and for their understanding and acceptance of the time commitment required of a career in academic medicine. In addition, I am thankful to Nicole and Kristen for creating some of the illustrations contained within this textbook. I also wish to express my thanks to my many mentors through the years. Among these individuals, the most influential person in my training and my career was Dr. Asa J. Wilbourn. He provided me with numerous academic opportunities and shared his early brachial plexus research with me, thereby allowing it to become our research. Ultimately, we advanced from a mentor–mentee relationship to a special friendship, as exemplified by his insistence that I allow him and his wife, Eileen, to care for me during my battle against cancer. Asa and I often spoke of authoring an EMG textbook together, but his untimely death eliminated that opportunity. Although this textbook would have been better with his involvement, I am hopeful that he would have

xii

been pleased with it nonetheless. I also wish to give special thanks to Dr. Randall B. King, my good friend, who introduced me to the EMG laboratory during my residency and, through his engineering background, has furthered my understanding of electricity and electronics through the years. I especially wish to pay homage to my close friend, Edward J. Milcarsky, MD, for proofing the first two chapters of this textbook for errors. His engineering background and his electronics laboratory provided valuable insights for both of these chapters. In addition, I thank Dr. Olivia Yambem for reviewing the case studies and Dr. Grace Madeiros for reviewing Chapters 1–20 for errors and for those areas needing further explanation. I would like to extend a special thanks to my EDX technicians, Billy Seay and Teresa James, for all of their long hours and extra effort, and to all of my predecessors who created and contributed to this field. Finally, I am especially thankful to all of the greater than 35,000 patients who have permitted me to perform EDX studies on them and for their willingness to allow small numbers of residents and fellows to be present in the EMG laboratory during their EDX testing.

Section

1

Introductory Chapters

Chapter

1

Basic Electricity, Electrical Concepts, and Circuits

Introduction Because electricity is at the core of electrodiagnostic (EDX) medicine, and because a strong understanding of this material generates higher-quality and more insightful EDX medicine studies and reports, it behooves practitioners of EDX medicine to possess a strong understanding of a number of electrical concepts (e.g., charge, current, voltage, resistance, and capacitance) pertinent to EDX medicine. This knowledge is mandatory for the proper performance of EDX studies and for a complete understanding of: (1) the instrumentation used in EDX medicine; (2) the generation and propagation of action potentials (APs); (3) the role of myelin to decrease capacitance and increase transmembrane resistance; (4) the EDX manifestations of demyelination; (5) the EDX effects of axon disruption and Wallerian degeneration; (6) the frequently encountered problems in the EMG laboratory (e.g., shock artifact) and how to overcome them; and (7) the proper interpretation of the EDX examination findings, as well as other pertinent issues (see Box 1.1). Box 1.1 Important electrical concepts and principles – Electrical circuits (Chapter 1) – Series circuits (resistors function as voltage dividers) – The importance of proper application of surface recording electrodes – The relationship between the recording electrodes and the amplifier – Parallel circuits (resistors function as current dividers) – Ion channels in membranes – Instrumentation (Chapter 2) – Filters

– Low-frequency, high-frequency, notch, and bandpass – Stimulators – Constant current versus constant voltage – Amplifiers – Monopolar versus differential – The importance of proper surface recording electrode position – Electrodes – Stimulating and recording – Membrane physiology (see Chapter 3) – Resistive and capacitive properties of nerve and muscle membranes – The length constant – The time constant – Resting membrane potential (voltage) of nerve and muscle membranes – How action potentials are generated and regenerated (capacitive current) – How action potentials are propagated (resistive current) – Important NCS principles and concepts (Chapters 6–8) – Motor NCS – The belly-tendon method for motor NCS – Sensory NCS – Orthodromic versus antidromic recording – The ideal interelectrode distance between the surface recording electrodes – Peak latency versus onset latency – Concentric versus monopolar needle recording electrodes (Chapter 13) – EDX pitfalls and troubleshooting (Chapter 18) – Electrical safety (Chapter 19)

3

Section 1: Introductory Chapters

This chapter includes a number of formulas. It is not necessary to memorize them; it is sufficient to simply understand the relationships between the variables composing them. For board examination preparation, however, it is important to memorize: (1) Ohm’s law (V = IR); (2) the formula for calculating the total resistance in a DC circuit in which the resistors are arranged in series with each other and with the voltage source; (3) the formula for calculating the voltage drop across a resistor in series with other resistors; and (4) the power formula (P = VI). These formulas and others, the variables composing them, and the mathematics required to utilize them are detailed in this chapter and the subsequent chapter. This chapter focuses on those electrical principles and concepts pertinent to EDX medicine, including charge, electrostatic force, current, resistance, and voltage. The mathematics provides insight and contributes to a better understanding of EDX instrumentation (discussed in Chapter 2). This chapter was authored for the reader with no significant understanding of electricity. Thus, all of the terms are defined when introduced, the electrical concepts and their relationships are explained in detail, and pertinent everyday examples are provided. Necessary redundancies serve to reiterate important concepts, where needed, and to contribute to the flow of the material being presented, thereby eliminating the need to review a previous section in order to comprehend the information contained in the current section. Although modern EMG machines utilize digital electronics, this chapter and the subsequent one reviews analog and digital electronics.

Electricity History Some of the properties of electricity were known to the ancient peoples of the world. For example, as indicated in ancient Egyptian texts dating back as far as the third millennium BC, people were aware of the propagating nature of the shocks received by electricity-generating electric fish (Moller and Kramer, 1991). In the second century, Galen applied a living torpedo fish to the head of a person suffering from a headache (Bonner and DevlescHoward, 1995). In addition, less ancient authors recognized that these shocks propagated along those objects capable of conducting (Bullock, 2005). Regarding static electricity,

4

the ancient Mediterranean people were aware that when amber (fossilized plant resin) was rubbed with wool or cat fur, it would subsequently attract lightweight objects, such as small pieces of dust or lint. When the amber is rubbed with fur, the fur transfers electrons to the amber, causing both of these neutral objects to become charged (the fur become positively charged and the amber becomes negatively charged). The lightweight objects attracted to the amber are actually neutral. The negatively charged amber causes them to become polarized (the positive charges in the lightweight particles move closer to the rod, whereas the negative charges move away from the rod). The term electricity, which was coined by William Gilbert, a seventeenth-century physician and scientist, is derived from the Greek word for amber, elektron. In 1745, the Leyden jar was developed (Bonner and DevlescHoward, 1995). This device could store electricity and, thus, was the precursor to the capacitor. Because it could store electricity, it made it easier to study this phenomenon. Benjamin Franklin invented the lightning rod in 1752, which allowed him to pass the electrical discharges of thunderstorms into a Leyden jar; introduced the concept of positive and negative into electricity; and invented the electrical battery (Bonner and DevlescHoward, 1995). Electricity, the movement of charged particles through a conductor, is characterized by its charge, current (rate of charge), voltage, and resistance. Consequently, these aspects of electricity are reviewed first, followed by other important concepts and principles pertinent to EDX medicine.

Charge To understand the principle of charge, we need to understand atomic structure. The universe is composed of two types of matter: dark matter (the majority) and visible matter. Dark matter, as the name implies, is invisible, and its presence is inferred from its effects on visible matter. Visible matter is composed of elements (e.g., copper). Elements, in turn, are composed of atoms. The term atom is derived from the Greek word, atomos, which indicates indivisible, although it subsequently became apparent that atoms consist of subatomic particles – protons, neutrons, and electrons – and, hence, are divisible. The protons and neutrons are centrally located and constitute the nucleus of the atom. Hence, they are also referred to as nucleons. These nucleons are held

Chapter 1: Basic Electricity, Electrical Concepts, and Circuits

together by the strong force (nuclear force) and, except in nuclear reactions, never separate. Protons and neutrons are composed of smaller particles, termed quarks; thus, protons and neutrons are not elementary particles. Electrons, like quarks, are elementary particles (i.e., they are not composed of smaller particles and, thus, are indivisible). According to current quantum electrodynamics, electrons are not strictly particles, but rather are both particles and waves. They are located at specific distances from the nucleus (referred to as shells or energy levels). Neutrons and protons have similar masses, whereas electrons have much smaller masses (the mass of an electron is 1/1836th of the mass of a proton or of a neutron). For this reason, essentially all of the mass of an atom is located within its nucleus. In addition to their mass differences, these subatomic particles also differ in their charge – protons are positively charged, electrons are negatively charged, and neutrons carry no electrical charge (i.e., they are neutral). The charge differences of protons and neutrons reflect their elementary particle composition. Thus, protons have a +1 charge because they are composed of 2 up-quarks and 1 down-quark. Because up-quarks carry a +2/3 charge and down-quarks carry a –1/3 charge, protons have a net +1 charge (2/3 + 2/3 + –1/3 = 4/3 – 1/3 = 1). Likewise, neutrons are neutral because they are composed of 2 down-quarks and 1 up-quark, which generates a zero charge (–1/3 + –1/3 + 2/3 = –2/3 + 2/3 = 0). The magnitude of the positive charge carried by a proton (+1) is equal to the magnitude of the negative charge carried by an electron (–1). The unit of electrical charge is the coulomb (C), where 1 C is equivalent to the charge of 6.24
1018 electrons. The quantity of charge possessed by a single electron or a single proton is extremely small and is referred to as an elementary charge unit (ECU); it has a charge value of 1.6022
10–19 C. For an electron, this value is negative (i.e., –1.6022
10–19 C), whereas for a proton, it is positive (+1.6022
10–19 C). Although the sum of positive and negative charges contained within a closed system cannot change (the law of conservation of charge), the charges are able to move throughout the system. Because atoms contain an equal number of protons and electrons, they have a neutral charge. The atomic number of an atom indicates the number of protons it contains (and the number of electrons). The mass number of an atom (i.e., its atomic mass) is equal to its weight, which essentially reflects the

Figure 1.1 The orbital depiction of an atom of carbon and its periodic table symbol, indicating its abbreviation (C), its mass number (12), and its atomic number (6).

6P 6N

12C 6

number of its protons plus its neutrons. The number of protons dictates the element. For example, if an element has an atomic number of 6, it is carbon. Thus, carbon contains 6 protons. Based on the preceding discussion, it also has 6 electrons (atoms are neutral). Its atomic mass is 12.001; the 12 indicates that it also contains 6 neutrons (because the number to the left of the decimal point indicates the number of protons and neutrons), whereas the digits beyond the decimal point represent the mass of its electrons. In periodic charts, the atomic mass of carbon is often rounded down to 12 (see Figure 1.1). Some atoms have more than one mass number, because the number of neutrons they contain can vary. These atomic variations are referred to as isotopes. For example, in addition to carbon-12 (6 protons and 6 neutrons), carbon-13 (6 protons and 7 neutrons) and carbon-14 (6 protons and 8 neutrons) also exist. Thus, carbon isotopes (e.g., carbon-12, carbon-13, and carbon-14) all contain 6 protons, which is what defines the atom as carbon. Atomic isotopes vary in the number of neutrons they contain. As stated earlier in the chapter, the electrons are located at various distances (energy levels) from its center, referred to as the 1 shell, 2 shell, 3 shell, 4 shell, 5 shell, 6 shell, and 7 shell. Shells contain subshells, each of which contains up to 2 electrons. The 2-electron per subshell limit reflects the Pauli exclusion principle, which states that electrons with the same spin cannot occupy the same subshell. Thus, because electrons can only spin in one of two directions, clockwise or counterclockwise, the maximum number of electrons per subshell is two. So as not to get shells and subshells confused, from this point forward, the shells will be referred to as energy levels. The number of subshells in each energy level varies. Energy level 1 contains only one subshell, termed the s subshell, which, by definition, can hold up to two electrons. The second energy level is composed of 4 subshells

5

Section 1: Introductory Chapters

(1 s subshell and 3 p subshells) and, thus, can hold up to 8 electrons (4
2 = 8). The third energy level is composed of 9 subshells (1 s subshell, 3 p subshells, and 5 d subshells) and, therefore, can hold up to 18 electrons (9
2 = 18). As indicated by the periodic chart, for atoms with an atomic number of 18 or less (i.e., 18 or less electrons), the electrons are distributed within the first 3 energy levels. Except for a few atoms (e.g., copper), atoms containing shells with higher energy levels are not pertinent to this discussion. The atomic number for copper is 29, and its 29 electrons are distributed among 4 energy levels as follows: 2 – 8 – 18 – 1. Because the negatively charged electrons are attracted to the positively charged nucleus, the innermost shell fills before the shell immediately surrounding it begins to fill. For example, copper, atomic number 29, has 29 electrons and, therefore, as stated above, an electron shell pattern of (2 – 8 – 18 –1). Other important electron shell patterns are sodium (atomic number 11 [2 – 8 – 1]), chlorine (atomic number 17 [2 – 8 – 7]), and potassium (atomic number 19 [2 – 8 – 8 – 1]). These electron shell patterns inform us that copper, sodium, and potassium have a single electron in their outer orbits, whereas chlorine has seven electrons in its outer orbit. The interactions of atoms to form molecules can be predicted by the number of electrons in their outer shell. As demonstrated through quantum mechanics, the electrons do not orbit the nucleus like planets orbit a star. Rather, their location is discontinuous (i.e., first they are at one point in their orbit and then they are at another, noncontiguous point). For this reason, the distribution of electrons around the nucleus is often referred to as an electron cloud. This is analogous to a football (the nucleus) located at the center of the 50-yard line within a stadium consisting of a number of rows of seats (energy levels). In this analogy, the camera flashes occurring during a great play represent the electrons coming in and out of existence at random, discontinuous points within their energy level rather than moving continuously along an orbital path within their energy level. When an atom gains or loses electrons, it becomes a charged particle (referred to as an ion). With electron loss (i.e., the loss of negativity), the atom becomes positively charged, whereas with electron gain (i.e., the acquisition of negativity), it becomes negatively charged. Positively charged ions are referred to as cations because they are attracted to

6

the battery cathode (the negative terminal of a battery), whereas negatively charged ions are referred to as anions because they are attracted to the battery anode (the positively charged terminal of a battery). This occurs because of a property of electricity that states that oppositely charged particles attract. Antimatter also bears charge, with each antiparticle having an equal and opposite charge to its corresponding matter particle (e.g., an anti-electron has a charge of +1.6022
1019 coulombs, similar to a proton). Whenever an object (living or nonliving) has an excess or a deficiency of electrons, that object has a charge. Because the object is not moving, the charge is said to be static and is also referred to as an electrostatic charge. Electrostatic charge accumulation usually occurs when two objects composed of electrically dissimilar materials are rubbed together so that one object transfers charge (e.g., electrons) to the other. The accumulated electrostatic charge is either positive (electron loss) or negative (electron gain). Objects with an electrostatic charge may attract each other (attractive force) or repel each other (repulsive force), depending on the similarity or dissimilarity of their charge. An electrostatic attractive force is observed when two objects of opposite charge (one negative and one positive) are brought together. Conversely, an electrostatic repulsive force occurs when two objects of like charge are brought together, such as two negatively charged objects or two positively charged ones. This phenomenon was known to ancient peoples (the frictive contact of amber and cat fur), although the mechanism underlying electrostatic force was not yet understood. In modern times, many of us have rubbed a glass rod with silk (this causes electrons to move from the glass rod to the silk) or caused a balloon to stick to the wall after rubbing it on a shirt. The rubbing process permits electrons to move from one object to the other, thereby charging the two objects (the object gaining electrons becomes negatively charged and the object losing electrons becomes positively charged). This is referred to as the triboelectric effect (the prefix triboderives from the Greek word for rub) and represents a type of contact electrification. Because the objects accumulate opposite charges, the two objects demonstrate an attractive electrostatic force. However, if two glass rods are rubbed with silk and then brought together, they would repel each other (repulsive electrostatic force) because they would then have the same positive charge. (The frictive force strips the

Chapter 1: Basic Electricity, Electrical Concepts, and Circuits

electrons from the glass rod, causing it to have a positive charge. The silk acquires these electrons and, thus, becomes negatively charged.) These attractive and repulsive electrostatic forces are proportional to the product of the charges (q) on the two objects and inversely proportional to the square of the distance (d) between their centers. This mathematical relationship, which resembles the attractive force of gravity (Newton’s inverse-square law of universal gravitation), is referred to as Coulomb’s inversesquare law, after Charles Augustin de Coulomb, who realized that electrostatic charges manifest in two opposing forms, attractive (objects with opposite charge) and repulsive (objects with like charge). These relationships are expressed by the following formula: F ¼ ke qlq2=d2 where F is the magnitude of the force of interaction between the two charged objects (in Newtons), q1 and q2 are the magnitudes of the charges of the two objects (in coulombs), d is the distance between the centers of the two charged objects (in meters), and ke is Coulomb’s constant (the dielectric constant), which has a value of 9.0
109 Nm2/C2 (newton-meters squared per coulombs squared). This formula can be used to calculate the magnitude of an electrostatic force, where positive values indicate repulsive forces and negative values indicate attractive forces. Consequently, and rather amazingly, two separate charges located at some distance from each other have an effect on each other. This phenomenon is referred to as action at a distance. According to the standard model of particle physics, the electrostatic force between two charged objects is expressed through subatomic particles termed photons. Photons move at the speed of light, which necessitates that they be massless (for the interested reader, gauge symmetry verifies that photons are massless). Because they are massless, they can never move at any speed other than the speed of light. According to quantum theory, photons, like other subatomic particles, are both particles and waves. When a charge moves, it creates ripples within its field (similar to the creation of ripples moving across the top of water by the jiggling of a bobber on the surface of the water). In this analogy, the bobber represents the moving charge and the ripples represent the massless photons. Because photons are massless, it is easy to generate electricity and electrical

forces. Indeed, all of the forces in the universe (e.g., gravitation) are manifested through such carriers, although not all carriers are massless. Electrostatic force is extremely strong. For example, the electrostatic force pushing two electrons apart is 1042 times stronger than the gravitational force pulling them together (Hawking, 1988). As another example, the electrical force between a proton and an electron is 1036 times greater than the force of gravity (Pollock, 2012). Although electrostatic force is much stronger than gravitational force, the force of gravity dominates throughout the universe because planets generally have no net charge and, therefore, no electrostatic charge. Electron surpluses and deficiencies underlie the mechanism of static electricity. As previously stated, when two electrically neutral materials are rubbed together, the transfer of electrons from one material to the other one causes the material receiving the electrons to become negatively charged and the material losing the electrons to become positively charged. This relationship also accounts for several common electrical phenomena, such as the electrostatic discharge (“shock”) that occurs when an individual with a net charge touches an object capable of conducting that charge or, equally as familiar, the movement of charge through the air (lightning) that occurs when the charge difference between a thundercloud and the ground surpasses the insulating ability of the air. In this setting, an electrical potential difference forms within the cloud, with its upper layers becoming positively charged and its lower layers becoming negatively charged. This phenomenon occurs when ice particles and water droplets rub against each other. The buildup of negative charges on the undersurface of the cloud causes the positive charges in the ground to become more superficial and the negative charges in the ground to move deeper into the ground (induction). Lightning occurs between the negatively charged undersurface of the cloud and the positively induced ground. Regarding the flow of electricity between individuals and conducting objects, when the surrounding air is dry (atmospheric humidity conducts electricity and tends to dissipate accumulating charge), insulators can hold a static charge for several minutes. Consequently, in a dry room, an individual wearing hard-soled shoes and walking across a carpeted floor can accumulate a charge difference of several thousand volts (the concept of voltage is discussed later in this chapter).

7

Section 1: Introductory Chapters

In this scenario, the friction between the carpet and the soles of the shoes generated by walking constitutes the “rubbing” or frictive force. The frictive force permits the exchange of electrons and, therefore, the accumulated charge (electrons accumulate on the body). This charge remains (is stored) on the surface of the body of the person until that person contacts a conductor (including another individual) or until the accumulated charge dissipates into the atmosphere. For example, should your hand approach a metal doorknob, the negative charges would flow from your hand to the doorknob (the electrons are attracted to the positive charges in the doorknob (created by polarization). Regarding lightning, there is a constant charge difference (voltage) between the ionosphere and the surface of the earth. The ionosphere is the ionized (by solar radiation) region of the Earth’s upper atmosphere that forms the inner edge of the magnetosphere. The lower atmosphere, which lies between the ionosphere and the surface of the Earth, acts an insulator. Accordingly, when considered together, these three layers (conducting–insulating– conducting) constitute an atmospheric capacitor. A capacitor is an electrical element capable of storing electrical charge. This device is discussed in detail later in the chapter. When the insulating layer (the lower atmosphere) breaks down, as it does when the charge difference across it becomes extremely large (the charge difference can reach millions of volts in magnitude), the charged particles pass through the air (lightning). Thus, lightning simply reflects the advancement of electrons through the air, away from the negatively charged object and toward the positively charged one.

Current Electric current (I) is defined as the movement of electric charge with respect to time (I = Q/t) across a given area. Thus, current expresses the rate of the flow of charge (charge per unit time). Current (charge) sources include free electrons (metals), ions (ionic solutions, such as the human body), combined electrons and ions (plasma), combined electrons and “holes” (semiconductors), and paired electrons (superconductors, which have near-zero resistance). Holes can be conceptualized as sites of relative positivity created by the displacement of an electron from the material. Thus, as the electrons flow in one direction, the holes flow in the opposite direction.

8

Current was originally defined as the flow of positive charges (typically referred to as conventional current and less commonly as hole current or hole flow) but was later realized to represent the flow of negative charges (true current). In some engineering projects, it is easier to think in terms of hole flow than it is to think in terms of electron flow. Electron flow is the focus of this chapter. In subsequent chapters, ion flow (tissue current), which is important to EDX medicine, will be introduced. The principles and concepts of electron flow discussed in this chapter are analogous to those of ion flow. The flow of charge occurs whenever a conductor is attached to two objects with unequal charges (e.g., electron imbalance). When in motion, charged particles (e.g., electrons) create an electric current. Like with charge, to best understand current, an understanding of atomic structure is again required. As previously discussed, the electrons composing the electron cloud surrounding the nucleus of an atom are located at various distances from the nucleus, termed energy levels. The outer energy level of an atom may be entirely filled with electrons or incompletely filled. Atoms that have a completely filled outer energy level neither accept nor donate electrons (i.e., they oppose electron movement) and, accordingly, are termed inert atoms. This type of atom is stable. Conversely, atoms with incompletely filled outer shells are less stable and, therefore, to enhance their stability, either gain electrons (to complete their outer energy level) or lose electrons (to eliminate their outer energy level). In exchange for stability, these atoms lose their neutrality and, thus, become ions. The conductivity of a material reflects the degree to which the charged particles composing it are free to move (free charge). The dielectric constant of a material indicates its propensity to conduct. When substances consist of atoms in which the outer electrons move among the atomic nuclei with relative ease (i.e., atoms that donate or accept electrons), they are referred to as conductors. Conversely, substances composed of atoms with electrons that move among the nuclei with difficulty or not at all (i.e., that are tightly bound) make poor conductors. Because inert substances (substances composed of atoms containing a completely full outer energy shell) do not exhibit electron movement, they make good insulators. When substances are composed of atoms with an outer shell that is half-filled (e.g., silicon and germanium), they are referred to as semiconductors

Chapter 1: Basic Electricity, Electrical Concepts, and Circuits

because they have both conducting and insulating properties. They conduct, but not as well as conductors, and they insulate, but not as well as insulators. In summary, the degree of freedom of the charges composing a material is what distinguishes conductors, semiconductors, and insulators. As discovered by Galvani in the late 18th century, current also flows between dissimilar metals connected through a conductor. Galvani utilized dissimilar metals (i.e., a brass hook attached to an iron railing) to generate the electrical current required in his frog muscle contraction demonstrations (Bonner and DevlescHoward, 1995). For this reason, the production of electricity, especially when it is generated through a chemical action, is termed galvanism. Metals, which contain large numbers of free electrons (free charge), are commonly employed as conductors. Copper, which has one electron is its outer shell (2 – 8 – 18 – 1), has a strong tendency to lose that electron. Consequently, the electrons contained within a copper wire are able to move among the copper nuclei composing that wire. As a result, when a copper wire is attached to a material containing excess electrons, the excess electrons move into the copper wire. However, the entry of electrons into the wire is limited because there is nowhere for the entering electrons to go. However, when a copper wire is connected between two objects, one of which contains excess electrons and the other of which contains a deficiency of electrons, the electrons flow from the electron-rich object to the electron-poor object via the copper atoms composing the wire. As the negatively charged electrons located within the outer energy levels of the copper atoms move toward the electron-poor object, the copper atoms losing their outer electrons become positively charged. This positive charge attracts a nearby electron, which then enters its outer shell, thereby restoring the atom to its neutrally charged state. At this point, the process repeats itself. Thus, the flow of electrons continues from the electron-rich object to the electron-poor object as long as a driving force remains (an electron imbalance between the two objects). Silver and aluminum are two other excellent conductors. The advantage of aluminum over copper is its light weight, which makes it the material of choice for the conducting wire attached to utility poles. Conductors do not have to be solid objects. Regarding liquids, elemental mercury is a good conductor, saltwater is a fair conductor, and distilled water is a poor

conductor. In general, gases are poor conductors because the electrons contained in the atoms or molecules composing the gas are extremely far apart. For this reason, most gases make good insulators. Other good insulators include plastic, paper, wood, rubber-like polymers, porcelain, and glass. When electrically charged particles move through a conductor, there is an electrical current (current equals the flow of charged particles). Examples include electrolysis (the movement of ions through liquid) and metallic conduction (the movement of electrons through a conductor). Although electric current consists of the flow of electrons from negative to positive (electron current), as previously stated, early scientists attributed electricity to the flow of positively charged particles, which is now referred to as conventional current (or, rarely, as theoretical current) and, consequently, is depicted as traveling from the most positive portion of a circuit to its most negative portion (i.e., circuit diagrams with arrows pointing in the positive to negative direction). As stated earlier, this is also referred to as “hole flow,” because the exiting electrons can be envisioned to leave positively charged “holes” in the atom from which they depart. In other words, as the electrons move from negative to positive, the holes are moving in the opposite direction. Although the electrons constituting the current move quite slowly (termed the drift velocity, which is measured in millimeters per hour and is proportional to the strength of the electric field), the speed of the electromagnetic waves traveling along the wire is much faster and depends on the dielectric constant of the material through which it is traveling. In a vacuum, the electromagnetic waves travel at the speed of light. Massless particles, termed photons, function as the energy carrier for electromagnetic waves (Pollock, 2003). As stated at the beginning of this chapter, electricity can be characterized in terms of its charge (Q), current (Q/t), voltage (V), and resistance (R). The unit of charge is the coulomb, symbolized as an uppercase C. It is defined as being equivalent to the charge on 6.24
1018 electrons. Current is defined as the flow of charge (electrons, ions, or any charged object) per unit of time (Q/t). The standard unit of electrical current is the ampere, which is defined as 1 coulomb per second and is symbolized by an uppercase A. Thus, 1 ampere of current is equivalent to 1 coulomb of charge passing a given point in 1 second. Current is typically expressed in

9

Section 1: Introductory Chapters

milliamperes (mA, where 1 milliampere equals one one-thousandth of an ampere). Examples of approximate current magnitudes include: power lines (1,000 A), household electricity (150 A), a 100-watt light bulb (1 A), cell phones (0.2 A or 200 mA for the iPhone 6), watches (1 microampere), and cell membranes (1 picoampere per receptor; a picoampere is one-trillionth of an ampere). The important relationships for current are: Current ¼ charge=time ðI ¼ Q=tÞ Ampere ¼ 1 coulomb=second

Voltage As previously stated, electron current passes through a copper wire when the wire is connected between one material with an electron surplus and another material with an electron deficit (i.e., there is an electron difference, or gradient, between the two materials). Thus, when a conductor is interposed between the two objects, the electrons flow down their concentration gradient (i.e., from the material with the greater number of electrons to the one with the lesser number). These two objects, which are separated in space, also have different charges (i.e., a charge gradient). The size of the charge difference between the two points (i.e., the electron gradient) represents the driving force that pushes the electrons through the copper wire. This driving force is referred to as an electromotive force (EMF) and its magnitude is expressed in volts (symbolized by an uppercase V). The EMF is inversely proportional to the distance between the two charges. One volt is the EMF (attractive or repellant) between 1 coulomb of charge located 1 meter from 1 coulomb of unlike [attractive] or like [repellant] charge). It requires 1 joule of energy to separate 1 coulomb of positive charge from 1 coulomb of negative charge by 1 meter (1 volt = 1 joule/1 coulomb). Thus, voltage quantifies the charge imbalance between two distinct sites and is a reflection of the energy per charge; its units are volts (joules per coulomb). For example, a 1.5-volt battery imparts 1.5 joules of energy to each coulomb of electrons exiting the terminal of the battery (e.g., a flashlight battery). In addition, as the electrons travel through the circuitry of the flashlight, some of the energy is lost (e.g., some of the electrical energy is transformed into heat energy) and some of the energy is utilized for the purpose of the circuit (e.g., to power the lightbulb).

10

In EDX medicine, voltage is commonly expressed in much smaller units – microvolts (μV) and millivolts (mV). In addition, because the units of EMF are termed volts, EMF is often simply referred to as voltage. Finally, because the EMF represents a charge difference, it is also referred to as a potential difference because the two separated charges have the potential to flow should the opportunity arise (i.e., should a conductor be placed between them). Consequently, voltage is an electrical form of potential energy. Because voltage is defined as the difference in energy per charge between two separate sites within a circuit, it can never exist at a single point. When voltage is reported (or measured), it is only meaningful when the two separate sites are specified. Often, for convenience, a reference point is assigned within a circuit and its voltage is declared to be zero (the circuit ground). Then we can report voltages at other points in the circuit as a number, and it is implied that the measurement is made with respect to the zero point or circuit ground. When current (charge/time) is multiplied by voltage (energy/charge), the product is power (energy/time), which is expressed in watts (W). Thus, electric power is the rate at which a system delivers, generates, or consumes electrical energy (P = IV, expressed in watts). Charge differences (EMF) between two points not connected by a conductor underlie electrostatic electricity (partially discussed above). Because there is no conductor, there is no flow, and consequently, the charges do not move (they are electrostatic). Again, they have the potential to move should a conductor be placed between them. These potential differences can be created at one site and then transported elsewhere to do work. Batteries, which convert chemical energy into electrical energy to maintain an essentially fixed voltage across its terminals, serve as transportable voltage sources. Devices that convert nonelectrical energy into electrical energy, such as batteries (convert chemical energy into electrical energy), are sometimes referred to as EMF sources. Batteries were originally referred to as galvanic cells, after Luigi Galvani. When a conductor (e.g., a conducting wire) is placed between the terminals of a battery, electrons flow from the negatively charged cathode to the positively charged anode until the electron imbalance no longer exists. The electron flow can be directed through a load (e.g., a lamp) to do work (e.g., illuminate a room). The voltage drop across the bulb reflects the

Chapter 1: Basic Electricity, Electrical Concepts, and Circuits

amount of electrical energy that is being converted into light energy. Similarly, the voltage drop across an electric iron represents the amount of electrical energy that is being converted into thermal energy. Other examples of voltage sources include electric generators and power supplies. The power available through the wall outlet is an example of a timevarying voltage source and is discussed later in this chapter (see alternating current). Although there are many different types of batteries, all batteries have two terminals and a voltage rating. For example, a lead-acid automotive battery consists of a group of electrochemical cells connected in series and located within a single container. Each electrochemical cell consists of two metal plates (one composed of lead and the other composed of leaddioxide) submerged in a solution, termed the electrolyte. In an automotive battery, the lead plate is the negative electrode, the lead-dioxide plate is the positive electrode, and sulfuric acid is the electrolyte solution. The solution may be thickened into a paste to lessen the likelihood of leakage. In this scenario, the electrolyte chemically reacts with the two submerged metal plates. The interaction causes charge accumulation on the metal plates, each of which protrudes from the cell. Because the charge quantities accumulating on the two plates differ, a charge difference (i.e., a potential difference or voltage) accumulates between them, with one of them becoming more negative (the lead plate) and the other becoming more positive (the lead-dioxide plate). These cells are attached in series (end-to-end) and connected to the two terminals of the battery. The more negative battery terminal (i.e., the one with an abundance of electrons) is called the cathode and the more positive one (i.e., the one with the relative deficiency of electrons) is termed the anode. In an automotive battery, the potential difference between the cathode and anode is 12 volts and is what is used to start the car. Although batteries are manufactured in many different forms, the principles underlying their performance are similar. With zinc-carbon batteries, the zinc forms the outer case and constitutes the negative terminal, a carbon rod constitutes the positive terminal, and the electrolyte is a paste composed of manganese dioxide and carbon. With alkaline batteries, the negative electrode is composed of granular zinc, a different substance (termed a polarizer) functions as the positive electrode, and the electrolyte is potassium hydroxide. Transistor batteries, such as the

box-shaped ones used in smoke detectors, are composed of six cells (zinc-carbon or alkaline) connected in series, each of which is 1.5 volts in magnitude, making the transistor battery 9 volts in magnitude (6
1.5 V = 9 V). Thus, batteries represent a form of electrical potential energy that can be used to drive electrons through a load. For example, when a battery or series of batteries are placed into the battery chamber of a battery-operated device, such as a flashlight, the cathode and anode of the battery (or of the series of batteries) contact the metal spring and the metal plate of the battery chamber, both of which are connected to the load (the flashlight bulb) via conducting wire. In this manner, the battery drives the electrons through the bulb, thereby generating light. Eventually, the chemical constituents comprising the chemical reaction of the battery are used up (i.e., the chemical energy is completely converted to electrical energy), and the potential difference between the two terminals drops to zero, at which point, the battery must be replaced or recharged. The batteries sold for usage in convenience items (e.g., flashlights, remote controls) are also termed electrochemical cells. They come in different sizes, including very small (AAA), small (AA), medium (C), and large (D). Despite their size differences, they are all 1.5 V in magnitude (i.e., they all deliver the same amount of energy per unit charge). The larger batteries deliver more energy to the battery-operated device and, consequently, are more expensive. Batteries are not perfect voltage sources. They possess an internal resistance that most strongly depends on the battery chemistry. Common alkaline cells have approximately 0.3 ohms of internal resistance, nickel-metal rechargeable cells have about 0.03 ohms of internal resistance, and lithium/iron disulfide cells have about 0.12 ohms of internal resistance. A lead acid battery has a very low internal resistance on the order of 0.003 ohms. The internal resistance of a battery is also dependent on its age, state of charge/ discharge, and temperature.

Resistance Because perfect conductors do not exist, there will always be opposition to the advancement of charge through matter. This opposition is referred to as resistance, which is symbolized by an uppercase R and is measured in ohms. Ohms are symbolized by the uppercase Greek letter omega (Ω). Resistance is

11

Section 1: Introductory Chapters

commonly expressed in kilohms (kΩ) or megohms (MΩ). The resistance of a conductor is a function of the material from which it is made (its resistivity) and its geometry (size and shape). For a cylindrically shaped wire or axon, it is directly proportional to its length and inversely proportional to its crosssectional area. (This relationship will become important when we discuss the passage of sodium current through an axon.) Because resistance (R) is the reciprocal of conductance (G), materials with high conductance, such as copper wires, have low resistance, whereas materials with low conductance, such as rubber, have high resistance. Regarding shape, wires composed of material with a long and thin shape have higher resistance than those composed of the same material with a short and thick shape. Again, voltage is the EMF that drives charge through matter. Resistance causes some of the electrical energy to be converted into thermal energy (heat). Consequently, resistance is best kept as low as possible unless heat generation is the goal (e.g., the heating element attached to the inside of the flat metal plate of an iron or to the metal coil [heating element] of a stovetop). Resistors are electrical circuit components that resist the flow of current and, thus, are used to reduce the flow of current to control the voltage level. Nowadays, resistors are microscopic in size and typically part of an integrated circuit (i.e., a chip). Integrated circuits are quite compact and may contain billions of electronic components in an area smaller than a dime. Resistors can also be much larger. For example, electric irons and the metal coils on stove tops are examples of large resistors. They are designed in a manner that is best suited to their purpose. These two examples are more accurately referred to as variable resistors because they have a control knob that allows the amount of resistance to be controlled, thereby permitting the amount of heat to be controlled.

Ohm’s Law Now that we have discussed current, voltage, and resistance, we can discuss their relationship to each other. As previously stated, for any resistive circuit, the amount of current flowing through the circuit is proportional to the voltage driving it and inversely proportional to the resistance impeding it. Again, current is the flow of charge passing through the circuit, voltage is the charge difference that exists

12

between two separate points within the circuit and that provides the driving force for the flow of charge, and resistance is the opposition of the circuit to charge flow. The German physicist, Georg Simon Ohm, established the relationship between these three variables and translated that relationship into mathematical terms (referred to as Ohm’s law): I ¼ V=R where I is the current through the circuit, V is the applied voltage, and R is the resistance. When the term electromotive force (EMF) is used, rather than voltage, the equation becomes E = IR. For this textbook, we will use the symbol for voltage rather than the symbol for EMF, as this is the symbol with which most readers are familiar. Ohm’s law shows that a higher voltage produces more current and a higher resistance limits the current. Algebraic permutations of this equation yield V = IR and R = V/I. For board examination purposes, whenever one of the three variables is held constant, it is easiest to utilize the Ohm’s law rearrangement that places the constant variable by itself (i.e., on the left side of the equal sign). In this manner, it can readily be determined what effect a change in one of the remaining two variables will have on the other variable. For example, if the current is held constant, then the ideal Ohm’s law rearrangement is I = V/R. Using this rearrangement, it is readily appreciated that the voltage and the resistance are directly related. Consequently, when the resistance doubles, the voltage doubles; and, conversely, when the resistance decreases by half, the voltage decreases by half. If the voltage is constant, then the V = IR rearrangement is utilized. Using this formula, it is clear that the current and the resistance are inversely related. In other words, the current is proportional to the reciprocal of the resistance. Thus, when the resistance increases, the current decreases; and when the resistance decreases, the current increases. Finally, when the resistance is constant, the best Ohm’s law permutation is R = V/I. From this rearrangement, it is apparent that the voltage and current are directly related. Consequently, a change in one leads to a proportional change in the other. Some individuals find it easier to conceptualize water flow rather than electron flow. This analogy is acceptable because the rules of electricity are similar to those of hydraulics. For example, imagine a container of water with an output pipe and faucet attached to its underside (see Figure 1.2).

Chapter 1: Basic Electricity, Electrical Concepts, and Circuits

Figure 1.2 The flow of water is analogous to the flow of charge. Voltage is represented by the pressure of the water (i.e., the weight of the water) on the floor of the container (the driving force, in this case is gravity), resistance is represented by the impedance to water flow generated by the faucet (in this case it is a variable resistor because the faucet permits more than one resistance setting), and current is represented by the flow of water. The pipe is the conductor (it also offers resistance to water flow). (With permission from Ferrante MA, What We Measure and What It Means, 2012.)

The pressure gradient of the water is the force that drives the water through the hose. It is analogous to the voltage (drives electrons through the conductor) and, like voltage, it represents a form of potential energy, the magnitude of which reflects the height difference of the water column (i.e., it is a pressure gradient). In this scenario, the hose represents the conductor and resists the flow of water. The water exiting the hose (i.e., the water flow) is analogous to the current and, like current, its units describe a rate of flow (e.g., gallons per second). Similar to Ohm’s law for charge flow (I = V/R), the formula for water flow is: Q ¼ ΔP=R where Q is flow, ΔP is the pressure gradient, and R is the resistance. Accordingly, if the quantity of water in the tank were increased, the pressure would increase (i.e., the weight of water on the floor of the tank would increase), and as a result, the flow of water would also increase. If, however, the quantity of water

in the tank were decreased, the pressure would decrease (i.e., the weight of water on the floor of the tank would decrease), and hence, the flow of water would also decrease. Because the faucet setting did not change, the resistance is constant (R = ΔP/Q). Therefore, an increase in pressure (ΔP) leads to an increase in flow (Q), and a decrease in pressure leads to a decrease in flow. This is analogous to the Ohm’s law rearrangement, R = V/I, with constant R, where an increase in voltage leads to an increase in current, and a decrease in voltage leads to a decrease in current. If the amount of water in the container is held constant (i.e., the water pressure is constant), then the best formula permutation is ΔP = QR. Using this rearrangement, if the faucet were set at a more open position, then the resistance to water flow would decrease and the flow (Q) of water would increase because flow and resistance are inversely related when the pressure gradient is constant (ΔP = QR). Conversely, if the faucet were set at a more closed position, then resistance would be increased and the flow of water would decrease. Again, these outcomes are analogous to Ohm’s law with constant voltage (V = IR). If the voltage is constant, a decrease in resistance results in an increase in current and an increase in resistance results in a decrease in current, given the inverse relationship between current and resistance when the voltage is constant. Finally, if the flow of water remains the same (Q = ΔP/R), then pressure and resistance are directly related. Consequently, a change in one leads to a proportional change in the other (similar to the I = V/R form of Ohm’s law).

Types of Circuits and Current An electrical circuit is the path through which current flows. In effect, it is the path by which the energy of the EMF source is delivered to the load. A simple circuit consists of an EMF source (to supply the energy necessary to advance the electrons through the circuit), a conductor through which the charges can flow (e.g., copper wire) from the EMF source to the load, a load (the electrical device providing the desired function, such as a bulb), and a return conductor for the current to travel back to the EMF source. The flow of current in a circuit varies with the manner in which the electrical components composing the circuit are interconnected. Of the numerous possible connections, the two simplest circuits are

13

Section 1: Introductory Chapters

termed series circuits and parallel circuits (discussed further on). Combinations of these two circuit types are commonly employed. These circuit arrangements can carry direct current (DC) or alternating current (AC). We will discuss DC first, as the concepts associated with it are simpler to comprehend.

Direct Current Whenever the voltage source of a circuit is DC, the circuit itself is referred to as a DC circuit. With DC, the flow of charge (current) is unidirectional and constant in magnitude. The most common DC voltage source is a battery. DC circuits can be arranged in series, in parallel, or in combinations of these two arrangements. Voltage can be positive or negative, depending on the orientation of the electron gradient. Likewise, a current can be positive or negative, depending on the direction in which the electrons are flowing. When working with circuits, it is helpful to be familiar with Kirchhoff’s current law (also known as Kirchhoff’s first law, the point rule, the junction rule, and the node rule) and Kirchhoff’s voltage law (also known as Kirchhoff’s second law, the loop rule, and the mesh rule). Kirchhoff’s current law, which states that the current flowing into a node equals the current flowing out of that node, is based on the principle of conservation of charge (see Figure 1.3). Again, these current concepts have water flow analogies. Thus, if two water pipes (A and B) are soldered together at one end and the soldered ends are then attached to a third pipe (C) – in other words, in a “Y-shaped” arrangement so that A and B join and continue as C – then the flow of water through pipes A and B equals the flow of water through pipe C. Kirchhoff’s voltage law states that the sum of all of the voltages contained within a closed loop equals

I1

I2

I3

Figure 1.3 As mentioned in the text, Kirchhoff’s current law states that the current flowing into a node equals the current flowing out of that node. This statement derives from the principle of conservation of charge. In this figure, based on Kirchhoff’s current law, i1 = i2 + i3.

14

zero volts. To better understand this concept, consider that the EMF source imparts a voltage rise to the electrons, whereas the loads produce voltage drops. Thus, the sum of the voltage rises equals the sum of the voltage drops. In other words, the equivalent voltage of the loop (the source voltage) and the individual voltages of the loads (the voltage drops) summate to zero.

Series DC Circuits In a series DC circuit, the electrical devices (i.e., the loads) within the circuit are connected in series (i.e., end-to-end) so that the output of one device is the input of the next one (i.e., the current flowing though one device has no option but to flow through the next device). Because all electrical devices offer resistance to electron flow, for discussion purposes, they can all be referred to as resistors. An example of a DC circuit connected in series is a flashlight. The electrons flow from a power source (the battery) through a conductor (a wire) to a load (the bulb) and then back to the power source via another conductor (a wire). When current flows through the circuit as intended, it is referred to as a closed circuit. If the resistance increases to a very high value at any point within this circuit, the current stops flowing (see Figure 1.4). This

Figure 1.4 A DC circuit with a single resistor connected in series with a battery. The plus and minus signs adjacent to the battery symbol are often omitted because they are indicated by the symbol itself (the side with the longer bar is positive). The resistor is symbolized by a saw-toothed line, labeled R1 in this diagram. Electron flow is depicted. In this simple circuit, the battery drives the electrons through the resistor and the resistor impedes the electron flow. If a voltmeter were placed across the leads of the resistor, it would measure the voltage gradient across the resistor (termed the voltage drop). The voltage drop across the resistor is also calculated by Ohm’s law (V = IR). Throughout this textbook, electron current is shown. However, essentially the entire electronics literature and realworld circuits use conventional current flow, in which case the arrow in the figure shown above would be oriented in the opposite direction.

Chapter 1: Basic Electricity, Electrical Concepts, and Circuits

situation is referred to as an open circuit, the utility of which is discussed later here. A third situation exists, referred to as a short circuit, when an unintended pathway of much lesser resistance (approaches zero) shunts current out of the circuit. Many circuits contain switches, which allow the circuit to shift between its open configuration and its closed configuration. In this way, current flow can be controlled (e.g., an onoff switch). When the switch is in the “on” position, the circuit is closed and, for example, the flashlight bulb produces light; when the switch is in the “off” position, the circuit is open and the bulb is dark. In the open position, current cannot flow because the contact points of the switch move apart (become separated by air, which is an excellent insulator). The household wall switches that are used to turn the ceiling lights on and off are another example of how switches are used to open and close a circuit (see Figure 1.5). Because there is only one circuit loop, all of the current in the circuit must flow through the attached load (i.e., there is no other pathway available between the cathode and the anode of the battery). The term load refers to the electrical component contained within the circuit, such as a bulb, and is represented by R1. In a series DC circuit, the current at one point is identical to the current at any other point, and consequently, each electrical component in the circuit carries the same amount of current (Kirchhoff’s current law). In other words, in a series DC circuit, the current running through the circuit is constant. Thus, the best Ohm’s law formula permutation is I = V/R. When the circuit contains more than one resistor, the

Figure 1.5 A DC circuit with a single resistor (e.g., a bulb) connected in series with a battery and a switch (shown in open [off] and closed [on] positions).

current must flow through each resistor in sequence in order to go from the battery cathode to the battery anode. Thus, the current is impeded by each resistor, and for this reason, the total resistance of the circuit is equal to the sum of the resistances of the individual resistors contained in the circuit. Simply stated, resistors in series simply add (see Figure 1.6). In Figure 1.6, two resistors are connected in series with a battery. Because electron current flows from the cathode to the anode, it must traverse both resistors. Thus, the resistance of the circuit (the total resistance [Rtotal] or the equivalent resistance [Req]) is equal to the sum of the two individual resistances. In mathematical terms, Req = Rtotal = R1 + R2. By Ohm’s law, I = V/Rtotal. Because Rtotal = R1 + R2, I = V/(R1 + R2). Because the current is constant in a series circuit, V/(R1 + R2) is the value of the current at every point in the circuit. Applying Ohm’s law, the voltage across R2 (termed V2) can be calculated (I = V2/R2, hence, V2= IR2). Because the current is constant throughout a series circuit, V/(R1 + R2) can be substituted for I and the equation simplified as follows: V2 V2 V2 V2

¼ IR2 ¼ ½V=ðR1 þ R2 Þ
R2 ¼ V
R 2 =ð R 1 þ R 2 Þ ¼ V
R2 =Rtota1

Thus, the voltage across a resistor is proportional to its percentage share of the total resistance (Rtotal). In other words, the percentage of voltage dropping at a resistor is equivalent to its percentage of the total resistance. This same principle applies to series circuits with greater than two resistors as well (see Figure 1.7). Because the wires connecting the components are considered to have zero resistance (in reality,

Figure 1.6 A DC circuit with two resistors connected in series with a battery. The sum of the voltage drops across the two resistors equals the voltage of the battery (Kirchhoff’s voltage law) and the current flowing through each of the resistors is equal (constant current).

15

Section 1: Introductory Chapters

Figure 1.7 A DC circuit with three resistors connected in series with a battery (their resistances are indicated). The sum of the voltage drops across the three resistors equals the voltage of the battery (Kirchhoff’s voltage law), and the current flowing through each of the resistors is equal (constant current).

this is not the case), there is no voltage drop across them. In Figure 1.7, three resistors are connected in series with a battery. Again, because electron current flows from the cathode to the anode, it must traverse all three resistors. Thus, the resistance of the circuit (the total resistance or equivalent resistance) is equal to the sum of the individual resistances. Therefore, in mathematical terms, Rtotal = R1 + R2 + R3. This formula can be derived through Kirchhoff’s voltage law. Vtota1 ¼ V1 þ V2 þ V3 Rtota1 ¼ Vtota1 =I hence, Rtota1 ¼ ðV1 þ V2 þ V3 Þ=I hence, Rtota1 ¼ V1 =I þ V2 =I þ V3 =I By Ohm’s law, V1 = IR1, V2 = IR2, and V3 = IR3; therefore, through substitution: Rtota1 ¼ IR1 =I þ IR2 =I þ IR3 =I Rtota1 ¼ R1 þ R2 þ R3 When electrical components are connected in series, the current is constant and produces voltage across each resistor it encounters. By Ohm’s law, the voltage produced is directly proportional to the resistance of the resistor (I = V/R). For this reason, the voltage is said to follow the resistance: when the resistance goes up, the voltage goes up; when the resistance goes down, the voltage goes down. Consequently, the voltage in a series DC circuit is distributed across the resistors in proportion to their resistance. For this reason, in a series DC circuit, the voltage that drops across a component is dictated by the resistance of that component in relation to the total resistance of the circuit: the higher the resistance of

16

the component, the higher the voltage across it. For this reason, resistors in series act as voltage dividers. In other words, the total voltage of the circuit is divided among the individual resistors in proportion to their resistance. This is probably the most important electrical concept for the EDX medicine practitioner to understand. This concept underlies the importance of proper electrode attachment, signal filtering, signal amplification, noise reduction, and a number of other topics, all of which are explained in detail in this textbook. Because of the importance of this concept, the simplicity of the calculations related to it, and the frequent presence of this type of calculation on neurophysiology board examinations, a few examples are provided here. Problem 1 In Figure 1.4, there is only a single resistor in the circuit. For this reason, the resistor represents 100% of the resistance of the circuit and, consequently, 100% of the voltage (all of the voltage drops across it). If the EMF of the battery is 12 volts and the resistance of the resistor is 5 ohms, what is the value of the current and what is the voltage across the resistor? Solution 1 Using Ohm’s law (I = V/R), the current (in amperes) traversing the circuit is equal to the voltage (in volts) divided by the resistance (in ohms); thus, the correct answer is 2.4 amperes, as shown: I ¼ 12V=5Ω ¼ 2:4A:

Because this is a single-resistor circuit, the single resistor accounts for all of the resistance of the circuit. Therefore, the voltage across it is 12 V (V = IR = 2.4
5 = 12). Thus, it receives 100% of the total voltage. Problem 2 In Figure 1.7, there are three resistors connected in series to a battery. Thus, the total (equivalent) resistance is 20 (5 + 10 + 5) ohms. As previously stated, these three resistors share the voltage in proportion to their percentage share of the total resistance of the circuit. If the EMF of the battery is 12 volts, what is the current of the circuit and what are the voltage drops across each resistor? Solution 2 The total voltage is 12 V and the total resistance is 20 ohms (5 + 10 + 5). Therefore, the current (I = V/R) is equal to 12/20 = 0.6 amperes, which is constant in a DC series circuit. The voltage across a resistor (its share of the voltage) is equal to its share of the resistance (its

Chapter 1: Basic Electricity, Electrical Concepts, and Circuits

resistance divided by the total resistance) multiplied by the total voltage. VR ¼ ðRR =Rtota1 ÞVtota1

Where VR is the voltage across a resistor, RR is the resistance of the resistor, Rtotal is the total resistance of the circuit, and Vtotal is the total voltage of the circuit. Stated another way, the resistor’s percentage of the total resistance dictates its percentage of the total voltage. For example, the voltage across R1 is: VR1 ¼ ðRR =Rtota1 ÞVtota1 ¼ 5=20
12 ¼ 3V Utilizing the same formula, the voltage across R2 is 10/20
12 = 6 V. Note that this is twice the voltage across R1, which is to be expected given that its resistance is twice the value of R1. Because R3 has the same resistance as R1, it receives 3 V. In summary, the total voltage (12 V) is distributed across the three resistors in proportion to their resistance ratio. Because their resistance ratios are 1:2:1, R1 receives 3 V, R2 receives 6 V, and R3 receives 3 V. Because resistors function as voltage dividers, it is also possible to create circuits that offer optional voltage outputs simply by modifying the resistance contributing to the output voltage (see Figure 1.8). When voltage sources (e.g., batteries) are connected in series, the total (equivalent) voltage of the circuit is their summed value, as long as their polarities are oriented identically. Thus, like resistors

Figure 1.8 In this figure, there are two output options, one across one resistor and one across both resistors. In this scenario, when the output is taken across both resistors (i.e., the total resistance of the circuit), the full 12 V is available. On the other hand, if the output is taken across just one of the resistors, then only half of the voltage is available (6 V), because each resistor represents 50% of the total resistance of the circuit. This circuit arrangement could permit a power tool to operate at two different voltages (6 V and 12 V).

connected in series, voltage sources connected in series are additive. This makes sense in that there is only one pathway for the voltage sources to “push” electrons and that is out from the negative terminal; thus, all of the batteries are pushing the electrons along the same wire and in the same direction. For this reason, the total voltage of the circuit is the sum of the voltages of the individual voltage sources (e.g., batteries). However, if one of the batteries were mistakenly connected with its polarity in reverse to the other batteries, then its voltage would be subtracted from the summed value of the others. This also makes sense because, with its polarity reversed, the erroneously positioned battery would actually “push” the electrons in the wrong (opposite) direction. For example, if a flashlight required two batteries of equal strength and one was mistakenly placed into the battery compartment “backwards” (i.e., with its polarity reversed with respect to the other one), the two batteries would negate each other and the equivalent voltage of the circuit would be zero. Consequently, the bulb would not illuminate when the flashlight switch was turned on. As a second and slightly more complex example, if the flashlight required four batteries of equal strength connected in series and one of the four was erroneously positioned with its polarity reversed, then the equivalent voltage would not be reduced by 25%, but rather by 50%, because the incorrectly positioned battery would negate the “push” of one of the other three correctly positioned batteries, leaving just two correctly positioned batteries to drive the electrons through the circuit. This point may be easier to envision if we imagine four men of equal strength pushing a stone block along a pathway. If one goes to the other side and pushes against the other three, then the strength of one of the three men on the other side is negated, and the block moves based on the force generated by only two of the men (see Figure 1.9). Similar to voltage sources connected in series, when resistors are connected in series, the equivalent resistance (i.e., the total resistance) of the circuit represents the sum of the individual resistances of each resistor. Unlike voltage sources connected in series, if we place a resistor into the circuit in reverse, the equivalent resistance remains the same because resistors do not have polarity. Thus, the resistance to DC generated by a resistor oriented in one direction is equal to its resistance when it is oriented in the opposite direction. As an example, if the resistor were

17

Section 1: Introductory Chapters

Figure 1.9 On the left, the two batteries are oriented similarly, and therefore, the potential difference of this circuit is 24 V, whereas the voltage on the right is zero because the batteries are oriented opposite to one another. Thus, the electrons are driven in opposite directions with no net effect.

Figure 1.10 A DC circuit with 3 resistors connected in parallel to a battery. The resistance of each resistor is also shown.

a bulb and the circuit wires attached to the bulb were disconnected and then reconnected at the opposite bulb attachment sites, the bulb would still light and would still consume the same amount of energy. The electron flow would simply move across the lightemitting filament in the opposite direction.

Parallel DC Circuits With parallel circuits, each device (load) in the circuit is directly connected to the voltage source, such as the cathode and anode of a battery. Thus, unlike a series circuit, with a parallel circuit there is more than one path through which the current may flow. Each connection is termed a circuit loop or a circuit branch (see Figure 1.10). Hence, each loop (and therefore each device) has the same voltage across it. In Figure 1.10, because each resistor is independently connected to the battery terminals, the voltage across each resistor is equal (Kirchoff’s voltage law) and current can flow through any of the three loops. Because the voltage is the same across each resistor (i.e., it is constant), the relationship between current and resistance is best determined by the V = IR

18

formula rearrangement of Ohm’s law. As indicated by V = IR, the current through a resistor varies inversely with its resistance. Stated another way, each component draws current from the battery, and the quantity of current drawn is inversely proportional to its resistance. Therefore, as shown in Figure 1.10, because it has the highest resistance, the current passing through R2 (termed, I2) is less than that passing through the other two resistors (I1 and I3). In other words, the current varies inversely with the resistance of the individual loop, and the total (equivalent) current drawn by the circuit equals the sum of the currents drawn by each loop. Thus, the adage that current takes the pathway of least resistance is perhaps more accurately stated as current predominates along the pathway of least resistance because it actually takes all of the available pathways, with the exact quantities traversing each pathway being inversely proportional to the resistance of the resistor (i.e., the pathway of least resistance gets the greatest share). In terms of water flow, if water is flowing from one tank to another via two garden hoses and a third hose is added, the total resistance to water flow from

Chapter 1: Basic Electricity, Electrical Concepts, and Circuits

the first tank to the other one decreases, but the pressure gradient between the two tanks (i.e., the push) is unchanged. As more hoses are added (i.e., as the total resistance to water flow lessens), more water (analogous to current) will flow (V = IR). Because the current traversing each resistor varies with its resistance, resistors arranged in parallel act as current dividers. Mathematically, we have: Ieq ¼ I1 þ I2 þ I3 where Ieq is the total current and I1, I2, and I3 are the currents through each of the three resistors. Because I = V/R, this formula can be rewritten with (V/R substituted for each I) as: Veq =Req ¼ V1 =R1 þ V2 =R2 þ V3 =R3 Because the voltage across each resistor is equal to the voltage of the voltage source (Veq = V1 = V2 = V3), the voltage terms in the above equation can be simplified to V, as shown: V=Req ¼ V=R1 þ V=R2 þ V=R3 By dividing both sides of the equation by V, the equation simplifies to: 1=Req ¼ 1=R1 þ 1=R2 þ 1=R3 From the final equation above, it can be seen that the reciprocal of the equivalent resistance (1/Req) is equal to the sum of the reciprocals of the individual resistances – or, stated another way, parallel resistors add reciprocally. As a result, when resistors are connected in parallel, the total resistance of the circuit decreases because there are more pathways available to the current. Thus, the total resistance of the circuit is lower than the resistance of any of its individual circuit loops. This concept may be more readily appreciated in terms of conductance. Recall that conductance is the inverse of resistance. When there are more pathways for current to traverse, conductance goes up (hence, resistance goes down). For example, the conductance is greater with four pathways simultaneously available than it would be with just one of them. Again, a water flow analogy may be more insightful. For example, if a reservoir of water has four output pipes attached to its undersurface, each with a valve allowing that pipe to be either open or closed, it is apparent that, as the number of pipes in the open state increases from one open pipe to four open pipes (i.e., decreasing resistance), the flow of

water (conductance) increases. An even simpler analogy is to a bucket of water in which a hole is drilled through the bottom of the bucket. The water flows through the hole in proportion to the size of the hole. When a second hole is placed in the bottom of the bucket (analogous to resistors in parallel), the amount of water flowing out of the bucket increases. As previously stated, for a constant voltage (V = IR), as the resistance decreases, the current increases. Thus, the resistance to water flow of a bucket containing two holes is lower than that of a bucket containing one hole. In other words, for a constant voltage (V = IR), as the resistance goes down, the current goes up. Mathematically, the formula used to calculate the total resistance (Req) of the circuit when the individual resistors are connected in parallel is: 1=Req ¼ 1=R1 þ 1=R2 þ 1=R
3 solving for Req Req ¼ 1=ð1=R1 þ 1=R2 þ 1=R3 Þ Problem 3 Calculate the total resistance of the circuit shown in Figure 1.10. Solution 3 The total resistance is 2 ohms: Req ¼ 1=ð1=5 þ 1=10 þ 1=5Þ ¼ 1=ð2=10 þ 1=10 þ 2=10Þ ¼ 1=ð5=10Þ ¼ 1=ð1=2Þ ¼ 2 ohms

Alternatively, using the less complex formula (1/Req = 1/R1 + 1/R2 + 1/R3): 1=Req ¼ 1=5 þ 1=10 þ 1=5 1 ¼ Req ¼ 2=10 þ 1=10 þ 2=10 1 ¼ Req ¼ 5=10 therefore, Req ¼ 10=5 ¼ 2

Note that the total resistance of this circuit is 2 ohms, which is lower than the resistance of any of the three resistors. The formula for calculating the total resistance of a circuit with two resistors oriented parallel to the battery simplifies to the product of the resistors divided by their sum: R T ¼ R 1 R 2 =ð R 1 þ R 2 Þ As shown by the formula, if two resistors of equal resistance are connected in parallel, the total

19

Section 1: Introductory Chapters

resistance is equal to half of the resistance of either one of them. This is in contrast to connecting them in series, in which case the total resistance would be twice either one of them. For example, if the resistance of the two resistors is 5 ohms, then the total resistance equals 2.5 ohms: Req ¼ ðR1 R2 Þ=ðR1 þR2 Þ¼ ð5
5Þ=ð5þ5Þ¼25=10¼2:5 A similar product-over-sums type formula for three resistors is also available, but is more complicated than the two provided above: Req ¼ R1 R2 R3 =ðR1 R2 þ R1 R3 þ R2 R3 Þ Using the values of the three resistors connected in parallel shown in Figure 1.10 (R1 = 5 ohms, R2 = 10 ohms, and R3 = 5 ohms): Req ¼ ð5
10
5Þ=½ð5
10Þ þ ð5
5Þ þ ð10
5Þ ¼ 250=ð50 þ 25 þ 50Þ ¼ 250=125 ¼ 2 This is the same answer as shown above in the solution for Problem 3. At this point, we have reviewed resistors connected in series, resistors connected in parallel, and voltage sources connected in series, but we have not discussed voltage sources connected in parallel. This is because, in general, voltage sources (e.g., batteries) are generally not connected in parallel because if they were to be and they were not of identical strength, their small differences would act to drive current from the stronger-voltage battery to the weaker one. Consequently, parallel DC circuits typically have only a single-voltage source. The arrangement of the elements contained within a circuit is important. Some arrangements may not work, and others may be dangerous. For example, if the flashlight components in Figure 1.11 were connected in parallel, rather than in series, a number of problems would result (see Figure 1.11). In Figure 1.11, when the switch is in the open position, all of the current travels through the loop with the bulb, and therefore, the bulb is continuously illuminated (switch open is “on” rather than “off”). When the switched is placed in the closed position, because the current prefers the pathway of least resistance, essentially all of the current will traverse the loop with the switch. As a result, there will be essentially no current available for the bulb and, consequently, it will not illuminate (switch closed is “off” rather than “on”). Although this problem could be

20

Figure 1.11 The battery, the switch, and the bulb (R1) are connected in parallel.

rectified by relabeling the switch, there is a much more serious problem. When the switch is closed, because the current travels down the pathway of least resistance, essentially all of the current will traverse the loop with the switch. As a result, this loop will behave as if a conducting wire were placed across the battery terminals, thereby creating a short circuit and potentially causing the conducting wires to heat up (fire hazard) or the battery to rupture or explode. As previously stated, a DC circuit may have some resistors oriented in series and others in parallel. Consequently, before leaving DC circuits, a summarizing exercise is provided here (see Figure 1.12). Problem. Calculate the equivalent resistance of the circuit shown in Figure 1.12. Solution. The first step is to simplify the circuit by calculating the equivalent resistance of the circuit portion containing R1 and R2, which are connected in parallel. Using the product over the sum formula for two resistors connected in parallel yields 24 ohms ([40
60]/[40 + 60] = 2,400/100 = 24). This value is then added to R3 and R4 (because these two resistors are in series with it). Thus, the equivalent resistance of resistors 1 through 4 is 124 ohms (24 + 20 + 80 = 124). This value is in parallel with R5, so again we use the product over the sum equation to generate the final answer – (124
100)/(124 + 100) = 12,400/224 = 55.4. Thus, the equivalent resistance of the circuit shown in Figure 1.12 is 55.4 ohms.

Capacitors In order to fully understand membrane potentials, the generation and propagation of action potentials, filtering, and many other issues pertinent to EDX

Chapter 1: Basic Electricity, Electrical Concepts, and Circuits

R3 20

R5

100

R1

40

R2

60

R4 80 Figure 1.12 A DC series-parallel circuit containing 5 resistors.

medicine, in addition to resistors, practitioners of EDX medicine must also understand capacitors and the important concepts and principles associated with them. A capacitor is an electrical device composed of three layers: a pair of conducting layers (two plates) separated by a central insulating layer. Unlike resistors, capacitors do not dissipate energy. Instead, they store charge and energy through charge separation (electrical potential energy; voltage). The phospholipid bilayers of cell membranes give them capacitance and permit them to accumulate charge. This is discussed in greater detail in Chapter 3. When a capacitor is connected between the cathode and the anode of a battery, the cathode drives electrons to the nearest capacitor plate, causing it to become negatively charged. Because of the insulating material between the two plates, the accumulating electrons are unable to advance to the opposing plate. As more and more electrons accumulate on this plate, the plate becomes more and more negatively charged. This buildup of negativity creates a repulsive force (Coulomb’s law) that drives the electrons located on the opposing plate toward the anode of the battery, causing the opposing plate to develop a positive charge. Therefore, with capacitive current, the current flows on both sides of the capacitor, but not through it. This type of current is also referred to as displacement current because charge must be displaced for current to flow. In this situation (a DC circuit), the capacitor plates continue to accumulate charge until the repulsive force across the two plates equals the electromotive force (voltage) of the battery. At this point, the capacitor is fully charged and further

charge accumulation cannot occur. For example, when a 12-volt battery is connected in series with a capacitor, the 12-volt battery charges the capacitor to 12 V (i.e., the charge difference between the two plates has a value of 12 V). Once fully charged, further charge accumulation is not possible. In other words, once the capacitor is fully charged, there is an electrical field created across the two plates of the capacitor that is equal to and opposite to the electric field of the battery, inhibiting further charge accumulation. When the capacitor is removed from the circuit, the stored charge remains with the capacitor and can be utilized elsewhere. For example, a defibrillator is simply a charged capacitor capable of discharging on demand. The amount of current flowing on the two sides of the capacitor reflects its capacitance (discussed in detail further on). The amount of charge that a capacitor can hold depends on a number of factors, including the surface area of the plate overlap, the distance between the two plates, the dielectric constant of the material separating the two plates, and the voltage of the source driving the current. With larger plates, more electrons (charge) are required to create a specified voltage or to remove it, whereas with smaller plates, fewer electrons are required to create a specified voltage. This fact explains why only a small number of ions need to traverse the cell membrane to establish a specified voltage. These concepts are important in understanding the effect of myelination on nerve conduction velocity values and are discussed later in this textbook (see Chapter 3). When a capacitor is connected in series with a battery, the rate at which electrons accumulate on the negative plate of the capacitor is dictated by the voltage and current capability of the battery. If a resistor is added to the circuit (in series with the battery and the capacitor), the time required to charge the negative plate is increased in proportion to the resistance of the resistor. In other words, the time required to charge the capacitor is inversely proportional to the rate of electron flow. It is important to understand that the charge accumulation on the negative plate does not occur in a linear fashion, but rather occurs exponentially. Because capacitors are able to store charge, they develop a voltage (charge separation; electrical potential energy) across their plates. Again, once fully charged, the voltage value across the capacitor equals the voltage of the battery driving the electrons. With a

21

Section 1: Introductory Chapters

battery, the cathode and anode are far enough apart that they do not affect each other. When a capacitor is connected in series with a battery, in effect, it is as if the battery terminals have extended to the capacitor plates, and because they are now adjacent to each other, they now affect each other (see Figure 1.21). In other words, the negatively charged capacitor plate becomes the negative battery post (the cathode) and the positively charged capacitor plate becomes the positive battery post (the anode). The amount of voltage (charge separation) that a capacitor can store is reflected by its capacitance (C), which is the ratio of the charge (Q) on either plate to the voltage (V) between the two plates (C = Q/V). The formula shows that the charge and voltage are proportional. Stated another way, the amount of charge that can be loaded onto a capacitor plate is proportional to the voltage and the capacitance of the capacitor (i.e., Q = VC). Hence, if you put a particular voltage across a capacitor, you get a particular charge on the plate. In other words, the capacitance is the amount of charge that you get for a given voltage. As the voltage increases, the charge increases. The Q:V ratio of a capacitor is constant and defines its capacitance, measured in farads (defined later in this chapter). The capacitance tells us how much charge we get for a given voltage. The amount of charge stored is proportional to the surface area of overlap (A) of the two plates of the capacitor (i.e., the larger the surface area of overlap of the plates, the greater the magnitude of charge that can be accommodated). As the distance (d) between the two plates increases, the effect that one plate has on the other one decreases. Consequently, capacitance is inversely proportional to the distance between the plates. Mathematically, C = A/d. In summary, capacitors with larger plates store more charge, and the effect that the two plates have on each other decreases as they are moved further apart. The unit of measurement for capacitance is the farad (named after Michael Faraday). A capacitance of 1 farad equals 1 coulomb per volt. In other words, 1 farad of capacitance produces 1 volt of potential when 1 coulomb of charge is stored on it. Capacitors commonly range in strength from picofarads to microfarads. Capacitor charging and discharging is not instantaneous – it takes time. In addition, charging and discharging are not linear, but rather occur at exponential rates. The time requirement for capacitor charging and discharging is reflected by its time

22

constant (T), which is equivalent to the product of the resistance and the capacitance (T = RC) and which is defined as the time required for the capacitor to charge or discharge by 63%. But why was 63% chosen as the time constant? The answer is indicated by the following equation: V ¼ 1  et=RC thus, V ¼ 1  et=T ðbecause T ¼ RCÞ where, V is voltage, e is the base of the natural logarithm (about 2.7), t is time, RC is the product of the resistance and capacitance, and T is the time constant. Note that when the time (t) equals the time constant (T), the equation simplifies to V = 1 – e1, which equals 0.632, as shown here: V ¼ 1  e1 V ¼ 1  1=e V ¼ 1  1=2:7 V ¼ 1  0:386 V ¼ 0:632 When t = T, the capacitor is 63.2% charged. Thus, for mathematical simplicity, the time constant is defined as the time required for the voltage of the capacitor to reach 63% of its full voltage. Not only is the time constant defined as the amount of time that is required for the capacitor to charge to 63% of its maximal value; it is also defined as the time required for it to discharge by 63% (i.e., to discharge to 37% of its fully charged value). Thus, it represents a 63% change in either direction. Once a capacitor is fully charged, its resistance to further current accumulation is essentially infinite. Consequently, it cannot pass additional current. There is no charge moving on or off the capacitor plates, and thus there is no current in the wires leading to and from the plates of the capacitor; the voltage is constant. Because capacitors resist changes in voltage, they are able to store this accumulated electrical charge. If the power source is then turned off, the electrons located on the plate connected to the cathode return to the cathode, and the electrons previously driven to the anode return to the capacitor plate connected to it. Thus, a displacement current occurs in the opposite direction and continues until the capacitor is fully discharged. Capacitors can be connected in series or in parallel with the voltage source. Mathematically, the total capacitance (equivalent capacitance, Ceq) of a DC circuit is calculated in

Chapter 1: Basic Electricity, Electrical Concepts, and Circuits

the opposite manner of that for total resistance. Thus, capacitors sum when connected in parallel and invert when connected in series. For example, for circuits containing three capacitors: For a parallel DC circuit, Ceq ¼ C1 þ C2 þ C3 For a series DC circuit, 1=Ceq ¼ 1=C1 þ 1=C2 þ 1=C3 With AC signal, however, because of its directionreversing nature, the capacitor successively charges and discharges as the AC signal changes its direction. This introduces a time delay into the circuit. These characteristics of capacitors are important for a full understanding of filters and are discussed in greater detail in the next chapter (see Chapter 2).

Alternating Current Unlike DC sources of electricity (e.g., batteries), in which the current and voltage are unidirectional and constant, with alternating current (AC), the current and voltage periodically reverse their direction and are constantly changing in magnitude. In other words, the electrons travel in one direction, slow to a stop, and then travel in the opposite direction. This cycle of events continuously repeats itself and the current and voltage are constantly changing throughout each cycle. As a result, there is essentially no negative or positive direction to the circuit as there is with DC sources. Rather, the electrons simply pulse back and forth over time without any net change in their location. The frequency of these reversals was previously expressed in cycles per second (CPS), but currently the term hertz (Hz) is utilized, where 1 Hz = 1 CPS. The frequency of AC in the United States is 60 Hz because generators in this country rotate at 60 rotations per second, and consequently, the generated current (and, thus, voltage) reverses its direction at a rate of 60 times per second (60 Hz). (In Europe, the AC frequency is 50 Hz.) Because of these continuous reversals and the continuously changing voltage and current values, AC is best depicted as a sinusoidal waveform, where the value of the amplitude (y-axis coordinate) reflects the voltage and the x-axis coordinate reflects the passage of time (see Figure 1.13). When depicted as a sinusoidal waveform, an AC signal can be described by its amplitude (in voltage), frequency (in hertz), and phase shift (horizontal shift expressed in degrees). By definition, the portions of the AC curve located above the x-axis are positive, the portions below the

+

Figure 1.13 A single cycle of AC signal.

0

– x-axis are negative, and the x-intercepts (i.e., where the wave crosses the x-axis) represent moments of current and voltage absence. As a result, the average value of AC over time is zero because the positive values and the negative values cancel each other out (summate to zero). Despite an average value of zero, AC delivers energy, first in one direction and then in the opposite direction. Thus, unlike DC, in which the voltage supply is constant, the voltage supply with AC is constantly changing. Ohms law can be extended to AC circuits using the formula, V = IZ, where V is voltage, I is current, and Z is impedance. Impedance, which is measured in ohms, consists of two quantities: resistance and reactance. It is a more general concept than resistance, but is required to reflect the non-dissipative opposition to current flow exhibited by reactive elements, such as capacitors and inductors. (Theoretically, reactive elements cannot dissipate energy; they can only store it.) The resistance component of impedance reflects the resistive elements (e.g., resistors) it contains. Resistive elements oppose all frequencies of current flow equally, and thus AC resistance is constant and similar to that observed in a DC circuit. Reactive elements are frequency-dependent devices. They resist current based on its specific frequency. Capacitors have already been discussed. An inductor (L) is a coil of wire through which a current is passing. As the current traverses one loop of the coil, it creates a magnetic field around the loop that induces current in the adjacent loop. This is termed electromagnetic induction. According to Lenz’s law, the electromotive force of the induced current opposes the electromotive force that produced it. As the frequency of the signal increases, the inductive reactance (i.e., the opposition to current flow)

23

Section 1: Introductory Chapters

Amplitude

Figure 1.14 AC depicted as a sinusoidal waveform.

Crest

+ Negative-bound x-intercept

0

Time

Positive-bound x-intercept

– Trough

increases exponentially. This is the opposite of what occurs with a capacitor. With a capacitor, capacitive reactance (i.e., the opposition to current flow) increases exponentially as the signal frequency decreases. For this reason, the total reactance in the circuit is composed of inductive reactance and capacitive reactance and these two entities oppose each other. The frequency dependence of capacitive and inductive reactance allows for the construction of electronic filters (i.e., electronic devices that are used to separate the signal based on its frequency). Stated another way, capacitors oppose lower frequencies more than higher frequencies, whereas inductors oppose higher frequencies more than lower frequencies. Therefore, at DC (0 Hz), inductors appear to be a short circuit (no opposition to current advancement) and capacitors appear to be an open circuit (complete opposition to current advancement). Conversely, at very high frequencies, inductors appear to be an open circuit and capacitors appear to be a short circuit. As previously stated, reactive elements do not dissipate energy, but rather store it. Capacitors store energy in an electrical field and inductors store energy in a magnetic field.

The Quantification of AC Signal A cycle of AC is defined as that portion of the wave that lies between two corresponding points. Of these, the easiest to identify are the crest, the trough, the negativebound x-intercept (zero point), and the positive-bound x-intercept (zero-point) (see Figure 1.14). The time interval to complete 1 cycle, measured in seconds, is termed the period of the wave, and is

24

symbolized by an uppercase, italic letter T. For example, the period of a 60 Hz AC wave is 1/60th of a second. Because the period of a sine wave and its frequency (symbolized by the lowercase, italic letter f) are inversely related (see Figure 1.15), they can be mathematically symbolized as follows: T ¼ 1=f f ¼ 1=T 1 ¼ fT For example, a computer with a microprocessor clock speed of 1 gigahertz (GHz) (1 billion cycles per second) has a frequency of 1 billion Hz and a period of 1/1
10–9 seconds (one-billionth of a second). As another example, we can calculate the wavelength of an AC signal. Anytime electrons move at varying speeds, they generate electromagnetic radiation (photons) that propagates through space at the speed of light (186,282 miles per second) and through air at nearly the speed of light. Because electromagnetic radiation oscillates as it propagates through space, it can also be characterized as a wave. Thus, electromagnetic radiation with a 60 Hz frequency has a wavelength of approximately 3,105 miles (186,282 miles/ second divided by 60 equals 3,104.7 miles). Thus, although the electrons oscillate back and forth over a distance of just a few micrometers (i.e., there is essentially no net electron movement), the electromagnetic radiation their movement generates travels at nearly the speed of light. Because of its sinusoidal nature, each cycle of AC can be divided into 360 equal parts (termed degrees or degrees of phase). Zero degrees is defined as the site at

Chapter 1: Basic Electricity, Electrical Concepts, and Circuits

Figure 1.15 The time period (T) of a sine wave as measured from the corresponding points of two cycles, in this case between two positive-bound x-intercept points.

+

0

Time

– Period

Figure 1.16 A cycle of AC signal depicted in degrees. Note that the 360-degree mark for the cycle is the 0-degree mark for the subsequent cycle. The time required for one full cycle of AC is termed the period. The frequency is 1 over the period (f = 1/T).

180° 90°

360° 270°

which the single cycle of the AC wave crosses the x-axis while moving in the upward (positive) direction. Consequently, the positive peak occurs at 90 degrees, the downward bound x-axis intercept occurs at 180 degrees, the negative peak occurs at 270 degrees, and the period ends at the upward bound x-axis intercept (360 degrees). Thus, the magnitude of the voltage is zero at 0, 180, and 360 degrees (see Figure 1.16). Amplitude and frequency characterize the AC signal. However, unlike with DC signal, which has a constant magnitude, quantifying the strength (amplitude) of an AC signal is challenging because the voltage is constantly changing. As stated earlier, the values above the x-axis are positive and those below the x-axis are negative, and therefore, AC signal cycles around zero. As a result, the mean value of the AC

wave is always zero, and thus the mean value of the AC signal does not reflect its strength. The term peak voltage (or peak current) can be used to indicate the value of the signal at the positive peak (referred to as the baseline-to-peak voltage [VB-P] or as the positive peak voltage [VP]). Of these two terms, VP is always used in electronics (and is used throughout the remainder of this chapter). The value of the signal at the trough is referred to as the negative peak voltage (VN). The term peak-to-peak voltage (VP-P) indicates the difference between the positive peak and negative peak voltage values. Importantly, because these values only reflect the maximum values of the cycle, they cannot be used to quantify the strength of the AC signal. The most commonly used approach to expressing the strength of an AC wave is its root-mean-square

25

Section 1: Introductory Chapters

(RMS) voltage value (VRMS), also referred to as its effective voltage or DC-equivalent voltage. These latter terms reflect the fact that the RMS voltage value is the energy equivalent of DC (i.e., 1 volt RMS equals 1 volt DC). The RMS voltage is 0.707
the peak voltage (0.707 is the value of the square root of ½). Because they are 90 degrees out of phase, they form a right triangle of equal sides. If the hypotenuse = 1 and the two sides are equal (e.g., X), then, by the Pythagorean theorem, X2 + X2 = 1. Thus, X = 1/ √2, which equals 0.707, as shown here: X þX ¼1 2X2 ¼ 1 X2 ¼ 1=2 X ¼ √1=2 ¼ 0:707 2

2

The RMS voltage also reflects the power formula (P = VI), discussed later in the book, which indicates that the power, expressed in watts, is proportional to the product of the voltage and the current. Through substitution (I = V/R) and algebraic manipulation, this formula can be rewritten as P = V2/R, which shows that the power is proportional to the square of the voltage or, stated another way, that the voltage is proportional to the square root of the power. Sometimes, rather than express the electrical quantity in volts, it is expressed as a proportionality to another value, either as a ratio, a percentage, or in decibels (dB). The bel (B), named after Alexander Graham Bell, is a unit used to compare two different power levels and is equal to the common logarithm (i.e., the base 10 logarithm) of the ratio of the two powers (P2/P1). Thus, B = log10 (P2/P1). The decibel (dB) is one-tenth of a bel. Thus, because there are 10 decibels in a bel, the decibel value is equivalent to 10 times the bel value : dB = 10 log10 (P2/P1). Based on the relationship between power and voltage – i.e., that power is proportional to the voltage squared (P = V2/R) – the voltage ratio (V2/V1) can also be expressed in decibels: dB = 10 log10 (V22/V12) = 10 log10 (V2/V1)2 = 2
10 log10 (V2/V1) = 20 log10 (V2/V1). Table 1.1 shows the relationship between specific power ratio and voltage ratio values and their decibel equivalents. Using the values shown in Table 1.1, 1 V RMS of AC delivers the same amount of power as 1 V DC. Put another way, RMS voltage is the voltage that an equivalent DC source would generate. Thus, the RMS voltage value of an ever-changing AC signal is equivalent to the strength of a continuous DC signal, which

26

Table 1.1 The values shown in this table are useful for calculating RMS values and the cutoff frequencies of analog filters. The first column shows the power ratio value, the second column shows the voltage ratio value (which is based on P = V2/R), and the third column shows the decibel value of these ratio values. For example, looking at the values in the first row, when the power ratio is 100, the decibel equivalent is 20, as it is when the voltage ratio is 10.

(P2/P1)

(V2/V1)

Decibels

100

10

20

2

1.414

3

1

1

0

½

0.707

–3

is why it is also referred to as the DC-equivalent voltage and the effective voltage. For example, household current is 117 V RMS and, therefore, is equivalent to the power generated by 117 V DC. (The 117 V value actually varies to a small degree with the distance between the transformer and the home and is slightly lower in value for longer distances and slightly greater in value for shorter distances.) The RMS voltage is calculated by initially squaring all of the values along the AC wave (in this manner, the negative values become positive), then averaging them, and finally taking the square root of the calculated average. In other words, the output value is the root of the mean of the squares (see Figure 1.17). The mathematical relationships between the RMS voltage value and the other previously mentioned voltage values are as follows (note that 1.414 equals the square root of 2): VRMS ¼ 0:3536ðVPP Þ VRMS ¼ VPP =2:828 VRMS ¼ VP =1:414 1:414ðVRMS Þ ¼ VP For example, given that household current equals 117 V RMS, then the positive peak voltage (VP) equals 165 V (1.414 times the RMS voltage), the negative peak voltage (VN) equals –165 V, and the peak-to-peak voltage (VP–P) equals 330 V.

Household Electricity A moving magnetic field generates current in a nearby wire conductor, and current conducting through a wire generates a magnetic field in the region surrounding the wire. Thus, moving magnets

Chapter 1: Basic Electricity, Electrical Concepts, and Circuits

1 0.707 B:P Voltage RMS Voltage

P:P Voltage

0

–1 Degrees 0 Cycles 0 Radians 0

90 1/4 p/2

180 1/2 p

270 3/4 3p/2

360 1 2p

Figure 1.17 A single cycle of AC depicting the constantly changing voltage (y-axis) over time (x-axis) and the relationship between RMS voltage, baseline-to-peak (B:P) voltage, and peak-to-peak (P:P) voltage.

generate electrical fields and moving charges generate magnetic fields. When the magnet moves near a wire, it induces a momentary current in one direction, and as the magnet moves back to its original position, the current induced moves in the opposite direction. Consequently, when a magnet is moved back and forth near a wire, the direction of the induced current alternates. As a result, AC electricity can be generated in a number of ways, including the rotation of a magnet within a fixed coil of wire or the rotation of a coil of wire within a fixed magnet or group of magnets. The amount of current generated can be increased by increasing the strength of the magnet, increasing the number of coils composing the coil of wire, or by increasing the rate of rotation. A mechanical force, such as falling water, can be used to drive the rotation of the coil of wire or of the magnet, which in turn generates the AC voltage; the latter appears at the opposite ends of the wire coil. Two reasons why AC power is preferred over DC power is that it is easy to create (e.g., falling water) and its magnitude is easily changed (power transformers). In the United States, the rotational speed of the coil of wire or of the magnet is 3,600 revolutions per minute, which yields an AC output frequency of 60 Hz (i.e., 60 revolutions per second). When the magnet is mounted on a rotating shaft, it is referred to as a rotor; the copper wire coil

surrounding it is referred to as a stator (see Figure 1.18). As the magnet rotates within the coil of copper wire, its magnetic field is continuously changing. The continuously changing magnetic field generates continuously changing current and voltage (AC signal). A major advantage of electrical power over other forms of power is that it is transportable from its point of origin to its point of usage through what is referred to as the grid. Just five electrical networks, called interconnections, serve the United States and Canada (Alaska interconnection, Quebec interconnection, Western interconnection, Texas interconnection, and Eastern interconnection). They have a hierarchy consisting of power plants, substations, and transmission lines. The power plants represent the points of origin of the electricity. Electricitygenerating power plants include coal-fixed and gasfixed power plants, nuclear power plants, and hydroelectric power plants. At these points of origin, there are adjacent switchyards where transformers (discussed further on) are used to step-up the voltage to very high levels (115,000–765,000 V). The electrical power then undergoes long-distance transport to regions where the power will be consumed (i.e., regional substations in cities, towns, and industrial parks) via transmission feeders. Although the resistance of aluminum is greater than that of copper, aluminum wiring is preferred because it is so much

27

Section 1: Introductory Chapters

Voltage Stator (copper wire coil) N O S

60-Hz AC power Rotor (rotating magnet)

0

0.5

1

1.5

Sine wave cycles

2

2.5 Time

Figure 1.18 The generation of AC power within a wire coil by varying the surrounding magnetic field. When a magnet located within a coil of wire is rotated at 60 revolutions per second, 60-Hz AC power is generated.

Stator (with 3 copper wire coils) Rotor (rotating magnet)

Figure 1.19 The three-phase system.

Voltage 2

1 N O S

1 2 3 Time

3

lighter than copper. A step-down transformer is used to further reduce the voltage (to 35,000–138,000 V) for subtransmission from the regional substations to local substations closer to the consumer, at which point it is stepped down again (2,400–25,000 V) and then transmitted along overhead lines on wooden poles for distribution to the consumer (e.g., residences). Just outside the home, it is stepped down to 120 V, which is the strength delivered to individual households. In actuality, there is a voltage drop related to the resistance of the wire between the transformer and the house of about 5 V and another small voltage drop from the electric panel to the receptacle. The technique used to distribute electrical power from its point of origin to its point of usage is currently much more efficient than it was in the past. Previously, electrical power was distributed using a single-phase system. With single-phase power, two conductors are required – one that delivers the power to the substation and another conductor that functions as a return path to complete the circuit. Currently, three-phase power is utilized. With threephase power, the magnet rotates within the stator as with single-phase power generation, but instead of a

28

single coil of wire, there are three coils equally spaced within the stator (i.e., they are mounted at exactly 120 degrees from each other) (see Figure 1.19). In this manner, each coil generates AC power with the same amplitude (voltage) and the same frequency (60 Hz). The electric power generated by each coil is termed a phase, and the three phases run along parallel conductors to the substation. Because the three coils are exactly 120 degrees apart, the 3 sine waves of AC power generated are 120 degrees out of phase with each other. As a result, at any point in time, the magnitudes of the individual currents (voltages) traversing the three parallel conductors summate to zero volts (see Figure 1.19). In this manner, there is no flashover (jumping of an electrical arc across the gap between the conductors related to large voltage differences). The risk of flashover increases as the voltage difference between the two conductors increases, and it decreases as the distance between the two conductors increases. Thus, when high voltages are transmitted along adjacent lines, this is a significant issue. The summation of the three lines causes the voltage to drop to zero, eliminating this concern.

Chapter 1: Basic Electricity, Electrical Concepts, and Circuits

In addition, the three-phase system is much more efficient than the former one-phase system because (1) it delivers three times the electrical power and (2) it does not require a separate conductor to complete the circuit (i.e., a separate return path), because the return path for each one of the three conductors is one or both of the other two conductors. Another major improvement to the three-phase system occurs when the number of magnets on the rotor is increased. As the number of magnets on the rotor increases, the required rotational frequency decreases. For example, when the rotor contains two magnets, 60 Hz AC power is generated with a rotor spin frequency of 30 Hz. Some large rotors contain 60 magnets, necessitating only a single spin per second (1 Hz) to generate 60 Hz AC power. The presence of the three-phase power grid is evident everywhere. For example, the number of transmission lines on a high voltage power tower is always 3 or a multiple of three (6, 9, or 12 lines), the substations have their electrical devices (e.g., transformers, circuit breakers, and other components) in groups of three, and the local utility poles carry three lines. Although four lines may be observed on the power towers, the upper line is a lightning conductor that functions to protect the system from lightning strikes. It is connected to the earth through the power tower and transmits any lightning safely to the earth. The high voltage of the three-phase power passes through a step-down transformer at the substation, after which it is transmitted along utility poles, via distribution feeders, to their point of usage. The wooden poles carry four lines, one for each phase (the three high-voltage distribution feeders) and a neutral wire. The neutral wire is required because the loads on each line are different (i.e., do not sum to zero). Thus, the unbalanced current is conducted along the neutral wire. The neutral wire is connected to ground at the substation and also at the circuit breaker panel at the point of usage. A single-phase transformer is mounted on the utility pole outside of the residence receiving the power. It is fed by one of the three distribution feeders traversing the utility pole. The transformer steps the voltage down to 240 volts. However, the secondary coil of the transformer is center tapped and the center tap is connected to a neutral wire. Three lines leave the transformer secondary – a neutral wire and two hot wires. The two hot wires are each 120 volts with respect to neutral

and 180 degrees out of phase with each other. These lines enter the circuit breaker panel of the home (after passing through the electrical meter outside the home). The neutral wire attaches to the safety ground of the circuit breaker panel and the two hot wires attach to the main circuit breaker of the panel. The main circuit breaker allows the electricity to the entire house to be interrupted via one switch. From this main circuit breaker, the hot wires are connected to individual circuit breakers. Because the hot wires are 180 degrees out of phase, there is 240 volts between them and they are used to power appliances requiring 240 volts, such as the hot water heater. The 240-volt circuits of the circuit breaker panel have a pair of circuit breakers, one for each hot wire, and are connected to each other so that they are always interrupted or engaged simultaneously. The National Electrical Code has provided terms for the various voltage ranges. The extremes are termed low voltage (1 million volts) (Koumbourlis, 2002). Although the voltage levels required by consumers are low, when electrical power is transmitted over large distances, it is transmitted at much higher voltage values. This avoids large power losses during transmission (due to the conversion of electrical energy into thermal energy). Recall that power is proportional to the product of the voltage and the current (P = VI). Because the power being transmitted is constant, V and I are inversely related (P = VI). Thus, as one goes up, the other goes down, and as one goes down, the other goes up. For example, for a constant power, an increase in voltage leads to a decrease in current; likewise, a decrease in voltage leads to an increase in current. Because V = IR (Ohm’s law), we can substitute IR for V in the power equation. Thus, P = V
I is equivalent to P = IR
I, which simplifies to P = I2R. The latter equation is referred to as the power loss equation. From this equation, it can be seen that the power loss is proportional to the square of the current. Thus, when the current is doubled, the power loss is quadrupled. Consequently, high current values result in high power losses. For this reason, high voltage transmission is preferred to high current transmission. This is best appreciated through an example. Given a constant power of 1,000,000 watts and a resistance of 25 ohms, when the power is transmitted at a strength of 10,000 V, the power loss is 250,000 watts, which is 25% of the power being transmitted.

29

Section 1: Introductory Chapters

This is calculated by first solving the power equation for the value of the current (I) and then plugging that value into the power loss equation: P ¼ VI, therefore I ¼ P=V I ¼ 1,000,000 watts=10,000 volts ¼ 100 amps P ¼ I2 R ¼ ð100Þ2
25 ¼ 10,000
25 ¼ 250,000 watts 250,000 watts=1,000,000 watts ¼ 0:25 ¼ 25% When this same magnitude of power is transmitted at high voltage (e.g., 250,000 V), the power loss is much lower: I ¼ P=V ¼ 1,000,000 watts=250,000 volts ¼ 4 amps P ¼ I2 R ¼ ð4Þ2
25 ¼ 16
25 ¼ 400 watts 400 watts=1,000,000 watts ¼ 0:0004 ¼ 0:04% Therefore, as demonstrated by this example, with high voltage transmission, the percentage of power loss drops significantly (from 25% to 0.04%). In the United States, the standard utility AC (i.e., household electricity) that is available via an electrical wall outlet ranges from 110 V RMS to 130 V RMS (defined as 117 V RMS but, for discussion purposes, is often rounded to 120 V). As previously stated, each house receives two 120 V wires and a neutral wire. The two 120 V lines are 180 degrees out of phase, and thus, when a device is connected across them, it receives 240 V (e.g., AC units, electric dryers, water heaters, electric ranges, and other high-power appliances). Each circuit from the electric box uses a 120V wire and a neutral wire. Each 3-prong receptacle is designed for a 120 V plug. Each 3-prong receptacle receives a 3-prong connector (i.e., the plug at the end of the power cord of the electrical device being powered) designed for 120 V. When the connector is plugged into the wall, it completes the circuit by connecting the hot wire and the neutral wire of the power cord with the hot wire and the neutral wire of the circuit box. The ground wire connection between the wall receptacle and the plug is for safety purposes and, under normal circumstances, carries no current.

Transformers Because of the magnetic field created by current flowing through a coil of wire, whenever two coils of wire are adjacent to each other, the magnetic field of one coil induces a current in the other coil. Transformers are electrical devices that contain two

30

adjacent coils of wire. They are used to convert AC sine waves from one RMS voltage value to another one through inductive coupling (electromagnetic induction). The frequency of the voltage does not change, only its magnitude. A transformer contains two windings, one on the left (the primary) and one on the right (the secondary), each of which consists of wire wrapped around a laminated iron core. The two windings are in close proximity and their iron cores serve to concentrate the magnetic field between them. The primary winding is the one to which the electricity is applied (the input), whereas the secondary winding is the one from which the electricity is derived (the output). Thus, the magnetic field created by the changing current in the primary winding induces a changing current in the secondary winding. The ratio between the number of turns of the two windings (the primary-to-secondary turns ratio) dictates the output voltage value (see Figure 1.20). With a step-down transformer (decreases the incoming voltage), the primary has more turns than the secondary, whereas with a step-up transformer (increases the incoming voltage) it has less. For example, if the turns ratio is 10:1 (i.e., a step-down transformer) and the AC input voltage across the primary winding is

V1

V2

N1/N2 = V1/V2 V2 = V1 ´ N2/N1 Figure 1.20 A transformer uses electromagnetic induction to transform voltage levels in AC circuits. The voltage across the primary and the secondary are represented by V1 and V2, respectively. The number of turns on the primary and secondary are represented by N1 and N2, respectively. The turns ratio is equal to the voltage ratio. The power is constant (P1 = P2). In this illustration, N1 = 7 and N2 = 5, so this is a step-down transformer where the output voltage (V2) is equal to 5/7 of the input voltage (V1).

Chapter 1: Basic Electricity, Electrical Concepts, and Circuits

e – e – e– e –

Electron flow e– e– e– e– e– e– e– e– e– e– e– e– e– e–

e– e–

+ – –

+ –

e– e– e– e

– – e– e e – e e– e– e– e– e– e– e– e– e– e– e– e– e– e– e– e– e– e– e– Electron flow

Figure 1.21 When a capacitor is connected in series with a battery, electrons accumulate on the plate attached to the cathode, causing that plate to become more and more negative as more and more electrons accumulate on it. The electron accumulation, in turn, drives electrons from the opposing plate to the anode, causing the opposing plate to become more and more positive as more and more electrons depart from it. This movement of electrons continues until the voltage across the capacitor plates equals the voltage of the battery, at which point electron movement ceases. (Modified with permission from Ferrante MA, What We Measure and What It Means, 2012.)

120 V RMS, then the output voltage is 12 V RMS. When the coils on the primary equal those on the secondary, the input voltage is equal to the output voltage. This is termed an isolation transformer and is used in medical equipment, such as EMG machines, to isolate the secondary from the primary. In this manner, the patient can be isolated (protected) from the primary circuit. There are other ways to isolate the circuitry connected to the patient from the main circuitry (discussed later). Transformers are also used within stimulators to separate the patient from the power supply (discussed in Chapter 2) and to lessen the degree of stimulus artifact (discussed in Chapter 18). Because transformers cannot pass DC signal, they can be used to remove DC signal from a signal containing both DC signal and AC signal.

Electromagnetic Radiation When charged particles, such as electrons, are at rest, they generate electric fields in the space surrounding them. These electric fields produce a force (termed an electrostatic force) on other charges located within the field. As previously discussed, the magnitude of this force is quantifiable through the application of Coulomb’s law. Because force is a vector, in addition to magnitude, electric fields have direction (i.e., they are vector fields). Although these fields extend to infinity,

they exponentially diminish with distance. Because the particles within an electric field have mass, they cannot move at the speed of light. Nonetheless, their effects within the electric field are exerted at the speed of light. When charged particles move, they generate magnetic fields. When the voltage difference between two objects is constant or when a wire carries a constant DC source, static electric and magnetic fields are produced that rapidly weaken as they extend into space. When the voltage between two neighboring objects varies or when the current or voltage in a wire either fluctuates in intensity or changes in direction, an electromagnetic field is produced. Therefore, all AC signal-carrying wires produce 60-Hz electromagnetic fields. In this setting, a changing electric field gives rise to a changing magnetic field, which in turn gives rise to another changing electric field, and so on. These continuous conversions advance at the speed of light and can extend great distances. Not only are electromagnetic fields created by charges, but electromagnetic fields act on charges. As a result of this latter property, electromagnetic fields can interfere with AC signal transmission. The level of electromagnetic emissions from the electrical circuitry within the walls and ceiling of a room can produce electromagnetic interference capable of detrimentally affecting electrical equipment within the room. To avoid such electromagnetic intrusions, cords and cables can be shielded. A typical electrical cord or cable consists of internal wire conductors, separated from each other by insulation, and wrapped in a jacket. By the addition of an external conducting layer (e.g., wire mesh), cords and cables can be shielded from such electromagnetic interference. Coaxial cable is an example of a shielded cable. Coaxial cable consists of a single, centrally located wire conductor (carries the AC signal), wrapped in insulation (or unwrapped, with air functioning as the insulator), and covered externally by a cylindrical conductor (the electromagnetic shield). Because the electromagnetic shield is grounded, it conducts external (i.e., unwanted) signals to ground, keeping them from interfering with the centrally located (desired) AC signal. It also serves to keep the transmitted signal from escaping into the environment. Signal interference by electromagnetic radiation is a major concern of practitioners of EDX medicine. This is discussed in much greater detail later in this textbook (see Chapter 18).

31

Section 1: Introductory Chapters

References Bonner FJ, DevlescHoward AB. AAEM minimonograph #45: the early development of electromyography. Muscle Nerve 1995;18: 825–853.

32

Koumbourlis AC. Electrical injuries. Crit Care Med 2002;30[suppl.]: S424–S430. Moller P, Kramer B. Review: electric fish. BioScience 1991;41:794–796. Pollock S. Particle physics for nonphysicists: a tour of the microcosmos. Chantilly, VA: The Teaching Company, 2003.

electromyography. Muscle Nerve 1991;14:937–946. Gibilisco S. Electricity demystified, 2nd ed. New York, McGraw Hill, 2012. Kling LB. Basic electricity. Piscataway, NJ: Research & Education Association, 2015.

Bullock TH. Electroreception. In Springer handbook of auditory research, vol. 21. Bullock TH, Hopkins CD, Popper AN, Fay RR (eds): 2005:5–7.

Recommended Reading

Misulis KE. Basic electronics for clinical neurophysiology. J Clin Neurophysiol 1989;6:41–74.

Hawking S. A brief history of time. Toronto: Bantam Press, 1988:77.

Barry ET. AAEM minimonograph #36: basic concepts of electricity and electronics in clinical

Wolfson R. Understanding modern electronics. Chantilly, VA: The Great Courses, 2014.

Chapter

2

Instrumentation

Introduction In this chapter, the topics reviewed in Chapter 1 are applied to the individual components of an EMG machine. The function of the electromyography (EMG) machine is to collect, amplify, and display the desired signal (for visual and auditory interpretation) while simultaneously removing (filtering out) the undesired signal. The focus of this chapter is on basic instrumentation as it relates to electrodiagnostic (EDX) testing. The first electronic component to be discussed is the filter. To fully understand signal filtering, a deep understanding of capacitance and impedance is mandatory, the basics of which were discussed in the previous chapter (Chapter 1). In this chapter, the discussion of capacitance and impedance is advanced. In Chapter 1, the two forms of the Ohm’s law equation were introduced: V = IR (for resistance in DC circuits) and V = IZ (for impedance in AC circuits), where V is voltage (in Volts), I is current (in Amperes), R is resistance (in Ohms), and Z is impedance (in Ohms). Although the V = IZ form of Ohm’s law was only discussed with AC circuits, it also applies to DC circuits. Recall that impedance (the opposition to current advancement) consists of two components: (1) resistance and (2) reactance. Because resistors do not have reactance, their impedance is limited to their DC resistance (in Ohms). As a result, the V = IR form of Ohm’s law is used. Because capacitors are frequencydependent resistors, their impedance to current flow consists solely of reactance (i.e., they do not have resistance). Capacitors only demonstrate DC resistance while they are charging (discussed below). Thus, once charged, the V = IZ form of Ohm’s law is used. Impedances combine in series and parallel in exactly the same manner as do resistances: series impedances add and parallel impedances add in reciprocal. Although some of the information included in the next two sections is beyond the scope of knowledge

necessary for the performance of quality EDX medicine, it is included here so that the interested reader can review it without the need to self-assemble it from a large number of complicated, jargon-filled, engineering sources and because it is essential in understanding environmental noise and its elimination (see Chapter 19).

Capacitance A capacitor has a transient (DC) response and a steady state (AC) response. The transient response refers to its impedance to current as it is charging (DC signal). Once it is charged, a capacitor offers infinite resistance to the passage of further current. The steady state response of a capacitor refers to its response to continuously changing voltage (AC signal). The basic relationship between the charge on a capacitor and the voltage across it is provided by the formula Q = CV, where Q is the total charge in the capacitor, C is the capacitance of the capacitor, and V is the voltage across the capacitor. The equation shows that the voltage across the capacitor is proportional to its charge. The rate of charge flow into the capacitor is the current. When we differentiate the Q = CV relationship with respect to time, a relationship for current through the capacitor is generated: Q ¼ CV d d Q ¼ CV dt dt Because the change in charge (dQ) divided by the change in time (dt) is the definition of current (I), the equation can be rewritten and then manipulated as follows: d CV dt d I¼C V dt I¼

33

Section 1: Introductory Chapters

I¼C

dV dt

where I is the instantaneous current through the capacitor (in amperes), C is the capacitance (in farads), and dV/dt is the instantaneous rate of voltage change (in volts per second). This equation represents Ohms law for a capacitor. The key feature of this equation is that current only flows when the voltage is changing (dV) over time. The voltage across a capacitor connected in series with a battery (a DC circuit) is only changing while it is charging and discharging. Once charged or discharged, there is no current flow. For this reason, in a DC circuit, the response of the capacitor is transient (temporary). Conversely, in an AC circuit, where the voltage across the capacitor is continuously changing, the response of the capacitor is continuous (steady state).

Impedance The impedance of a circuit simply reflects its opposition to current flow. Impedance is a complex quantity that has magnitude and phase. Complex arithmetic is required to preserve the phase relationship between current and voltage in the circuit. Complex numbers are expressed as vectors in a twodimensional plane. Impedance, Z, equals a + ib, where i is a mathematical operator that indicates a 90-degree phase shift. To avoid it being confused with current, the letter i is replaced by the letter j. For a resistor, current and voltage are in phase, whereas for a capacitor, current and voltage are 90 degrees out of phase. Thus, for a capacitor, the impedance is –j/ωC, where –j indicates that the voltage waveform (the voltage sine wave) lags the current waveform (the current sine wave) by 90 degrees and 1/ωC represents the magnitude of the capacitive impedance, where ω is the angular velocity (which equals 2πf ) and C is the capacitance (in Farads). Thus, the impedance of a capacitor can be written as –j/2πfC. Because the frequency is in the denominator, the magnitude of the impedance of a capacitor decreases with increasing frequency (i.e., at higher frequencies, it offers less opposition to current flow) and increases with decreasing frequency. Thus, at low frequencies, the capacitive impedance is very large (approaches an open circuit); it is infinite at 0 Hz (DC signal). Conversely, at high frequencies, the capacitive impedance becomes very small (approaches a short circuit).

34

A resistor in series with a capacitor has a total impedance of Z = R – j/ωC, where R is the magnitude of the resistive impedance and 1/ωC is the magnitude of the capacitive impedance. For the purposes of this textbook, we are concerned only with the magnitude of the impedance and not with the phase shifts. For this reason, we can ignore the j variable. In the same manner that resistors are placed in series to form voltage dividers for DC signals, resistors and capacitors are placed in series to form voltage dividers for AC signals. Because the impedance of a capacitor, unlike that of a resistor, is frequency-dependent, its ability to divide the voltage is also frequency dependent. It is this frequency dependence that is the basis for an analog filter.

Filters It is important for EDX providers to understand filters because their inappropriate use has negative consequences. As previously stated, the capacitor has a transient (DC) response and a steady state (AC) response. Because a capacitor is a linear element, these two responses sum to form its complete response. To simplify the material contained in this section, the frequency-dependent behavior of a capacitor is reviewed based on its transient response rather than on its steady state response. Although this more simplistic approach is not wrong, it is not completely accurate. Its usage here is to make filtering more readily understood. Unlike a sine wave, which has a single frequency, such as the unwanted 60 Hz AC environmental signal (see Chapter 1), the compound bioelectrical signals collected during EDX testing are composed of a mixture of low-, middle-, and high-frequency components. These compound bioelectrical signals represent summations of individual actions potentials (APs). In addition, even the individual APs composing these compound potentials are themselves composed of low-, middle-, and high-frequency components. All waveforms, regardless of their complexity, can be mathematically broken down into their subcomponent sine waves, all of which have uniform frequencies and amplitudes. The compound potentials collected during EDX studies are displayed in two dimensions: voltage on the y-axis and time on the x-axis. Therefore, the slope of the signal is expressed as the change in voltage divided by the change in time (Δy/Δx). Simply stated,

Chapter 2: Instrumentation

those portions of the recorded compound electrical potential demonstrating the fastest changes in voltage per unit time (i.e., those with the steepest slopes, whether positive or negative) contain the faster signal components (i.e., the higher frequency components), whereas those regions with less significant changes in voltage over time represent regions containing more of the slower signal components (i.e., the slower frequencies). Because the period of a sine wave and its frequency are inversely related, as shown by the formula f = 1/T, the frequency of the sine wave is easily determined from its period (see Chapter 1). Using this same approach, the frequency of any portion of a compound electrical potential can be calculated from the duration of that portion of the potential. For example, if the rise time of a collected waveform is 0.8 msec, then the frequency of the rise time is 1,250 Hz (1 cycle/0.8 msec  1,000 msec/1 sec = 1,250 cycles per second = 1,250 Hz). Because there are 1,000 msec in 1 second, the frequency of a response component (e.g., the rise time) can be determined simply by dividing 1,000 msec by its duration (in msec). The importance of this example will become clear later in the textbook when the ideal interelectrode distance for sensory response recording is discussed. The signals recorded during an EDX study are composed of a range of frequencies, some of which are desired (e.g., desired signal) and some of which are undesired (undesired signal). The undesired signal is typically referred to as noise, an example of which is the 60 Hz power line artifact related to electromagnetic induction from the environment. Thus, we need electrical circuits that are able to remove the undesired signals (i.e., circuits with filtering ability). By incorporating filtering circuits, we are able to remove undesired signal frequencies (low, intermediate, or high) from the original signal, leaving the desired signal behind. We often remove frequencies from both ends of the original signal by running the original signal through both a low-frequency filter and a high-frequency filter. In this manner, a desired range of frequencies (termed a bandwidth of frequencies) is collected and the undesired frequencies are removed. In this manner, the signal-to-noise ratio is enhanced. Ideally, we do not want to remove any of the composite frequencies of the desired signal because this will distort its morphology and potentially generate misleading results. To avoid this error,

the effects of filtering must be intimately understood by the EDX provider. In modern-day electronics, analog filters (e.g., those composed of resistors and capacitors [termed RC circuits]) are no longer utilized. Instead, analog signal (continuously varying signal) is passed through transistors and digitized using complicated mathematical formulas. Following digitization, the digitized signal is manipulated (e.g., filtered). Digital signal consists of a series of on-off pulses representing ones and zeroes. In this manner, the frequencies outside of the frequency range of the desired signal (i.e., the unwanted signal) are removed. By amplifying the desired signal and eliminating the undesired signal, the signal-to-noise ratio is enhanced. Although no longer a prominent component of modern electronics, the RC circuit – which consists of a resistor (a resistive electrical device, providing equal resistance to all signal frequencies) and a capacitor (a reactive electrical device, providing greater resistance to the lower frequencies) connected in series with a power source – illustrates several important electrical concepts and remains a frequent source of neurophysiology board examination questions (see Figure 2.1). Thus, analog filtering is discussed in detail in this section. RC circuits remove (filter out) specific frequencies (undesired signal) from the recorded signal. Because all electrical components within a circuit offer resistance to electron advancement, the capacitor of an RC circuit can be thought of as a resistor and, therefore,

Figure 2.1 An RC circuit. In this circuit, a resistor and a capacitor are connected in series with a power source. When the power source is a battery, as current flows through the circuit, it produces a voltage drop across the resistor and it charges the capacitor. Once the capacitor is fully charged, current flow ceases and there is no longer a voltage drop across the resistor. When the battery is replaced by a source of alternating current, the direction-changing nature of the alternating current (AC) causes the capacitor to charge and discharge repeatedly. When the frequency of the AC signal exceeds 1/T, the capacitor never fully charges before the AC signal changes its direction, whereas when the frequency of the AC signal is below 1/T, the capacitor fully charges prior to the direction change of the AC signal. This concept is discussed in detail in the text.

35

Section 1: Introductory Chapters

as a voltage divider (for those frequencies that it resists). For this reason, the capacitor and the resistor in an RC circuit divide the voltage in proportion to their resistance. The voltage relationship between resistors in series was discussed in Chapter 1. Briefly, an RC circuit (resistor + capacitor) can be compared to a series circuit with two resistors (resistor + resistor), where both the capacitor and the resistor function as voltage dividers, dividing up the total voltage of the circuit between them based on their percentage share of the total resistance. According to Kirchhoff’s current law (i.e., resistors in series have the same current), in an RC circuit, the current across the resistor equals the current across the capacitor. This law reflects conservation of charge: at any point in the circuit, the incoming charge (current) equals the outgoing charge (current). Because resistors impede current at all frequencies, they are classified as frequency-independent resistors. The degree of resistance offered by a capacitor depends on its degree of charge. As they charge, capacitors offer more and more resistance to current advancement. Once fully charged, the capacitor offers infinite resistance to current flow. With direction-reversing current (AC signal), because the current continuously changes direction, the capacitor continuously charges and discharges. As discussed earlier in this chapter, the magnitude of the current on the capacitor depends on the rate of the change of the voltage: I¼C

dV dt

where I is the instantaneous current through the capacitor (in amperes), C is the capacitance (in farads), and dV/dt is the instantaneous rate of voltage change (in volts per second). Again, this equation represents Ohms law for a capacitor. As shown in the equation, current only flows when the voltage is changing (dV). To appreciate how capacitors function as filters, we need to recall the inverse relationship between frequency and the time period of the wave (i.e., its wavelength), the formula for the time constant (T = RC for DC signal and T = ZC for AC signal), that the length of a single cycle of AC signal is 2π, and the capacitor voltage formula (see Chapter 1): Recall that f = 1/T and T = ZC therefore, f = 1/ZC

36

Because a single cycle of AC signal = 2π, f ¼ 1=2πZC and f ¼ 1=2πT V ¼ 1  et=ZC V ¼ 1  et=T where f is the frequency, T is the time constant, R is resistance, Z is impedance, π is approximately 3.14, V is voltage, e is approximately 2.7, and t is the time in seconds. Because f = 1/T, whenever the AC signal frequency exceeds 1/T (f > 1/T), the AC signal reverses (changes direction) before the capacitor is fully charged. As a result, the capacitor is unable to offer resistance to frequencies faster than 1/T. However, when the frequency is less than 1/T, the capacitor fully charges before the AC signal reverses and, thus, offers infinite resistance to the signal prior to the reversal. Stated another way, the capacitor acts like a short circuit for higher frequencies (>1/T), permitting their unimpeded advancement, and behaves like an open circuit for lower frequencies (1/T) have been removed (available at the capacitor); and one from which the slower frequencies ( RMP).

Depolarization and Repolarization There are three gates involved in depolarization and repolarization: two sodium channel gates (one for activation and one for inactivation) and one potassium gate (for both activation and inactivation). All three of these gates open in response to different voltage values (i.e., they have different activation thresholds). Unlike the nongated channels, which dictate the RMP, the voltage-gated channels are responsible for membrane depolarization and repolarization and, hence, for action potential generation and

57

Section 1: Introductory Chapters

propagation. At this point, the basic information discussed earlier in this chapter can be integrated with new information to better describe the membrane changes associated with depolarization and repolarization. At rest, the voltage-gated sodium channels (VGNCs) and the voltage-gated potassium channels (VGKCs) are closed and the nongated sodium and potassium channels are open. Because the nongated potassium channels far outnumber the nongated sodium channels, the permeability of the membrane toward K+ is much higher than it is toward Na+. As a result, the RMP is dominated by K+ flow, and therefore, the TMP value of –70 mV is much closer to the Nernst equilibrium potential for K+ (–98 mV) than it is to the Nernst equilibrium potential for Na+ (+67 mV). The VGNC has two gates, only one of which needs to be closed for Na+ influx to be blocked. At the RMP, the Na+ activation gates are closed and the Na+ inactivation gates are open. As the polarity of the membrane decreases (depolarization), the activation gates of some of the VGNCs open, permitting Na+ entry. The degree of Na+ entry is limited, and only a small membrane depolarization results (a subthreshold depolarization) that is offset by an equal amount of positive charge exit (K+ efflux) via the nongated potassium channels. With larger subthreshold depolarizations, a larger number of activation gates open, and thus there is more Na+ influx and greater membrane depolarization. This causes more VGNCs to open and further Na+ entry. This positive feedback process continues until Na+ permeability exceeds K+ permeability, termed the depolarization threshold because, at this level of depolarization, all of the remaining VGNCs open and a large depolarization results. This occurs in the –55 to –50 mV range. At the depolarization threshold, gate opening is rapid, occurring within a few ten-thousandths of a second (the Na+ activation gates are fast gates, unlike the Na+ deactivation gates and the gates of the voltage gated K+ channels). Once all of the VGNCs are open, the Na+ permeability of the membrane is approximately 5,000-fold greater than it was at rest, and consequently, the TMP rapidly moves toward the Na+ equilibrium potential (+67 mV). However, the TMP does not reach +67 mV because the open sodium channel configuration is time-dependent. At about 1 millisecond, the inactivation gates assume their closed configuration and Na+ influx ceases. This occurs around +30 mV. At the same

58

time, the voltage-gated K+ channels (VGKCs) open. At this point, the inside of the cell is positive with respect to the outside of the cell. Thus, the electrical and chemical forces on K+ are both outward, the nongated and gated K+ channels are in their open configuration, and the VGNCs are closed. For this reason, K+ flux is even greater than it is at rest and rapid repolarization occurs. Unlike VGNCs, which demonstrate a time-dependent open state, VGKCs demonstrate a voltage-dependent open state. Thus, they remain open until the membrane voltage triggers their closure. Indeed, the potential of the membrane transiently moves beyond its resting value of –70 mV and even closer to the K+ equilibrium potential (i.e., this phenomenon is termed hyperpolarization). The point at which K+ permeability exceeds Na+ permeability represents the peak of the rising (depolarization) phase. Thus, selective permeability leads to the two phases of the action potential: depolarization (VGNC opening) and repolarization (VGKC opening).

The Absolute Refractory Period and the Relative Refractory Period There are important time periods during repolarization that require further discussion. During repolarization, the sodium activation gates close, at which point the Na+ activation and inactivation gates are both closed. In this configuration, a second action potential cannot be generated. This time period is referred to as the absolute refractory period. Because some of the VGKCs stay open for several milliseconds, the membrane is hyperpolarized (about –90 mV). During this period of membrane hyperpolarization, the Na+ channels return to their resting configuration (activation channel closed and inactivation channel open). In this configuration, an action potential can again be generated. However, because the membrane is hyperpolarized, a greater amount of depolarization is required to reach the depolarization threshold value. This time period is referred to as the relative refractory period. Thus, an action potential generated during the relative refractory period takes longer to reach threshold. The absolute refractory period serves to maintain unidirectional AP propagation. This is necessary because opening of the VGNCs results in focal Na+ entry with bidirectional Na+ flow along the axon. If the recently depolarized more proximal segment could be immediately depolarized, further Na+ entry would occur, and this process would

Chapter 3: Anatomy and Physiology of Neurons

Figure 3.3 The relationship between sodium conductance, potassium conductance, and the action potential.

continue until the Na+ gradient was exhausted, leaving the membrane inexcitable. Following repolarization, Na+ and K+ are moved against their concentration gradients, thereby returning their concentration gradients to their baseline values (see Figure 3.3). Again, because the actual number of Na+ and K+ ions traversing the membrane during depolarization and repolarization is miniscule in comparison to the total number of ions within each compartment, there is no substantial change in the compartment concentrations of either ion during these events. Consequently, these concentration gradient changes across the membrane are immeasurable.

Action Potential Generation Action potential generation occurs when the sum of incoming excitatory and inhibitory local potentials reaches the threshold value for depolarization. This occurs at membrane regions with a high VGNC density, such as the axon hillock and the initial segment of the axon. The axon hillock is the distalmost portion of

the cell body, and the initial axon segment is the thick, unmyelinated initial segment of axon that projects from the axon hillock. Action potentials are more frequently generated at the initial axon segment than at the axon hillock. Sites with greater VGNC density generate denser Na+ currents. With greater degrees of Na+ current density, the rate of depolarization is quicker (termed accommodation). The VGNC density of sensory neurons is about 1 channel per square micrometer, whereas it is approximately 150 channels per square micrometer at the axon hillock and the initial axon segment. It is approximately 1,500 channels per square micrometers at the nodes of Ranvier, which ensures that the action potential will be propagated past the node (discussed later here).

Local Responses As stated earlier, because of its phospholipid bilayer structure, the membrane separates charges. This charge-separating ability gives the membrane both

59

Section 1: Introductory Chapters

Extracellular fluid +

gNa+

TMP

ENa+

gK+

gLeakage

EK+

ELeak

+ –

TMC

gNa+

ENa+

– Intracellular fluid

gK +

EK+

gLeakage

+ –

TM

ELeak

N

Figure 3.4 The electrical circuit equivalent of the membrane.

capacitance and voltage. The total capacitance of the membrane, termed total membrane capacitance, is proportional to the amount of charge stored on it. The charge storing ability of the membrane is directly proportional its surface area and inversely proportional to its thickness, as indicated by the formula for capacitance, C = A/d, where C is capacitance, A is the area of overlap, and d is the distance between the charges (see Chapter 1). The resistance of the membrane to ion flow reflects the number of open ion channels, all of which are perpendicular to the membrane and, thus, parallel to each other. As the permeability of the membrane toward an ion increases (more open channels), the total membrane resistance toward that ion decreases. Because of its capacitive and parallel resistive properties, the cell membrane can be modeled as an RC circuit where the capacitor and resistor are oriented in parallel (see Figure 3.4; WF Brown, 1984, p. 9). The term cable properties refers to those electrical properties of the cell that are passive in nature (i.e., unrelated to the dynamic changes that are associated with changes in ionic permeability). Cable properties are also termed electrotonic properties. When a square wave current pulse is delivered to the cell membrane (termed a transmembrane current), the voltage across the membrane (the transmembrane voltage or TMP) increases, as determined by its resistance and its capacitance. If the membrane only had resistive properties, the voltage generated by the transmembrane current would be instantaneously identical to the square wave current pulse. However, because of its capacitive properties, the voltage builds up slowly rather than instantaneously.

60

Figure 3.5 Transmembrane capacitance.

Recall from Chapter 1 that the time constant (T) of a capacitor is the time required for it to charge or discharge by 63%. Because the membrane has capacitance, it also has a time constant. Like other capacitors, the time constant of the membrane is defined as the time it takes for the membrane to reach 63% of its final voltage value (TMP) following the application of the square wave current pulse (or to decrease by 63% of its final value when the square wave current pulse ceases) (see Figure 3.5). As discussed in Chapter 1, the time constant of the membrane is equal to the product of its resistance and its capacitance (T = RC). Because larger nerve fibers have larger membrane surface areas, they hold more charge and, thus, have larger capacitances; consequently, they have larger time constants. As discussed in Chapter 1, charge does not flow across a capacitor, but rather the buildup of ions on one side displaces the ions on the other side (capacitive current; displacement current). Thus, a large charge on one side of the membrane requires a large charge on the other side of the membrane to displace it.

Chapter 3: Anatomy and Physiology of Neurons

Local potentials (also termed electrotonic potentials or graded potentials) are small, non-propagated potentials that are generated at a single site. Their magnitude reflects the density of the ion channels leading to their formation. Although these potentials conduct away from the generation site, they decay over short distances (a few millimeters). Thus, they are not propagated along the membrane like an action potential. Examples of local potentials are excitatory postsynaptic potentials (EPSPs) and inhibitory postsynaptic potentials (IPSPs) that are generated at excitatory and inhibitory synapses, respectively. The EPSPs and IPSPs summate with each other, both spatially and temporally. Through summation, these potentials increase in size, which, in turn, allows them to advance further prior to their decay. When one of these summated potentials reaches an area of membrane with a high concentration of VGNCs (e.g., the axon hillock or initial axon segment of an AHC) and is large enough to depolarize the membrane to its depolarization threshold, VGNC opening occurs and an action potential is generated. This is also what occurs at the neuromuscular junction. The opening of the acetylcholine receptor channel allows Na+ to traverse the membrane, thereby locally depolarizing the postsynaptic muscle membrane. If the depolarization reaches the depolarization threshold of the postsynaptic muscle membrane, an action potential is generated that propagates bidirectionally from that site (see Chapter 4). But how does the AP propagate along the axon? VGNC opening leads to a local increase in Na+ ions. The incoming positivity of the Na+ moves away from its entry point, thereby neutralizing some of the negative charges along the internal aspect of the membrane. This releases some of the positive charges located along the external aspect of the membrane. In other words, the Na+ entry increases the internal positivity, which, in turn, displaces the external positivity (i.e., capacitive current; displacement current). In this manner, the membrane segment adjacent to the initial site of depolarization becomes depolarized, opening its VGNCs and permitting the entry of Na+ at the adjacent segment. In other words, depolarization of one segment triggers the depolarization of the next segment. In this manner, the internal Na+ current is continuously rejuvenated as it propagates along the axon. This type of action potential propagation is termed continuous propagation because the action potential advances in a continuous manner (see Figure 3.6).

++ +++

++ +++

– +

+ + ++ + +

– +

++ +++

++ +++

+ +++ + + +++ + +++ + +++

Figure 3.6 Action potential propagation via continuous conduction. In the illustration, when the sodium channel opens (indicated by the inwardly directed arrow-shaped structure), sodium ions rush in, causing the intracellular space to become locally more positive and the extracellular space to become locally more negative. This extracellular charge attracts adjacent sodium ions, causing the adjacent membrane to become relatively more negative, depolarizing the adjacent membrane, and opening the adjacent sodium channels. This process continues unidirectionally (e.g., following sensory nerve fiber activation via a sensory receptor) or bidirectionally (e.g., when depolarization is produced in the EMG laboratory through the stimulating electrodes of a handheld stimulator), such as is illustrated here.

The Advantage of Myelin Nerve segments with higher capacitance (higher charge) require more time for the charge to be displaced (i.e., more time to discharge), whereas membrane segments with lower capacitance require less time. Thus, membrane capacitance dictates the time to threshold (i.e., the time to VGNC opening), not only at the site of initial action potential generation, but also at the subsequent sites. The speed of action potential propagation depends on many factors, but primarily on whether the nerve is myelinated or unmyelinated (discussed in detail later in this chapter). Myelination of the axon increases action potential propagation speed. With unmyelinated nerve fibers, the action potential propagates via continuous conduction (about 5 m/sec), whereas with myelinated nerve fibers, conduction is discontinuous, because only the nodal membrane must be discharged. The myelinated segments (internodes) do not have opposing charges across them because these segments are too thick (the ions are too far apart to affect each other). Thus, unlike with unmyelinated nerve fibers, in which the dielectric of the capacitor is the center of the membrane (i.e., the lipid layer), with myelinated nerve fibers, the myelin is the primary dielectric. The type of action potential propagation associated with myelinated nerve fibers is termed saltatory conduction (i.e., leaping conduction) because the charge displacements along the axon occur discontinuously (only at the nodes of Ranvier). Saltatory conduction was first recognized in the 1940s (Falck and Stalberg, 1995). The myelinated segments (internodes) range in length from 0.2 to 2.0 mm (Falck and

61

Section 1: Introductory Chapters

Stalberg, 1995). For a heavily myelinated axon, they are about 1–2 mm in length. The unmyelinated segments (nodes of Ranvier) are 0.001–0.002 mm in length. Consequently, the ratio of myelinated membrane (i.e., internodal membrane) to unmyelinated membrane (nodal membrane) of a more heavily myelinated nerve fiber is at least 1:0.002 (500:1). Thus, among nerve fibers with this ratio of myelinated membrane to nodal membrane, the capacitance is reduced by 99.8%. The time constant of a capacitor (i.e., the time required for it to charge or discharge by 63%) is proportional to the product of its resistance and capacitance (T = RC) (see Chapter 1). By decreasing membrane capacitance, myelination lowers the time constant of the membrane, allowing it to charge and discharge much quicker, thereby increasing the speed of action potential propagation along the axon. Because open channels permit transmembrane current leakage, the TMP decreases in size (decays) as it propagates along the axon. Should it decay to a value below the depolarization threshold, it will stop advancing. Because myelin also functions as an insulator, it decreases the amount of current that leaks across the membrane (termed transverse current or transmembrane current), thereby permitting the internal current to advance further down the axon. The length constant (also called the space constant) is another concept pertinent to action potential propagation. Like the time constant, the length constant is defined mathematically – the distance at which the potential decays to 37% of its original value (i.e., the distance at which it has decreased by 63%). The length constant, which is symbolized by the Greek letter lambda (λ), is equal to the square root of the membrane resistance (Rmembrane) divided by the axon resistance (Raxon): λ¼

pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Rmembrane =Raxon

As demonstrated by the above equation, the length constant is proportional to the membrane resistance (the higher the membrane resistance, the lower the current leakage, and the further down the axon the current advances) and inversely proportional to the resistance of the axon (axon resistance impedes current advancement). When the transmembrane resistance is much higher than the resistance of the axon, current advances further down the axon. The resistance of the axon to current advancement is

62

inversely proportional to its diameter (just like water flowing through a pipe) and directly proportional to the resistivity of its axoplasm and its length. Thus, axons of larger diameter have lower axon resistance (and, hence, longer length constants), whereas thinner diameter axons have higher axon resistance (and, thus, shorter length constants). In summary, the current travels further down the axon when its membrane resistance is high (impedes current leak) and its axon resistance is low (facilitates current advancement).

Action Potential Propagation Speed Unmyelinated fibers are less than 1.5 micrometers in diameter and have conduction velocities in the 0.5–2.0 m/sec range. Lightly myelinated fibers with diameters up to 3 micrometers have conduction velocities in the 3–15 m/sec range, whereas those with diameters up to 6 micrometers demonstrate conduction velocities in the 5–30 m/sec range. Heavily myelinated fibers have diameters of 12–20 micrometers and have conduction velocities in the 70–120 m/sec range. The latter include type 1a afferent fibers from muscle spindles and Golgi tendon organs and the efferent fibers innervating skeletal muscle. Herbert Gasser originally reported this relationship between nerve fiber diameter and nerve conduction velocity in 1939 (Gasser, 1939). Clinically, the smaller-diameter nerve fibers (the slowly conducting unmyelinated and lightly myelinated fibers) convey pain, temperature, and crude touch and belong to the exteroceptive system (vital sensation). The larger-diameter fibers (the faster conducting, more heavily myelinated fibers) convey proprioception (Golgi tendon organs), vibration (Pacinian corpuscles), and discriminative touch (Meissner corpuscles) and belong to the proprioceptive system (gnostic sensation). Among myelinated fibers, as the thickness of the myelin increases, the capacitance of the membrane decreases, generating faster action potential propagation speeds. Certain nerve fiber parameters are associated with optimal action potential propagation speed (conduction velocity). Regarding the ratio of the axon diameter to the nerve fiber diameter (i.e., the relationship between the axon diameter and the thickness of the myelin), the optimal conduction velocity occurs when the diameter of the axon is 60% of the diameter of the nerve fiber (i.e., when the ratio is 0.6). Regarding the radius, this would be a 3:2 (axon: myelin) relationship

Chapter 3: Anatomy and Physiology of Neurons

20%

60%

20%

Figure 3.7 The relationship between axon diameter and nerve fiber diameter that generates the optimal conduction velocity occurs when the axon diameter is approximately 60% of the nerve fiber diameter. Electrically, this ratio (0.6) reflects the ideal relationship between the time constant and the length constant.

(see Figure 3.7). The conduction velocity of a myelinated nerve fiber (in meters/second) can be estimated based on its diameter in micrometers – the nerve conduction velocity (in meters/second) is about 5.5 times its diameter (Waxman and Bennett, 1972). Thus, a heavily myelinated axon with a diameter of 12 micrometers would have an approximate conduction velocity of 66 m/sec (5.5  12 = 66), which is equivalent to approximately 160 miles per hour (66 m/sec  3,600 sec/h  1 km/1,000 m  1 mile/ 1.61 km = 148 miles/h). A more heavily myelinated axon with a diameter of 20 micrometers would have an approximate conduction velocity of 110 m/sec, which is equivalent to 246 miles per hour. The distance between nodes (i.e., the internodal distance) also contributes to action potential propagation speed. Nerve fibers with longer internodal segments conduct faster because their ratio of internodal membrane to nodal membrane is lower. When considered with respect to nerve fiber dimeter, the optimal conduction velocity occurs when the length of the internode is about 100-fold greater than the outer diameter of the nerve fiber (i.e., a ratio of 100:1) (Rasminski, 1971). Shorter depolarization threshold times (i.e., the time it takes to move the membrane potential from its resting value to its depolarization threshold value) are associated with faster nerve conduction velocity values. Thus, as the VGNC density of the membrane (i.e., the number of VGNCs per unit area of membrane) increases, the depolarization threshold is reached earlier, thereby increasing the nerve conduction velocity. The conduction along unmyelinated axons is continuous and occurs at a velocity of about 5 m/sec. Once a region of membrane is depolarized, sodium ions enter and move down the axon (the internal longitudinal current). The action potential decays as its passes down the axon due to membrane capacitance, incomplete membrane resistance, and axon

resistance. Because the sodium current decays over short distances, it must be rejuvenated as it advances. When the Na+ current reaches the adjacent membrane segment, it induces a capacitive current that depolarizes the membrane just in front of it (i.e., its positivity displaces positive charges along the external aspect of the membrane). This occurs along the entire axon in an all-or-nothing manner and is referred to as continuous conduction. An example of an unmyelinated nerve fiber is a C fiber (e.g., temperature perception). The action potential is generated at the distal end of the axon through receptor activation, after which it propagates proximally along the axon to the sensory neuron. Because the axon is unmyelinated, action potential propagation occurs in a continuous manner and requires that the capacitance of the entire axonal membrane be discharged. For this reason, the time required to traverse the axon will depend on the surface area of the axon membrane (the product of the circumference of the axon [2πr] and its length [L]), the resistance of the membrane to current leakage, and the resistance of the axon to current advancement. As discussed earlier in this chapter, the membrane of unmyelinated axons (and muscle fibers) contains uniformly distributed VGNCs and VGKCs that are responsible for the depolarization and repolarization phases of the action potential. With saltatory conduction, the AP propagates at a much faster rate, with intermittent delays occurring at the nodal membrane due to the need for nodal membrane discharge (i.e., the nodal membrane has capacitance) and the time required to depolarize the nodal membrane from its resting potential to its depolarization threshold potential (see Figure 3.8). Unlike the uniform distribution of the VGNCs and the VGKCs of unmyelinated axons, the voltage-gated channels of myelinated axons are not distributed uniformly. The nodal segments of the axon contain high densities of VGNCs, whereas the internodal membrane segments contain higher densities of VGKCs, especially in the paranodal/juxtaparanodal regions (see Figure 3.9). Although we discussed saltatory conduction and the jumping of the action potential from one node to the next node, an action potential encompasses more than one node simultaneously. Using the formula for distance (distance = velocity  time), for a myelinated nerve fiber with a conduction velocity of 60 m/sec (this is identical to 60 mm/msec) and an action potential duration of 0.2 msec, the distance is 12 mm

63

Section 1: Introductory Chapters

C

A

B

Myelin

Myelin

Unidirectionally vs bidirectionally propagatingAPs

B

Myelin

+30 –55 –70 Myelin

C A Figure 3.8 Action potential propagation via saltatory conduction. As the sodium current passes distally down the axolemma, its arrival at a segment of nodal membrane (the site with a high density of VGNCs) causes the internal aspect of the axolemma to become more positive than the external aspect, thereby reversing the transmembrane potential (TMP). At rest, the TMP is approximately –70 mV, but with the opening of the VGNCs, the resistance to Na+ influx dramatically increases and the TMP moves toward the equilibrium potential of Na+. Because the VGNCs only open transiently, the TMP does not reach the latter value. It usually peaks at about +30 mV. As the TMP passes from –70 mV to +30 mV, it passes through –55 mV, which is the membrane potential that triggers VGNC opening (i.e., the membrane depolarization value). The sodium current rushes in and moves proximally and distally along the axon. However, due to the absolute refractory period of the internodal membrane proximal to the node, action potential propagation only occurs distally. Because the nerve is located within a volume conductor, the sodium ions flow along an arcing pathway to enter the axon. The surface recording electrode senses four sequential sodium current flows: (1) Na+ moving toward the electrode (produces a downward deflection on the monitor), (2) Na+ moving away from the electrode (produces an upward deflection on the monitor), (3) Na+ moving toward the electrode (produces a downward deflection on the monitor), and (4) Na+ moving away from the electrode (produces an upward deflection on the monitor). These four deflections produce the triphasic wave (positive–negative–positive). The peaks and troughs are designated as A, B, and C in the figure.

Nodal membrane

Internodal membrane

Internodal membrane

Figure 3.9 Nodal and internodal membrane segments. The ion channel density is greatest at the nodal membrane, whereas the potassium channel density is greatest along the internodal membrane, especially adjacent to the nodal membrane (i.e., along the paranodal membrane).

(60 mm/msec  0.2 msec). For an internode length of 1 mm, this distance is equivalent to 12 myelinated segments. If the duration of the action potential is 0.5 msec, then the axon length involved is 30 mm (i.e., 30 internodes). Consequently, as action potentials propagate along the nerve fiber, they simultaneously encompass many nodes. The duration of the action potential includes both depolarization and repolarization. Because we are interested in depolarization, we will focus on the

64

duration of the rise time later in this textbook (see Chapter 8). In addition to conduction velocity differences between unmyelinated and myelinated axons, there are other differences demonstrated by the larger, more heavily myelinated nerve fibers that lessen the surface area of their nodal membranes, further increasing conduction velocity. As stated earlier, the axon is cylindrically shaped, and thus the surface area (SA) of the nodal membrane is approximated by the product of the circumference of the axon (πd) and the length of the node (L) as shown by the following formula: SA ¼ πdL The nodes of the larger, more heavily myelinated nerve fibers are shorter in length. This lessens the membrane surface area of the node, thereby decreasing its capacitance. In addition, the axons of the larger, more heavily myelinated fibers narrow at the node to about one-third of the internodal axon diameter (Reles and Friede, 1991). This also decreases the

Chapter 3: Anatomy and Physiology of Neurons

nodal membrane surface area. Finally, and not apparent in the above formula, the nodal membrane has a high VGNC density, thereby increasing the density of the sodium current and, consequently, lessening the time to reach depolarization threshold.

Connective Tissue Elements of Nerve Trunks The connective tissue elements of the peripheral nerve trunks include the endoneurium, perineurium, and epineurium. The endoneurium is the innermost layer. It surrounds the myelin sheath (and, hence, the Schwann cells), which, in turn, surrounds the axons. This delicate connective tissue layer is also referred to as the endoneurial sheath, Henle’s sheath, and the endoneurial tube. The endoneurium is composed of fibroblasts, loose collagen, capillaries (but no lymphatics), a liquid referred to as endoneurial fluid, and the basal lamina (see Figure 3.10). Historically, the endoneurial sheath and the myelin sheath were erroneously thought to represent a single structure, termed the neurilemma. Terms derived from this misconception (e.g., neurilemmoma) are outdated and should be avoided. The endoneurial tubes are bound into fascicles by another layer of connective tissue, referred to as the perineurium. The perineurium consists of several layers of perineurial cells. It is the neural component that predominantly provides the elasticity and tensile strength to the nerve trunk and, therefore, provides protection against stretch injuries (Sunderland, 1990). This layer also functions as a diffusion barrier (the blood–nerve barrier). The number of fascicles contained within a nerve varies along the length of the nerve from 1 to more than 100, depending on the specific nerve and segment assessed (Sunderland, 1990). In addition, there is marked fascicular rearrangement along the length of the nerve as small

groups of nerve fibers move from one fascicle to another fascicle with distal advancement (Sunderland, 1990). As nerve branches approach their exit site, they begin to occupy a more localized position in the nerve and a lesser number of fascicles, ultimately becoming superficial at the exit site for the neural branch. Although the nerve fibers ultimately join to form nerve branches at or below the elbow and knee, at the more proximal segments of the peripheral nervous system, they are more widely distributed, involving many and possibly all of the fascicles (Sunderland, 1990). In general, the number of fascicles contained within a nerve is proportional to the amount of mechanical stretch to which the nerve is exposed. Thus, for a given cross-sectional area of nerve, its tensile strength is proportional to the number of fascicles at that level (Sunderland, 1990). Thus, the number of fascicles present indicates the amount of mechanical stretch to which the nerve is exposed. The outermost layer of connective tissue is the epineurium. It lies between the fascicles (interfascicular epineurium) and around the entire nerve trunk (epifascicular epineurium) (Sunderland, 1978). The epineural layer is the most superficial (adventitial) layer and contains blood vessels and adipose tissue (see Figure 3.11). The epineurium serves to protect the fascicles from compressive injury. The relative cross-sectional area proportions of the fascicular tissue and the epineurial tissue vary along the length of the nerve. Overall, the fascicular contribution varies from 25% to 65%. The exception to this statement is along the gluteal segment of the

Endoneurium Endoneurial fluid Myelin Axon

Figure 3.10 Illustration of an endoneurial tube. The endoneurial tube is the endoneurium and its contents, in this case the endoneurial fluid and a myelinated axon.

Figure 3.11 The connective tissue layers composing a nerve fiber include the endoneurium, which surrounds the individual axons; the perineurium, which surrounds groups of axons (termed fascicles); and the epineurium, which lies between and around the fascicles.

65

Section 1: Introductory Chapters

sciatic nerve, where the fascicular contribution does not exceed 30% (Sunderland, 1990). Where nerves cross joints, they are more susceptible to injury (e.g., traction and compression). Consequently, the epineurial tissue is surrounded by loose connective tissue (provides mobility to the nerve trunks and lessens the likelihood of a stretch injury) (Ferrante and Wilbourn, 2015). In addition, across a joint, the number of fascicles increases and their size decreases, thereby rendering them less susceptible to compression (Sunderland, 1990). A series of nutrient arteries supply blood to nerves along their course. There is an extensive internal anastomosing network of blood vessels and an extensive external one that, together, render nerves resistant to ischemic insult related to loss of a regional nutrient artery. Within a fascicle, the largest caliber

References Bonner FJ, DevlescHoward AB. AAEM minimonography #45: the early development of electromyography. Muscle Nerve 1995;18:825–853. Brandstater ME, Lambert EH. Motor unit anatomy. Type and spatial arrangement of muscle fibers. In Desmedt JE, editor, New developments in electromyography and clinical neurophysiology. Bassel: Karger, 1973:14–22.

findings. J Clin Neurophysiol 1995;12:254–279.

ipsilateral hindlimb reflexes in cat. J Neurophysiol 1943;6:293–315.

Feinstein B, Lindegaard B, Nyman E, Wohlfart G. Morphologic studies of motor units in normal human muscles. Acta Anat 1955;23:127–142.

Lodish H, Berk A, Kaiser CA, Krieger M, Scott MP, Bretscher A, Ploegh H, Matsudaira P. Molecular cell biology, 6th ed. New York: WH Freeman and Co., 2007.

Ferrante MA, Wilbourn AJ. The electrodiagnostic examination of peripheral nerve injuries. In Mackinnon SE, editor, Nerve surgery. New York: Thieme Medical Publishers, 2015:59–74.

McComas AJ. Invited review: motor unit estimation: methods, results, and present status. Muscle Nerve 1991;14:585–597.

Brown PB. The electrochemical basis of neuronal integration. In Haines DE, editor, Fundamental neuroscience. New York: Churchill Livingstone, 1997:31–49.

Gasser HS, Grundfest H. Axon diameters in relation to spike dimensions and conduction velocity in mammalian fibers. Am J Physiol 1939;127:393–414.

Brown WF. The physiological and technical basis of electromyography. London: Butterworth, 1984:8.

Gath I, Stalberg E. In situ measurement of the innervation ratio of motor units in human muscles. Exp Brain Res 1981;43:377–382.

Dodge FA, Cooley JW. Action potential of the motor neuron. IBM J Res Develop 1973;17:219–229.

Gwathmey KG. Sensory polyneuropathies. Continuum 2017;23:1411–1436.

Eccles JC, Sherrington CS. Numbers and contraction values of individual motor units examined in some muscles of the limb. Proc R Soc Biol 1930;106:326–357.

Kugelberg E. Properties of rat hindlimb motor units. In Desmedt JE, editor, New developments in electromyography and clinical neurophysiology. Bassel: Karger, 1973:2–13.

Falck B, Stalberg E. Motor nerve conduction studies: measurement principles and interpretation of

66

vessel is a capillary. Unlike most capillaries, the endothelial cells lining these vessels are joined by tight junctions. This results in a continuous layer of endothelial cells that functions to separate the plasma from the extracellular fluid of the nerve. In addition to this diffusion barrier, the innermost cell layer of the perineurium also serves as a barrier to protect the endoneurial environment (e.g., the axons and Schwann cells) from potential extrafascicular threats. Together, these two barriers are referred to as the blood–nerve barrier. The blood–nerve barrier is less competent at the roots, dorsal root ganglia, autonomic ganglia, and the terminal motor branches innervating individual muscle fibers. The VGNC density of the nodal membrane is about ten-fold greater than that of the internodal membrane (Dodge and Cooley, 1973).

Lloyd DPC. Neuron patterns controlling transmission of

Rasminski M, Sears TA. Internodal conduction in undissected demyelinated nerve fibers. J Physiol 1971;22:323–350. Reles A, Friede RL. Axonal cytoskeleton at the nodes of Ranvier. J Neurocytol 1991;20:450–458. Rotshenker S. Traumatic injury to peripheral nerves. In Tubbs RS, Rizk E, Shoja MM, Loukas M, Barbaro N, Spinner RJ, editors, Nerves and nerve injuries. Amsterdam: Elsevier; 2015:611–628. Schnepp P, Schnepp G. Faseranalytische untersuchung an peripherin nerven bei tieren vershciedenergrosse. Z Zellforsch 1971;119:99–114. Stalberg E, Antoni L. Electrophysiological cross section of

Chapter 3: Anatomy and Physiology of Neurons

the motor unit. J Neurol Neurosurg Psychiatry 1980;43:469–474. Stevens JC, Lofgren EP, Dyck PJ. Histometric evaluation of branches of peroneal nerve: technique for combined biopsy of muscle nerve and cutaneous nerve. Brain Res 1973;52:37–59. Sunderland S. Nerve and nerve injuries, 2nd ed. Edinburgh: Churchill Livingstone, 1978.

Sunderland S. The anatomy and physiology of nerve injury. Muscle Nerve 1990;13: 771–784. Sunderland S, Bradley KC. The cross-sectional area of peripheral nerve trunks devoted to nerve fibers. Brain 1949;72: 37–59. Waxman SG, Bennett MV. Relative conduction velocities of

small myelinated and unmyelinated fibres in the central nervous system. Nature 1972;238:217–219. Woodbury JW. Action potential: properties of excitable membranes. In Ruch TC, Patton HD, editors, Neurophysiology. Philadelphia: WB Saunders, 1965:26–72.

67

Chapter

4

Anatomy and Physiology of the Neuromuscular Junction

Introduction Cells composing neural circuits communicate with each other at contact sites, termed synapses, the majority of which are chemical in nature (i.e., chemical synapses). A synapse is a unidirectional structure composed of the membranes of two cells in juxtaposition, including the space between the two membranes (termed the synaptic space or synaptic cleft). The membrane contribution from the first cell is termed the presynaptic membrane, and the membrane contribution from the second cell is referred to as the postsynaptic membrane. With chemical synapses, communication occurs via chemicals, termed neurotransmitters. The neurotransmitter is packaged into membrane-bound vesicles (sacs), some of which are transported to presynaptic membrane regions termed active release zones. At these zones, there are rows of voltage-gated calcium channels (VGCCs) oriented in pairs. In response to membrane depolarization, these channels open (i.e., they are voltage-gated), allowing Ca++ to enter the presynaptic cell. The resultant increase in intracellular Ca++ concentration facilitates exocytosis of the neurotransmitter into the synapse. (Magnesium has the opposite effect, impeding neurotransmitter release.) The released neurotransmitter diffuses across the synaptic space and binds to a specific receptor located on the postsynaptic membrane. This causes the receptor to undergo a conformational change that opens a channel within it that allows ions to traverse it, thereby locally changing the transmembrane potential. Depending on the type of ion channel, these synapses may be inhibitory or excitatory on the postsynaptic cell. For these reasons, a synapse can be thought of as an energy transducer, converting electrical energy into chemical energy and then back to electrical energy. The chemical synapses between the peripheral nervous system (PNS) and the muscular system are termed neuromuscular junctions (NMJs). Neuromuscular

68

junctions are excitatory in nature and utilize acetylcholine (ACh) as the chemical neurotransmitter. The ACh molecules released from the presynaptic cell bind to nicotinic ACh receptors (AChRs) on the postsynaptic cell, causing a conformational change in the latter that results in the opening of a centrally located channel through which ions flow. Because the conformational change results in the opening of an ion channel, AChRs are classified as ionotropic receptors. Because they open in response to ACh binding, these channels are also classified as ligand-gated. In general, ionotropic receptors are located within 1 micrometer of the neurotransmitter release sites, which allows for the rapid generation of postsynaptic potentials. In general, ionotropic receptors have low affinity for the neurotransmitter, which limits the duration of receptor activation (Swanson, 2000). The motor axon given off by the anterior horn cell (AHC) enters the muscle and arborizes into many terminal branches, each of which innervates a single muscle fiber through a single contact site. The exact number of terminal branches depends on the innervation ratio of the muscle (see Chapter 3). The contact sites between the terminal motor nerve branches and the muscle fibers are termed NMJs. The presynaptic cell membrane of the NMJ represents an enlargement of the terminal branch. These enlargements are termed axon terminals or terminal boutons. The distal portion of the terminal branch is not myelinated and, thus, has higher membrane capacitance. The NMJ is also vulnerable to dysfunction because it lacks a blood–nerve barrier and because NMJ transmission requires the involvement of a substantial number of synaptic proteins. The postsynaptic cell membrane is provided by a depression into the muscle fiber that is referred to as the endplate region (or simply the endplate). The endplate is a specialized portion of the muscle fiber membrane. Infoldings from the primary synaptic cleft form secondary synaptic clefts

Chapter 4: Anatomy and Physiology of the Neuromuscular Junction

(junctional folds), which serve to increase the surface area of the NMJ. The major events that occur at each of these three regions can now be discussed in more detail.

Presynaptic Region The presynaptic region is involved in many activities, including ACh synthesis, ACh storage, and ACh release. The enzyme, choline acetyltransferase, synthesizes ACh from two substrates: choline and acetyl-CoA. This occurs in the cytoplasm. The synthesized ACh is stored in membrane-bound vesicles. The vesicular ACh transporter slowly packages the synthesized ACh into the vesicles. Each vesicle contains up to 10,000 molecules of ACh. The number of ACh molecules contained within a single vesicle is termed a quantum; the total number of vesicles released with axon terminal depolarization is termed the quantal content. Many other proteins are involved in ACh vesicle release, including MUNC13 (docking and priming of vesicles), SNAP25B (a SNARE protein essential for exocytosis), and synaptotagmin II (also involved in exocytosis) (Aran et al., 2017). The vesicles of ACh behave as if they were located in three separate intracellular compartments, termed the main reserve (the largest compartment; about 250,000 vesicles), the mobilization store (the intermediate compartment; about 10,000 vesicles), and the immediate release pool (the smallest compartment; about 1,000 vesicles) (Keesey, 1989). The main reserve supplies the mobilization store and the mobilization store supplies the immediate release pool. The main reserve is composed of ungrouped ACh vesicles within the cytoplasm of the terminal bouton, the mobilization store includes those vesicles of ACh near the active release zone, and the immediate release pool consists of those ACh vesicles in the active release zones. Exocytosis of the ACh-containing vesicles is calcium-dependent, which is why the presynaptic membrane contains parallel double rows of VGCCs (Kandel et al., 2000). It is adjacent to these sites that vesicles of ACh ready for immediate release accumulate. ACh release is a multistep process: 1. depolarization of the axon terminal by the incoming nerve fiber AP 2. opening of the voltage-gated calcium channels 3. Ca++ influx from the extracellular space into the axon terminal

4. binding of calcium ions to proteins located on ACh-containing vesicles 5. transport of ACh vesicles to the presynaptic membrane 6. fusion of the vesicle membrane with the presynaptic membrane 7. exocytosis of ACh molecules (up to 10,000 molecules per vesicle) Increases in Ca++ influx result in the release of a larger number of vesicles of ACh (increased quantal content) (Hubbard and Schmidt, 1963). The calcium ions diffuse out of the cell within 200 msec (this fact will be of importance later when repetitive nerve stimulation is discussed). Step 6, membrane fusion, results in the release of the ACh molecules into the synaptic space (termed exocytosis). The presynaptic membranes form bulges (termed terminal boutons) that project into the postsynaptic membrane folds. This lessens the width of the synaptic space, thereby decreasing the distance that the ACh molecule must traverse to bind to a receptor. In this manner the likelihood of an ACh–ACh receptor interaction is increased.

Synaptic Space The synaptic space is about 500 Angstroms across. The released ACh molecules diffuse across this space in as little as 50 microseconds (Brown, 1984), at which point they bind to the ACh receptors protruding from the postsynaptic membrane. Within a few milliseconds, those ACh molecules not binding to an ACh receptor (about 30–50% of the molecules) drift into the postsynaptic folds and are rapidly hydrolyzed into acetate and choline by the enzyme acetylcholinesterase (AChE). AChE, the primary cholinesterase in the body, is anchored to the postsynaptic basal lamina by a collagen-like subunit tail. Following ACh hydrolysis, the acetate molecule diffuses away, but, due to limited quantities of choline in the axon terminal, the molecule of choline undergoes reuptake into the axon terminal. The rate of choline reuptake increases with axon terminal depolarization. Following reuptake, the molecule of choline is attached to another molecule of acetyl-CoA, by choline acetyltransferase, resulting in a new molecule of ACh. Following their release from the AChR, the bound ACh molecules undergo hydrolysis to acetate and choline as described above.

69

Section 1: Introductory Chapters

Postsynaptic Region The postsynaptic region is often referred to as the endplate. The endplates are situated roughly in the center of the muscle fiber, between its origin and its insertion. During development, clusters of ACh receptors accumulate in the central region of muscle fibers. This process is termed prepatterning and is essential for effective neurotransmission. This process begins with the release of agrin from the developing lower motor neuron. Agrin enhances the kinase activity of muscle-specific kinase (MuSK), through the agrin receptor, LRP4 (low-density lipoprotein receptorrelated protein 4). MuSK is coactivated by Dok-7. Activated MuSK promotes AChR clustering through rapsyn. The release of ACh by the developing lower motor neurons reinforces the central accumulation of AChRs. However, because this process requires ACh receptor-mediated postsynaptic membrane potentials, postsynaptic specializations must occur prior to nerve terminal differentiation (Witzemann, 2006). Although the endplate region is centered along the individual muscle fibers, the arrangement of the muscle fibers within the muscle dictates the innervation zone of the muscle. With longitudinal muscles, the innervation zone is central, but with other muscle fiber configurations (e.g., pennate), it is more dispersed (see Chapter 5). The endplate region is involved in a number of activities, including: 1. 2. 3. 4.

opening of the ACh receptor channels nonselective cation influx (predominantly Na+) localized endplate depolarization generation of a muscle fiber action potential

The binding of two ACh molecules to the alpha subunits of the ACh receptor results in channel opening (discussed in greater detail below). Although the ACh receptor channel is nonselective, it predominantly transmits Na+ (Na+  K+ > Ca++ and Mg++). The actual ratio of transmitted Na+ to transmitted K+ is about 30: 1 (Brown, 1984). The molecule of ACh remains bound to the ACh receptor for about 100–200 microseconds, at which point it is released. If the released ACh is not enzymatically broken down by AChE, it may bind to another ACh receptor (thereby increasing the endplate current) or it may diffuse out of the synapse (Katz and Miledi, 1973). The endplate region of the muscle fiber is remarkable for a large number of invaginations (termed postsynaptic folds). There is a major fold,

70

termed the primary synaptic cleft, which contains a number of its own clefts, termed secondary clefts. The postsynaptic folding serves to increase the surface area of the synapse, thereby permitting a greater number of potential ACh–ACh receptor interactions. The surface area of the axon terminal is about 2,000–6,000 µ2 and contains about 300 active zones, whereas that of an excitatory afferent from a dorsal root ganglion cell forms about 4 synapses, each of which is about 2 µ2 and contains only a single active zone (Kandel et al., 2000). The larger-diameter, fastertwitch muscle fibers have larger synaptic surface areas than do the smaller-diameter, slower-twitch muscle fibers (Duchen, 1971). It has been suggested that this increase in surface area is necessary to accommodate the higher current requirements associated with the lower resistance of these larger-diameter muscle fibers (Kuno et al., 1971). The number of ACh receptors per NMJ is approximately 1 to 5  107 (Fambrough et al., 1973). The NMJ can be thought of as an energy transducer because it converts electrical energy (the action potential of the presynaptic membrane) into chemical energy (ACh) and then back into electrical energy (the endplate potential of the postsynaptic membrane). Another anatomic strategy that enhances the probability of ACh–ACh receptor interaction is related to the distribution of the ACh receptors – the peaks of the folds have the greatest density of ACh receptors, decreasing in density away from the peaks. The peak density has been estimated to be about 10,000 ACh receptors per µ2 (Sine, 2012). The ACh receptor densities are considerably less within the membrane region outside the junction (i.e., extrajunctional membrane). The troughs of the folds have the greatest concentration of acetylcholinesterase, the enzyme responsible for breaking ACh into its component parts (acetate and choline). ACh receptors are composed of five subunits – two alpha, one beta, one epsilon (gamma in fetal ACh receptors), and one delta – arranged to form a channel through the postsynaptic membrane. The two alpha subunits lie adjacent to each other and each has a binding site for ACh. When ACh binds to one of these two alpha subunits, it induces a conformational change in the other alpha subunit that increases its affinity for ACh binding (termed cooperative binding) (Sine, 2012). Once a second molecule of ACh binds to the other alpha subunit, the ion channel opens, and cation influx occurs. The positively charged

Chapter 4: Anatomy and Physiology of the Neuromuscular Junction

cations depolarize the postsynaptic membrane. The endplate depolarization generated is termed an endplate potential. It is not regenerated and, thus, decays over time and distance. Because EPPs are electrotonic potentials, they summate in space and time like EPSPs and IPSPs. The likelihood that summation will occur depends on the resistance and capacitance of the membrane, as reflected by the values of its length constant (λ) and its time constant (T) (see Chapter 3). In general, EPP summation is more than sufficient (45–75 mV) to depolarize the muscle membrane to its depolarization threshold value (15 mV) and precipitate a bidirectionally propagating muscle fiber action potential.

Postsynaptic Membrane Depolarizations At rest, single vesicles of ACh randomly fuse to the presynaptic membrane, releasing their contents into the synapse. Each vesicle contains up to about 10,000 molecules of ACh, some of which bind to the postsynaptic ACh receptors. The contents of a single vesicle of ACh results in the opening of approximately 1,000 ACh receptor channels (Brown, 1984). The cation influx associated with the release of a single vesicle of ACh generates a small postsynaptic membrane depolarization, termed a miniature endplate potential (MEPP). MEPPs demonstrate a single negative phase (i.e., they have a monophasic morphology that is deflected upward) and, because of their small size (about 1 mV), are unable to trigger muscle fiber depolarization (requires about 15 mV) (Engel et al., 1977). The EDX characteristics of MEPPs are discussed in greater detail later in this textbook (see Chapter 13). The utility of these small depolarizations may be to maintain the structural integrity of the endplate zone because without ongoing ACh release, the endplate degenerates. With axon terminal depolarization, however, 50–150 ACh vesicles are released in near-synchrony, which is much higher than the 1–10 quanta released at most CNS synapses (Kendall, 2000). The actual number of ACh vesicles released is termed the quantal content. The ACh molecules bind to the postsynaptic ACh receptors, causing them to open, thereby permitting the entry of positive current (mostly Na+). Each vesicle generates an MEPP, and the generated MEPPs summate (in time and space) to form an EPP. The amplitude of the EPP is 45–75 mV.

Because only a 15-mV depolarization is required to trigger muscle membrane depolarization, the EPP is 3–5 times larger than the requirement. This excess depolarization, which is termed the safety factor of neuromuscular junction transmission, reflects the large overage of ACh molecules released and ACh receptors available, which results in a far larger number of ACh–ACh receptor interactions than is required for membrane depolarization. In addition to excess ACh release and overabundance of ACh receptors, this overage of ACh–ACh receptor interactions also reflects other aspects of the NMJ junction, including: (1) the higher density of ACh receptors on the peaks of the folds, (2) the higher density of AChE in the troughs, (3) the narrow width of the synapse, and (4) the postjunctional folding of the membrane. Mathematically, if 100 ACh vesicles are released, each containing 10,000 ACh molecules, then 1,000,000 ACh molecules of ACh are released per axon terminal depolarization. Hence, the NMJ can be thought of as an electrical amplifier with a gain that exceeds 100 (see Chapter 2). This gain overcomes the large electrical mismatch between the small axon terminal and the much larger muscle fiber (Keesey, 1989). The two main factors determining the amount of ACh released into the synapse are: (1) the number of vesicles in the active release zone and (2) the intracellular Ca++ concentration. As previously stated, intracellular calcium ions facilitate ACh release. The effect lasts until the Ca++ is sequestered into the mitochondria (about 100–200 msec). Both factors are affected by repetitive depolarization of the axon terminal. When repetitive axon terminal depolarization occurs at rates below 5–10 Hz, the number of ACh vesicles in the immediately available ACh pool decreases. As a result, the quantal content of each subsequent stimulus also decreases. As previously stated, the ACh vesicles of the immediately available pool are replenished through the other two compartments. When the number of incoming ACh vesicles approximates the number of ACh vesicles being released (usually by 7–8 stimulations), the EPP value levels off (steady state). Although this steady state value is below the normal value, it remains above the depolarization threshold value (i.e., it remains suprathreshold). Consequently, all stimuli generate suprathreshold EPPs and, thus, a muscle fiber action potential. Any process that decreases the number of ACh–ACh receptor interactions lowers the magnitude of the EPP and,

71

Section 1: Introductory Chapters

hence, lowers the safety factor. As the safety factor diminishes, it becomes more likely that the EPP will be subthreshold and fail to generate a muscle fiber action potential. When repetitive axon depolarizations occur at frequencies above 5–10 Hz, calcium

References Aran A, Segel R, Kaneshige K, Gulsuner S, Renbaum P, Oliphant S, Meirson T, Weinberg-Shukron A, Hershkovitz Y, Zeligson S, Lee MK, Samson AO, Parsons SM, King M-C, Levy-Lahad E, Walsh T. Vesicular acetylcholine transporter defect underlies devastating congenital myasthenia syndrome. Neurology 2017;88:1021–1028. Brown WF. The physiological and technical basis of electromyography. London: Butterworth, 1984:372–375. Duchen LW. An electron microscopic comparison of motor en-plates of slow and fast skeletal muscle fibers of the mouse. J Neurol Sci 1971;14:37–45. Engel AG, Lambert EH, Gomez MR. A new myasthenic syndrome with

72

ion accumulation occurs in the axon terminal, enhancing ACh vesicle release. These concepts are important to repetitive nerve stimulation studies and are discussed in more detail later in this textbook (see Chapter 12).

end-plate acetylcholinesterase deficiency, small nerve terminals, and reduced acetylcholine release. Ann Neurol 1977;1:315–330. Fambrough DM, Drachman DC, Satyamurti S. Neuromuscular junction in myasthenia gravis: decreased acetylcholine receptors. Science 1973;182:293–295.

Keesey JC. AAEE minimonograph #33: electrodiagnostic approach to defects of neuromuscular transmission. Muscle Nerve 1989;12:613–626. Kuno M, Turkanis SA, Weakly JN. Correlation between nerve terminal size and transmitter release at the neuromuscular junction of the frog. J Physiol 1971;213:545–556.

Hubbard JI, Schmidt RF. An electrophysiological investigation of mammalian motor nerve terminals. J Physiol 1963;166:145–167.

Sine SM. End-plate acetylcholine receptor: structure, mechanism, pharmacology, and disease. Physiol Rev 2012;92:1189–1234.

Kandel ER, Schwartz JH, Jessell TM. Principles of neural science, 4th ed. New York: McGraw-Hill, 2000.

Swanson TH. Synaptic transmission. In Levin KH, Luders HO, editors, Comprehensive clinical neurophysiology. Philadelphia: WB Saunders 2000:57–68.

Katz B, Miledi R. The binding of acetylcholine to receptors and its removal from the synaptic cleft. J Physiol 1973;231:549–574.

Witzemann V. Development of the neuromuscular junction. Cell Tissue Res 2006;326:263–271.

Chapter

5

Anatomy and Physiology of Muscle

The Organization of Muscle Tissue Motor Units Approximately 250 million muscle fibers compose the 434 skeletal muscles of the human body, accounting for 40–45% of total body mass (Lai and Mitsumoto, 2000). Functionally, the muscle fibers are organized into motor units, which represent the muscle fibers innervated by a single anterior horn cell (AHC). The number of motor units within a muscle is equivalent to the number of motor axons innervating the muscle and varies among different muscles. In general, larger muscles contain a larger number of motor units. In 1955, based on anatomical studies, Feinstein et al. (1955) reported that the first dorsal interosseous muscle contained 119 motor units, the tibialis anterior contained 445 motor units, and the gastrocnemius muscle contained 579 motor units. In 1971, using a motor unit number estimation technique, McComas et al. (1971) estimated that the extensor digitorum brevis muscle contains 210 motor units, the muscles composing the thenar eminence contain 342 motor units, and the muscles composing the hypothenar eminence contain 390 motor units.

Muscle Innervation Ratios The number of muscle fibers innervated by a single AHC (also termed a lower motor neuron [LMN]) dictates the innervation ratio of that muscle. This value is equivalent to the total number of muscle fibers composing the muscle divided by the total number of motor axons innervating it. This value also varies among different muscles but is roughly constant for a given muscle (i.e., the motor units composing a given muscle have roughly the same number of muscle fibers). As was originally argued by Clark in 1931, muscles responsible for delicate movements, such as extraocular muscles, have lower innervation ratio values than do muscles having less precise movements

(Weddell et al., 1944). For example, the innervation ratios for the gastrocnemius, tibialis anterior, and first dorsal interosseous are 1900, 562, and 340, respectively, whereas extraocular muscles average just 9 muscle fibers per motor unit (Feinstein et al., 1955). The innervation ratio can be used to calculate the number of motor units composing a muscle through substitution, because the total number of axons innervating the muscle is equal to the total number of motor units composing the muscle, as follows: Innervation ratio ¼ total # muscle fibers= total # motor axons Total # of motor units ¼ total # of muscle fibers= innervation ratio

Structural Organization of Muscle The structural organization of muscle is similar to that of peripheral nerves. The individual muscle fibers composing the skeletal muscle are surrounded by endomysium (as compared to individual axons being surrounded by endoneurium), collections of muscle fibers are grouped into fascicles by perimysium (as compared to collections of axons being grouped into fascicles by perineurium), and muscle fascicles are surrounded by epimysium (as compared to fascicles of axons being surrounded by epineurium). Unlike nerve fibers, the connective tissue elements at the ends of the muscle fibers coalesce to form tendons, which, in turn, attach to bone and permit muscle contraction–induced skeletal movements. The membrane surrounding the muscle fiber is termed the sarcolemma (as opposed to the axolemma for nerve fibers). The cytoplasm of muscle fibers is termed sarcoplasm or myoplasm (as opposed to axoplasm). The sarcolemma is surrounded by the basement membrane. Unlike neurons, muscle fibers contain multiple nuclei (syncytial tissue), the overwhelming majority of which are subsarcolemmal. Satellite cells

73

Section 1: Introductory Chapters

are located external to the sarcolemma and internal to the basement membrane. Each muscle fiber is composed of a collection of longitudinally oriented cylindrical structures termed myofibrils. Each myofibril is, in turn, composed of segments, termed sarcomeres, which are connected in series and that shorten in response to muscle fiber membrane depolarization (i.e., contraction). Thus, sarcomeres are the functional units of muscle contraction (discussed later here). Through the musculoskeletal system, the nervous system is able to generate movement, thereby enabling movement within the environment.

Excitation–Contraction Coupling Action Potential Propagation Like the axolemma (nerve membrane), the sarcolemma (muscle membrane) contains ion channels that generate a resting membrane potential (RMP). Also like the axolemma, the presence of voltage-gated channels studding the sarcolemma permits the generation and propagation of action potentials. Despite these similarities, muscle fibers are quite different from nerve fibers. The membrane potentials that follow ACh receptor opening and positive current entry (mostly Na+) are local potentials incapable of propagation. These potentials summate in space and time to form an endplate potential. When the latter is large enough, it triggers sarcolemmal depolarization, at which point the voltage-gated sodium channels (VGNCs) of the muscle membrane open and sodium entry occurs, generating an action potential that bidirectionally propagates away from the endplate zone. Because the diameter of muscle fibers is much larger than that of nerve fibers, the internal resistance of muscle fibers to action potential advancement is much lower. However, because the muscle membrane is not myelinated, action potential propagation occurs in a continuous manner. In addition, the sarcolemma has invaginations, termed transverse tubules or T tubules (to indicate their perpendicular orientation to the long axis of the muscle fiber). These structures allow the muscle fiber action potential to be transmitted into the depths of the muscle fiber. Because these membranous invaginations significantly increase the surface area of the sarcolemma, the capacitance of the muscle membrane is significantly increased. In fact, due to the presence of the transverse tubules, the

74

capacitance of the sarcolemma more than doubles (Brown, 1984). As a result, the time constant of the muscle fiber membrane is much larger, and for this reason, action potential propagation speed is significantly slowed (3–5 m/sec). Other significant differences between muscle membrane and nerve membrane include: (1) muscle membrane repolarization occurs quicker because of its higher K+ permeability; and (2) muscle membrane demonstrates a longer hyperpolarization phase and, thus, a longer relative refractory period (Brown, 1984).

The Sarcoplasmic Reticulum In addition to the transverse tubules, muscle contains another network of tubular structures, referred to as the sarcoplasmic reticulum. As the name implies, this network is located within the sarcoplasm and, unlike the transverse tubules, the sarcoplasmic reticulum surrounds the myofibrils and is oriented parallel to them. For this reason, these structures are also referred to as the longitudinal tubules. Their ends, which are dilated, are termed terminal cisternae and function in the uptake, storage, and release of calcium ions. Each transverse tubule is situated between two terminal cisternae, and the three together compose a structure termed a triad (sarcoplasmic reticulum-T tubule-sarcoplasmic reticulum). The triad allows the transverse tubule depolarization to be transmitted to the sarcoplasmic reticulum, which triggers Ca++ release and excitation–contraction coupling (discussed later in this chapter). These triads are ideally positioned over the sarcomere where the A bands and I bands abut each other (discussed further on). Faster-contracting muscles have a greater number of contacts between the transverse tubule and the sarcoplasmic reticulum (Brown, 1984).

The Sarcomere The sarcomere contains a number of filamentous proteins that are involved in muscle contraction, including actin, troponin, tropomyosin, and myosin. Actin monomers, each of which has a binding site for myosin, polymerize into actin filaments. Two actin filaments rotate around each other to form a final helical structure. Tropomyosin, similar to actin, is formed by two chains wound into a helical configuration. Tropomyosin spans 7 actin monomers and has a single molecule of troponin bound to it. Troponin is composed of 3 subunits: troponin T (binds

Chapter 5: Anatomy and Physiology of Muscle

tropomyosin), troponin C (binds calcium ions), and troponin I (inhibits actin-myosin interaction). Actin, troponin, and tropomyosin combine to create the thin filaments of the sarcomere. The thick filaments of the sarcomere are composed of myosin. Myosin has two distinct regions: a straight portion and a globular portion. The straight portion, which is composed of light meromyosin, is referred to as the tail, whereas the globular portion, which is composed of heavy meromyosin, is referred to as the head. The head region contains two binding sites, one for actin and one for ATP. The portion where the tail and head join is referred to as the hinge. The thin and thick filaments are the major contractile elements of the myofibril. The thin filaments are distally located within the sarcomere, extending from the terminal Z disks toward its center, which they do not reach. The center of the sarcomere is identified by a structure termed the M line. The thick filaments, which are centrally located within the sarcomere, extend toward the Z disks at the two ends of the sarcomere but do not reach them. Consequently, at rest, the center of the sarcomere is composed solely of thick filaments, its distal extremes are composed solely of thin filaments, and the segment between these two regions is composed of overlapping thin and thick filaments. This arrangement causes the myofibrils to have a banded appearance when viewed under polarized light. The dark bands represent regions of thick and thin filament overlap, and the lighter areas represent segments without overlap. The A band (the anisotropic band) extends from one end of the thick filament to the other end. As a result, the A band is darker at its distal segments (i.e., where the thick filaments overlap with the thin filaments) and lighter centrally (i.e., where there is no overlap). This centrally located lighter region is referred to as the H band. The M-line is centrally located within the H band. The I bands (isotropic bands) are situated between the A bands. These bands are light in color because they represent segments containing only light filaments. The Z-line, which demarcates the sarcomeres, is centrally located within the A band. In the presence of calcium ions, the thick filaments pull the thin filaments centrally, causing them to slide past each other (the sliding filament theory) (Huxley and Hanson, 1954; Huxley and Niedergerke, 1954). This shortens the length of the sarcomere, thereby contracting the myofibril and, thus, the muscle fiber to which it belongs (muscle fiber contraction).

As the sarcomere progressively contracts, the thick and thin filaments overlap each other more and more. As the overlap increases, the lengths of the H band (nonoverlapping thick fibers) and the I band (nonoverlapping thin fibers) decrease. At maximal contraction, there is complete overlap of the thick and thin filaments (i.e., the thick and thin filaments span the length of the sarcomere, and there are no longer nonoverlapping regions). Because the overlap is complete, the I bands and the H bands disappear. The length of the A band is unchanged, but it differs in that it no longer demonstrates a central H band. The mnemonic used to recall which bands disappear with contraction is “Hi, bye,” to indicate that you say goodbye to the H and I bands when the overlap is complete. Importantly, the force of the muscle contraction is proportional to the degree of the thin and thick filament overlap (i.e., to the number of crossbridges between the thick and thin filaments). The biomechanics underlying the sliding filament theory are straightforward. At rest, ADP is bound to the myosin head (at its ATP binding site) and the myosin head is in its cocked position. With depolarization of the transverse tubule, the adjacent sarcoplasmic reticulum releases calcium ions into the sarcoplasm surrounding the myofibrils. The calcium ions bind to the troponin C subunit of troponin, inducing a conformational change that exposes the myosin-binding site of the actin filament that was previously blocked by its troponin I subunit. This exposure permits cross-bridging of the thick and thin filaments and generates flexion of the hinge portion of the myosin filament (the power stroke). This flexion of the myosin head results in the pulling of the actin filament centrally, thereby increasing the degree of thin and thick filament overlap and reducing the length of the sarcomere and its H and I bands. Following this movement, the ADP and Pi are released from the myosin and a new molecule of ATP binds to the ATP-binding site. Hydrolysis of the new molecule of ATP (ATP ! ADP + Pi) dissociates the thin and thick filaments and resets the myosin head to its cocked position for the next actin monomer. The ADP remains bound to the myosin head and the Pi diffuses away. It is important to realize, therefore, that filament dissociation is an energy dependent event. This process – actin-myosin binding, sliding, and ATP hydrolysis-induced unbinding – repeats itself until free calcium ions are no longer present. Once membrane repolarization occurs, Ca++

75

Section 1: Introductory Chapters

reuptake into the sarcoplasmic reticulum follows and the thick and thin filaments return to their resting states. Because thick filament–thin filament dissociation is energy dependent, once the ATP supply is exhausted (a few hours after death), the thick and thin filament cross-bridges cannot dissociate. This accounts for the phenomenon of rigor mortis (stiffness of death).

Neural Control of Muscle Introduction It was not always known that muscle contracts in response to nerve activation. In 1658, Swammerdam showed that frog muscle changed its shape in response to mechanical stimulation of the innervating nerve rather than by inflation related to the inflow of a gaseous or liquid substance, which was the prevailing theory at that time (Bonner and DevlescHoward, 1995). In 1791, Galvani reported the phenomenon of animal electricity. While dissecting a frog, a spark from a conductive device unintentionally contacted his scalpel and caused the musculature of the frog to instantly contract, allowing him to conclude that the nerves were good conductors of electricity (Bonner and DevlescHoward, 1995).

The Final Common Path Lower motor neurons receive input from cortical motor neurons (the upper motor neurons) through the corticospinal tract, as well as from neurons located in other areas of the central nervous system, including the red nucleus (through the rubrospinal tract), the vestibular nuclei (through the medial and lateral vestibulospinal tracts), the tectum (through the tectospinal tract), and the reticular formation (through the reticulospinal tract). The rubrospinal tract, the main component of the lateral descending motor system, controls primarily flexor muscles of the upper extremity, whereas the medial and lateral vestibulospinal, reticulospinal, and tectospinal tracts, which represent the medial pathways of motor control, are the oldest (phylogenetically) elements of the descending motor system. These tracts terminate predominantly on LMNs innervating the axial and proximal extremity muscles. These parallel systems have a hierarchal relationship with each other, with the cortical motor neurons dominating. When the input of the corticospinal tract to the LMNs is lost, the rubrospinal tract dominates

76

and increases the flexor tone of the upper extremity muscles while the other tracts increase the extensor tone of the lower extremity muscles. When input from the rubrospinal tract also is lost, the remaining tracts cause the extensor tone of all four extremities to be increased. This is the pathophysiology underlying decorticate and decerebrate posturing, respectively. The LMNs of the brainstem (motor cranial nuclei) and spinal cord (anterior horn cells) innervate a large number of muscle fibers via a single motor axon. This element (the LMN and the muscle fibers it innervates), termed the motor unit, is the smallest element capable of generating muscle force. Thus, it is the fundamental unit of skeletal muscle contraction. This concept was first conceived by Liddell and Sherrington in 1925 and defined by Sherrington in 1926 (Liddell and Sherrington, 1925; Sherrington, 1926), although the original definition did not include the cell body or dendrites of the lower motor neuron (Burke, 1980). Because it is the only pathway by which the CNS can communicate with the muscle fibers, it is commonly referred to as the final common path (Sherrington, 1929).

Types of Lower Motor Neurons There are three types of lower motor neuron: alpha, beta, and gamma; the alpha and gamma motor neurons are the main types. The alpha motor neurons produce muscle fiber contraction and contractile force through extrafusal muscle fiber innervation. Because they innervate skeletal muscle fibers, they are also referred to as skeletomotor neurons. Alpha motor neurons have larger cell bodies (40–80 micrometers) and are faster conducting than gamma motor neurons. Gamma motor neurons innervate the intrafusal muscle fibers located within muscle spindles (discussed below) and, for this reason, are also called fusimotor neurons. They have smaller cell bodies (20–40 micrometers) and conduct more slowly than alpha motor neurons. Beta neurons innervate both skeletal and intrafusal muscle fibers and, therefore, are also called skeletofusimotor neurons.

The CNS Influence The CNS influences the skeletal muscle fibers through the alpha motor neurons. The alpha motor neurons receive direct input from the corticospinal tract and from Ia sensory axons. The Ia sensory inputs create specific monosynaptic reflexes at the brainstem and

Chapter 5: Anatomy and Physiology of Muscle

spinal cord levels, such as the muscle stretch reflex. These reflexes are integrated by descending CNS inputs into a variety of automated movements that control posture and locomotion (Kandell and Schwartz, 1991). Thus, when muscle stretching activates the monosynaptic reflex, it concomitantly activates other neuronal circuits that activate synergistic muscles and that inhibit antagonistic muscles. The concomitantly activated circuits are polysynaptic and serve to facilitate the monosynaptic muscle stretch reflex. As stated earlier, skeletal muscles are referred to extrafusal muscle fibers; this distinguishes them from the intrafusal muscle fibers contained within muscle spindles (muscle spindles are discussed below). Unlike extrafusal muscle fibers, intrafusal muscles fibers do not contain myofibrils and, therefore, do not contract. Intrafusal muscle fibers function to maintain muscle spindle sensitivity. When the CNS generates extrafusal muscle fiber contraction (through alpha motor neuron activation), it simultaneously activates gamma motor neurons. This is termed alpha gamma coactivation. The level of tension generated by a muscle at rest, termed muscle tone, depends primarily on alpha motor neuron activity. Gamma motor neurons have an indirect effect on muscle tone through their Ia sensory fiber connections to alpha motor neurons (discussed later here). With upper motor neuron disruption, the gamma motor neurons become more active (termed gamma gain). The heightened activity of gamma motor neurons causes increased alpha motor neuron activity (produces hypertonia) and increased muscle spindle sensitivity to stretch (produces hyperreflexia). The latter two clinical features (hypertonia and hyperreflexia) are termed UMN features.

Muscle Spindle Anatomy Muscle spindles are encapsulated sensory receptors (stretch receptors) that are embedded in the extrafusal muscle fibers of muscle and that consist of three main components: (1) a group of 3–12 intrafusal muscle fibers, (2) sensory nerve fibers, and (3) gamma motor nerve fibers. The sensory nerve fibers supply the central regions of the intrafusal muscle fibers (the noncontractile regions), whereas the motor nerve fibers supply the terminal regions of the intrafusal fibers (the contractile regions). The muscle fiber contraction of intrafusal muscle fibers does not contribute to the contractile force of the muscle.

The intrafusal muscle fibers, which are parallel to the extrafusal muscle fibers, are of two types, based on the arrangement of their nuclei: nuclear bag fibers (nuclei are centrally located) and nuclear chain fibers (nuclei are located in a single row). The nuclear chain fibers are about twice as prevalent. The sensory axons from the central region are of two types. The larger, more heavily myelinated axons (group Ia fibers), termed primary endings, supply the central regions of the intrafusal muscle fibers; they convey dynamic information (the rate of change of the length of the intrafusal muscle fibers). The smaller and less heavily myelinated axons, termed secondary endings, supply the paracentral regions of the intrafusal muscle fiber; they convey static information (absolute length of the intrafusal muscle fibers).

Muscle Spindle Physiology The sensory nerve fibers of the muscle spindle primarily convey information about extrafusal muscle fiber length – important for the perception of limb position (e.g., proprioception) and the velocity of limb movement – and the gamma motor nerve fibers of the muscle spindle serve to maintain its sensitivity.

Muscle Lengthening The muscle stretch reflexes elicited during the neurological examination reflect muscle spindle function in response to muscle lengthening – tendon tapping stretches the tendon, which stretches the extrafusal (skeletal) muscle fibers, which also stretches the intrafusal muscle fibers. The latter activates the Ia sensory axons, increasing their firing rate and exciting the LMNs to which they are attached. This, in turn, leads to contraction of the originally stretched muscle (and, through interneurons, inhibition of antagonistic muscles). The muscle stretch also stimulates Golgi tendon organs (located at the junction of the muscle and its tendon), resulting in the activation of group IIb sensory axons. This polysynaptic pathway results in inhibition of the stretched muscle and activation of the antagonistic muscles, thereby terminating the muscle spindle-induced contraction.

Muscle Shortening When skeletal muscle fibers contract (shorten), the sensory nerve fibers are stimulated and activate gamma motor neurons within the anterior horn. This causes the polar regions of the intrafusal muscle fibers

77

Section 1: Introductory Chapters

to contract, which stretches the noncontractile central region. This increases the level of activation of the sensory endings innervating the central region, causing the gamma motor neurons to increase their activity, thereby shortening the polar regions of the intrafusal muscle fiber. This restores the sensitivity to stretch of the muscle spindle at its new length. Thus, the gamma motor neurons maintain the sensitivity of the muscle spindle to stretch. Gamma motor axons represent about 30% of the motor axons reaching the muscle from the spinal cord (Hunt, 1951).

Motor Unit and Muscle Fiber Types Based on the time required to reach maximal force and on how easily they fatigue, motor units can be classified into two major types – type I and type II – although there is some overlap. The type I motor units have smaller cell bodies, thinner axons, smaller endplate zones, and thinner muscle fibers, whereas the type II motor units have larger cell bodies, thicker axons, larger endplate zones, and thicker muscle fibers. Type I motor units are more resistant to fatigue, whereas the type II motor units conduct faster (because of their larger-diameter axons) and generate larger twitch tensions (their large-diameter muscle fibers contain more myofibrils). Type II motor units are further divided into type IIA motor units and type IIB motor units. Both types demonstrate fast contraction times. Type IIA motor units generate less tension and are fatigue resistant, whereas type IIB motor units generate higher tension and are less fatigue resistant. In humans, both motor unit types are present in every muscle, and the muscle fibers of the two types intermingle with each other.

Motor Unit and Muscle Fiber Physiology, Biochemistry, and Metabolism The muscle fibers composing type I motor units differ in their rates of glycolytic and oxidative activity from the muscle fibers composing type II motor units. These differences cause them to have differing histochemical characteristics. Many of these histochemical differences are brought out by various stains. For example, the adenosine triphosphatase (ATPase) stain is performed at three different pH levels, each of which optimizes the visualization of a specific muscle fiber type. The ATPase of type I muscle fibers is activated at low pH (acidic) values, whereas it is

78

activated at higher pH (alkaline) values among type II muscle fibers. Because the muscle fibers of type I motor units are more dependent on aerobic (oxidative) metabolism for ATP synthesis, they require a richer capillary supply, have more mitochondria, and contain higher levels of oxidative enzymes (e.g., succinic dehydrogenase; cytochrome oxidase). With aerobic metabolism, 38 molecules of ATP are generated per molecule of glucose. The muscle fibers of type IIB motor units, which depend on anaerobic metabolism for immediate energy, have more abundant glycogen stores and higher levels of phosphorylase. With anaerobic metabolism, each molecule of glucose generates only two molecules of ATP. The glucose is derived from intramuscular glycogen stores and the blood supply to the muscle. Type IIA muscle fibers are intermediate to these two extremes – they have a rich capillary supply and intermediate levels of oxidative enzymes, as well as high glycogen and phosphorylase levels. There are several important biochemical pathways involved in the generation of contractile force. Phosphocreatine is an immediate energy source stored in muscle tissue. During muscle contraction, it is used to convert adenosine diphosphate (ADP) to adenosine triphosphate (ATP), as follows: ADP þ phosphocreatine ! ATP þ creatine This reaction is catalyzed by a cytosolic enzyme, creatine kinase. Although phosphocreatine stores are high in resting muscle, they are consumed within the first 10 seconds of high-intensity muscle contraction. During rest, the reaction runs in the opposite direction and the dephosphorylated creatine is rephosphorylated back to phosphocreatine via aerobically derived ATP (Tarnopolsky, 2016). Another important biochemical pathway involves the enzyme adenylate kinase, which catalyzes the conversion of two molecules of ADP to one molecule of ATP and one molecule of adenosine monophosphate (AMP), as follows: 2 ADP ! ATP þ AMP This reaction is “pulled” forward by the removal of AMP, which is converted to inosine monophosphate (IMP) by the enzyme myoadenylate deaminase. Ultimately, IMP is converted to xanthine and, under aerobic conditions, xanthine is converted to urate via xanthine oxidase. Muscle contraction is fueled by the breakdown of glycogen (glycogenolysis) and glucose (glycolysis)

Chapter 5: Anatomy and Physiology of Muscle

and through the oxidation of free fatty acids. Intramuscular glycogen serves as the primary source of carbohydrate during exercise, whereas the source of free fatty acids is derived from the blood and from intramuscular lipids. Although amino acids also contribute to energy production, their contribution is minimal (2–5%) (Tarnopolsky, 2016). Although carbohydrate breakdown can be done anaerobically or aerobically, fatty acid oxidation can only be done through aerobic metabolism. The point at which a person changes from aerobic metabolism to anaerobic metabolism (the anaerobic threshold) varies with the type, intensity, and duration of the activity, the training status of the individual, and other factors. It is usually described in terms of maximum oxygen consumption (VO2max). In general, for most untrained individuals, free fatty acid oxidation predominates over carbohydrate metabolism at exercise levels below 50% VO2max. At higher levels of effort, the contribution of carbohydrates to energy production increases. Anaerobic metabolism is required once the anaerobic threshold is reached. The anaerobic threshold is around 70% VO2max in untrained individuals and can approach 85% VO2max in welltrained individuals (Tarnopolsky, 2016). At all levels of VO2max, endurance exercise training increases the proportion of free fatty acids utilized in the production of energy.

Motor Unit Force Generation The force generated by a motor unit primarily reflects the cross-sectional area of its muscle fibers, which correlates with the number of contractile units (myofibrils) it contains. Therefore, the force generated by type IIB motor units is about 100 times greater than that generated by type I motor units. The degree of fatigue resistance primarily reflects the oxidative capacity of the muscle fiber, which is greatest in the type I motor units. Based on the above, type I fibers (slowoxidative [SO] fibers) can generate low levels of force for sustained periods (their rich capillary networks permit continuous delivery of substrate). Type IIB fibers (fast-glycolytic [FG] fibers) can generate higher levels of force, but because their glycogen supply is limited, the higher force levels cannot be sustained (i.e., these muscle fibers fatigue). Type IIA fibers (fast, oxidative, glycolytic [FOG] fibers) generate energy through both aerobic and anaerobic mechanisms and, consequently, are intermediate to type I and

type IIB muscle fibers in their ability to resist fatigue (Brooke and Kaiser, 1970; Brown, 1984; Amato, 2008). Training produces changes in the properties of motor units, including increased capillarization, increased oxidative enzyme content, and increased resistance to fatigue, which may cause some type IIB motor units to convert to type IIA motor units. Unlike individual animal muscles, which contain predominantly type I muscle fibers (e.g., the leg muscles of chicken [drumsticks and thighs]; dark meat) or type II muscle fibers (e.g., the breast muscles of chicken; white meat), human muscles contain a mixture of all three muscle fiber types, although axial postural muscles may contain more type I fibers (for the generation of sustained, low-tension force) and phasic extremity muscles may contain more type II muscle fibers (for the generation of rapid, hightension force). The three muscle fiber types are intermingled with each other and demonstrate a checkerboard pattern when stained with muscle fiber typing stains, such as ATPase. The red color of muscle is generated by myoglobin, which is a heme protein that binds a single oxygen molecule. It is important for oxidative metabolism in exercising muscle where it enhances the movement of oxygen down its concentration gradient. The muscle fibers of an individual motor unit are of the same type, as dictated by the LMN innervating them. Following collateral sprouting (distal axon sprouting), whenever a motor neuron of one type reinnervates denervated muscle fibers of a different type, the reinnervated muscle fibers are transformed from their original type to that of the adopting motor unit. Thus, in addition to muscle fiber–type transformation related to changes in activity, muscle fiber–type transformation also follows reinnervation (Adams, 1975).

Motor Unit Size and Distribution As stated throughout this textbook, the motor unit is defined as a single AHC, the muscle fibers it innervates, and the intervening NMJs. The total number of muscle fibers belonging to an individual motor unit (the innervation ratio) dictates the size of the motor unit. The size of the motor unit is inversely proportional to the control required of that muscle. Thus, most motor units contain several hundred muscle fibers: those requiring higher degrees of control (e.g., extraocular muscles) have lower values, and those requiring lower degrees of control

79

Section 1: Introductory Chapters

(e.g., gastrocnemius) have higher values. The muscle fibers innervated by a single AHC, although confined to a small area of the muscle (the territory of the motor unit), are not adjacent to each other, but rather are intermingled with the muscle fibers of approximately 30 other motor units. The density of the muscle fibers belonging to the same motor unit is slightly greater at the center of its territory (Buchthal, 1980). The transverse diameters of the muscle fibers belonging to a motor unit also vary in size, depending on gender and the specific muscle. For example, in adult males, the transverse diameters of quadriceps muscle fibers range from 40 to 80 micrometers, whereas in adult females, they range from 30 to 70 micrometers (Dubowitz, 2013). The NMJ of an individual muscle fiber is centrally located. Therefore, the arrangement of the muscle fibers composing a muscle, termed its muscle architecture, dictates the orientation and distribution of the endplate zones of that muscle. When the muscle fibers are oriented parallel to the axis of force of the muscle, the muscle is said to be a longitudinal muscle (e.g., the biceps). With longitudinal muscles, the endplate zone will have a linear shape and run perpendicular to the muscle fibers (i.e., across the belly of the muscle). However, when a muscle has a more complex arrangement of its muscle fibers, such as with pennate muscles, then the distribution of its endplate zone is more complex. A pennate muscle (from the Latin word, pinnatus, for feathered wing) is one in which its muscle fibers are obliquely oriented to the tendon rather than longitudinally oriented. Depending on the intramuscular tendon architecture, there are unipennate muscles, bipennate muscles, and multipennate muscles. When the muscle fibers lie on one side of the tendon, the muscle is unipennate, whereas when the muscle fibers lie on both sides of a central tendon, it is referred to as bipennate (e.g., the rectus femoris). If the central tendon branches, the muscle is termed

References Adams, RD. Diseases of muscle: a study in pathology, 3rd ed. Hagerstown, MD: Harper & Row, 1975. Amato AA, Russell JA. Neuromuscular disorders. New York: McGraw-Hill, 2008. Bonner FJ, DevlescHoward AB. AAEM minimonography #45: the early

80

multipennate (e.g., the deltoid). Pennate muscles generate higher levels of force (because there are a greater number of muscle fibers oriented parallel to each other) but have smaller ranges of motion (because they are at an angle to the longitudinal axis [the direction of action] of the muscle). In addition, muscle fibers composing pennate muscles tend to be shorter. However, the shorter muscle fiber length does not reduce its ability to generate force, because the maximal (tetanic) force a muscle fiber generates reflects its cross-sectional area and the muscle fiber type. Shorter muscle fibers contract more slowly because they have a smaller number of sarcomeres per myofibril. (Recall that muscles are composed of extrafusal muscle fibers that are composed of myofibrils, which, in turn, are composed of actin and myosin filaments arranged in contractile units termed sarcomeres.) The purpose of understanding muscle fiber architecture is that is makes it easier to comprehend endplate zone distributions. During motor NCS, as will be discussed in Chapter 7, the belly-tendon method of electrode placement is used. As the name implies, with this method of electrode placement, the E1 surface electrode is placed over the belly of the muscle and the E2 electrode is placed over the tendon. Thus, for small muscles with longitudinally oriented fibers, such as the extensor digitorum brevis muscle of the foot, it is easy to place E1 over the endplate zone. However, for bipennate muscles, such as the rectus femoris, there is no single position where the E1 electrode can be centered over the entire endplate zone of the muscle. Another issue concerns the hypothenar and thenar eminences, which are composed of multiple muscles. When recording from these sites, unavoidable endplate zone overlap occurs. For these reasons, it is important to place the E1 electrode in the exact position defined by the technique used to collect the normal values.

development of electromyography. Muscle Nerve 1995;18:825–853. Brooke MH, Kaiser KK. Muscle fiber types: how many and what kind? Arch Neurol 1970;23:369–379. Brown WF. The physiological and technical basis of electromyography. London: Butterworth, 1984.

Buchthal F. The general concept of the motor unit. Publ Assoc Nerve Ment Dis 1961;38:28–30. Buchthal F. Motor unit of mammalian muscle. Physiol Rev 1980;60:90–142. Burke RE. Motor units in mammalian muscles. In Sumner AJ, editor, The physiology of peripheral nerve disease. Philadelphia: WB Saunders Company. 1980:133–194.

Chapter 5: Anatomy and Physiology of Muscle

Dubowitz V, Sewry CA, Oldfors A. Muscle biopsy: a practical approach, 4th ed. London: WB Saunders, 2013. Feinstein B, Lindegard B, Nyman E, Wohlfart G. Morphologic studies of motor units in normal human muscles. Acta Anat 1955;23:127–142. Galvani L. De viribus electricitatis in motu musculari. Commentarius. De Bononiensi Scientiarum et Artium Instituto atque Academia Commentarii 1791;7:363–418. Ghez C. The control of movement. In Kandel ER, Schwartz JH, Jessell TM, editors, Principles of neural science, 3rd ed. Norwalk: Appleton & Lange 1991:533–547. Hunt C. The reflex activity of mammalian small-nerve

fibres. J Physiol 1951;115 (4):456–469. Huxley HE, Hanson J. Changes in the cross-striations of muscle during contraction and stretch and their structural interpretation. Nature 1954;173:973–976. Huxley AF, Niedergerke R. Structural changes in muscle during contraction: interference microscopy of living muscle fibres. Nature 1954;173:971–973. Lai H, Mitsumoto H. Muscle physiology and pathophysiology. In Levin KH, Luders HO, editors, Comprehensive clinical neurophysiology. Philadelphia: WB Saunders 2000:81–87. Liddell EGT, Sherrington CS. Recruitment and some other factors of reflex inhibition. Proc R Soc B 1925;97:488–518.

McComas AJ, Fawcett PRW, Campbell LMJ, Sica REP. Electrophysiological estimation of the number of motor units within a human muscle. J Neurol Neurosurg Psychiatry 1971;34:121–131. Sherrington CS. Remarks on some aspects of reflex inhibition. Proc R Soc B 1926;97:519–547. Sherrington CS. Some functional problems attaching to convergence. Proc R Soc B 1929;105:332–362. Tarnopolsky MA. Metabolic myopathies. Continuum 2016;22:1829–1851. Wedell G, Feinstein B, Pattle RE. The electrical activity of voluntary muscle in man under normal and pathological conditions. Brain 1944;67:178–257.

81

Section

2

Nerve Conduction Studies

Chapter

6

Electrodes and Nerve Conduction Study Basics

History Historically, studies of motor nerves date back to the middle of the 19th century. In 1852, Herman von Helmholtz made the first measurements of median motor nerve conduction velocity, and in 1867, in collaboration with Baxt, he accurately estimated the median motor nerve conduction velocity to be 61.0 m/sec (this was based on the mechanical response of the muscle rather than on the collection of a motor response) (Bonner and DevlescHoward, 1995; Falck and Stalberg, 1995). The pioneering work of von Helmholtz led to the development of the string galvanometer. This device, which is also known as the Einthoven galvanometer, was invented by Einthoven in 1901 to record the electrical potentials in heart muscle associated with its contractions. Einthoven termed the recording an electrocardiogram (Einthoven, 1901). In 1908, using a string galvanometer, Piper recorded a median motor response and estimated its conduction velocity as 117 m/sec (Falck and Stalberg, 1995). He also authored the first textbook on electromyography, which was published in German in 1912 (Piper, 1912). In 1941, Harvey and Masland described the value of motor NCS for studying neuromuscular junction transmission (using the belly-tendon method; see Chapter 7), and a few years later, Harvey and colleagues reported its utility for the study of motor nerves (Wilbourn and Ferrante, 1997). Importantly, unlike earlier reports, which focused on motor nerve conduction velocity values, these latter two manuscripts focused on the amplitude values of the recorded motor responses. In 1948, Hodes, Larrabee, and German published their techniques for performing various motor NCS, which included amplitude, latency, and conduction velocity measurements (Hodes et al., 1948). A year later, Dawson and Scott reported median and ulnar NCS with the stimulating electrodes placed at the wrist and the recording surface

electrodes situated more proximally (Dawson and Scott, 1949). With this technique, the recorded median and ulnar responses were mixed (i.e., compound electrical potentials composed of motor and sensory nerve fiber action potentials. In 1956, three key papers were published that, together, dictated the routine application of various NCS in the EMG laboratory. First, Dawson described a technique for recording sensory responses in humans (Dawson, 1956). By stimulating the median or ulnar sensory nerve fibers at the appropriate digit and recording more proximally (wrist and elbow), these techniques eliminated the motor nerve fiber action potentials from the recording (i.e., they were sensory responses). Second, Simpson identified focal conduction slowing in the setting of median neuropathies at the wrist and ulnar neuropathies along the elbow segment (Wilbourn, 1994). Third, Lambert documented that certain peripheral nerve disorders affected the motor nerve conduction velocity values, whereas anterior horn cell disorders, muscle disorders, and certain other peripheral nerve disorders did not (Wilbourn, 1994). Two years later, in 1958, Gilliatt and Sears reported the use of median and ulnar sensory NCS to study peripheral nerve lesions in humans (Gilliatt and Sears, 1958). In 1959, Sears introduced an antidromic technique whereby the median or ulnar sensory fibers were stimulated at the wrist while recording at the appropriate digit (Sears, 1959). In 1970, the sural sensory NCS technique was introduced (Wilbourn, 1994).

Introduction In this chapter, several basic principles and concepts related to the performance of nerve conduction studies (NCS) are reviewed. This general discussion is applicable to all three types of NCS: motor, sensory, and mixed. Because these three types of NCS are performed using different techniques, the details

85

Section 2: Nerve Conduction Studies

specific to each are discussed in subsequent chapters: motor NCS in Chapter 7 and sensory and mixed NCS in Chapter 8. How the NCS measurements are affected by various disease processes is covered in Chapter 9. The appropriate application of EDX testing, which depends on the clinical scenario, permits the accurate localization and characterization of lesions involving the peripheral neuromuscular system, which, in turn, contributes to patient management. We discussed the generation and propagation of individual action potentials (APs) in Section 1 (see Chapters 3–5). From this point forward, however, we will be focusing on the elicitation and recording of semi-synchronous groups of APs (compound action potentials; compound electrical potentials), which are the ones collected during the EDX examination. The source of the individual APs comprising these compound electrical potentials varies with the particular study type being performed. With sensory and mixed NCS, the compound electrical potentials collected are composed of groups of nerve fiber APs, whereas with motor NCS (and needle EMG studies), the compound electrical potentials collected are composed of groups of muscle fiber APs. The compound electrical potentials recorded during the sensory NCS are termed compound sensory nerve APs (SNAPs; sensory responses); those collected during mixed NCS are termed mixed nerve APs (or mixed responses); and those collected during motor NCS are termed compound muscle action potentials (CMAPs; motor responses). In this textbook, the terms “motor response” and “CMAP” are used interchangeably, as are the terms “sensory response” and “SNAP.” Although the compound APs collected during the needle EMG examination and the motor NCS are both composed of muscle fiber APs, they are not equivalent (discussed further on). With the needle EMG study, the groups of muscle fiber APs collected belong to a single motor unit and, thus, are referred to as motor unit action potentials (MUAPs) (see Chapter 13). On motor NCS, however, because all of the motor axons in the nerve trunk under study are stimulated, the CMAP is composed of all of the MUAPs of the muscle under study, each of which, in turn, is composed of individual muscle fiber APs. Thus, the MUAPs collected during the needle EMG reflect the muscle fibers of individual motor units, whereas the CMAPs collected during the motor NCS reflect the muscle fibers of the motor units of the entire muscle.

86

Several attributes of these compound electrical potentials (i.e., SNAPs, CMAPs, and MUAPs) – which reflect only the larger-diameter, more heavily myelinated nerve fibers – are measured and analyzed. The smaller-diameter, lightly myelinated and unmyelinated nerve fibers do not contribute to these measurements. Thus, in clinical terms, only the sensory fibers conveying position, vibration, and discriminative touch and only the motor fibers innervating the extrafusal muscle fibers are assessed during the EDX examination. Consequently, when an individual with a stocking distribution of sensory complaints is referred for suspected polyneuropathy, a normal EDX study does not exclude the possibility of a small fiber sensory polyneuropathy (ideally, this should be stated in the report). The motor and sensory responses provide important information about the number of functioning nerve fibers within the nerve under study (amplitude and negative area under the curve measurements) and the maximum AP propagation speed of the nerve fiber group (nerve conduction velocity). Of these two types of information, the former is by far the most important, as will be repeatedly stated in this section of the textbook and throughout it.

Electrodes History During the early 20th century, when vacuum tubes were utilized to control the amount of current flowing through a circuit, the surface electrodes were referred to as the G1 and G2 electrodes. The letter “G” referred to the grid (the control electrode) of the vacuum tube. The vacuum tube (electron tube), which was invented by Fleming in 1904, controls the flow of current between two electrodes. It is an evacuated container (a glass tube) that contains metal electrodes. With this device, electrons flow in the evacuated space between the electrodes. By applying different electrical signals to the grid, the amount of current flowing through the vacuum tube is controlled. In the 1950s, transistors were invented. These components did the same thing that vacuum tubes did, but they were solid-state devices. Because they were built from solid materials (no evacuated glass space), they were less fragile. In addition, they were much smaller. Thus, vacuum tubes became obsolete, although they are still used in satellite amplifiers because they are generally more

Chapter 6: Electrodes and Nerve Conduction Study Basics

effective in high-powered, high-frequency applications. In the late 1960s, a large number of transistors and other electronic components were interconnected on a thin wafer of silicon to form an integrated circuit (i.e., a chip). Some integrated circuits contain billions of individual transistors. Because vacuum tubes are of limited utility (e.g., satellite amplifiers), the terms “G1” and “G2” are obsolete. Thus, in this textbook, the surface recording electrodes will be termed the E1 and E2 electrodes.

Surface Recording Electrodes For NCS, surface electrodes are utilized. There are two types: surface recording electrodes and surface stimulating electrodes. The surface recording electrodes are designated the E1 and E2 electrodes. Importantly, because both electrodes are identical in their composition, they are both equally capable of recording electrical signals. These electrodes are part of the circuit. One end is inserted into the amplifier box (referred to as the preamplifier by some EDX providers) and the other end is attached to the patient. In addition to the connections at its two ends, each recording electrode also contains internal connections between its ends. Importantly, when these junctions are composed of differing metals, a static voltage may develop (similar to a capacitor). Consequently, it is important that the E1 and E2 electrodes be identical in their composition so that their internal resistances are equal. This avoids impedance mismatch. Impedance mismatch results in the amplification of common mode signal (discussed later). Also, at the interface between the electrode and the patient, an electrochemical half-cell potential (voltage) is created that also has capacitive properties. Half-cells result when a conductive electrode is placed into a conductive lotion allowing electric charge to move between the two, thereby creating a potential difference between them (voltage; charge separation). When the half-cell potentials of the two surface recording electrodes are of differing values (i.e., a half-cell potential imbalance), the difference between the two potentials is amplified by the amplifier. Because half-cell potentials can range from 1 to 10 mV, it is possible for a half-cell potential imbalance to be larger than the collected motor response and, thus, far larger than the collected sensory response (Netherton, 2010). Thus, these issues must be considered in order to minimize the amplification of unwanted

signal (e.g., environmental noise). At this point, we can turn our attention to the desired signal.

Proper Placement of the Recording Electrodes The E1 electrode is placed as close as possible to the desired electrical activity (the desired signal). The E1 electrode is also referred to as the active electrode. Conversely, the E2 electrode is intentionally located away from the desired electrical activity (but not too far from the E1 electrode) and, therefore, is also referred to as the inactive electrode and the reference electrode. The latter term reflects the fact that the E1 electrode is referenced to the E2 electrode (i.e., the signal recorded by the E1 electrode is compared to the signal recorded by the E2 electrode). The ideal E2 electrode would include all of the environmental noise and none of the desired signal, but this is never the case. Because the E2 electrode generates electrical activity (it records the environmental undesired signal and, typically, some of the desired signal), it is best not referred to as the inactive electrode. If it were truly silent, the environmental noise signal could not be subtracted out by differential amplification. Using this approach, the difference between the two recorded signals represents the desired signal (because the desired signal is only present at the E1 electrode). This signal difference is amplified and is referred to as the differential signal. The undesired signal (e.g., the environmental noise), which is identical at the two electrodes, is eliminated through common signal rejection. Because the E2 electrode is active and, consequently, never has a zero potential, it influences the morphology of the recorded response (this is discussed in more detail later in this chapter and in the next two chapters). In addition, the distance between the E1 and E2 electrodes is important. When the E2 electrode is positioned too close to the E1 electrode, it simultaneously records some of the electrical activity being recorded at the E1 electrode. When this happens, the simultaneously recorded signal is subtracted out. Thus, there is signal loss. As a result, the recorded response amplitude is lower and, because there is less total signal, the recorded response peaks earlier (i.e., the peak latency value is decreased). The proximity of the E1 and E2 surface recording electrodes is especially important for the sensory mixed NCS because they are both positioned over the nerve fibers under

87

Section 2: Nerve Conduction Studies

study. Thus, to avoid the loss of the desired signal, they should be far enough apart so that the E2 electrode is not simultaneously recording the same electrical activity that the E1 electrode is recording. With motor NCS, however, because the muscle belly is so much larger than the surface recording electrodes, there is essentially no position on the muscle belly at which the E2 electrode can be positioned where it would not record, through volume conduction, some of the electrical activity being recorded at the E1 electrode. Consequently, with motor NCS, the E2 electrode is located off of the muscle belly. Surface recording electrode positioning pertinent to the specific type of NCS being performed is discussed further in Chapters 7 and 8. Another issue occurs when the E2 electrode is positioned too far from the E1 electrode. This causes the E2 electrode to have a different vantage point than the E1 electrode with respect to the undesired environmental signal. As a result, the undesired (common) signal is recorded differently by the two electrodes, and the difference is amplified. In other words, differences in the environmental electrical signal related to different electrode locations may be amplified as differential signal. Thus, the E2 electrode should not be too far from the E1 electrode. The bottom line is that there is no ideal site for the E2 electrode with respect to the E1 electrode. For this reason, it is important that it be situated at the exact same site that was used during the collection of the normal values employed by the EMG laboratory.

Bipolar versus Referential Recordings NCS recordings can be bipolar or monopolar. With bipolar recordings, the two electrodes (E1 and E2) are both active, although E1 is typically more active than E2, whereas with referential recordings, only E1 is active. For example, stimulating the superficial radial sensory nerve at the distal forearm segment and recording from the nerve near the snuff box with the E1 and E2 electrodes placed 3 centimeters apart (or using a bar electrode) is an example of a bipolar recording technique. Even motor responses, which are collected using the belly-tendon technique with the E1 and E2 electrodes much farther apart, are bipolar recordings, because the E2 electrode overlying the tendon affects the morphology of the waveform, especially the terminal (repolarization) portion of the waveform. However, when the E2 electrode is

88

positioned far enough away from the E1 electrode so that it does not influence the morphology of the recorded response, the recording is a referential (monopolar) recording.

The Surface Stimulating Electrodes The surface stimulating electrodes are used to stimulate the nerve under study and typically are fashioned as two prongs (the cathode [attracts cations because of its negative charge] and the anode [attracts anions because of its positive charge]) protruding from the undersurface of the stimulator, which is typically a handheld device. The function of the stimulator was previously discussed in the instrumentation chapter (see Chapter 2). The two prongs of the stimulator are positioned 2.5 cm–3 cm apart. The stimulator is oriented so that the cathode and anode both overlie the nerve with the cathode nearest to the E1 recording electrode (discussed later here). The cathode functions to depolarize the nerve fibers composing the nerve under study. The anode is located over the nerve as well, but further from the E1 electrode. However, to lessen the amplitude of the stimulus artifact, the anode is often rotated around the cathode so that it is positioned off of the nerve. This orientation accelerates the return of the stimulus-displaced trace to baseline, thereby minimizing its effect on the subsequently recorded response. The ideal degree of anodal rotation depends on the orientation of the shock artifact being generated by the stimulator and is discussed in significant detail later in this textbook (see Chapter 18). In addition to the E1 and E2 surface recording electrodes, a ground electrode is also attached to the patient and connected to the preamplifier. It generates the zero potential to which the E1 and E2 electrodes are compared. It is best positioned between the stimulating and recording electrode pairs. In this position, it reduces the stimulus artifact. However, the use of isogrounds limits this attribute (see Chapter 19).

The Basic Technique The surface recording electrodes are affixed to the skin overlying the muscle (motor NCS; discussed further on) or the nerve (sensory and mixed NCS; discussed in Chapter 8) either by taping them in place or by using adhesive electrodes. They must be properly secured so that they do not become dislodged when stimulation produces patient movement. Prior

Chapter 6: Electrodes and Nerve Conduction Study Basics

to their application, the skin must be prepared. First, the skin is cleansed of any loose debris (alcohol wipe) and any rough skin is removed (either using an abrasive cream or abrasive sandpaper-like material that is typically dispensed in a roll). This step lessens the impedance between the patient and the electrode. To avoid impedance mismatch, both sites must be equally prepared. Second, the recording electrodes are attached to the patient. There are two general types of surface recording electrodes, nonadhesive and adhesive. Nonadhesive surface recording electrodes are taped to the skin after a conductive paste (such as an electrolyte cream or lotion) is applied to them. These electrodes are made of a conducting metal that is attached to an insulated wire lead that is plugged into the preamplifier at its other end. The E1 and E2 electrodes must be identical in all aspects, including their metallic composition, and they must be equally clean in order to avoid impedance mismatch related to resistance differences between them. As previously stated, impedance mismatch causes the like signal (e.g., 60-Hz artifact) to be amplified because when the resistances of the two leads differs, the like signal produces different voltage drops across them, thereby mimicking differential signal (see Chapter 2). As their name implies, adhesive electrodes do not require tape to secure them. They also do not require the application of conductive paste (the adhesive contains conducting material). Thus, they are much more time efficient. These electrodes are usually sold in sheets of individual electrodes, which are peeled off of the sheet and affixed to the skin. They have a tab to which an alligator clip is attached. When adhesive electrodes are used, a separate lead is required for each electrode. Often, it consists of three leads (E1, E2, and ground) contained within a single insulated cable with the distal portion of the leads exposed so that the distal end of the three leads (the end with the alligator clips) can move away from each other. The other end of the cable is plugged into the preamplifier. Three separate leads with attached alligator clips can also be used, but this introduces more 60 Hz artifact (see Chapter 18). The nonadhesive electrodes must be cleansed between patients, whereas the adhesive electrodes are simply discarded. In addition, some recording electrodes are designed for a specific type of recording technique. For example, in our EMG laboratory, we prefer to use curved clip electrodes to study the digital nerves. With this type of recording electrode, a separate ground

electrode is required. It can be adhesive or nonadhesive in type. At this point, an electrolyte cream is applied to the stimulating electrodes (discussed later here) and the nerve under study is stimulated. The stimulus is incrementally increased until a response is evoked. The stimulus strength is then progressively increased further until the generated response stops increasing in size (termed a maximum response). To ensure that the response is indeed maximal, the intensity of the stimulus is increased slightly further. This final increment should be limited to 10–20% so that the nerve segment is not stimulated distal to the cathode. This phenomenon, termed stimulus lead, occurs because as the stimulus strength is increased, the strength of the electric field beneath the cathode increases radially, causing the region of depolarization to also expand. As a result, the nerve trunk is stimulated distal to the cathode (i.e., closer to the E1 electrode), causing the onset latency value to decrease. When this occurs, the onset latency value of the recorded response will be falsely decreased. When no further response increment occurs, the response is accepted as maximal. Importantly, although the stimulus strength is supramaximal, the response is not. In other words, the maximal response is elicited by a supramaximal stimulus. It is inaccurate to refer to the maximal response as supramaximal, because the response itself can never be more than maximal (see Figure 6.1). The morphology of the recorded compound electrical potentials reflects the voltage value changes over time. There is an initial departure from the baseline (the onset) and an early ascent to the peak of the response (the rise time). Both of these two events (the onset and the rise time) occur over short time periods, and consequently, the subcomponent frequencies composing them are high. The descent time is also relatively rapid. The frequency of any portion of the response is simply the number of times it can repeat in 1 second. For example, when a portion of the recorded response occurs over a 100 msec period, it has a frequency of 10 Hz, whereas a portion occurring over a 1 msec time period has a frequency of 1,000 Hz. In addition, the inflection points (i.e., the peaks and troughs) of the response demonstrate slope reversals (i.e., a change in direction from ascending to descending or from descending to ascending, respectively), and thus contain higher frequencies. However, other than at

89

Section 2: Nerve Conduction Studies

5 mV 5 ms

1

2

3

4

5

6

7

Figure 6.1 As the stimulus intensity is increased, more and more motor axons are stimulated. This continues until all of the motor axons composing the nerve under study are stimulated, at which point further increases in stimulus intensity do not generate further increases in motor response size. In this figure, the response increases in size from 1 to 7, but further increases in stimulus intensity do not result in a larger motor response. The size of response 8 is identical to that of response 7. Hence, regarding response 8, the stimulus intensity is supramaximal and the motor response is maximal.

8

the inflection points, the peaks and troughs are flatter (more horizontally oriented) than the rest of the response, and thus the frequencies composing them are overall lower. It is important to understand that different parts of the response reflect different subcomponent frequency compositions because it allows one to envision the response morphology changes that occur when the low frequency and high frequency filters are manipulated. The deleterious effects of overfiltering are discussed in more detail later in the textbook (see Chapter 18).

Important Basic Concepts Volume Conduction Unlike the electrons conducting through an insulated copper wire, which are confined to the copper wire, the ionic currents propagating along the axolemma traverse the membrane (exiting and entering). Examples of ions crossing the axolemma include the

90

inward flow of Na+ when the VGNCs open (positive current flows inward), the outward flow of K+ when the VGKCs open (positive current flows outward), or the transverse leakage current (i.e., the leakage of sodium current as it advances along the axon). Another major difference, and the one underlying the principle of volume conduction, concerns the tissue surrounding the nerve trunk. Unlike the copper wire, which is surrounded by an insulator (e.g., plastic), which, in turn, is surrounded by another insulator (air), the tissue surrounding nerve fibers is a conductor rather than an insulator. Thus, although pure water is a poor conductor, solutions containing charged ions (i.e., electrolytic solutions), such as body fluids, readily conduct electricity. For this reason, the extracellular fluid (ECF) surrounding nerve fibers is an excellent conductor. Consequently, the human body functions as a volume conductor because it is capable of conducting electricity. In the laboratory, when the action potential propagating along the axon is measured using two recording electrodes in which one is placed within the axon and the other is located outside the axon, a monophasic event is recorded. From the point of view of the intracellular electrode, it is a positive event (positivity moving toward it), whereas it is a negative event from the point of view of the extracellular electrode (positivity moving away from it). Because the human body is a volume conductor, the current traveling along the external aspect of the nerve extends out into the ECF. As a result, when the aforementioned monophasic action potential is recorded from the body using an extracellular recording electrode, it has a triphasic morphology (discussed in detail further on). Another major difference is that the value of the recorded response is of considerably lower magnitude. Coulomb’s law, which measures the magnitude of the effect that two charges removed in space have on each other, indicates that the magnitude of their effect decreases with the square of the distance. Thus, the magnitude of the charge decreases exponentially over distance away from the axon (i.e., the magnitude of the current is greatest along the axon and exponentially decreases as the distance from the axon increases). A dipole is two point charges of equal magnitude and opposite sign (i.e., no net charge). Because an action potential represents two dipoles (+/– followed by –/+), it is in effect a quadrupole. As a quadrupole, its magnitude decreases more rapidly over distance (i.e., it is inversely proportional to the

Chapter 6: Electrodes and Nerve Conduction Study Basics

C

B

A

Myelin

Myelin

Unidirectionally vs bidirectionally propagating APs B

Myelin

A

+30 –55 –70

Myelin

Figure 6.2 The relationship between the action potential and the recording electrode. The plus signs depict sodium ions, and the three dotted lines (labeled A, B, and C) represent reversals in the direction of flow of the sodium ions. Thus, as the action potential propagates past the recording electrode, the electrode “observes” four sequential events: (1) sodium ions moving toward it (between the A line and the membrane in the illustration), (2) sodium ions moving away from it (between the A and B lines in the illustration), (3) sodium ions moving toward it (between the B and C lines in the illustration), and (4) sodium ions moving away from it (between the C line and the membrane in the illustration).

C

cube of the distance rather than to the square of the distance) (Lagerlund, 1996). As previously stated, when a propagating action potential is collected using a pair of surface recording electrodes, the morphology of the recorded waveform changes from monophasic (negative) to triphasic (positive-negative-positive) (see Figure 6.2). In electrical terms, there is a negative sink (positivity moving away from the observer) and adjacent current sources (positivity moving toward the negative sink). It is important to understand these terms, as they permeate discussions of action potentials. The term sink is applied to situations in which, from the vantage point of an observer, the current moves from a more visible compartment to a less visible one (analogous to water moving from the sink to the drain). Thus, when current enters a cell, because the observer is outside of the cell, the point of observation is also outside the cell. Thus, the cell represents the sink for any positive charges flowing into it and the source for any positive charges flowing out of it. (These terms would be reversed if negative charges were being considered, but that does not apply to a propagating action potential.) In addition, because positivity is moving away from the observer, it registers as a negative event, and hence the phenomenon is more accurately termed a negative sink. Because the action potential is moving, the current sources on the two sides of the negative sink are termed the leading source current and the trailing source current. The

potential recorded reflects the leading source current, the negative sink, and the trailing source current and is triphasic. The first positive phase is referred to as the initial positive phase and the second positive phase is referred to as the terminal positive phase. Between these two positive phases is the negative phase. In other words, as the Na+ enters the axoplasm and moves bidirectionally along the axon, it drives the positive charges located along the ECF surface of the axolemma away from the axolemma (capacitive current) and toward the negative sink created at the Na+ entry site. Because the ECF conducts electricity, the positive ions displaced from the external aspect of the axonal membrane follow an arc-shaped course toward the negative sink. Consequently, two loops of positive ionic current, both capacitive in nature, are formed on each side of the negative sink. By definition, because positivity is moving toward the observer, these two regions are referred to as current sources. It is important to understand what these events represent with respect to ionic conductance. The propagating action potential can be thought of as a propagating negative sink that is preceded and followed by a positive source current. With respect to a surface recording electrode positioned over the nerve, the positive leading source current precedes the negative sink, which, in turn, precedes the positive trailing source current (+/–/+). Thus, although the action potential is actually two dipoles in sequence

91

Section 2: Nerve Conduction Studies

(leading positive current source–negative sink [+/–] and negative sink–positive trailing current source [–/+]), it appears as a propagating tripole to the recording electrode (+/– –/+ appears as +/–/+). By convention, negativity generates an upward deflection and positivity causes a downward deflection. Thus, the electrode records a positive phase, followed by a negative phase, followed by a positive phase (i.e., a triphasic waveform). Thus, four deflections result in three phases (see Figure 6.2). In some settings, however, a triphasic waveform morphology is not observed. For example, when the potential is recorded at its inception, it will not have an approaching phase (initial positive phase) and, thus, will be biphasic (– +). An example of a biphasic electrical potential is a motor response (discussed in Chapter 7). As another example, when the propagating potential is unable to conduct past the recording electrode, it will not have a trailing phase (terminal positive phase) and, hence, will be biphasic (+ –). An example of such a potential is a positive sharp wave (see Chapter 14). The two positive phases of the triphasic potential are not identical. The leading positive source current is denser because it is on the side of the negative sink in which the action potential is advancing, whereas the trailing positive source current is less dense because it is on the opposite side of the negative sink. This is analogous to the Doppler effect that a moving vehicle has on the surrounding sound waves it creates as it passes a specific point in space. At the front of the vehicle, the sound waves are more compressed, whereas they are less compressed behind the vehicle. The higher density of the more compressed sound waves at the front of the vehicle creates a more intense sound than does the lower density of the less compressed sound waves trailing the vehicle. In a similar manner, the higher charge density of the leading positive source current causes the positive phase associated with it to be steeper and deeper than that of the less dense trailing positive source current. Thus, the two positive phases are not symmetric. The first one is more abrupt in onset and larger in amplitude, whereas the last one is smaller in amplitude and blunter in shape. The triphasic appearance of the action potential that results from recording it in a volume conductor can also be explained using solid angle analysis of the charge changes as the potential conducts past the recording electrodes (Kimura, 2001). The efflux of K+ occurring with membrane

92

repolarization adds to the final positive phase (positivity moving toward the recording electrode).

Orthodromic versus Antidromic The terms orthodromic and antidromic refer to the direction of AP propagation. With orthodromic conduction, the APs conduct as they would physiologically (i.e., in the distal direction [centrifugally] for motor axons and in the proximal direction [centripetally] for sensory axons). When APs propagate in the direction opposite to their physiologic direction (i.e., centripetally along motor axons and centrifugally along sensory axons), the term “antidromic” is applied. Physiologically generated nerve fiber APs are unidirectional, because the absolute refractory period inhibits the entering sodium currents from re-exciting axon segments that were just depolarized. Thus, for example, when an AP reaches a node and triggers Na+ influx, although the Na+ current advances bidirectionally into the axoplasm, only the distally advancing Na+ current continues down the nerve, because the proximal axon segment is inexcitable (absolute refractory period) (see Chapter 3). Thus, the absolute refractory period maintains unidirectional AP propagation among physiologically generated APs. With APs generated by transcutaneous stimulation, the axons are excited by the cathode of the stimulator. As a result, when the sodium current enters the axoplasm and advances bidirectionally, it does not encounter an inexcitable membrane segment on either side of its entry point. Thus, the APs generated propagate bidirectionally. The stimulated segment becomes inexcitable, so the bidirectionally propagating APs cannot back-activate this segment. When we perform motor NCS, we record from the muscle and stimulate the nerve innervating that muscle. Thus, the evoked action potentials move distally along the motor axons (orthodromically). When we perform sensory NCS, it is possible to stimulate the sensory axons distally while recording proximally (orthodromically) or to stimulate them proximally while recording distally (antidromically). For example, a pure sensory nerve can be stimulated proximally and recorded distally (antidromic technique). An example of this technique is the superficial radial sensory NCS. As another example, a mixed nerve can be stimulated proximally (generating sensory and motor nerve fiber action potentials) while

Chapter 6: Electrodes and Nerve Conduction Study Basics

recording distally over isolated sensory nerve fibers (only sensory nerve fiber action potentials are collected). This would be an antidromic sensory NCS technique, an example of which is the median sensory NCS recording from one of the digits. Finally, the mixed nerve could be stimulated distally at a site containing only sensory axons while recording more proximally over a site containing sensory and motor nerve fibers. In this setting, only sensory nerve fiber action potentials would be generated and collected. This would be an orthodromic technique, an example of which would be the median sensory NCS stimulating one of the digits while recording over the wrist. Lastly, with mixed NCS, the mixed nerve is stimulated distally and the evoked action potentials are collected more proximally. Examples of mixed NCS are the median palmar, ulnar palmar, medial plantar, and lateral plantar mixed NCS. With mixed NCS techniques, the sensory response component is orthodromic and the motor response component is antidromic. Orthodromic and antidromic sensory NCS are discussed in detail in Chapter 8.

The Stimulating Electrodes and Their Proper Placement Physiologically, APs are generated proximally for motor neurons (axon hillock; initial axon segment) and distally for sensory neurons (at the receptor). The APs are generated by a reversal of polarity at the AP generation site. The precipitated AP then propagates unidirectionally in an all-or-nothing manner. The direction of physiologic AP propagation is termed orthodromic (centrifugal for motor axons and centripetal for sensory axons). In the research laboratory, an action potential can be generated by the direct passage of current through the membrane; a microelectrode is placed within the axon (intracellular) and another one is placed extracellularly, with the intracellular one relatively more positive than the extracellular one (i.e., the opposite polarity of the resting membrane). In this manner the negative electron current is responsible for axon depolarization. In this laboratory setting, the electrical potential recorded exhibits a single phase (monophasic), because the axon is surrounded by a nonconducting medium, such as oil or air, limiting the ability of the current to extend out from the nerve trunk. In the EMG laboratory, the motor, sensory, and mixed NCS are performed using extracellular

electrodes for stimulation (the prongs of the stimulator), termed the cathode (attracts cations because of its negative charge) and the anode (attracts anions because of its positive charge). The stimulator is positioned over the nerve with the cathode closest to the E1 electrode and the anode furthest from the E1 electrode. A stimulus is delivered and incrementally increased until a response is evoked. The stimulus strength is then progressively increased further until the generated response stops increasing in size (termed a maximum response). At this point, the intensity of the stimulus is increased slightly further to ensure that the response does not increase further. If no further increment occurs, the response is accepted. Although the stimulus strength is supramaximal, the response is not. In other words, the maximal response is elicited by a supramaximal stimulus (ensures it is indeed maximal). It is inaccurate to refer to the response as supramaximal. The stimulus strength increment should be limited to no more than 20% to avoid depolarizing the nerve segment distal to the cathode, which would result in an erroneously shorter latency value (Buchthal and Rosenfalck, 1966). As the stimulus strength is increased, the strength of the electric field beneath the cathode increases radially, causing the region of depolarization to expand also. As a result, the nerve trunk under study may be stimulated distal to the cathode (i.e., closer to the E1 electrode), causing the onset latency value to decrease. This phenomenon is referred to as stimulus lead. Initially, the electrons flow from the cathode to the anode, via the ECF, and do not affect the nerve under study. However, as the stimulus intensity increases, the negativity surrounding the cathode increases, making the tissue external to the axolemma more and more negative. As this occurs, the positive charges along the external aspect of the axolemma move toward the cathode (attractive electrostatic force), thereby releasing the negative ions from the internal aspect of the axolemma. This causes the polarity across the axolemma to diminish (i.e., the –70 mV TMP becomes less and less negative). When the TMP reaches the depolarization threshold (–55 mV to –40 mV), the VGNCs open and Na+ enters the axoplasm. The nerve fibers with the largest diameter have the lowest depolarization threshold and, therefore, are favored to depolarize first. However, the nerve

93

Section 2: Nerve Conduction Studies

fibers within the nerve trunk located nearest to the cathode will be depolarized at lower stimulus intensities than those located more distant from the cathode (i.e., the deeper ones). Thus, nerve fiber recruitment is not always from largest to smallest.

References Bonner FJ, DevlescHoward AB. AAEM minimonograph #45: the early development of electromyography. Muscle Nerve 1995;18:825–853. Buchthal F, Rosenfalck A. Evoked action potentials and conduction velocity in human sensory nerves. Brain Res 1966;3:1–119. Dawson GD. A summation technique for the detection of sensory and motor nerve fibres in man. J Physiol (Lond) 1956;131:436–451. Dawson GD, Scott JW. The recording of nerve action potentials through sin in man. J Neurol Neurosurg Psychiatry 1949;12:259–267.

94

Whether activated or not, the slowly conducting thinly myelinated and unmyelinated nerve fibers do not contribute to the NCS responses recorded. Thus, only the larger, more heavily myelinated nerve fibers are studied by NCS.

findings. J Clin Neurophysiol 1995;12:254–279. Gilliatt RW, Sears TA. Sensory nerve action potentials in patients with peripheral nerve lesions. J Neurol Neurosurg Psychiatry 1958;21:109–118. Hodes R, Larrabee MG, German W. The human electromyogram in response to nerve stimulation and the conduction velocity of motor axons. Arch Neurol Psychiatry 1948; 60:340–365. Kimura J. Electrodiagnosis in disease of nerve and muscle: principles and practice. Oxford, Oxford University Press, 2001.

Einthoven W. Ein neues galvanometer. Drudes Ann Physik, 1901.

Lagerlund TD. Volume conduction. In Daube JR, editor, Clinical neurophysiology. Philadelphia, FA Davis Company, 1996:29–39.

Falck B, Stalberg E. Motor nerve conduction studies: measurement principles and interpretation of

Netherton BL. Electrodiagnostic instrumentation: understand and manage it. In Neurophysiology and

Instrumentation Course. 57th Annual Meeting of the AANEM, Quebec City, Quebec, Canada, 2010:21–28. Piper H. Elektrophysiologie menschlicher Muskeln. Berlin, Verlag Julius Springer, 1912. Sears TA. Action potentials evoked in digital nerves by stimulation of mechanoreceptors in the human finger. J Physiol (Lond) 1959;148:30–31. Wilbourn AJ. How can electromyography help you? Postgrad Med 1983;73:187–195. Wilbourn AJ. Sensory nerve conduction studies. J Clin Neurophysiol 1994;11:584–601. Wilbourn AJ, Ferrante MA. Clinical electromyography. In Joynt RJ, Griggs RC, editors, Clinical neurology, Philadelphia: LippincottRaven, 1997:1–76.

Chapter

7

Motor Nerve Conduction Studies

Introduction The motor NCS were the first type of NCS to be applied clinically. They are performed by recording the motor response from a muscle while stimulating the nerve innervating that muscle at one or more sites. The stimulated nerve can be either a pure motor nerve (e.g., the suprascapular nerve) or a mixed nerve (e.g., the median nerve). Because the recording electrodes are positioned over the muscle, the motor response is composed of the individual muscle fiber APs. For this reason, the motor response, or M-wave, is referred to as a compound muscle action potential (CMAP). In the normal state, it reflects the muscle fibers composing the muscle and, indirectly, the motor axons composing the nerve branch innervating the muscle. Because, in general, each motor axon innervates hundreds of muscle fibers, a large number of muscle fiber APs contribute to the CMAP. As a result, it is much larger than the sensory response (see Chapter 8). Because of its larger size, the motor response is much less vulnerable to physiologic temporal dispersion. For this reason, much longer segments of the nerve trunk can be assessed by motor NCS than by sensory NCS. In addition, their large size also renders them less susceptible to physical factors, such as those related to longer distances between the recording electrodes and the AP generation site (e.g., obesity; edema), and to previous trivial trauma. Unlike sensory NCS, which only assess sensory nerve fibers, the motor NCS assess the neuromuscular junctions and muscle fibers, in addition to the motor nerve fibers. These concepts are discussed in more detail in this chapter.

Recording Technique Unlike the sensory and mixed NCS (see Chapter 8), the motor NCS assess all portions of the lower motor neuron: the cell body, the axon, and the terminal branches. In addition, because the surface recording

electrodes are situated over the muscle, they also assess the muscle fibers innervated by the activated motor axons and the intervening neuromuscular junctions (NMJs).

Belly-Tendon Method Motor NCS are performed using surface recording electrodes and the belly-tendon method. It is termed the belly-tendon method because the E1 surface recording electrode is affixed over the center of the muscle belly (i.e., over the motor point of the muscle) and the E2 electrode is situated over the tendon of the muscle (Isley et al., 1993) (see Figure 7.1). The motor points of the body, which were initially mapped out by Von Sie Siemssen in the mid-19th century, are the sites where the nerve branches innervating the muscles enter the muscle tissue, and consequently are also the sites at which the endplates of the individual muscle fibers are located (Bonner and DevlescHoward, 1995). For this reason, a motor point is also referred to as an endplate region. This is the site at which the electrical activity of the individual muscle fiber APs is generated (i.e., the negative sinks of the muscle fibers composing the muscle). Because the stimulating electrodes are positioned over the nerve, proximally, and the recording electrodes are positioned over the muscle-tendon unit, distally, this is an orthodromic technique (i.e., the evoked APs traverse the nerve fibers in their physiological direction, from proximal to distal).

Biphasic Morphology Because the E1 electrode is placed over the motor point of the muscle, the muscle fiber APs are generated below it, and the CMAP recorded demonstrates its shortest rise time and highest amplitude. Thus, the CMAP is recorded at its inception, and hence there is no positive leading phase (initial source current). For this reason, the motor response has

95

Section 2: Nerve Conduction Studies

E2 E0 E1

Anode

(+)

Cathode

(–)

Stimulator

Figure 7.1 The belly-tendon method. As shown in the illustration, when the belly-tendon method is used to record the motor response, the E1 electrode is positioned over the belly of the muscle and the E2 electrode is positioned over the tendon. In this position, the E2 electrode, which is always somewhat active (i.e., it is never completely inactive), contributes minimally to the configuration of the recorded waveform, especially its repolarization component. For this reason, it must be placed in the exact position that was used to collect the normal control values of the technique being performed. Because the E1 electrode overlies the endplate region (i.e., the site at which the action potentials are generated), the E1 electrode does not record an approaching phase. As a result, the motor response is biphasic. (Modified with permission, Isley MR, et al. Electromyography/Electroencephalography, SpaceLabs Medical, Biophysical Measurement Series, 1993.)

10 mV 5 ms

Figure 7.2 Biphasic motor response.

a biphasic configuration (negative-positive) (see Figure 7.2). The initial negative phase is approximately twice the size of the subsequent positive phase (Lee et al., 1975). These two phases represent the negative sink current (initial negative phase) and the terminal source current (subsequent positive phase) of the activated muscle fibers. In other words, because the inception of the CMAP is below the E1 electrode, there is no approaching phase (initial source current) visualized. For this reason, whenever the motor response has an initial positive phase (i.e., a triphasic appearance), the E1 electrode should be repositioned to ensure that it overlies the motor point (discussed below).

96

The Influence of E2 Placement The E2 Electrode Is Not Inactive The E1 and E2 electrodes were previously referred to as the active (E1) and inactive (E2) electrodes. However, the E2 position, although it is positioned over the tendon, is not inactive because it records, via volume conduction, the muscle fiber APs (Kincaid et al., 1993; Falck and Stalberg, 1995). Therefore, the E2 electrode contributes to the morphology of the waveform, especially its repolarization side. In general, the E1 electrode is positioned over the motor point of the muscle (typically over the center of the belly of the muscle) where the negative phase is the largest, and the E2 electrode is positioned off of the muscle, where the negative phase is less visible (e.g., overlying the tendon). Notably, even off of the muscle, the E2 electrode is never completely inactive (i.e., it always sees some of the motor response through volume conduction and thus is never at zero potential). Because every remote site is active to some degree, it is important that the E2 electrode be situated at the exact site it was placed when the normal values for the technique being utilized were collected. This is especially true for the ulnar motor response, recording from the hypothenar eminence, where the E2 electrode causes the recorded motor

Chapter 7: Motor Nerve Conduction Studies

5 mV

5 ms 5 mV

5 ms

Figure 7.3 Ulnar motor response with two peaks. Note that the cursors are misplaced at the onset of the trace.

5 mV

5 ms

Figure 7.4 Low motor response related to E2 electrode malpositioning. In the top trace, the belly-tendon method is used to record the ulnar motor response (hypothenar eminence), whereas in the bottom trace, the E2 electrode is placed over the hypothenar eminence, distally. With the latter placement, the amplitude and the negative area under the curve values are diminished.

5 mV

Figure 7.5 Misplacement of the E2 electrode at a more inactive location may generate a larger motor response and result in a falsely negative study. In the illustration, the top trace is recorded with the E2 electrode correctly placed along the medial aspect of the proximal phalange of the index finger, resulting in an ulnar motor response (recording first dorsal interosseous) with an amplitude value of 10.3 mV. In the bottom trace, the E2 electrode is placed along the medial aspect of the third digit (the long finger), resulting in an amplitude value of 15.7 mV.

response to have a second peak (Kincaid et al., 1993) (see Figure 7.3).

Improper E2 Placement Affects Amplitude Because it is not electrically silent, precise placement of the E2 electrode is just as important as that of the E1 electrode. When the E2 electrode is positioned too close to the E1 electrode, such that it overlies the distal portion of the muscle belly, some of the desired muscle signal will be lost and the response amplitude will be submaximal (see Figure 7.4). When the E2 electrode is not positioned at the site used to collect the normal values, but at a less active site, the quantity of subtracted signal is reduced and

5 ms

Figure 7.6 Ulnar motor response (recording hypothenar eminence) waveform changes related to E1 placement site. The top trace is recorded with the E1 electrode in the center of the muscle belly, the center trace with the E1 electrode positioned 2 cm more medially, and the bottom trace with the E2 electrode positioned 2 cm more laterally.

the amplitude is falsely increased, which can result in a falsely negative study (see Figure 7.5). Errors in E1 placement produce even greater changes in waveform morphology (see Figure 7.6).

Positive Dip When the E1 electrode is improperly positioned (i.e., not situated over the motor point), a positive deflection appears at the onset of the motor response, termed a positive dip. In other words, the CMAP has

97

Section 2: Nerve Conduction Studies

5 mV

5 ms

Figure 7.7 A motor response demonstrating a positive dip.

a triphasic morphology rather than a biphasic one (see Figure 7.7). This occurs because the muscle fiber APs are generated away from the E1 electrode and propagate toward it (i.e., there is an approaching phase). Hence, whenever a positive deflection is noted at the onset of the motor response, the E1 electrode should be repositioned and the stimulus repeated. The duration of the positive phase is proportional to the distance of the E1 electrode misplacement from the endplate region. There are other reasons for the presence of a positive dip, such as muscle atrophy, the activation of an adjacent muscle, and the presence of an innervation anomaly. In this setting of an atrophic muscle, relocation of the E1 electrode does not lead to disappearance of the positive dip. When a positive dip is observed with proximal stimulation but not with distal stimulation, excessive stimulus current may have resulted in the inadvertent excitation of an adjacent nerve at the proximal stimulation site, causing the activation of muscles in the vicinity of the E1 electrode. The latter propagates toward the E1 electrode, producing the positive dip. The opposite phenomenon can occur (positive dip with distal stimulation but not with proximal stimulation) when the distal stimulation site stimulates more than one nerve (e.g., stimulation at the wrist resulting in median and ulnar nerve activation). A positive dip involving only the proximal motor response may also be observed with a MartinGruber anastomosis when there is concomitant carpal tunnel syndrome (the crossover fibers bypass the carpal tunnel and reach the E1 electrode before those traversing the carpal tunnel). This is discussed in more detail in Chapter 18.

98

Using a Needle Electrode for E2 In the past, some electromyographers used needle recording electrodes for collecting the motor responses, because the pickup area of the needle electrode could be placed much closer to the muscle fibers. Although this technique more accurately reflects the onset latency of the motor response, the amplitude value recorded only reflects those muscle fibers in the immediate vicinity of the needle electrode (typically only 1–3 muscle fibers). With needle electrode recording, the CMAP amplitude correlates with the distance between the pickup area and the contributing muscle fibers rather than with the total number of functioning muscle fibers. Because the amplitude measurement is much more informative than is the latency measurement, this technique is rarely indicated (e.g., to verify nerve continuity when volitional MUAPs cannot be recruited).

Nerve Stimulation For all of the basic motor NCS performed (recording distal muscles), the nerve is stimulated at two sites, one distal and one proximal, each of which generates a motor response, termed the distal motor response and the proximal motor response, respectively. The stimulation sites are chosen based on where the nerve under study is most superficial (typically at or near joints), so that it is amenable to transcutaneous stimulation. For example, for the median motor NCS, recording from the thenar eminence, the wrist serves as the distal stimulation site and the elbow serves as the proximal stimulation site. The nerve portion between the two stimulation sites is referred to as a nerve segment. Thus, for the median nerve study just described, the nerve segment between the stimulation sites is referred to as the elbow-wrist segment or as the forearm segment. The distal motor response is collected first. The stimulator is set on low stimulus duration and low stimulus intensity, and the parameters are increased until a motor response is observed, and then further increased until the maximum response is obtained. Once the maximum response is collected, the stimulator is turned up a little more (no more than 10–20%) and a final stimulus applied. This is to ensure that the maximum response has been collected (see Chapter 6, Figure 6.1). Although the final stimulus is a supramaximal stimulus, the response it evokes can never be supramaximal.

Chapter 7: Motor Nerve Conduction Studies

10 mV 5 ms Stimulus site

Lat ms

Amp mV

1. Wrist

3.6

13.7

2. Elbow

7.0

12.8

Figure 7.8 The waveform morphology of the proximal median motor response, recording thenar eminence, is essentially identical in appearance to that of the distal median motor response. There is a slight increase in the duration of the negative phase and a small reduction in response amplitude. The only major difference between the two responses is the onset latency value, which reflects the different stimulation sites of the two responses (wrist stimulation versus elbow stimulation).

Thus, it is erroneous to refer to the maximal response as a supramaximal response. Throughout this period of stimulation, it is important to monitor the morphology of the waveform. The waveform morphology of all of the collected responses should be similar in appearance albeit of varying size. If its morphology suddenly changes, it is likely that an adjacent nerve was inadvertently stimulated with the addition of a separate motor response from a different muscle in the vicinity of the recording electrode pair. It is also important to monitor the muscle under study to ensure that it is the one being activated. For example, when stimulating the median nerve at the wrist while recording from the thenar eminence, the thumb should abduct with each stimulation. The sudden addition of another movement (e.g., finger abduction) suggests inadvertent stimulation of an adjacent nerve (e.g., the ulnar nerve). Once the maximum distal response is collected, the same technique is applied to the nerve more proximally. Except for a large onset latency difference, these two responses should be nearly identical in appearance. Due to physiologic temporal dispersion (discussed later here), the proximal response is slightly longer in duration and, due to phase interactions among the contributing muscle fiber APs, often slightly lower in amplitude (see Figure 7.8). If the responses are of different sizes, then it may be that the smaller response is submaximal or that the larger response is related to volume conduction related to inadvertent stimulation of an adjacent nerve (stimulation of the median and ulnar nerves rather than just the desired nerve). Usually this is

associated with a sudden change in waveform morphology and is readily identified. Pathological reasons for differences between these two responses (e.g., demyelinating conduction block and early axon loss) are discussed in Chapter 9. Differences may also be observed with median-to-ulnar crossovers (referred to as Martin-Gruber anastomoses) (see Chapter 18). For those motor NCS techniques in which the only stimulation site is the supraclavicular fossa (e.g., suprascapular nerve, recording infraspinatus; axillary nerve, recording deltoid), only a single motor response is collected.

Physiologic Temporal Dispersion Temporal dispersion refers to the arrival time difference of the fastest and slowest muscle fiber APs composing the CMAP. There are physiological reasons for these differences, including differences in their nerve and terminal nerve conduction times (primarily due to the different lengths of the terminal nerves), NMJ transmission times, and muscle fiber conduction times (due to muscle fiber diameter differences and differences in distance between the different muscle fibers and the E1 electrode). Because the individual muscle fiber APs composing the CMAP do not arrive at the E1 electrode simultaneously, there may be overlap between their negative and positive phases. This muscle fiber AP overlap results in signal cancellation (amplitude decrement). As the nerve is activated more and more proximally, the distance between the cathode and the E1 electrode increases. This results in loss of synchrony between the individual nerve fiber APs, which in turn increases the duration of the response recorded and the amount of phase cancellation occurring between the muscle fiber APs. This concept is important to appreciate because it is a normal phenomenon and must not be misinterpreted as pathological. It is analogous to two vehicles driving on the interstate at speeds that differ by 10 mph (e.g., 40 mph and 50 mph). Despite their constant speeds, the faster vehicle gets further and further ahead of the slower vehicle over time (every hour the faster vehicle is 10 miles further from the slower vehicle). Thus, at 1 hour, the faster car is 10 miles ahead of the slower vehicle, whereas at 2 hours they are 20 miles apart. Analogously, the individual nerve fiber APs traveling along the nerve become more and more separated as they propagate over longer and longer nerve segments. This

99

Section 2: Nerve Conduction Studies

dispersion (loss of synchrony) results in greater degrees of negative and positive phase interaction, with resultant signal cancellation. The degree of signal loss increases with greater distances. Because of its larger amplitude (mV), longer negative phase, and the narrower range of motor conduction velocities, motor NCS are less susceptible to physiologic temporal dispersion than are the sensory NCS. This permits much longer segments of nerve to be studied during motor NCS than during sensory NCS. As a general rule, for motor NCS, when the interstimulus distance is less than 20 cm, the degree of amplitude loss related to physiologic temporal dispersion should not exceed 20% (Dumitru, 1994). In our EMG laboratory, we refer to this as the 20:20 rule. As an aside, computer simulation studies have suggested that physiologic temporal dispersion can produce up to a 50% amplitude decrement (this would require maximal phase cancelation if the negative phase to positive phase ratio was 2:1) (Rhee et al., 1990). Not infrequently, however, the ratio is lower, in which case greater degrees of phase cancelation are theoretically possible. Among the routine motor NCS, normal tibial motor responses often show significant amplitude decrements (20–50%) (see Figure 7.9). When attempting to determine whether a large drop in tibial response amplitude is related to focal demyelinating conduction block (discussed in detail in Chapter 9) or to physiologic temporal dispersion, the motor nerve is stimulated at multiple sites along its course, from the distal stimulation site toward the

proximal stimulation site, seeking a specific site at which the motor response suddenly drops. When this is identified, the responsible process is focal demyelinating conduction block. When there is uniform decrement over distance, physiologic temporal dispersion is indicated. The needle EMG study may also be helpful in this discrimination. With a pathologic demyelinating conduction block of 50%, a neurogenic MUAP recruitment pattern may be observed (see Chapter 14). In addition, because a lesion capable of producing demyelination conduction block in 50% of the motor axons would likely disrupt at least a few axons, depending on the timing of the study, at least some fibrillation potentials are expected. This is one of the reasons why all three components of the EDX examination are always performed. In conclusion, if the amplitude decrement exceeds 50%, it is most likely pathological in nature. When the decrement is lower in magnitude but is demonstrated over a short distance, it also is more likely to be pathological in nature. However, when the amplitude drop occurs over a longer distance, physiologic temporal dispersion must be considered, especially when the distance between the distal and proximal stimulation sites is long (see Figure 7.10). 5 mV

5 ms

5 mV

Figure 7.9 Typical proximal tibial motor response amplitude decrement. In this illustration, the proximal response amplitude is 33% smaller than the distal response amplitude (5.8 mV versus 8.6 mV). The negative AUC values are more similar (22.5 mV-msec versus 20.8 mV-msec). In this example, there is only an 8% negative area under the curve decrement, consistent with the fact that negative area under the curve is more resistant to physiological temporal dispersion than is amplitude. Overall, the waveform morphology is similar, arguing against a technical error.

100

Figure 7.10 Normal peroneal motor study showing an amplitude decrement between the ankle and below-fibular head stimulation sites. The three peroneal responses were recorded with ankle, below-fibular head, and above-fibular head stimulation. The amplitude values are 5.6 mV, 3.3 mV, and 3.3 mV, respectively. Thus, the response recorded with below-fibular head stimulation is nearly 40% lower than the response recorded with ankle stimulation. Comparing the negative AUC values (16.5 mV-msec, 13.2 mV-msec, and 13.2 mV-msec, respectively) shows the difference to be small, indicating temporal dispersion. Overall, the waveform morphology is similar, arguing against pathologic temporal dispersion. Clinically, this was the asymptomatic lower extremity, and the other EDX studies performed on this limb were normal. An accessory peroneal nerve was not present (the technique of identifying an accessory peroneal nerve is discussed in Chapter 18).

Chapter 7: Motor Nerve Conduction Studies

In the latter situation, the percentage of negative area under the curve decrement between the distal and proximal motor responses with respect to the degree of negative phase duration prolongation can be used (Van Asseldonk et al., 2006).

Standard and Nonstandard Motor NCS In most EMG laboratories, the median (recording thenar eminence), ulnar (recording hypothenar eminence), peroneal (recording extensor digitorum brevis muscle), and tibial (recording abductor hallucis muscle) comprise the standard motor NCS. Other reliable motor NCS include the suprascapular (recording infraspinatus), axillary (recording deltoid), musculocutaneous (recording biceps), proximal radial (recording from the extensor aspect of the forearm, proximally), distal radial (recording from the extensor aspect of the forearm, distally), ulnar (recording first dorsal interosseous [FDI]), median (recording second lumbrical), femoral (recording rectus femoris), peroneal (recording tibialis anterior), and tibial (recording ADQP). The techniques for performing these NCS and their age-adjusted normal values are provided in Section 6 (see Appendix 5 and Appendix 6).

What We Measure and What It Means From each motor response collected, specific measurements are made. Each measurement provides unique information about the neuromuscular elements under study. In order to accurately interpret these measurements, EDX medicine providers must understand what each measurement represents. Following the two-point stimulation of the motor nerve, the distal and proximal motor responses are assessed. The important motor response measurements are: the amplitude, the negative area under the curve (AUC), the distal latency, the nerve conduction velocity (CV), and the negative phase duration. Normally, the amplitude, negative AUC, and duration values of the two responses are similar, whereas the onset latencies are quite different. The waveform morphologies of the two motor responses are compared and should be similar in appearance. Of all of the measurements made, the amplitude and the negative area under the curve (AUC) are by far the most important because they reflect the number of functioning (innervated) muscle fibers and, therefore, indirectly, the number of functioning

motor axons. These values have significant utility in that they enable to EDX provider to estimate the percentage of affected muscle fibers, thereby approximating the percentage of affected motor axons (because the innervation ratio is constant for a given muscle). Thus, because percentage is a 100-point scale, these values are much more specific than the 6-point MRC scale. With the latter, 0 is defined as no visible muscle movement, 1 is defined as muscle movement without joint movement, 2 is defined as joint movement with gravity removed, 3 is defined as joint movement against gravity, 4 is defined as joint movement against gravity plus some resistance, and 5 is defined as normal strength. When this is graphed (point score on the y-axis and percentage of strength on the x-axis), the line generated is nonlinear. Although 4 of the 5 scores represent specific points on the line, a score of 4 represents a range that encompasses about 80% of the graph (approximately 20–99%). In other words, a score of 3 indicates 20% strength and a score of 5 indicates 100%, leaving a score of 4 to be somewhere in between. For this reason, many neurologists add 4– and 4+ for a total of 8 different scores (0, 1, 2, 3, 4–, 4, 4+, and 5). With this approach, 4+ represents mild weakness, 4 represents moderate weakness, and 4– represents severe weakness. Unlike the 6-point MRC scale, in the acute to subacute time frame prior to reinnervation by collateral sprouting (i.e., when the innervation ratio of the muscle is uniform), the ratio of the distal CMAP of the affected limb to the distal CMAP of the unaffected limb generates a percentage, permitting the percentage of muscle fibers affected to be calculated on a 100-point scale that is also a linear scale (the utility of this approach is discussed in detail and demonstrated in Chapter 9 and in pertinent cases in Section 5). This same approach can be used in the setting of focal demyelinating conduction block to estimate the percentage of the block (see Chapter 9 and Section 5). Clinical strength assessment is less useful for approximating the degree of motor axon loss because it is time dependent. Although it is often stated that weakness is not recognized until approximately 50% of the motor axons of a nerve are involved, in addition to severity, the timing of the loss also dictates its perception. For example, acute losses of 25% are typically recognized, whereas very slow losses of 75% or more may be asymptomatic. Although the clinical assessment of strength has drawbacks as an indicator of the severity of motor

101

Section 2: Nerve Conduction Studies

axon loss, it is ideal for identifying the distribution (pattern) of weakness. In addition to the total number of functioning muscle fibers, the amplitude and negative AUC values are influenced by a number of other factors, including the diameters of the muscle fibers contributing to the response, the orientation of the surface recording electrodes with respect to the longitudinal orientation of the contributing muscle fibers, the distance between the surface recording electrodes and the motor nerve fibers generating the APs, and the synchrony of the propagating APs. Regarding muscle fiber diameter, larger-diameter muscle fibers generate greater currents and, thus, generate larger APs, whereas atrophic muscle fibers generate smaller currents and, thus, generate smaller APs. However, with severe disuse atrophy of proximal limb muscles, such as the deltoid muscle following a rotator cuff tear or the quadriceps after knee surgery, the amplitude of the recorded CMAP often is normal (Wilbourn, 1983). This phenomenon may reflect the fact that disuse leads to less internal cytoskeletal elements per myofibril but not to a decrease in the number of myofibrils. The fact that muscle fiber AP amplitude is proportional to muscle fiber diameter is more important in the assessment of fibrillation potentials, with higher-amplitude fibrillation potentials signifying recent denervation and lower-amplitude fibrillation potentials signifying chronicity (see Chapter 14). Regarding the depth of the motor unit within the muscle belly, the deeper muscle fibers of deeper motor units contribute less current to the CMAP than do the more superficial muscle fibers of the more superficial motor units. Because the E1 electrode cannot be equidistant from all of the motor units, there is always some dyssynchrony and resultant phase cancellation. The distal latency and conduction velocity values only reflect the fastest conducting fibers (i.e., a very small minority of the contributing fibers). The negative phase duration and the morphology of the waveform reflect the range of AP propagation speeds among the contributing motor nerve fiber APs. These parameters will now be discussed in greater detail.

Amplitude As previously discussed, compound electrical potentials have a triphasic morphology when recorded in a volume conductor – the initial positive phase

102

represents the APs propagating toward the E1 electrode (the initial positive current source), the initial negative phase represents their presence below the E1 electrode (the negative sink), and the terminal positive phase represents their propagation away from the E1 electrode (the terminal positive current source). However, because the muscle fiber APs are generated below the E1 electrode, there is no initial positive current source (the initial positive phase is absent). Thus, the waveform morphology of the motor response is biphasic. Whenever a motor response is recorded with an initial positive phase, the E1 electrode is likely not properly positioned over the innervation zone and, hence, should be repositioned and a new motor response recorded. There are a number of reasons why relocating the E1 electrode may not overcome this issue (e.g., activation of a neighboring muscle from excessive stimulation; anomalous innervation), all of which are discussed in Chapter 18. The amplitude of the motor response is measured from the baseline of the recording to the first negative peak. It is the voltage difference between these two points. Because a single nerve fiber AP generates a very large number of muscle fiber APs, the motor response is quite a bit larger than a nerve fiber AP and, hence, is measured in millivolts (mV). Thus, in this regard, the motor unit can be thought of as a motor response amplifier. The amplitude of the motor response is proportional to the number of muscle fiber APs activated, and because the innervation ratio is constant for a given muscle, the amplitude is also proportional to the number of motor nerve fibers. When the recorded amplitude value from the symptomatic side is below the normal control value of the EMG laboratory, the amplitude is labeled absolutely abnormal, whereas when the value is less than 50% of the value recorded from the contralateral (healthy) side, it is designated relatively abnormal. In general, in the setting of unilateral disease, the value recorded from the contralateral asymptomatic limb is a better indicator of normal than is the population-derived normal control value of that study. Thus, unless the response is well within normal limits, the contralateral side should be studied to look for a relative abnormality whenever the nerve under study is suspected to be abnormal, especially when the response is near the lower limit of absolute normal. For example, when a patient with ulnar distribution deficits has normal ulnar findings, the contralateral

Chapter 7: Motor Nerve Conduction Studies

ulnar nerve should also be assessed (e.g., ulnar sensory and motor NCS and needle EMG of ulnar nerve-innervated muscles). In this manner, relative abnormalities will not be missed. The amplitude is the most valuable measurement because it is a semi-quantitative indicator of the number of motor axons able to conduct APs from the stimulation site to the recording site (i.e., the functioning motor axons). This is true because the innervation ratio of a muscle is constant. Thus, the motor response is proportional to the number of functioning (i.e., innervated) muscle fibers. Because the ratio between the motor axons and the muscle fibers is constant, the motor response is also proportional to the number of functioning motor nerve fibers. As a result of this relationship, the motor response amplitude provides indirect evidence about the number of functioning motor axons and, hence, can be used as a semiquantitative indicator of lesion severity prior to reinnervation by collateral sprouting (see Chapter 9). In addition to reflecting the number of muscle fibers, the amplitude also reflects the synchrony of the contributing muscle fiber APs. The latter is related to the synchrony of conduction along the motor axons, the nerve segment distance between the stimulating electrodes and the motor endplate regions, and the synchrony of conduction along the terminal nerve branches, across the NMJs, and along the muscle fibers. As muscle fiber AP synchrony is lost, the amplitude of the response decreases. Loss of synchrony is termed temporal dispersion and can be physiologic or pathologic. Pathologic temporal dispersion reflects focal nonuniform demyelinating conduction slowing and is discussed in Chapter 9. In a volume conductor, the muscle fiber AP is proportional to the diameter of the muscle fiber generating it (Hakansson, 1956). The motor response amplitude (CMAP amplitude), in addition to muscle fiber diameter, is affected by the relationship between the surface recording electrodes and the endplate zone of the muscle (i.e., the endplates of the individual muscle fibers from which the recording is being made). When these sites are in the same vicinity, their negative sinks summate and the amplitude is maximal. For more dispersed endplates, however, the electrode records an initial positive current for some of the muscle fiber APs, which acts to negate the summated negative phases (i.e., phase cancellation among the contributing muscle fiber APs due to loss

of synchrony). Dispersion of the motor point can be secondary to skeletal changes, such as those occurring with arthritis. The amplitude is also diminished when the E2 electrode overlies muscle tissue (see Chapters 6 and 18). The amplitude is also diminished by intervening body tissue located between the muscle fibers and the recording electrodes, but this is less of a problem for motor NCS than for sensory NCS. Thus, for a given motor unit size, its amplitude decreases with distance (depth) away from the surface recording electrode. In addition, for axons of identical excitability, those nearest the cathode are preferentially excited because they are closest to the stronger stimulator currents (stimulator current is strongest closest to the cathode). Finally, distant innervation zones also negatively affect the innervation zone under study. For example, with the E1 and E2 electrodes positioned over the thenar eminence (i.e., median motor NCS recording thenar eminence), median nerve stimulation at the wrist activates not only the thenar eminence muscles but also the more distant lumbrical muscles, which influence the recorded waveform. The E1 electrode records the postsynaptic potential onsets of the motor units directly below it (e.g., the thenar eminence muscles) as a negative potential, but the motor unit onsets more distant (e.g., lumbrical motor units) as a positive potential or, if it reaches the E1 electrode recording site, as a positive-negative biphasic potential. This same phenomenon occurs with ulnar motor studies recording from the hypothenar eminence (contamination via medial lumbrical muscle stimulation), with peroneal motor studies recording from the extensor digitorum brevis muscle (contamination via dorsal interossei muscles of the foot), and with inadvertent stimulation of neighboring nerves, such as the ulnar nerve (contamination via adductor pollicis muscle stimulation). The CMAP amplitude has several important uses in EDX medicine. First, the CMAP amplitude allows for lesion severity estimation. As has been previously stated, prior to collateral sprouting, because the innervation ratio is fairly constant for a given muscle, the percent decrement of the CMAP (reflects functioning muscle fibers) is proportional to the percentage decrement of the motor axons composing the nerve under study. Thus, when the distal motor response recorded from the symptomatic side is 50% smaller than the homologous response recorded from

103

Section 2: Nerve Conduction Studies

the contralateral, asymptomatic side, approximately 50% of the motor fibers are involved. This percentage underestimates lesion severity following collateral sprouting because the innervation ratios of the reinnervating neurons are increased. Second, the CMAP amplitude can determine the underlying pathophysiology. Thus, once enough time has passed for Wallerian degeneration to occur (about 7 days for motor nerve fibers), demyelinating conduction block and axon loss can be differentiated (see Chapter 9). Third, because of the essential lack of physiologic temporal dispersion, the motor NCS can assess longer lengths of nerve and, thus, have lesion-localizing ability. Thus, once enough time has passed for Wallerian degeneration to have occurred, any significant discrepancy between the distal motor response amplitude and the proximal motor response amplitude identifies a demyelinating conduction block lesion located somewhere between the two stimulation sites. By stimulating between these two points, the actual lesion site can be better determined. For example, if during the ulnar motor NCS, recording hypothenar eminence, the distal motor response (stimulating at the wrist) is 8 mV and the proximal motor response (stimulating above the elbow) is 2 mV, then a demyelinating conduction block must exist somewhere between the above-elbow and wrist stimulation sites. Consequently, stimulation below the elbow would be performed and, if the response amplitude were similar to the value recorded with stimulation at the wrist (i.e., 8 mV), then the lesion must be located somewhere between the above-elbow and below-elbow stimulation sites (i.e., an elbow segment lesion). By adding more stimulation sites (e.g., inching), the site of the lesion could be further localized. Prior to the completion of Wallerian degeneration, early axon loss lesions can be localized in the same manner. In summary, the amplitude of the motor response reflects the number of functioning motor axons and the synchrony of motor nerve fiber conduction. Thus, they are of value in quantifying focal axon loss and focal demyelinating conduction block and identifying focal demyelinating conduction slowing, respectively. The amount of focal axon loss is estimated by comparing the amplitude of the distal motor response collected from the symptomatic side to the distal motor response from the asymptomatic side. For example, with a traumatic right median neuropathy of 4 weeks’ duration, if the distal median motor response (recording thenar eminence) is 4 mV on

104

the symptomatic side and 8 mV on the asymptomatic side, then about 50% of the motor axons innervating this muscle group are involved. The amount of focal demyelinating conduction block is estimated by comparing the motor response recorded with stimulation distal to the lesion with the motor response recorded with stimulation proximal to the lesion. For example, in an individual with right ulnar distribution numbness and weakness of 4 weeks’ duration, if the ulnar motor response (recording hypothenar eminence) is 8 mV with stimulation below the elbow and 4 mV with stimulation above the elbow, then about 50% of the motor axons innervating this muscle group are involved. This same approach is used for lesions with both axon loss and demyelinating conduction block. For example, if an individual with a right foot drop is studied 4 weeks after the onset of the foot drop and the peroneal motor response (recording tibialis anterior) is 3 mV with stimulation below the fibular head, 1.5 mV with stimulation above the fibular head, and the contralateral response is 6 mV with stimulation below the fibular head, then the lesion involves 75% of the motor fibers – 50% are axon loss (3 mV versus 6 mV) and 25% are demyelinating conduction block (3 mV versus 1.5 mV). Only 25% of the motor fibers are normal. In addition, they can be performed within the first three days of an injury to identify preexisting motor axon loss.

Negative AUC The negative AUC is the integrated area between the negative phase of the signal and the baseline. Because it represents an area, it has a two-dimensional unit of measurement, similar to carpeting (square feet) or torque (foot-pounds). Because the y-axis represents its amplitude, in millivolts (mV), and the x-axis represents its duration, in milliseconds (msec), the value of the negative AUC is expressed in mV-msec. The negative AUC is calculated by the EMG machine through the application of calculus (see Figure 7.11). Like the amplitude value, the negative AUC value reflects all of the innervated muscle fibers, not just the ones with the most common arrival time at the recording electrodes (i.e., the ones generating the peak of the negative phase). As a result, the negative area under the curve measurement is much less susceptible to pathological dispersion. For this reason, it is more accurate than amplitude for estimating the number of

Chapter 7: Motor Nerve Conduction Studies

10 mV

5 msec

Figure 7.11 The negative area under the curve is the twodimensional area between the negative phase of the motor response and the baseline.

conducting motor nerve fibers (as previously depicted in Figures 7.9 and 7.10). Loss of nerve fiber action potential synchrony results in pathological dispersion, which, in turn, results in more interaction between the negative and positive phases of the nerve fiber action potentials contributing to the CMAP. As a result, the negative AUC value is reduced, although typically to a much lesser degree than is the amplitude. Waveform dispersion is only observed with nonuniform demyelinating conduction slowing. Because this pathophysiology is the least symptomatic of the nerve fiber pathophysiologies, when it is observed in the EDX laboratory, demyelinating conduction block or axon loss are usually also noted (i.e., because patients with only demyelinating conduction slowing tend to be asymptomatic and, consequently, tend not to be referred to the EDX laboratory). Because uniform demyelinating conduction slowing (this pathophysiology is most commonly encountered among patients with mild carpal tunnel syndrome) does not change the morphology of the waveform, it does not produce pathologic dispersion and, consequently, does not affect the value of the negative AUC. Conversely, because both demyelinating conduction block and axon disruption block the propagating nerve fiber action potentials from reaching the recording electrodes, the value of the negative AUC is reduced. In summary, in the setting of pathological dispersion (i.e., nonuniform demyelinating conduction slowing), the negative AUC is a much more accurate indicator of the number of functioning muscle fibers than is the amplitude. Otherwise, they are rather similar estimators of the number of functioning muscle fibers. In our EDX laboratory we routinely record both and typically use the value of the negative AUC to calculate the percentage of involved motor axons (it is divided by the value recorded from the

10 mV 5 ms

Figure 7.12 The onset latency of the top trace is referred to as the distal latency, whereas the onset latency of the lower trace is referred to as the proximal latency. The latter is used to calculate the conduction velocity and, thus, is not typically reported.

asymptomatic contralateral side, then multiplied by 100). For example, if negative AUC value of the distal motor response of the median nerve (recording thenar eminence) equals 4 mV and the value recorded from the contralateral side is 8 mV, then approximately 50% (4/8  100) of the median motor axons innervating the thenar eminence are involved (assuming reinnervation via collateral sprouting has not yet occurred, which underestimates the severity of the lesion).

Distal Latency The onset latency is the time from the stimulus to the first deflection of the motor response (i.e., the point at which the trace first deviates [either upward or downward] from the baseline). The onset latency of the distal motor response is referred to as the distal latency, whereas the onset latency of the proximal response is called the proximal latency (see Figure 7.12). Both of these values represent the time elapsed, in milliseconds, between the moment that the stimulator fires and the onset of the CMAP. The proximal and distal onset latency values reflect the summation of the nerve activation time, nerve conduction time, terminal branch conduction time, NMJ transmission time, muscle fiber activation time, muscle fiber conduction time, and the time to reach the recording electrode (discussed in more detail later in this chapter). Although the onset latency of the proximal response is collected, it is typically not reported because its sole purpose is to calculate the nerve conduction velocity (discussed further on). Unlike the values of the amplitude and the negative AUC, which correlate with the number of functioning motor nerve fibers within the nerve, the distal latency value reflects the AP propagation speed of the fastest motor axons contributing to the CMAP.

105

Section 2: Nerve Conduction Studies

As previously discussed, motor responses are biphasic in morphology in a volume conductor, because the compound electrical potential is generated below the E1 electrode (i.e., the recorded potential is not triphasic because the first positive phase [the approaching phase] does not occur because the potentials originate below the recording electrode). For this reason, whenever the CMAP morphology is triphasic, it implies that the E1 electrode is not situated over the motor point of the muscle and, hence, should be repositioned. On occasion, this positive dip persists (discussed in detail later in Chapter 18). In this setting, the latency can be taken at the point of initial downward deflection from the baseline as long as this point is utilized as the onset latency of the motor responses collected at the other stimulation sites. Although a negative deflection may be obtained by using a needle recording electrode rather than a surface recording electrode, the amplitude value of the motor response will be of no utility because it will reflect only the adjacent muscle fibers (Horning et al., 1972). The majority of the motor axons contained within a nerve trunk are smaller in diameter (and hence more slowly conducting). Thus, the fastest fibers represent only a small minority of the population of motor fibers contributing to the CMAP. Consequently, the distal latency value provides no information about the overwhelming majority of the motor fibers contributing to the CMAP. As an example, and using mathematically friendly numbers for calculation simplification purposes, if the motor branch to a muscle contains 1,000 axons and has an innervation ratio of 1,000: 1, then of the 1,000,000 muscle fibers contributing to the recorded CMAP, the value of the distal latency reflects only 1,000 muscle fibers (0.1%). Thus, 99.9% of the CMAP is unrecognized by the value of the distal latency. Even with a muscle with a much lower innervation ratio (e.g., 100:1) and a lesser number of motor axons (e.g., 100), the distal latency would only reflect 100 of the 10,000 muscle fibers (1%), and thus, 99% of the contributing muscle fibers would be unrecognized by the recorded distal latency value. The point is that the distal latency value is extremely insensitive to focal axon loss. Since focal axon loss represents the overwhelming type of pathology associated with lesions of the PNS, this is unfortunate. On the other hand, it is quite sensitive to isolated focal demyelination. Although this type of pathology is uncommon, it is

106

7 Stimulator (Skin) 1 2

6

3 1. Nerve fiber activation time (~0.15 msec) -Tissue transit time -Depolarization threshold times 2. Nerve fiber conduction time 3. Terminal nerve fiber conduction time (much slower) 4. NMJ transmission time (~1 msec) 5. Muscle fiber activation time (threshold time) 6. Muscle fiber conduction time (3–5 m/sec) 7. Tissue transit time

4 5

Figure 7.13 The elements contributing to the CMAP onset latency time. These events are discussed in detail in the text.

the initial pathology occurring in the setting of median nerve entrapment at the wrist (i.e., carpal tunnel syndrome). Consequently, because this is by far the most common mononeuropathy encountered in the EMG laboratory, measurement of the distal latency has an important role. The Contributors to the Distal Latency Time The time elapsed, in milliseconds, between nerve stimulation and the onset of the distal motor response (i.e., the distal latency) reflects a number of events in addition to the motor nerve fiber conduction time, including: (1) nerve fiber activation time (tissue transit time and threshold time), (2) nerve conduction time, (3) terminal nerve branch conduction time, (4) neuromuscular junction transmission time, (5) muscle fiber activation time, (6) muscle fiber conduction time, and (7) tissue transit time (muscle fiber to E1) (see Figure 7.13). These individual times will now be discussed separately. The time from stimulator discharge at the skin surface to nerve activation is referred to as the nerve activation time. The nerve activation time consists of two separate components: tissue transit time and threshold time. The tissue transit time is the time it takes for the current to traverse the tissue located between the stimulator and the nerve. The threshold time is the time required for the membrane potential of the nerve to change from its resting membrane potential to its depolarization threshold potential. The nerve activation time has been estimated to be 0.15 milliseconds (Krarup et al., 1992). Once the nerve is activated, the APs generated propagate along

Chapter 7: Motor Nerve Conduction Studies

the motor axons to the arborization point where the motor axon divides into its terminal branches (the exact number depends on the innervation ratio of the muscle). This is the actual nerve conduction time. The APs then propagate along the terminal branches, which is referred to as the terminal branch conduction time. The terminal branches conduct APs much slower than do the parent motor axons. (The axons thin as they approach the muscle, so even proximal to the terminal branches there is some unwanted slowing.) Next is the NMJ transmission time, which is estimated to be about 1 millisecond. This is followed by the muscle activation time (threshold time). Unlike nerve activation time, which consists of two time components (tissue transit time and threshold time), muscle activation time only includes threshold time because there is no tissue to transit (i.e., NMJ transmission time includes the postsynaptic membrane, which is part of the muscle tissue; thus, there is no tissue transit time). Next, the muscle fiber APs generated at the postsynaptic membrane of the NMJ propagate bidirectionally along the muscle fibers. This is termed muscle fiber conduction time, which is in the 3–5 m/sec range. Finally, the time it takes for these potentials to reach the E1 electrode is termed the muscle tissue transit time. Thus, there are a number of events that are included in the onset latency time in addition to the actual motor nerve conduction time. If this time were used to calculate the motor nerve conduction velocity (NCV), the calculated value would underestimate the true value because of the transmission delays (activation and threshold times and NMJ transmission time) and the slower conduction velocities (terminal branch CV and muscle fiber CV). Thus, of these seven times, we are interested only in nerve conduction time. Because the sole purpose of the proximal latency value is to calculate the conduction velocity (discussed next), it typically is not reported.

Conduction Velocity The conduction velocity, like any velocity, is the change in distance divided by the change in time (e.g., miles per hour; meters per second). The onset latency reflects the fastest fibers to reach the recording electrodes from the stimulation site. Because the onset latency value reflects, in addition to nerve conduction time, the NMJ transmission time, the muscle fiber conduction time, and other times, all of which are

7 S1

S2

1 2

6

3 4 5

Figure 7.14 Illustration of a single motor axon depicting the effect of calculating the motor nerve conduction velocity along the nerve between the two stimulation points. By subtracting the onset latency of the distal response from that of the proximal response, elements 1 and 3 through 7 are eliminated, as well as the overlapping portion of element 2. The time difference is illustrated by the thick line between the two stimulation sites (S1 and S2). This time difference represents the time required for the action potential to propagate from the proximal stimulation site to the distal stimulation site. Once the distance between the two sites is measured, a conduction velocity can be calculated.

much slower than the nerve conduction time, it is not the true nerve conduction time and, hence, cannot be utilized to accurately estimate the nerve conduction time. Thus, in order to avoid underestimating the motor NCV value, the time intervals other than actual nerve conduction time need to be eliminated (i.e., 1 and 3–7). This is accomplished by stimulating the nerve at two sites (the distal and the proximal motor responses) and then subtracting the value of the distal onset latency from that of the proximal latency. This eliminates time intervals 1 and 3–7. It also eliminates the nerve conduction time over the distal aspect of the nerve distal to the distal stimulation site (i.e., the overlapping segment of time interval 2). Thus, the difference represents the conduction time between the proximal and distal stimulation sites (see Figure 7.14). This value and the surface distance between the two stimulation sites are used to calculate the motor nerve conduction velocity between the two stimulation sites. Although this calculation is true for a single motor axon, it is not always true when a population of motor axons is being stimulated, because the calculated conduction velocity represents the first fiber to reach the E1 electrode, not the first fiber to reach the distal stimulation site. For example, imagine a nerve composed of fibers with three different conduction velocity ranges: fast, intermediate, and slow. If the fast

107

Section 2: Nerve Conduction Studies

fibers are delayed at some site distal to the distal stimulation site so that fibers conducting at intermediate velocities reach the motor point first, then the calculated NCV of the forearm segment actually represents the intermediate fibers rather than the fastest fibers. For example, in this scenario, the calculated CV of the forearm segment following elbow and wrist stimulation will be spuriously low whenever the fastest fibers are selectively delayed distal to the wrist. In other words, when multiple motor axons are stimulated, although we stimulate at two sites and calculate the CV for the nerve segment between those two sites, the calculation actually reflects the first fibers to reach the motor point. Thus, although the fastest fibers may reach the wrist first (following elbow stimulation), they do not dictate the conduction velocity if they are selectively held up distal to the wrist. Thus, in this setting, the calculated conduction velocity value is an underestimation. It is important to be familiar with this concept, because spurious slowing of the forearm segment is frequently observed in the setting of carpal tunnel syndrome when the focal demyelinating conduction slowing is limited to the fastest conducting fibers, as often occurs early in the disease process (see Figure 7.15). Like onset latency, the conduction velocity only reflects the action potential propagation speed of the fastest conducting motor axons. This measurement therefore does not provide any information about the other motor axons, which represent the overwhelming majority of the motor axons composing the nerve under study. For this reason, the latency and conduction velocity values are insensitive to axon loss. It is important to understand the following statement: Although we stimulate at two sites along the nerve (e.g., the wrist and the elbow for the median nerve) and calculate the nerve conduction velocity using the onset latencies of the two responses, we are actually determining the conduction velocity of the fastest motor axons to travel from the proximal stimulation site to the recording electrodes, not from the proximal stimulation site to the distal stimulation site. This would only be the case when the fastest conducting fibers not only reach the distal stimulation site first but also reach the recording electrodes first, or if the nerve only contained a single motor axon. For example, it is not uncommon to encounter patients with carpal tunnel syndrome who exhibit a

108

Wrist

Elbow IFS

IFS 4.2 Wrist stimulation

9.6

Dd = 250 mm Dt = 5.4 msec

Elbow stimulation F I S

CV = Dd/Dt = 46.3 m/sec Figure 7.15 Spuriously slow forearm conduction velocity due to the selective slowing of the fastest fibers, distally. The onset latency values of the distal and proximal motor responses are 4.2 msec and 9.6 msec, respectively. The conduction velocities of the contributing motor axons are grouped into fast (F), intermediate (I), and slow (S). The distal and proximal motor responses indicate that the intermediate fibers arrived prior to the fast fibers due to selective focal demyelinating conduction slowing distal to the wrist. Thus, although the fastest fibers to reach the wrist stimulation site from the elbow stimulation site were the fastest fibers, the calculation actually reflects the intermediate fibers because they arrived at the recording electrodes before the fastest fibers. This phenomenon is observed in patients with carpal tunnel syndrome when the more heavily demyelinated fibers are affected out of proportion to the other fibers, as may occur in early carpal tunnel syndrome.

slowed motor conduction velocity as calculated from the onset latencies with elbow and wrist stimulation, both of which are proximal to the carpal tunnel. It is not that median nerve damage in the carpal tunnel causes conduction velocity slowing more proximally, but rather that the more heavily myelinated axons (i.e., the fastest conducting motor axons) are being selectively delayed in the carpal tunnel and the slightly slower ones (i.e., those that are slightly less myelinated) are passing through the carpal tunnel unimpeded. Consequently, in this setting – where the fastest motor axons are selectively delayed in the carpal tunnel – the fastest conducting axons do not reach the muscle belly first (and, hence, do not reach the E1 electrode first). Instead, motor axons with somewhat lower conduction velocities reach the E1 electrode first. As a result, the onset latency of the proximal motor response reflects the muscle fiber action potentials of somewhat slower motor axons. Thus, although the more heavily myelinated axons arrive at the wrist first, because they are subsequently delayed at the carpal tunnel, they do not reach the E1 electrode first. Thus, the fastest fibers are not reflected in the calculated motor nerve conduction velocity (see Figure 7.15).

Chapter 7: Motor Nerve Conduction Studies

Negative Phase Duration Although it is easy to identify the onset point of the motor response, it can be quite difficult to define its termination point. Thus, the total duration of the motor response is typically not measured. Instead, the duration of the negative phase is measured because its onset and termination points are easily identified. The duration of the negative phase is defined as the time interval, in milliseconds, between the onset and termination points of the negative phase of the CMAP. It reflects the summation of the individual motor unit APs (MUAPs) and, hence, the conduction velocity differences between the fastest and slowest conducting motor axons contributing to the motor response. It is increased (termed response dispersion) when there is nonuniform slowing (such as what occurs with nonuniform demyelinating conduction slowing; see Chapter 9) among the fastest and slowest conducting fibers. Response dispersion also occurs physiologically as the distance between the stimulating and recording electrodes is increased. This is much like what occurs between two vehicles traveling along the interstate at different speeds (i.e., one slower vehicle and one faster vehicle). As time passes, the distance between the two vehicles continuously increases. Over greater and greater distances, their difference in velocity becomes more and more apparent. Another analogy is among two runners

References Bonner FJ, DevlescHoward AB. AAEM minimonograph #45: the early development of electromyography. Muscle Nerve 1995;18:825–853. Dawson GD. The relative excitability and conduction velocity of sensory and motor nerve fibers in man. J Physiol (Lond) 1956;131:436–451. Dumitru D. Electrodiagnostic medicine. Philadelphia, Hanley and Belphus, 1994. Falck B, Stalberg E. Motor nerve conduction studies: measurement principles and interpretation of findings. J Clin Neurophysiol 1995;12:254–279. Hakansson CH. Conduction velocities and amplitude of the action

racing. In a short race (e.g., down the driveway), the faster runner wins by a shorter amount of time than in a longer race (e.g., around the block). The increasing dispersion among the individual muscle fiber APs causes increasing overlap of the positive and negative phases, thereby resulting in negative phase cancellation. For a number of reasons, motor responses are much more resistant to amplitude decrement by physiologic dispersion than are sensory responses. This primarily reflects the fact that the duration of the negative phase of the motor response is so much greater than that of the sensory response. For example, the duration of the negative phase of the median motor response is 5–6 msec (Lee et al., 1975). In addition, motor NCS (1) are biphasic (i.e., less positive phase-negative phase interactions), (2) are larger in size (more negative area under the curve to begin with), and (3) have a narrower range of conduction velocities (less dispersion over distance) (Kimura et al., 1986). These differences are discussed in more detail in the chapter on sensory NCS (see Chapter 8). This resistance to physiologic dispersioninduced amplitude decrement allows much longer nerve fiber segments to be studied with motor NCS. Nonetheless, it is seldom necessary to stimulate proximal to the elbow level during upper extremity assessments or proximal to the popliteal fossa level during lower extremity assessments.

potential as related to circumference in the isolated frog muscle fiber. Acta Physiol Scand 1956;37:14–34. Hodes R, Larrabee MG, German W. The human electromyogram in response to nerve stimulation and the conduction velocity of motor axons. Arch Neurol Psychiatry 1948; 60:340–365. Horning MR, Kraft GH, Guy A. Latencies recorded by intramuscular needle electrodes in different portions of a muscle: variation and comparison with surface electrodes. Arch Phys Med Rehabil 1972;53:206–210. Isley MR, Krauss GL, Levin KH, Litt B, Shields RW Jr, Wilbourn AJ. Electromyography and electroencephalography. Biophysical

Measurement Series. Redmond, SpaceLabs Medical, Inc, 1993. Kimura J, Machida M, Ishida T, Yamada T, Rodnitzky RL, Kudo Y, Suzuki S. Relationship between size of compound sensory or muscle action potentials and length of nerve segment. Neurology 1986;36:647–652. Kincaid JC, Brasher A, Markand ON. The influence of the reference electrode on CMAP configuration. Muscle Nerve 1993;16:392–396. Krarup C, Horowitz SH, Dahl K. The influence of the stimulus on normal sural nerve conduction velocity: a study of the latency of activation. Muscle Nerve 1992;15:813–821. Lambert EH. Electromyography and electrical stimulation of peripheral nerves and muscles. In Mayo Clinic

109

Section 2: Nerve Conduction Studies

staff, editors, Clinical examinations in neurology. Philadelphia, WB Saunders, 1956:287–317. Lee RG, Ashby P, White DG, Aguayo AJ. Analysis of motor conduction velocity in human median nerve by computer simulation of compound muscle action potentials. Electroencephalogr Clin Neurophys 1975;39:225–237. Rhee EK, England JD, Sumner AJ. A computer simulation of

110

conduction block: effects produced by actual block versus interphase cancellation. Ann Neurol 1990;28:146–156.

block base on computer simulation studies of nerve conduction with human data obtained in the forearm segment of the median nerve. Brain 2006;129:2447–2460.

Simpson JA. Electrical signs in the diagnosis of carpal tunnel syndrome and related syndromes. J Neurol Neurosurg Psychiatry 1956;19:275–280.

Wilbourn AJ. How can electromyography help you? Postgrad Med 1983;73:187–195.

Van Asseldonk JTH, Van den Berg LH, Wieneke GH, Wokke JHJ, Franssen H. Criteria for conduction

Wilbourn AJ. Sensory nerve conduction studies. J Clin Neurophysiol 1994;11:584–601.

Chapter

8

Sensory Nerve Conduction Studies

Introduction Historically, because the magnitude of the sensory responses was lower than the magnitude of the noise generated by the amplifier, the technology utilized to collect motor responses was unable to collect sensory responses. Once technological advancements were introduced, such as digital delays (permitted the stimulus and sweep to be triggered independently and, thus, superimposed upon one another) and averaging (to increase the signal to noise ratio), sensory responses could be more easily collected (Pitman, 1958; Buller and Styles, 1959; Gilliatt et al., 1965). The sensory NCS are an indispensable part of the EDX assessment for a number of reasons. First, they are the only portion of the EDX examination that assesses the sensory fibers of the PNS. Consequently, sensory neuronopathies and sensory neuropathies are only identifiable by this portion of the study. Second, they have localizing value. Because the sensory neurons are located outside of the intraspinal canal, they permit ganglionic and postganglionic disorders to be differentiated from preganglionic lesions (e.g., radiculopathies). Third, in the setting of a mixed lesion (i.e., one that involves sensory and motor axons), the sensory responses are more sensitive to axon loss than are the motor NCS. Fourth, following a mild lesion in which the motor axons have reinnervated the denervated muscles fibers, thereby normalizing the motor NCS, the sensory responses may be the only NCS abnormality. Thus, without the inclusion of the sensory NCS, the EDX examination is incomplete. Physiologically, sensory nerve fiber APs are generated by sensory receptor stimulation, peripherally. Once generated, the APs propagate centrally to their cell bodies of origin in the DRG. Thus, they normally conduct centripetally. As with motor NCS, however, the stimulator-induced APs propagate bidirectionally from the stimulation site because there is no nerve

segment in its absolute refractory period to impede Na+ current advancement. Also like the motor response, the sensory response is a compound electrical potential. It is composed of the APs of the sensory nerve fibers contained within the nerve under study and, thus, is referred to as a compound sensory nerve action potential (SNAP). When sensory NCS are performed with the stimulating electrodes located distal to the recording electrodes, they are termed orthodromic (the APs composing the SNAP propagate in the physiological direction), whereas those performed with the surface recording electrodes oriented in the opposite direction are termed antidromic (the APs generating the SNAP propagate in the opposite direction). Unlike motor NCS, which are performed in just one way (i.e., the orthodromic belly-tendon method), sensory NCS can be performed by stimulating and recording a pure sensory nerve (using an orthodromic electrode arrangement or an antidromic one), by stimulating a mixed nerve while recording from one of its cutaneous branches more distally (an antidromic technique), or by stimulating a cutaneous nerve while recording more proximally from a segment containing both sensory and motor axons (an orthodromic technique). The advantages and disadvantages of the orthodromic and antidromic techniques are discussed below.

Technique Like the motor NCS, with the sensory NCS, the nerve is supramaximally stimulated to generate a maximal sensory response and the response amplitude and the response latency are measured. However, other than that, they are quite different. With sensory NCS, the recording electrodes overlie the sensory nerve fibers composing the nerve under study. Thus, there is no amplification effect, and consequently, sensory responses are much smaller in size (measured in microvolts) than are motor responses (measured in

111

Section 2: Nerve Conduction Studies

20 mV

20 mV

2 ms

Lat ms

Amp mV

4.8

31.0

Lat ms

Amp mV

10.3

14.0

1 ms

Figure 8.2 The effect of E1 electrode misplacement during the sensory NCS. The upper trace demonstrates a superficial radial sensory response recording with the E1 electrode overlying the superficial radial nerve as it crosses the extensor pollicis longus tendon. The lower trace shows the effect of misplacing the E1 electrode 1 cm laterally.

Figure 8.1 The median sensory response, recording second digit, following wrist stimulation (top trace) and following elbow stimulation (bottom trace). As depicted in the illustration, there is considerable amplitude loss with proximal stimulation. Thus, sensory responses are less useful for assessing large lengths of sensory axons.

millivolts). As a result of their small size, sensory responses are significantly diminished by volume conduction losses as they traverse the tissue to reach the recording electrodes (i.e., physiologic temporal dispersion). Consequently, proximal stimulation sites have a much greater effect on the sensory response amplitude than they do on the motor response amplitude (see Figure 8.1). For this reason, proximal sensory responses are less frequently recorded. When they are performed, they should be compared to the contralateral asymptomatic side. Because of their smaller size, a much greater degree of expertise is required to elicit sensory responses. Thus, unlike the motor NCS, which, in the setting of technique errors, are relatively forgiving, the sensory NCS are quite unforgiving. For example, when the E1 electrode is misplaced slightly away from the nerve rather than directly over it, the morphology of the wave may change (the initial positivity may be lost) and the amplitude may be significantly reduced (up to 70%) (Raynor et al., 1997) (see Figure 8.2). In addition to technical errors, physiologic factors, trivial trauma, and other minor pathologic issues can seriously hinder the collection of sensory NCS. In our EMG laboratories, we typically begin the EDX study with the sensory NCS because they are more sensitive to focal axon loss lesions than are the motor NCS.

Measurements Because the recorded sensory response is composed of sensory nerve fiber APs, sensory NCS are usually performed with stimulation at just a single site.

112

Although the conduction time of sensory responses is not delayed by conduction slowing related to slowly conducting terminal branches, NMJ transmission, or slowly conducting muscle fibers, it is delayed by tissue transition time and nerve fiber activation time (discussed further on). With a single stimulation site, there is only a single response. Thus, only a single waveform requires measurement. The important sensory response measurements are: the amplitude, the latency, and the calculated conduction velocity, as well as the waveform morphology. The negative AUC value, which is measured in microvolt-msec, is typically not reported. The amplitude of the response is measured based on the morphology of the recorded waveform. When the response is biphasic, the amplitude is measured from the baseline to the negative peak, whereas when it is triphasic, it is measured from the first positive peak to the first negative peak. Other sensory response measurements vary, depending on the preferences of the EMG laboratory. The latency can be taken at the onset of the response (termed the onset latency) or from the peak of the first negative phase (termed the peak latency). The conduction velocity, when calculated, can be based on 1-point or 2-point stimulation (discussed later). Similar to motor responses, the onset latency and conduction velocity values reflect the AP propagation speeds of the fastest conducting sensory axons. Hence, these values represent only a small minority of the total number of sensory axons in the nerve. The meaning of these measurements and the pros and cons of the measurement options are discussed in detail below. Also, because the individual sensory nerve fiber APs composing the SNAP are less than 2 msec in duration, physiological temporal dispersion has a much

Chapter 8: Sensory Nerve Conduction Studies

greater effect on the amplitude of the sensory response than it does on the amplitude of the motor response, thereby limiting the length of nerve segment that can be studied by the sensory NCS. For this reason, proximal stimulation (e.g., elbow stimulation) sites are typically not utilized when performing sensory NCS. When they are performed, they should always be compared to the response taken from the contralateral, asymptomatic side. In our EMG laboratories, we consider the following sensory NCS to be standard: the median (recording second digit), the ulnar (recording fifth digit), the superficial radial (recording dorsum of the hand), the sural (recording adjacent to the lateral malleolus), and the superficial peroneal (recording dorsum of ankle). We also consider the following sensory NCS to be reliable and perform them frequently: the median (recording first, third, or fourth digit), the ulnar (recording digit 4), the dorsal ulnar cutaneous (recording dorsum of hand), the lateral antebrachial cutaneous (recording volar aspect of forearm), and the medial antebrachial cutaneous (recording volar aspect of forearm). The techniques for performing these NCS and their age-adjusted normal values are provided in Section 6 (see Appendix 5). Although we occasionally perform the saphenous NCS, we consider it an unreliable study because they can be difficult to elicit, even from the asymptomatic limbs of young individuals. For similar reasons, we do not perform the posterior femoral cutaneous or the lateral femoral cutaneous NCS.

Amplitude Like the motor response amplitude, the amplitude of the sensory response reflects the number of functioning sensory axons and their synchrony, as well as technical factors such as the filter settings and the distance between the nerve fibers and the recording electrodes. Unlike the motor NCS, in which the muscle fiber APs are generated directly below the E1 electrode, with the sensory NCS, the sensory nerve fiber APs are generated below the cathode of the stimulator. Thus, the generated nerve fiber APs travel toward, below, and away from the E1 electrode. As a result, in the presence of volume conduction, they typically have a triphasic appearance. Unlike the motor responses, which use the belly-tendon method to locate the E1 electrode in the most active area and the E2 electrode in a relatively inactive area, with the

sensory responses, the E1 and E2 electrodes are both situated over the nerve being studied. Consequently, both record the sensory response (at slightly different times) and, because this is a bipolar recording montage, both contribute to the recorded sensory response (i.e., the monitor displays the summation of the two responses). For this reason, the distance between the E1 and E2 electrodes has a major effect on the response morphology, especially its amplitude (discussed further on). When the sensory response has a biphasic morphology, the amplitude is measured from the baseline to the negative peak (termed the baseline-topeak amplitude). When the sensory response has a triphasic morphology, the amplitude is measured from the peak of the first positive phase (first peak; first positive peak) to the peak of the first negative phase (second peak; first negative peak). This is termed the peak-to-peak amplitude or the P1:P2 amplitude. The baseline-to-peak amplitude and the peakto-peak amplitude described above both reflect the depolarization side of the recorded potential. The peak of the first positive wave is where the trace goes from moving downward (related to the approaching positivity of the sodium current) to moving upward (the onset of negative sink related to sodium ion entry into the nerve) and, consequently, represents the onset of the sensory response. Thus, the onset of the sensory response is the peak of the first positive phase, not where the response curve crosses the baseline (i.e., not the onset of the negative phase). Some EMG laboratories measure the peak-to-peak amplitude from the peak of the first negative phase to the peak of the second positive phase (i.e., the P2:P3 amplitude). However, this side of the response represents the repolarization phase of the potential and, moreover, is the side where the E2 electrode has its greatest contribution. Consequently, the P2:P3 amplitude measurement can be misleading and is best avoided (Dumitru. Electrodiagnostic Medicine, 1995; p 132). Unlike motor response amplitudes, which are measured in millivolts, because of their small size, sensory response amplitudes are measured in microvolts. The sensory responses are much more susceptible to physiologic temporal dispersion than are the motor responses. There are four reasons for this: their smaller response size, their shorter negative phase duration, their wider range of conduction velocities, and, regarding triphasic sensory responses, their greater number of phases. First, because sensory

113

Section 2: Nerve Conduction Studies

responses (measured in microvolts) are so much smaller than motor responses (measured in mV), their negative AUC values are much lower. For this reason, even small amounts of positive and negative phase overlap among the contributing sensory nerve fiber APs produce significant decrements in amplitude and negative AUC. Second, because the negative phase duration of sensory responses is much shorter than that of the motor responses, phase overlap is greater, resulting in amplitude and negative AUC value decrements (Wilbourn and Ferrante, 1997). For this reason, even small degrees of physiologic temporal dispersion result in significant response reduction. For this reason, sensory NCS are performed over limited nerve segment lengths. Third, the range of conduction velocities (i.e., the difference in conduction velocity between the fastest and the slowest conducting axons) is wider than with motor responses, a reflection of the wider range of nerve fiber diameters among sensory nerve fibers. The conduction velocity range with sensory responses is about 25 m/sec, which is about twice the value of the conduction velocity range of the motor fibers composing motor responses (12 m/sec) (Dorfman et al., 1982; Dorfman, 1984; Kimura, 1986). Finally, the motor responses are biphasic, as they are generated below the recording electrode, whereas most sensory responses are triphasic (positive-negative-positive). This increases the degree of overlap between the negative and positive phases and, hence, the amount of negative phase cancellation. As a result, the study of longer segments through the application of more proximal stimulation sites results in significant amplitude loss. Thus, unlike the motor NCS, the sensory NCS cannot be used to screen long segments of nerve for focal demyelinating conduction block (see Chapter 9). In addition to the number of functioning axons, the amplitude of the sensory response is affected by the body tissue situated between the sensory nerve under study and the recording electrodes. This is true because body tissue has a high frequency filtering effect on the recorded signal. Because sensory responses contain more high-frequency components than do motor responses, they are even more susceptible to this tissue-filtering phenomenon. As the higher frequencies are lost, the rising phase of the electrical potential is delayed. Thus, the loss of the faster frequencies not only diminishes the amplitude of the response, it also delays the formation of its peak. This is most evident among the antidromic digital sensory responses, in

114

which digit girth is inversely related to amplitude size (Bolton and Carter, 1980). This explains why the digital sensory response amplitude values recorded from individuals with thinner digits are so much higher than those recorded from thicker digits and why the digital sensory response amplitude values tend to be larger among females than among males (females tend to have thinner digits than males). Indeed, not infrequently, asymptomatic men with thick digits have digital sensory response amplitude values slightly below the lower limit of normal. When the surface recording electrodes are not positioned directly over the nerve, this loss of amplitude related to tissue filtering effect is even more pronounced. When the recording electrodes are not positioned directly over the nerve, the distance between the nerve under study and the recording electrodes is greater, thereby increasing the amount of signal loss related to the highfrequency filtering effect of body tissue. In addition to the filtering effect of normal body tissue, abnormal body tissue, such as edema, also causes amplitude loss when it is located between the electrical source and the recording electrodes.

Orthodromic versus Antidromic Techniques Unlike the motor NCS, which can only be performed in an orthodromic manner (because the muscle always lies distal to the nerve), the sensory nerve fiber APs can be collected using an orthodromic technique or an antidromic technique. The term orthodromic is used to characterize a sensory NCS technique in which the collected action potentials propagate along the nerve under study in the same direction as they do physiologically (centripetally), whereas the term antidromic is used to characterize a sensory NCS technique in which the elicited action potentials propagate in the opposite direction to their physiologic direction (centrifugally). Thus, for example, the median sensory response can be collected with wrist stimulation while recording from the index finger (antidromic technique) or with index finger stimulation while recording from the wrist (orthodromic technique). In the majority of EMG laboratories, except for the digital sensory NCS, all of the sensory NCS techniques utilize an antidromic approach. Regarding the median and ulnar digital sensory NCS, however, some laboratories prefer the antidromic technique for digital sensory NCS, whereas other laboratories prefer the orthodromic technique.

Chapter 8: Sensory Nerve Conduction Studies

Motor branch to lumbrical muscles Recurrent thenar motor branch

Median nerve Digital sensory branches

20 mV

1 ms

Figure 8.3 Distal advancement of the E1 and E2 ring recording electrodes eliminates the lumbrical motor response artifact when it interferes with the collection of digital sensory responses. The upper panel shows distal advancement of the ring recording electrodes along the second digit when the lateral lumbrical muscles are activated by median nerve stimulation. The lower panel shows separation of the ulnar sensory response, recording fifth digit, after the ring recording electrodes were advanced 0.5 cm along the little finger. At this point, the ulnar sensory response can be clearly differentiated from the much larger motor response artifact.

Historically, despite an identical distance between the stimulating and recording electrodes, it was thought that the peak latency value of the digital response recorded using the antidromic technique was greater than the value obtained with an orthodromic technique (Murai and Sanderson, 1975; Chodoroff, 1985). However, this phenomenon was later shown to be related to differences in the distance between the E1 and E2 electrodes (Cohn et al., 1990). Thus, when the interelectrode distance between E1 and E2 was identical, there was no peak latency

difference between the techniques. There are advantages and disadvantages to both techniques, and it is important for the EDX medicine provider to be aware of these. The major advantage of the antidromic approach reflects the recording site (the recording electrodes are on the digit). Because the tissue between the nerve fibers and the recording electrodes is minimal at this site, the amplitude of the response is much greater. Because the value of the response amplitude is the most important response parameter collected, this is the major advantage of the antidromic technique. Another reported advantage of the antidromic technique is that it is less painful (Wilbourn and Ferrante, 1997), presumably because the density of pain receptors is lower at the wrist than at the digit. The primary disadvantage of the antidromic technique reflects the site of stimulation. Because the nerve is stimulated at the wrist, both sensory and motor nerve fibers are stimulated. As a result, a volume-conducted motor response related to lumbrical muscle activation is generated that may interfere with collection of the median sensory response (motor artifact). Typically, when this occurs, it can readily be overcome by shifting the digital recording electrodes 0.5 centimeters more distally. This will eliminate the motor lumbrical response or eliminate its overlap with the sensory response so that it can be properly measured (see Figure 8.3). Because the motor artifact can easily be eliminated, this drawback is really just a minor inconvenience. On the rare occasion when it cannot be overcome, an orthodromic study can be utilized, although we have never had to resort to this tactic in our EMG laboratories. The major advantage of the orthodromic approach is that the nerve is stimulated at the digit. For this reason, only sensory fibers are activated, and hence there is no motor response artifact. The major disadvantage of the orthodromic approach is the greater quantity of subcutaneous tissue between the nerve fibers and the recording electrodes, which exponentially diminishes the amplitude of the response, sometimes making it challenging to record. In addition, the additional tissue through which the APs are recorded increases the degree of volume conduction and may cause the waveform morphology to be triphasic rather than biphasic. That this is indeed the explanation for the triphasic waveform morphology can be demonstrated by applying pressure to the bar electrode, thereby bringing the recording electrodes

115

Section 2: Nerve Conduction Studies

closer to the nerve fibers, which lessens or eliminates the initial positive phase of the triphasic waveform morphology. In our EMG laboratory, because of the importance of the amplitude value, we utilize an antidromic approach and reposition the recording electrodes more distally whenever an unwanted motor artifact interferes with the collection of the sensory response. Because shifting the recording electrodes distally increases the distance between the stimulator and the recording electrodes, the latency value will also be increased. To determine whether the recorded latency value is normal, we either perform the study on the contralateral, asymptomatic side and use the contralateral latency value as our normal value, or we increase the upper limit of normal of our control value using a conversion factor of 0.2 milliseconds per centimeter of distance added. This correction factor (0.2 msec/cm) is calculated using the lower limit of normal for upper extremity conduction velocity, which is 50 m/sec, as follows. velocity ¼ distance=time time ¼ distance=velocity ðrearrangingÞ ¼ 1 cm=50 meters per second ¼ 10 mm divided by 50 mm=msec ðconverting to mm and msecÞ ¼ 0:2 msec For example, when the distance is increased by 1.0 cm, the upper limit of normal is increased by 0.2 msec, and when the distance is increased by 0.5 cm, the upper limit of normal is increased by 0.1 msec.

Latency Of all of the measured sensory response values, the latency value has the greatest variation among EMG laboratories. This variation reflects a number of important technique issues, including: (1) whether the distance between the cathode and E1 electrode is fixed (i.e., the same for everyone) or landmark-based (i.e., not the same for everyone), (2) whether peak latencies or onset latencies are recorded, and (3) whether the latency value is reported or a 1-point conduction velocity value is calculated from the latency value and reported. It is important to realize that the latency values and the conduction velocity values are simply different measurements of the same thing – the speed of AP propagation of the fastest conducting fibers. This is similar to the common descriptors of speed – when

116

the distance is set, we describe speed as the time elapsed to traverse the distance, whereas when the distance varies, we report speed as a ratio of distance traveled to time elapsed. For example, in track-andfield competitions, when the distance is set, we report only the elapsed time. For example, the winner of the Boston marathon is reported with the winning time, not a calculated average speed. As another example, regarding the 1-mile run, if the winning time was 4 minutes, we simply report that time (the runner ran a 4-minute mile). We do not convert it into an average speed value (the runner averaged 15 miles per hour). Conversely, when the distance is not known, we report the speed: such as, the top speed along the straightaway portion of a racetrack among several cars was 200 miles per hour. We do not report a time quantity to traverse the straightaway because different tracks have different straightaway lengths, and hence the value would not convey the speed. In a similar manner, some EMG laboratories report the latency value, whereas other EMG laboratories convert the latency value into a conduction velocity value and report that value instead. However, for a number of reasons, these two values are not identical. Prior to discussing this concept, it is important to understand the time periods composing the onset latency of the sensory response.

The Time Periods Composing the SNAP Onset Latency The onset latency value of the SNAP reflects the time elapsed from the discharge of the cathode at the skin surface to the arrival of the initial sensory nerve fiber APs at the E1 electrode. Like the onset latency of the CMAP, the onset latency of the SNAP reflects more than just nerve conduction time. The onset latency value of the SNAP equals the sum of the nerve activation time (tissue transit time + depolarization threshold time), the nerve conduction time, and the tissue transit time from the nerve fiber APs to the E1 electrode. Thus, like the motor NCS, the onset latency time reflects more than just nerve conduction time. Consequently, when the onset latency value is used to calculate the nerve conduction velocity, it will be an underestimate.

Fixed Landmarks versus Fixed Distances With fixed landmarks, the distance between the stimulating and recording electrodes is different for each individual. For example, when performing the median motor response recording from the thenar

Chapter 8: Sensory Nerve Conduction Studies

eminence, the recording electrodes are positioned over the thenar eminence using the belly-tendon technique. When the distal wrist crease landmark is used to dictate placement of the cathode, the distance between the cathode and the E1 electrode will vary among individuals. As a result, normal control values must be acquired for every possible distance. To circumvent this problem, some EMG laboratories convert the latency value into a conduction velocity value by measuring the distance in millimeters from the distal wrist crease to the E1 electrode and dividing this value by the recorded latency value (i.e., distance/ time equals velocity). This approach only requires a single control value. However, because the calculated CV is derived from a single stimulation point, it underestimates the true conduction value because it does not account for nerve activation time, tissue transit time, or physiologic slowing (discussed later here). In addition, when the distance varies, the amplitude is affected due to physiologic dispersion. Because the amplitude is the most important measurement made, anything that affects it is best avoided. With a fixed distance, however, a single control value is determined for a predetermined distance, which includes the nerve conduction time plus the other times listed above. With this approach, the AP propagation speed of the fastest fibers is reported as a time value (in msec) rather than as a velocity value (meters/second).

Peak Latencies versus Onset Latencies Because of the technical limitations of historical EMG machines and the inability to accurately identify the onset latency of the sensory response (due to baseline noise), peak latencies were initially utilized. With respect to onset latencies, peak latencies are more easily identified, and there is less variation in cursor placement among different practitioners. It is extremely important to realize, however, that peak latency values do not represent the fastest conducting fibers. Instead, they more closely reflect the average nerve conduction velocity among the activated nerve fibers. For this reason, when peak latencies are converted into conduction velocities, they more accurately represent the average conduction velocity of the activated nerve fibers rather than the fastest fibers. It is important to understand this concept because most EMG machines provide both a latency value and a conduction velocity value, where the conduction velocity value is calculated from the peak

latency value. Conduction velocities calculated in this manner always underestimate the conduction velocity of the fastest fibers. As EMG machine technology advanced and the onset latency became more apparent, some laboratories began to record the onset latency rather than the peak latency, because the onset latency more accurately reflects the fastest conducting nerve fibers, which are the ones that most accurately identify demyelinating conduction slowing. However, even with modern EMG machines, for a number of reasons, the true onset may be challenging or impossible to identify, including baseline instability, shock artifact, and background noise. Even among the most expensive EMG machines, baseline fluctuations often obscure identification of the onset site, resulting in greater interobserver variability in cursor placement. For these reasons, the variability of cursor placement among practitioners is not negligible. Some practitioners determine the onset latency by drawing a horizontal line of best fit from the baseline on the repolarization side of the response. This approach may overestimate or underestimate the fastest fibers. Other practitioners simply using the peak of the first positive phase, which is much more readily identified. However, this is not the onset of the first positive phase. Additionally, some EMG machine manufacturers set the filters to overfilter the lower frequencies so that the onset latency is more apparent. However, this technique affects the morphology of the waveform by subtracting out the lower frequencies of the desired signal. Finally, even when the onset of the positive phase is clear, its value includes the nerve activation time (about 0.15 msec) and the distal tissue transit time to reach the recording electrode. Thus, the calculated conduction velocity with single-site stimulation underestimates the true conduction velocity calculated with two-point stimulation. This difference is much less pronounced with sensory NCS than with motor NCS due to the delay related to neuromuscular junction transmission (about 1 msec) (see Chapter 7). Another problem occurs with antidromically collected biphasic digital sensory responses (i.e., negative phase–positive phase). This occurs because of the increase in current density as Na+ current advances from the hand compartment (a larger tissue volume) into the digital compartment (a smaller tissue volume). Because the hand compartment volume is much larger than the digital compartment volume,

117

Section 2: Nerve Conduction Studies

the density of the current is greater in the smaller, digital compartment. In other words, the water is saltier when 2 teaspoons of salt are added to a glass of water than when they are added to a bucket of water. The increase in current density as the Na+ current moves from the hand compartment to the digital compartment triggers a far-field potential (i.e., one that moves at the speed of light). For this reason, the E1 and E2 electrodes simultaneously “see” the first positive phase of the sensory response. Thus, the “like signal” (the initial positive phase) is subtracted out through differential amplification (see Chapter 2) (Dumitru, 1995). The result is that the measured onset latency value of the biphasic response is greater than that of the triphasic response. Although it has been argued that the absence of the initial positive phase is related to the limited amount of tissue between the nerve fibers and the recording electrodes, the presence of a terminal positive phase argues against this explanation, because less tissue volume should make the triphasic response appear monophasic, not biphasic. In conclusion, for the reasons discussed above and because no study has ever shown onset latencies (which reflect the fastest conducting fibers) to be superior to peak latencies (Wilbourn, 1994), in our EMG laboratories, we continue to use peak latencies. Moreover, in theory, peak latency values, which represent the summation of all of the conducting sensory nerve fibers, may actually be more sensitive than onset latency values (Nandedkar, 2010).

Converting Latency Values into Conduction Velocity Values As stated earlier, when SNAP onset latency values are converted into CV values, they underestimate the true CV because they contaminate the true nerve conduction time with tissue transit time and nerve activation times. Moreover, when the value of the peak latency is used in the equation for the conduction velocity (rather than the onset latency), the calculated nerve conduction velocity can be significantly underestimated. Unlike the onset latency, which better reflects the CV of the fastest conducting fibers, the peak latency more closely approximates the average CV of the conducting fibers. The range of CVs among the sensory fibers contributing to the sensory response is about 25 m/sec (Kimura et al., 1986). Thus, the difference between the fastest conducting

118

fibers and the average conducting fibers is approximately 12.5 m/sec. In addition, there is the error introduced by the proximal tissue transit time (cathode to nerve), the nerve activation time, and the distal tissue transit time (nerve to E1 electrode). Consequently, when peak latency values are converted into CV values, the calculated nerve conduction velocity value may suggest demyelination and lead to erroneous management (e.g., a misdiagnosis of carpal tunnel syndrome and an unnecessary release procedure). For example, the author has observed EMG reports in which the diagnosis of carpal tunnel syndrome was delivered based on the median palmar response demonstrating an “abnormal” calculated CV (47 m/sec), despite a normal peak latency response (2.1 msec) and a normal interpeak latency difference between the median and ulnar palmar responses (0.2 msec). This same EMG laboratory diagnosed a demyelinating sensory polyneuropathy based on “abnormal” calculated CVs of the median palmar and ulnar palmar responses, despite normal peak latencies. These misdiagnoses may result in mismanagement and, when this results in a bad outcome, in potential litigation.

Conduction Velocity In addition to the issues related to the onset and peak latency already discussed, there are a number of other issues related to the sensory nerve CV value that require discussion, including the number of stimulation sites (1-point versus 2-point stimulation) and the normal physiologic slowing that occurs along the distal segment of the nerve. Regarding the number of stimulation sites utilized, unlike with motor NCS, which use two separate stimulation sites to eliminate the latency components unrelated to nerve conduction time (i.e., the times related to nerve and muscle fiber activation, terminal nerve branch and muscle fiber conduction, and NMJ transmission), sensory NCS are usually performed with stimulation limited to a single site. Thus, in addition to nerve conduction time, there are other time elements that diminish the calculated nerve conduction velocity value, including nerve activation time (tissue transit time and depolarization threshold time) and distal tissue transit time (from the nerve fibers to the E1 electrode). Because of these unrelated times, whenever the nerve CV value is calculated from the latency value, the calculated value will be falsely lowered. To eliminate these dilutions, some have added a second stimulation site (as with

Chapter 8: Sensory Nerve Conduction Studies

the motor NCS). For example, the sural nerve CV can be calculated by stimulating the sural nerve at 7 centimeters proximal to the E1 electrode and at 21 centimeters proximal to the E1 electrode, thereby generating a distance of 14 centimeters. Although this eliminated the problem, the sensitivity was not significantly greater than simply comparing the recorded latency to the control value. Thus, in our EMG laboratories, we no longer do that. However, many EMG machines automatically calculate and display the nerve CV value for the sensory response obtained through 1point stimulation. Again, when the peak latency value is utilized to calculate the nerve CV, the calculated value may be well below the lower limit of normal, generating an erroneous conclusion and potential mismanagement. In summary, the peak latency value should never be used to calculate the nerve CV, as it is more reflective of the average CV rather than the fastest CV. A final issue concerns the physiological slowing related to the distal segment of sensory nerves. This physiologic slowing is due to thinner axons (increases the resistance to current advancement), thinner myelin coatings (increases current leakage), shorter myelin segments (increases membrane capacitance because the nodal membrane to internodal membrane ratio is higher), and cooler temperatures (Gilliatt and Thomas, 1960). Taller individuals also demonstrate slower nerve CV values (Campbell, 1981).

Advantages and Disadvantages of Sensory NCS As discussed earlier, because of their smaller size (lower amplitude and shorter negative phase duration), greater number of phases, and wider range of conduction velocities, the sensory responses are much more susceptible to physiologic dispersion than are the motor responses. Interestingly, these very features also generate certain advantages. The advantages of the sensory NCS include their sensitivity to focal axon loss pathology, their tendency toward incomplete response recovery (even when there is apparent complete clinical recovery), their strong contribution to lesion localization, and, of course, their ability to assess the sensory nerve fibers (i.e., this is the only portion of the EDX examination that assesses the sensory nerve fibers). First, their sensitivity to pathology reflects their much smaller size and the wider range of conduction velocities among the axons contributing to the sensory

response. Second, because the sensory axons do not have the same capacity for reinnervation that the motor axons do, the sensory responses do not always normalize following nerve fiber disruption. For this reason they frequently permit the recognition of remote lesions no longer appreciable clinically, by motor NCS (reinnervation via collateral sprouting normalizes the motor response; see Chapter 9), or by needle EMG (especially if the lesion was only mild in severity). Third, the sensory NCS contribute significantly to lesion localization, especially in the differentiation of preganglionic versus postganglionic/ ganglionic lesions. For example, in the setting of generalized weakness, sensory response involvement excludes anterior horn cell, NMJ, and muscle disorders. Fourth, only the sensory NCS assess the sensory nerve fibers. Thus, without sensory NCS, sensory neuropathies (e.g., superficial radial neuropathies), sensory polyneuropathies, and sensory neuronopathies would not be recognizable. There are a number of limitations associated with the sensory responses. First, because of their small size, they overestimate the lesion. Consequently, they are not useful for severity estimation. In general, when a mixed nerve is approximately 50% disrupted, the motor response is about 50% reduced in amplitude (compared to the asymptomatic contralateral side), whereas the sensory response is typically very low or absent. Other limitations related to their small size include greater challenge in their collection and greater susceptibility to physical and technical issues, such as intervening body tissue (finger girth; adipose; edema) and electrical artifacts (e.g., shock artifact; motor response artifact). Second, some elderly individuals show bilaterally unelicitable lower extremity sensory responses. In one study, it was concluded that individuals under the age of 75 years should have recordable lower extremity sensory responses (Tavee, 2014). Third, because the cutaneous nerves generating these responses are superficial, they may be absent due to previous trivial trauma. Fourth, the sensory responses do not assess the distalmost aspects of the sensory axons (i.e., the segment distal to the recording electrodes).

Other Important Sensory NCS Issues Effects of Filtering Also because of their small size, sensory responses are more susceptible to filtering. With both high-

119

Section 2: Nerve Conduction Studies

frequency and low-frequency filtering, signal is removed. Thus, the amplitude and negative AUC are diminished with both types of filtering. Highfrequency filtering and low-frequency filtering have opposing effects on the value of the peak latency. Because the rise time of the sensory response is predominantly composed of higher frequencies, as more and more high frequencies are removed through high-frequency filtering, the rise time is prolonged, causing the peak latency of the curve to be delayed. With excessive high-frequency filtering, the degree of peak latency delay is greater than the degree of amplitude decrement. Because the majority of components of the sensory response are of lower frequency, excessive low-frequency filtering significantly reduces the total signal recorded. As a result, the peak latency occurs earlier and the response terminates earlier. The onset latency is unaffected because it is a high-frequency component. Thus, the negative phase duration value is also lower. With excessive low-frequency filtering, the degree of amplitude loss is more pronounced than the degree of phase lead and can be dramatic enough to suggest an axon loss process. Importantly, with digital filtering, unlike with analog filtering, the peak latency shifts reflect signal loss (true signal filtering) rather than a phase shift. These concepts are discussed in greater detail later in this textbook (see Chapter 18).

Ideal Interelectrode Distance Although with a monopolar recording technique (i.e., the use of just one surface recording electrode) the recorded waveform is simpler to interpret, it is much more affected by unwanted signal (noise) than it is when a bipolar recording technique (the use of two surface recording electrodes, E1 and E2) is utilized. However, although a bipolar recording technique significantly lessens the undesired signal (because the compound electrical potential passes under two recording electrodes sequentially), interactions between the two recorded signals can potentially complicate waveform response interpretation. In a 1965 paper addressing this issue, it was pointed out that because the bipolar technique affects the amplitude and duration of the recorded response, the interelectrode distance utilized (the distance between the E1 and E2 electrodes) must be reported with the normal control values (Gilliatt et al., 1965).

120

As discussed in Chapter 2, with differential amplification, the response recorded by the E2 electrode is subtracted (inverted and then added) from the response recorded at the E1 electrode, with the goal being to subtract out the like signal (the undesired noise) and to amplify the desired signal (the compound AP). When the two surface recording electrodes are too close to each other, the E2 electrode records some of the desired signal (some of the negative phase) while it is still being recorded at the E1 electrode. As a result, it is inverted and added to the E1 input and lost. When significant, a submaximal amplitude is recorded and, when unrecognized, may be erroneously interpreted as axon loss, generating a falsely positive interpretation. In addition, when the peak of the response occurs earlier (because there is less signal overall), it may normalize a mildly delayed response, causing focal demyelinating conduction slowing to go unappreciated and, hence, generating a falsely negative conclusion (see Chapter 9). When the two electrodes are situated too far apart, their different recording environments might cause the common signal to appear differently (i.e., as differential signal), permitting it to be amplified. Because the amplitude value is the most important parameter measured, and because the rise time is the primary determinant of the amplitude, it is important to ensure that the E1 and E2 electrodes do not record the rise time simultaneously. To avoid this, the rise time should be fully recorded by the E1 electrode before any of it reaches the E2 electrode. Assuming a rise time of 0.8 milliseconds and a conduction velocity of 50 m/sec, the ideal interelectrode distance is 4 centimeters (Eduardo and Burke, 1988): velocity ¼ distance=time distance ¼ velocity  time ðrearrangingÞ ¼ 50 m=sec  0:8 msec ¼ 50 mm=msec  0:8 msec ¼ 40 mm ¼ 4 cm The rise time is akin to the depolarization time (the time from VGNC opening to the time of VGNC closure). In other words, when the wave of depolarization advances at 50 m/sec, the distance that the AP propagates in 0.8 msec is 4 cm. Thus, 4 cm is the nerve distance over which the rise time (depolarization phase) occurs. In this manner, the depolarization phase is just finishing at the E1 electrode when it is just arriving at the E2 electrode. Hence, the desired signal is only recorded at one electrode at a

Chapter 8: Sensory Nerve Conduction Studies

E1 E1

E2

E1

E2

E2

(a)

(b)

(c)

Figure 8.4 The effect of varying distance between the E1 and E2 recording electrodes. A. When the distance between the E1 and E2 electrodes is too small, the recorded depolarizations overlap and desired signal is lost, resulting in response amplitude reduction. B. In the ideal situation, the E2 electrode records the depolarization just as the E1 electrode finishes recording it. The lack of depolarization overlap maximizes the recorded response amplitude. C. When the distance between the E1 and E2 electrodes is too great, the undesired environmental noise appears differently and, thus, is amplified thereby contaminating the recording.

time, limiting inadvertent common mode rejection of differential signal. When the conduction velocity is greater than 50 m/sec, then an even greater interelectrode distance is required to avoid signal loss, whereas when it is less than 50 m/sec, a lesser interelectrode distance is acceptable. When shorter interelectrode distances are used, desired signal is lost, and when longer interelectrode distances are used, noise may be amplified (see Figure 8.4). In the above calculation, it was shown that the ideal separation between the E1 and E2 recording electrodes is 4 cm to avoid rise time overlap and signal loss. However, because the rise time is a small portion of the total response, the more distal portions of the waveform (those occurring after the rise time) will overlap. Thus, the early repolarization phase at E1 (a negative deflection) will overlap with the early depolarization at E2 (also a negative deflection). This will result in some signal loss and a steeper descent. When the E1 electrode is recording the postdepolarization positivity (i.e., its terminal positive phase), the E2 electrode is still recording depolarization (negativity). Thus, the E2 negativity will be inverted (made positive) and added to the positive E1 signal. In this manner, the positive phase increases in amplitude. For this reason, the repolarization side of the response, which is the portion that is most affected by the E2 electrode, is the most unreliable portion. Consequently, it is imperative that the E1 and E2 electrodes be placed in the exact positions used to collect the normal values of the technique and that the amplitude not be measured from the negative peak to the subsequent positive peak (P2: P3). Because the E1 electrode always begins to record the response before the E2 electrode, the onset latency is never affected by the interelectrode distance.

Finally, a 4-cm distance can be problematic when individuals with shorter digits are studied. For this reason, many EMG laboratories use shorter interelectrode distances than ideal and accept some signal loss. Thus, the ideal interelectrode distance is the distance that was used to collect the normal control values for the particular NCS. In our EMG laboratories, we use a 3-centimeter interelectrode distance because that was the distance used in the collection of the normal control values used in our EMG laboratory (see Appendix 6).

Mixed Nerve Conduction Studies With mixed NCS, the stimulating electrodes are positioned over the sensory and motor nerve fibers at one site along the nerve and the recording electrodes are situated at a more proximal site. Like the sensory and motor responses, these mixed responses are compound electrical potentials, the composition of which reflects proximally conducting sensory and motor nerve fiber APs. Thus, the SNAP component is orthodromic and the CMAP component is antidromic. Also, like sensory responses, because mixed responses are composed solely of nerve fiber APs, there is no amplification effect (the motor nerve fibers are conducting proximally). Mixed responses (e.g., the median palmar response) tend to be larger than the sensory responses (e.g., the median sensory response recording second digit) because they reflect the nerve fiber APs of the sensory nerve fibers and the motor nerve fibers, rather than just a subset of the sensory nerve fibers (e.g., those from the second digit). In general, mixed responses have a triphasic waveform morphology because the APs are generated away from the recording electrodes. However, they

121

Section 2: Nerve Conduction Studies

may also have a biphasic morphology. The amplitude and latency measurement values are obtained in the same way that they are for sensory responses – baseline to peak amplitudes for biphasic responses and the peak of first positive phase to the peak of first negative phase (P1:P2) for triphasic responses; we collect the peak latency rather than the onset latency. Mixed NCS commonly used are the median and ulnar palmar NCS in the upper extremity and the medial and lateral plantar NCS in the lower extremity. The plantar sensory responses are helpful with identifying early polyneuropathy and in determining whether the absence of the routine lower extremity sensory responses is abnormal or related to adipose tissue. Thus, when the routine lower extremity sensory responses are absent in an obese individual, the question arises as to

References Bolton CF, Carter K. Human sensory nerve compound action potential amplitude. Variation with sex and finger circumference. J Neurol Neurosurg Psychiatry 1980;43:925–928. Buller AJ, Styles PR. A new averaging technique for improving the signalto-noise ratio of evoked potentials. J Physiol (Lond) 1959;149:65P. Campbell WW, Ward LC, Swift TR. Nerve conduction velocity varies inversely with height. Muscle Nerve 1981;4:520–523. Chodoroff G, Tashjian EA, Ellenberg MR. Orthodromic versus antidromic sensory nerve latencies in healthy persons. Arch Phys Med Rehabil 1985;66:589–591. Cohn TG, Wertsch JJ, Pasupuleti DV, Loftsgaarden BS, Schenk VA. Nerve conduction studies: orthodromic versus antidromic latencies. Arch Phys Med Rehabil 1990;71:579–582. Dorfman LJ. The distribution of conduction velocities (DCV) in peripheral nerves: a review. Muscle Nerve 1984;7:2–11. Dorfman LJ, Cummins KL, Abraham GS. Conduction velocity distributions of the human median

122

whether their absence reflects a sensory polyneuropathy or a technical limitation related to body habitus. Because individuals do not store significant amounts of adipose tissue below the medial malleolus, the plantar NCS can be used to differentiate these two possibilities. Because of the stocking distribution of polyneuropathies, the presence of normal plantar mixed responses indicates that the absent sural and superficial peroneal responses do not represent a polyneuropathy and, consequently, more likely represent the technical impediments related to excess adipose. Unfortunately, the plantar responses may be absent among normal individuals over the age of 45 years. The techniques for performing these NCS and their age-adjusted normal values are provided in Section 6 (see Appendixes 5 and 6).

nerve: comparison of methods. Muscle Nerve 1982;5:S148–S153. Dumitru D. Electrodiagnostic medicine. Philadelphia: Hanley & Belfus, 1995. Eduardo E, Burke D. The optimal recording electrode configuration for compound sensory action potentials. J Neurol Neurosurg Psychiatry 1988;51:684–687. Gilliatt RW, Melville ID, Velate AS, Eillison RG. A study of normal nerve action potentials using an averaging technique (barrier gri storage tube). J Neurol Neurosurg Psychiatry 1965;28:191–200. Gilliatt RW, Thomas PK. Changes in nerve conduction with ulnar lesions at the elbow. J Neurol Neurosurg Psychiatry 1960;23:312–320. Kimura J, Machida M, Ishida T, Yamada T, Rodnitzky RL, Kudo Y, Suzuki S. Relationship between size of compound sensory or muscle action potentials and length of nerve segment. Neurology 1986;36:647–652. Murai Y, Sanderson I. Studies of sensory conduction. Comparison of latencies of orthodromic and antidromic sensory potentials. J Neurol Neurosurg Psychiatry 1975;38:1187–1189.

Nandedkar S. Motor and sensory nerve conduction: technique, measurements, and anatomic correlation. In Neurophysiology and Instrumentation, 57th Annual Meeting of the American Association of Neuromuscular and Electrodiagnostic Medicine, Quebec City, Quebec, Canada, 2010:1–8. Pitman JR. A digital delay and timescale generator. J Physiol (Lond) 1958;148:30–31P. Raynor EM, Preston DC, Logigian EL. Influence of surface recording electrode placement on nerve action potentials. Muscle Nerve 1997;20:361–363. Tavee JO, Polston D, Zhou L, Shields RW, Butler RS, Levin KH. Sural sensory nerve action potential, epidermal nerve fiber density, and quantitative sudomotor axon reflex in the healthy elderly. Muscle Nerve 2014;49:564–569. Wilbourn AJ. Sensory nerve conduction studies. J Clin Neurophysiol 1994;11:584–601. Wilbourn AJ, Ferrante MA. Clinical electromyography. In Joynt RJ, Griggs RC, editors, Clinical neurology, Philadelphia: LippincottRaven, 1997:1–76.

Chapter

9

The NCS Manifestations of Various Pathologies

Introduction The peripheral nervous system is composed of nerve fibers, each of which is composed of a central axon and its coverings. Schwann cells surround the axons. Some Schwann cells wrap around multiple axons (termed unmyelinated axons), whereas others only wrap around a single axon (termed myelinated axons). With myelinated axons, the myelin is multilayered, and because each Schwann cell surrounds just a small segment of the axon, the myelin sheath is said to be segmental. Between these segments of myelinated axon is a small region of unmyelinated axon (the node of Ranvier). This information was covered in detail in Chapter 3. EMG testing only assesses the heavily myelinated axons. Consequently, the EDX manifestations of all nerve fiber disorders can be divided into two pathologic types: (1) disorders resulting in myelin disruption (termed demyelination) and (2) disorders resulting in axon disruption (termed axon loss). Because myelin is segmentally arranged, its focal loss is referred to as segmental demyelination. When segmental demyelination is confined to the internodal membrane region adjacent to the node, it is referred to as paranodal demyelination. Paranodal demyelination often precedes segmental demyelination. Demyelination also occurs with diseases of the Schwann cell. With axon disruption, Wallerian degeneration (also termed axonal degeneration) follows. Many electromyographers refer to disorders of axon disruption as axon loss disorders. Wallerian degeneration also results from disorders of the cell body (lower motor neuron disorders; sensory neuron disorders). Of these two pathologies (demyelination and axon loss), axon loss is by far the more common. These two pathologies (demyelination and axon loss) are associated with three distinct pathophysiologies: (1) demyelinating conduction slowing, (2) demyelinating conduction block, and (3) axonal

conduction failure. All three of these pathophysiologies are demonstrable by NCS and each one has unique NCS manifestations. To understand the associated NCS manifestations, the EDX provider must understand the effect that these two pathologies have on action potential propagation.

Effects of Focal Demyelination on Action Potentials As previously reviewed, AP propagation along myelinated nerve fibers occurs through the regeneration of Na+ current via sequential depolarization of the nodal regions (see Chapter 3). Because AP propagation along myelinated axons only requires membrane discharge at the nodes, it “leaps” over the internodal (myelinated) segments. This type of AP propagation is termed saltatory conduction. Focal demyelination is associated with two types of pathophysiology: demyelinating conduction slowing and demyelinating conduction block. With lesser degrees of demyelination, conduction slowing occurs across the lesion site, whereas with greater degrees of myelin loss, conduction across the lesion is blocked. Recall from Chapter 3 that one of the major functions of myelin is to increase the distance between the inner surface of the axolemma and the extracellular fluid. As a result, charge does not accumulate on the two sides of the axolemma, and the capacitance of the membrane is significantly reduced, thereby increasing the speed of AP propagation. With myelin loss, exposure of the axolemma results, thereby increasing the capacitance of the membrane (i.e., the internodal membrane behaves like nodal membrane when the myelin is removed). In other words, the loss of myelin lessens the distance between the inner surface of the axolemma and extracellular fluid, thereby permitting charge accumulation along the two surfaces of the axolemma. This charge accumulation must be removed (i.e., the membrane must be discharged)

123

Section 2: Nerve Conduction Studies

before the AP can traverse it. This converts saltatory conduction to continuous conduction, significantly slowing the speed of AP propagation (i.e., the NCV). Because the VGNC density of the nodal membrane is approximately 10-fold greater than that of the internodal membrane, exposure of the internodal membrane, in addition to increasing membrane capacitance, decreases the density of the Na+ current. As a result, it takes slightly longer to reach the voltage threshold for depolarization. This also slows the speed of AP propagation. When demyelination produces slowing of AP propagation, the term demyelinating conduction slowing is applied. The slowing can be uniform (all of the nerve fibers are equally affected) or nonuniform (the nerve fibers are unequally affected). Uniform and nonuniform slowing is discussed in more detail later here. Clinically, because the APs ultimately reach their targets, negative symptoms, such as weakness or numbness, do not occur. Instead, these lesions are either asymptomatic or generate transient positive symptoms, such as episodic tingling. With greater degrees of demyelination, the APs are unable to traverse the lesion. Because AP propagation is blocked, the term demyelinating conduction block is applied. Clinically, because the APs do not reach their target, negative symptoms (e.g., weakness, numbness) result. In addition to focal demyelination, demyelination of nerve fibers may be multifocal, such as with acquired disorders producing generalized demyelination (e.g., acute inflammatory demyelinating polyneuropathy). One of the most important concepts in EDX medicine is that the effects of focal demyelination are local (they do not result in changes distant to the lesion site). As a result, in order for focal demyelination to be identified by NCS, current must pass through the lesion. Therefore, the stimulating and recording electrodes must straddle the lesion (i.e., be located on opposite sides of the lesion). This is discussed in greater detail further on.

Effects of Focal Axon Loss on Action Potentials When the axon is disrupted, the segment of axon distal to the site of transection degenerates through a process termed Wallerian degeneration. This is also sometimes referred to as distal stump degeneration. Initially, there is degradation of the axon, the

124

axolemma, and the myelin sheath. This is accompanied by Schwann cell proliferation. The endoneurial sheath fills with mobile Schwann cells (Hanke Bungner bands). (In the past, the endoneurial sheath and the myelin sheath were thought to represent a single layer, the neurilemma, which is now an outdated term.) The degradation products are subsequently removed by infiltrating macrophages and Schwann cells. Within hours of the disruption, the axons within the proximal stump sprout collaterals that then advance distally. Depending on the distance between the proximal and distal nerve stumps and the degree of connective tissue involvement, the regenerating axons (1) enter their original endoneurial tube, (2) enter a different endoneurial tube, or (3) fail to enter an endoneurial tube and exit the fascicle. This is discussed in greater detail in Chapter 16. Thus, unlike with focal demyelination, which has only local effects, with focal axon disruption, there are both local (at the injury site) and distant (distal to the injury site) effects. The local effects begin immediately, but the distant effects are delayed while Wallerian degeneration is occurring. Once the axon is transected, the AP cannot traverse it, and conduction fails. This occurs immediately. Prior to Wallerian degeneration, the axon segment distal to the transection remains able to conduct APs. Thus, immediately following axon disruption, the stimulating and recording electrodes must be on opposite sides of the lesion in order for the lesion to be appreciable, because current must traverse the lesion for it to be visible. Once Wallerian degeneration occurs, the lesion is visible as long as current traverses the lesion or the degenerated distal segment. As stated earlier, the pathophysiology associated with axon disruption is termed conduction failure, axonal conduction failure, or, more simply, axon loss. In addition to distal stump degeneration, the cell body also undergoes pathologic changes, such as chromatolysis and an increase in the synthesis of RNA and protein. These latter changes do not have NCS manifestations and, hence, are not discussed further.

The Motor NCS Manifestations of Pathology and Pathophysiology Introduction The motor unit is defined as a single AHC and all of the muscle fibers it innervates, including the

Chapter 9: The NCS Manifestations of Various Pathologies

intervening neuromuscular junctions (NMJs). When a motor unit is activated, its muscle fibers contract and contractile force is generated. The motor unit is the smallest unit of contractile muscle force. The magnitude of the force generated by a motor unit is proportional to its innervation ratio (i.e., the greater the number of muscle fibers, the greater the amount of contractile force generated per activation) and to its firing frequency (i.e., the faster the activation frequency, the greater the contractile force). Although the innervation ratio of the motor unit is invariable, the firing frequency varies (typically in the 5–50 Hz range). The magnitude of contractile force generated by the muscle belly is proportional to its crosssectional area (πr2). Different neuromuscular disorders affect different regions of the motor unit (cell body, axon, myelin sheath, terminal nerve branches, NMJs, muscle fibers) and in different ways. The severity and potential for recovery reflect the underlying disorder. With focal demyelination, recovery occurs through remyelination, which typically occurs within three months (and frequently within just a few weeks). With axon disruption, reinnervation occurs through one of two mechanisms: proximodistal axon advancement and collateral sprouting. With proximodistal axon advancement, axon sprouts emerge from the proximal end of the severed axons. These then grow distally to reinnervate the denervated muscle fibers. With collateral sprouting, the unaffected motor axons within the muscle respond to the denervated muscle fibers by sprouting collaterals (branches) that then grow to reinnervate the denervated muscle fibers. With the latter mechanism, the distance from the motor axon to the denervated muscle fiber is short because it occurs intramuscularly. Thus, this mechanism of reinnervation can occur quite quickly. Unfortunately, collateral sprouting only occurs when the lesion is incomplete, because it requires unaffected intramuscular axons to generate the collateral sprouts. With cell body loss (motor neuronopathies, such as amyotrophic lateral sclerosis), proximodistal axon advancement cannot occur because the entire lower motor neuron is lost. In this setting (e.g., amyotrophic lateral sclerosis), the denervated muscle fibers can only be reinnervated through collateral sprouting. As more and more neurons are lost, the likelihood of reinnervation through this mechanism diminishes. Reinnervation is discussed in detail later in this textbook (see Chapter 16).

The Pathology of Focal Demyelination The motor NCS manifestations of focal demyelination are easily understood when the two primary functions of myelin are understood. First, as previously discussed (see Chapter 3), the myelin sheath eliminates the membrane capacitance (the accumulated charges on the two sides of the axolemma), thereby decreasing the time constant (the time required to displace the charges, referred to as the discharge time) of the membrane. This increases the speed of AP propagation because only a minority of the membrane requires discharging (the nodes of Ranvier). Second, the myelin sheath insulates the axon, which lessens the amount of leakage current (transverse membrane current), thereby increasing its length constant (the distance the AP can travel without rejuvenation). The ratio of the distance the AP can propagate to the distance required for rejuvenation is referred to as the safety factor of AP propagation. The safety factor can also be expressed in terms of current as the ratio of the current available to the current required to discharge the capacitance of the next nerve segment (current available: current required). The safety factor for larger, more heavily myelinated axons can be as high as 7 (i.e., the available current is 7 times greater than the required amount) (Campbell, 1999). Hence, the focal loss of myelin results in a focal increase in membrane capacitance and focal increase in leakage current. The increase in membrane capacitance increases the discharge time of the membrane, thereby slowing the speed of AP propagation (demyelinating conduction slowing, DMCS), whereas the increase in transverse current leakage decreases the distance along the axolemma that the AP can conduct without rejuvenation. When it is so short as to not reach the next rejuvenation site (i.e., the next node), the AP is lost (termed demyelinating conduction block, DMCB). Two other effects of focal demyelination also slow the AP propagation speed, albeit to lesser degrees: (1) increased K+ efflux and (2) decreased Na+ current density. First, because the nodal and internodal membranes have different ion channel distributions – the nodal membrane contains more sodium channels and the internodal membrane contains more potassium channels (especially the paranodal membrane) – focal demyelination increases the number of exposed potassium channels, and hence, K+ efflux increases. The increase in K+

125

Section 2: Nerve Conduction Studies

efflux polarizes the membrane, thereby impeding its depolarization and slowing AP propagation speed. Second, because myelin loss in effect increases the surface area of the unmyelinated membrane (i.e., the surface area of the “nodal” membrane is larger), the density of the VGNCs (the number of VGNCs per unit area) is reduced. As a result, the density of the incoming Na+ current is reduced. For this reason, more time is required to reach the membrane depolarization value (about –55 mV), which also contributes to a slower AP propagation speed. Certain sites within the PNS are more susceptible to AP propagation failure than others. For example, impedance mismatch occurs at segments along the neuron where there is a transition between myelinated and unmyelinated nerve segments (e.g., the initial axon segment and the terminal axon segment) and branch points result in current dilution. The safety factor for AP propagation (available current/ required current) is diminished at these sites. This may explain why early Guillain-Barre syndrome has a tendency to involve proximal and distal sites (in addition to sites of common entrapment).

The Pathophysiology of Focal Demyelination Focal Demyelinating Conduction Slowing With the loss of myelin, the charges in the extracellular fluid and along the inner surface of the axolemma can again interact, causing a return of the capacitance of the membrane at the demyelinated segment. As a result, AP propagation through the region changes from saltatory to continuous, which is much slower. As expected, the NCS manifestations associated with focal demyelination reflect this slowing of AP propagation speed. Again, these features are only demonstrable on NCS when the stimulating and recording electrodes straddle the lesion. In other words, the NCS manifestations of focal demyelination are only apparent when current passes through the lesion. Consequently, when a focus of demyelination is situated proximal to the stimulating and recording electrodes, it is not discernible. Likewise, when a focus of demyelination is situated distal to the stimulating and recording electrodes, it is also not discernible In regard to the site of focal demyelination and its relationship to the stimulating and recording electrodes, there are two important features unique to motor NCS that the EDX provider must understand. First, because motor NCS are performed in an

126

orthodromic manner with the recording electrodes overlying the muscle belly (E1) and the tendon (E2), the recording electrodes are always distal to the nerve and, therefore, they are always distal to the focal demyelination. In other words, unlike with sensory NCS, the focus of demyelination can never be distal to the recording electrodes. The other unique feature of motor NCS is that the nerve is stimulated at two different sites, one proximal and one distal. Thus, there are three possible relationships between the two stimulation sites and the lesion. The lesion may be situated distal to both stimulation sites, it may be located between the two stimulation sites, or it may lie proximal to both stimulation sites. With focal lesions producing DMCS, the NCS manifestations vary with the uniformity of involvement of the affected fibers, the completeness of the lesion, and the degree of involvement of the fastest conducting fibers. Focal DMCS is uniform when the nerve fibers are equally slowed as they traverse the lesion. When the nerve fibers are unequally slowed at the lesion site, the term nonuniform DMCS is applied. Uniform Demyelinating Conduction Slowing––With uniform demyelinating conduction slowing, which is also referred to as synchronized slowing, the nerve fibers are equally slowed. Consequently, the individual motor nerve fiber action potentials composing the recorded motor response maintain their temporal relationship to each other. Consequently, the morphology of the motor response waveform appears normal. For this reason, with uniform DMCS, the onset latency of the motor response is delayed but the waveform morphology is normal or nearly so. Carpal tunnel syndrome. This type of pathophysiology is typically seen in early carpal tunnel syndrome (i.e., prior to the development of axon loss). With carpal tunnel syndrome, the APs propagating down the affected axons are slowed as they pass through the carpal tunnel. Because the carpal tunnel lies distal to both stimulation sites, the focus lies between the stimulating and recording electrodes for both motor responses. Thus, both motor responses show the delay, which is equal in degree (see Figure 9.1). In the figure, the calculated motor conduction velocity is normal. Even when an axon loss lesion is severe in degree, until all of the heavily myelinated fibers are affected, the calculated conduction velocity remains normal (see Figure 9.2).

Chapter 9: The NCS Manifestations of Various Pathologies

5 mV

5 ms Lat ms

Amp Dist mV mm

12.0

3.6

17.3

3.5

C.V. m/s

50 273

51.1

Figure 9.1 Focal demyelinating conduction slowing with carpal tunnel syndrome. Both motor responses are equally delayed in onset, indicating focal demyelinating conduction slowing distal to the wrist. Because the waveform morphologies are preserved, this is termed uniform DMCS. The calculated conduction velocity is normal. In this case, the amplitudes of the two responses are equally reduced, indicating concomitant axon loss. Although axon loss can result in conduction slowing when it involves all of the heavily myelinated fibers, the degree of slowing illustrated here is too pronounced to be accounted for by the concomitant axon loss, and moreover, the calculated nerve conduction velocity is normal, indicating that at least some of the heavily myelinated fibers are conducting in the normal range. This is an example of advanced carpal tunnel syndrome, with features of both demyelinating conduction slowing and axon loss.

5 mV

5 ms Lat ms

Amp mV

Dist mm

C.V. m/s

15.3

0.750 50 256 0.750

53.8

20.0

Figure 9.2 Severe carpal tunnel syndrome associated with a normal forearm conduction velocity value. Until all of the heavily myelinated fibers are affected, the calculated forearm conduction velocity remains normal.

Wrist

I

It is important to understand that the calculated forearm segment conduction velocity reflects the first action potentials to reach the recording electrodes, not the distal stimulation site. Thus, the calculated forearm conduction velocity does not actually reflect the fastest fibers between the proximal and distal stimulation sites (i.e., the forearm segment). Whenever all of the fastest conducting forearm fibers are delayed at the carpal tunnel, the calculated forearm segment conduction velocity will reflect a slower nerve fiber population. In other words, the first nerve fiber action potentials to reach the recording electrodes are not the fastest conducting ones of the forearm segment of the nerve (see Figure 9.3). Nonuniform Demyelinating Conduction Slowing When the demyelinating lesion lies between the two stimulation sites, the amplitude and latency values of the distal motor response are normal because current does not traverse the lesion when the nerve is distal to it. Thus, for the lesion to be identified, the nerve must be stimulated proximal to the lesion. When the lesion involves all of the fastest conducting fibers, the latency of the proximal motor response is delayed, and consequently, the calculated conduction velocity value is reduced. When the demyelinated nerve fibers are affected to different degrees, they are slowed to different degrees. As a result, the APs composing the motor response lose their temporal relationship to

Elbow

F S

4.2

I

F S

9.6

Wrist stimulation

Elbow stimulation

F I S

Figure 9.3 Spurious forearm segment slowing associated with carpal tunnel syndrome. In the illustration, the fastest fibers (F) are selectively delayed within the carpal tunnel, allowing the intermediate fibers (I) to reach the recording electrodes first. Thus, it is important to realize that the calculated forearm segment conduction velocity reflects the first fibers to reach the recording electrodes, not the distal stimulation site. Thus, whenever the fastest fibers are selectively slowed distal to the forearm segment (e.g., within the carpal tunnel), the calculated forearm segment conduction velocity is spuriously lowered. In this illustration, the intermediate fibers (I) reach the recording electrodes before the fastest (F) and slowest (S) conducting fibers and, consequently, dictate the calculated forearm segment conduction velocity.

Dd = 250 mm Dt = 5.4 msec CV = Dd/Dt = 46.3 m/sec Normal forearm segment CV (intermediate fibers)

127

Section 2: Nerve Conduction Studies

each other, and consequently, the morphology of the recorded waveform is distorted (i.e., it is dispersed due to the loss of synchrony). This form of pathological temporal dispersion is referred to as nonuniform slowing or differential slowing. In this setting, the waveform conformation of the proximal motor response differs from that of the distal motor response. Because all of the individual motor nerve fiber action potentials reach the recording electrodes (i.e., the degree of individual nerve fiber demyelination among the affected nerve fibers is insufficient to block action potential conduction), the negative area under the curve value remains unchanged (see Figure 9.4). Nonuniform slowing is commonly observed with ulnar neuropathies at the elbow segment. With ulnar neuropathies at the elbow, stimulation below the lesion (e.g., at the wrist and the below-elbow stimulation sites) is unaffected because current is not traversing the lesion. However, with stimulation above the lesion, current traverses the lesion, and therefore, pathological temporal dispersion is apparent. Focal Demyelinating Conduction Block Regarding the other major function of myelin – axon insulation (resistance to transverse current flow) – the loss of myelin results in loss of axon insulation. This permits current leakage through the demyelinated segments as the sodium current advances along the nerve. As a result, the length constant is shortened 5 mV

(the Na+ current decays quicker due to current leakage during advancement through the demyelinated region), and hence, the current does not propagate as far down the axon. When the length constant is reduced to the point that the magnitude of the advancing Na+ current is below the depolarization threshold value of the subsequent region of membrane, the VGNCs do not open, and as a result, sodium current rejuvenation of the AP does not occur. Consequently, the AP ceases to exist. When demyelination results in this outcome, it is termed demyelinating conduction block (DMCB). Like focal DMCS, focal DMCB is only appreciable when current runs through the lesion. Unlike focal DMCS, there is no slowing or dispersion with pure DMCB. With DMCB, there is a loss of current (signal) from the recorded response (e.g., decreased amplitude and negative AUC). Again, this occurs with stimulation proximal to the lesion but not with stimulation distal to the lesion. This difference in motor response size (CMAP size) at different stimulation sites is what allows the DMCB to be localized and quantified. These lesions are more common across the spiral groove and the elbow segment (see Figures 9.5 and 9.6). The amount of response amplitude or negative area under the curve loss that confidently identifies a DMCB lesion is challenging to define, primarily because the decrement reflects more than one variable – decrement related to the lesion (pathological decrement from demyelinating conduction block or nonuniform demyelinating conduction slowing) and

5 ms 5 mV

Figure 9.4 Nonuniform demyelinating conduction slowing. In the illustration, stimulation below the lesion generates the top trace and stimulation above the lesion generates the bottom trace. With nerve stimulation below the lesion, current does not traverse it, and thus, the response has a normal appearance. With stimulation above the lesion, current traverses the lesion. In the illustration, the waveform conformation of the proximal motor response differs from that of the distal motor response. Preservation of the negative area under the curve indicates that the change reflects nonuniform demyelinating conduction slowing. With uniform slowing, the waveform conformation would not change, and with demyelinating conduction block (discussed below), the negative area under the curve would be reduced. In this example, the calculated conduction velocity was 55 m/sec. Thus, at least one of the more heavily myelinated nerve fibers is conducting normally.

128

5 ms Lat ms

Amp mV

3.4

6.7

4.3

7.3

9.0

0.333

Figure 9.5 Demyelinating conduction block of the radial nerve across the spiral groove. The radial motor response, recording extensor digitorum, is significantly reduced with stimulation above the spiral groove (the lower trace) as compared to the responses recorded with stimulation at the elbow (upper trace) and below the spiral groove (middle trace). This amplitude discrepancy indicates a focus of demyelinating conduction block between the below-spiralgroove and above-spiral-groove stimulation sites (i.e., within the spiral groove segment of the radial nerve). This location is a common site of compression with radial nerve lesions.

Chapter 9: The NCS Manifestations of Various Pathologies

5 mV

5 ms

Figure 9.6 Demyelinating conduction block of the ulnar nerve across the ulnar groove. The ulnar motor response, recording abductor digiti minimi, is significantly reduced with stimulation above the elbow (the lower trace) as compared to the responses recorded with stimulation below the elbow (middle trace) and at the wrist (upper trace). This amplitude discrepancy indicates a focus of demyelinating conduction block between the above-elbow and below-elbow stimulation sites (i.e., across the elbow segment). This location is a common site of compression with ulnar nerve lesions.

decrement related to distance between the stimulating and recording electrodes (physiologic decrement related to temporal dispersion). As previously discussed, with physiological temporal dispersion, the decreased synchrony allows for greater interaction between the positive and negative phases of the individual motor axon APs comprising the CMAP, thereby increasing signal cancellation. Assuming a 2:1 ratio between the negative AUC of the first phase and the positive AUC of the second phase, the theoretical maximum response decrement possible from physiological temporal dispersion is 50% of the negative area under the curve value. Indeed, computer simulation studies have shown that physiologic dispersion alone can cause the proximal motor response to be 50% lower in amplitude than the distal motor response (Rhee et al., 1990). Because motor response decrement related to physiologic temporal dispersion is distance dependent and occurs uniformly over distance, whenever a decrement is sudden or is observed over a short distance, the decrement is likely to be pathologic. For example, an amplitude drop exceeding 20% over a 5-centimeter distance is likely abnormal, whereas the same amount of amplitude decrement over a 20-centimeter distance could be normal. Because temporal dispersion causes the duration of the negative phase to increase, the negative phase duration is a third variable to consider (i.e., degree of decrement, distance over which the decrement occurs, and change in the duration of the negative phase). For example, when the phase

duration increases by less than 15%, an amplitude or negative AUC decrement of more than 20% defines DMCB, whereas when the negative phase duration increases by over 15%, the amplitude or negative AUC decrement must exceed 50% to meet the criteria for DMCB (Ad Hoc Subcommittee of the American Academy of Neurology AIDS Task Force, 1991). When Uncini and colleagues applied these durationdependent definitions, they reported several false positive results using the first definition (>20% decrement with a negative phase duration of less than 15%) and more accurate results applying the second definition (Uncini et al., 1993). Other considerations include advanced nonuniform demyelinating conduction slowing and regeneration of motor axons (Cornblath et al., 1991). Moreover, with significant nonuniform demyelination conduction slowing, as the response becomes more polyphasic, the degree of negative and positive phase interaction increases further, causing focal nonuniform demyelinating conduction slowing to mimic focal demyelinating conduction block. Finally, in the setting of concomitant motor axon loss, the conduction velocity slowing related to motor axon regeneration also contributes. Regarding the NCS manifestations of focal DMCB, similar to focal DMCS, because routine motor NCS stimulate the nerve trunk at two sites, one proximal and one distal, there are three possible relationships between the two stimulation sites and the lesion site: the lesion may be situated distal to the two stimulation sites, it may be located between the two stimulation sites, or it may lie proximal to the two stimulation sites. First, when the lesion lies distal to both stimulation sites, the distal and proximal motor responses are equally reduced in size (equally decreased amplitude and negative AUC values). This pattern of equally low amplitude distal and proximal motor responses mimics an axon loss process (discussed further on). When the two responses are significantly reduced and the needle EMG shows normal-appearing MUAPs (assuming enough time has elapsed for reinnervation via collateral sprouting to have occurred), a DMCB is suggested. The presence of fibrillation potentials should not be used to differentiate between these two possibilities. For example, assuming an innervation ratio of 400, disruption of a single axon generates 400 fibrillation potentials. The concomitant disruption of a single motor axon would not be uncommon in the setting of a focus of DMCB.

129

Section 2: Nerve Conduction Studies

5 mV

5 ms Lat ms

Amp mV

3.4

6.7

4.3

7.3

9.0

0.333

5 mV

Figure 9.7 Radial neuropathy due to demyelinating conduction block across the spiral groove. The amplitude discrepancy can be used to estimate the percentage of blocked fibers. In this case, the percentage of fibers traversing the lesion is 0.333/7.3  100%, which equals 5%. Thus, the lesion involves 95% of the heavily myelinated nerve fibers.

Second, when the lesion lies between the distal and proximal stimulation sites, the distal motor response is normal (current does not pass through the lesion), whereas the proximal motor response is reduced in size (current passes through the lesion). The size discrepancy between the two responses (i.e., the amplitude and negative AUC values) approximates the percentage of blocked motor nerve fibers. These percentages are easily calculated, as shown in Figure 9.7 and in the Case Studies Section of this textbook (see Figure 9.7 and Section 5). Third, when the lesion lies proximal to both stimulation sites, it is unrecognized, because neither stimulation site results in current traversing the lesion. However, its presence is inferred whenever a neurogenic MUAP recruitment pattern is observed during needle EMG of a muscle from which a normal or near-normal motor response was recorded. This is true because a neurogenic MUAP recruitment pattern only occurs with axon loss and DMCB (i.e., the two pathophysiologies that result in AP loss). Because the motor response is normal or near-normal, axon loss is excluded, leaving only one possibility – DMCB. Because it was not identified during the routine motor NCS, it must lie proximal to the previous stimulation sites. Consequently, the study is repeated with stimulation applied at more proximal levels (see Figure 9.8). For the upper extremity, the most proximal stimulation site is the supraclavicular fossa, which stimulates axons at the trunk level of the brachial plexus. For the lower extremity, the most proximal stimulation site is the popliteal fossa. When proximal stimulation does not disclose the lesion, it must lie

130

5 ms

Stimulus site 1. 2. 3. 4. 5.

Wrist B. elbow A. elbow Axilla Erb’s

Lat ms

Amp mV

3.8 8.2 10.4 12.8 17.7

14.7 11.5 11.5 10.9 2.8

Figure 9.8 Demyelinating conduction block between the axilla and supraclavicular (Erb’s point) stimulation sites. This 48-year-old male presented with ulnar nerve distribution weakness and had normal routine ulnar motor NCS. During the needle EMG study, a neurogenic MUAP recruitment pattern was noted in the first dorsal interosseous muscle, indicating that the responsible lesion is either axon loss or demyelinating conduction block. Because the distal ulnar motor response (recording from the first dorsal interosseous muscle) is normal, an axon loss process cannot be responsible and, thus, can be excluded. Consequently, a demyelinating conduction block proximal to the distal stimulation site is the only possible explanation. For this reason, the study is repeated and more proximal nerve stimulation is performed (seeking to identify the lesion by stimulating proximal to it). Stimulation at the below-elbow, above-elbow, and axilla sites did not show a significant decrement in response size, indicating that the lesion lies proximal to the axilla. Stimulation in the supraclavicular fossa showed a significant motor response decrement, localizing the lesion to the supraclavicular fossa-axilla segment of the ulnar motor fibers to the first dorsal interosseous as they traverse the brachial plexus. The decrement was reproducible and was not observed on the contralateral side. This patient had other upper extremity demyelinating conduction blocks, consistent with multifocal motor neuropathy.

proximal to the most proximal stimulation site stimulated (see Section 5, Case 49). On the other hand, whenever a neurogenic MUAP recruitment pattern is observed and the MUAPs are increased in size, it can be concluded that axon disruption with subsequent reinnervation via collateral sprouting has occurred. When denervation proceeds slowly, reinnervation is able to keep pace with it. In this setting, the motor response may be normal despite severe motor axon loss. This is also true for the strength assessment clinically. Both the motor response and the strength of a muscle reflect the number of functioning muscle fibers, not the number of functioning nerve fibers. Thus, as long as muscle fiber reinnervation keeps pace with muscle fiber denervation, the motor response and the contractile force of the muscle remain normal. Fortunately, reinnervation via collateral sprouting is readily

Chapter 9: The NCS Manifestations of Various Pathologies

recognized during the needle EMG examination (long-duration MUAPs). This is one reason why a needle EMG study of the abductor pollicis brevis muscle should be performed with carpal tunnel syndrome, even when the amplitude is normal. There may be significant motor axon loss not visible on the motor NCS but readily apparent on the needle EMG study. One of the major advantages of motor NCS is its resistance to physiological temporal dispersion. This allows the motor NCS to assess long segments of the nerve trunk and identify areas of focal demyelination. In addition, where the segment of nerve under study is superficial, the stimulator can be “inched” (usually in increments of 1–2 cm) along the nerve, looking for sudden changes in the morphology of the waveform. In this manner, the focus of demyelination is localized. As discussed earlier, with nonuniform DMCS, the negative AUC value is relatively preserved, whereas it is diminished with DMCB.

Axon Loss Pathology and Conduction Failure Pathophysiology Conduction Failure When the axon is disrupted, the term axon loss is applied. Unlike focal demyelination, which produces changes confined to the lesion site, axon disruption, in addition to the changes occurring at the lesion site, triggers changes distant to the lesion site as well. Because the axon is a cytoplasmic extension of the cell body, its disruption separates the axon distal to the lesion (the distal segment or stump) from the proximal axon segment and cell body. Because the distal axon segment is no longer nourished, it undergoes degeneration. Distal axon degeneration, which was first described in the mid-1800s by Waller, is termed Wallerian degeneration and consists of many steps (Waller, 1850). The finer details of this process have been recently reviewed (Rotshenker, 2015). Wallerian degeneration involves the distal portions of the disrupted nerve fibers. Within the first 24 hours, macrophages accumulate at the lesion site, and the Schwann cells within the distal axon segments detach themselves from their myelin sheaths. The distal axon segments and the detached myelin sheaths then undergo degeneration. Macrophages are recruited from the circulation and enter the lesion site through ruptured vasculature and by diapedesis (transmigration

through unruptured vasculature). The recruited macrophages and the resident Schwann cells clear the degenerated axons and myelin sheaths through internalization and metabolic degradation. The time required for axon clearance is proportional to the volume of the axon (i.e., to the diameter and length of the distal segment). Interestingly, the mechanism of injury contributes to the onset site and direction of axon breakdown. For example, with transection injuries, axon breakdown starts at the proximal aspect of the distal segment and proceeds distally, whereas with crush injuries, it starts distally and moves proximally (Rotshenker, 2015). The resident Schwann cells proliferate and reorganize themselves into Bunger bands. Fibroblasts also proliferate. The Schwann cells and fibroblasts of the distal nerve segment promote regeneration through the release of neurotrophic factors that interact with the advancing axons from the proximal axon segment. Two cytokines, interleukin-6 (produced by fibroblasts and recruited macrophages) and LIF (produced by fibroblasts and Schwann cells), promote axon advancement. Because Wallerian degeneration does not occur instantaneously, the distal stump remains capable of conducting action potentials for several days. This has an effect on the motor NCS. With stimulation above the lesion, APs are generated in all of the motor axons of the nerve trunk. However, only those APs propagating along the unaffected motor axons will be able to traverse the lesion site to reach the recording electrodes. Thus, the CMAP recorded with stimulation above the lesion is diminished in size in proportion to the number of affected axons. With stimulation below the lesion, APs are elicited in the unaffected and affected (but not yet degenerated) motor axons. Hence, the motor response recorded with stimulation below the lesion is of normal size, whereas the motor response recorded with stimulation above the lesion is of reduced size (similar to the pattern observed with DMCB). Around day 3, some of the affected motor axons of the distal stump stop conducting, and as a result, the size discrepancy decreases. Between days 3 and 7, progressively more and more motor axons become incapable of conducting. By day 7, the distal stumps of all of the severed motor axons are inexcitable, and the distal and proximal motor responses are equivalent. At this point, the underlying pathophysiology is apparent (conduction failure due to axon loss). Thus, the conduction block pattern associated with axon loss is transient and no longer present

131

Section 2: Nerve Conduction Studies

beyond day 7. In this textbook, this phenomenon will be referred to as transient axonal conduction block. Importantly, unless the degree of axon loss is fairly severe, the motor response measurements that reflect AP propagation speed (latency; conduction velocity) are usually normal. Even when the lesion is severe enough to affect the latency and conduction velocity values, the effects are small in comparison to the effects on the amplitude and negative AUC values. This is true because the latency and conduction velocity values only reflect the fastest conducting fibers contributing to the recorded response, whereas the amplitude and negative AUC values reflect all of the contributing motor fibers. Consequently, as long as a few of the fastest fibers are unaffected, the latency and conduction velocity values remain normal. For this reason, the latency and conduction velocity values are insensitive in the identification of axon loss. In a study of ALS patients, measurements of AP propagation speed (e.g., conduction velocity) remained normal until severe motor response amplitude decrement was present (Lambert and Mulder, 1957). ALS patients ideally address this issue because motor neuron loss produces pure motor axon loss. With myopathies, there is a loss of muscle fibers and, hence, of muscle fiber APs. As a result, the MUAPs generated are smaller in amplitude and shorter in duration. Thus, the smaller MUAPs result in smaller CMAPs. However, because most myopathies affect proximally located muscles, the routine motor NCS typically are normal (e.g., median, recording thenar eminence; ulnar, recording hypothenar eminence). When proximal motor NCS are added, such as the suprascapular (recording infraspinatus), the axillary (recording deltoid), and the musculocutaneous (recording biceps), the yield increases. However, because myopathies are typically symmetric, side-to-side comparisons do not identify relative abnormalities. Thus, the severity must be greater for the motor NCS to be abnormal.

The Major Advantages and Disadvantages of the Motor NCS The major advantages of the motor NCS include: (1) the magnification effect provided by the innervation ratio (generates much larger responses that are easier to acquire and less susceptible to trivial trauma); (2) their relative ease of performance (the belly-tendon method can be applied to most muscles);

132

(3) their ability to identify the underlying pathology and pathophysiology (especially their ability to screen long segments of nerve fibers for demyelinating conduction block); (4) their ability to differentiate acquired and hereditary demyelination; (5) their ability to assist in the identification of malingering and hysteria/conversion; and (6) their ability to estimate the severity of the lesion. The motor response reflects all of the muscle fibers capable of generating an AP. Therefore, because the innervation ratio of a muscle is constant, the percentage decrement of the muscle fiber APs is equal to the percentage decrement of the MUAPs (i.e., the percentage decrement of the motor axons composing the nerve). Thus, by comparing the motor response of the symptomatic limb with the same motor response from the asymptomatic limb, the percentage decrement estimates the percentage loss of motor axons. There are two shortcomings associated with the motor NCS. First, low-amplitude motor responses are nonspecific – they are associated with neuropathies (from the cell body of the lower motor neuron to its terminal branches), neuromuscular junction transmission disorders (e.g., Lambert-Eaton myasthenic syndrome), and myopathies (especially distal myopathies). However, because motor NCS are not performed in isolation, this shortcoming is overcome by the concomitant performance of the sensory NCS and the needle EMG examination. A second shortcoming, and the most important one for the EDX medicine provider to understand, is that reinnervation of denervated muscle fibers by collateral sprouting (discussed in detail in Chapter 16) results in normalization of the motor response, which, in turn, causes the severity of the lesion to be underestimated. Indeed, in the setting of a slowly progressive lesion, which allows reinnervation to keep pace with denervation, the motor response and clinical strength may remain essentially normal despite significant motor axon (or motor neuron) loss. This is also true when a patient is studied following reinnervation via collateral sprouting (see Figure 9.9). To further illustrate these relationships, the following hypothetical example is offered. When a patient with a musculocutaneous neuropathy related to a knife injury in the distal axilla is studied on day 11 by motor NCS and the musculocutaneous motor response, recording from the biceps muscle, is 2 mV, compared to 5 mV on the contralateral (normal) side, it can be estimated that approximately 60% of the

Chapter 9: The NCS Manifestations of Various Pathologies

10

10

10

CMAP = 30 MFAPs = c/l side

10

10

10

CMAP = 20 MFAPs = 2/3 of c/l side (30) 1/3 of nerve disrupted 15

15

Normal CMAP (30 MFAPs = c/l side) Figure 9.9 Motor response normalization through collateral sprouting. Top panel: Three motor neurons compose a motor nerve, each of which innervates 10 muscle fibers (i.e., their innervation ratio is 10). A maximal motor response (CMAP) of this hypothetical nerve will be composed of 30 muscle fiber action potentials (MFAPs) and will be similar to the CMAP recorded from the contralateral side. Middle panel: Following axon disruption and Wallerian degeneration, the maximal motor response only reflects 20 MFAPs. Thus, when the patient is studied in the acute or subacute phase, prior to reinnervation by collateral sprouting, the severity of the lesion can be accurately estimated (one-third of the motor axons are disrupted). Bottom panel: Following reinnervation via collateral sprouting, the maximal motor response returns to normal because the CMAP reflects the number of innervated muscle fibers (i.e., the number of MFAPs), not the number of functioning motor axons. Thus, when the patient is studied after collateral sprouting, the severity of the lesion is underestimated or even missed altogether (as in this hypothetical case).

muscle fibers composing the biceps are no longer contributing to the motor response. Clinically, the patient would only generate 40% of the contractile force generated on the contralateral side. Because

the innervation ratio (muscle fibers per motor axon) is constant, the nerve lesion involves approximately 60% of the motor axons innervating the biceps muscle. Thus, in this example, it can be concluded that the musculocutaneous neuropathy is axon loss in nature and involves 60% of the motor axons to the biceps muscle. If the patient is restudied in 3 months and the motor response is then 4 mV, it could be concluded that only 20% of the muscle fibers remain without innervation. Thus, the unaffected motor axons (40%) now innervate 80% of the muscle fibers. Now suppose this patient had presented to the EMG laboratory at the 3-month mark; the initial motor response would suggest a 20% lesion rather than a 60% lesion. In other words, following reinnervation via collateral sprouting, the motor response underestimates the degree of motor axon loss. In reality, the needle EMG study of the biceps would have identified that the lesion was indeed much more severe than 20% because it would show long-duration MUAPs and a neurogenic pattern of MUAP recruitment; the latter typically is not observed unless at least 50% of the motor axons composing the nerve are dysfunctional (see Chapter 14). Finally, suppose that rather than an acute process, the patient had a slowly progressive disorder causing the musculocutaneous motor axon loss. Because reinnervation would keep pace with denervation, he might not ever be aware of the motor axon loss.

The Sensory NCS Manifestations of Pathology and Pathophysiology Unlike motor axons, which innervate multiple muscle fibers, sensory axons innervate a single receptor. Consequently, unlike what is observed among motor axons, there is no terminal axonal arborization and no neuromuscular junctions. The cell bodies of origin of the sensory axons are located within the dorsal root ganglia (DRG) and are termed DRG cells or sensory neurons. Unlike motor neurons, which reside within the intraspinal canal (the anterior horn of the spinal cord), the sensory neurons reside within the intervertebral foramina of the spinal column. Unlike motor neurons, which only give off a single peripherally directed axon, sensory neurons give off two axons, one that is peripherally directed (toward the receptor) and one that is centrally directed (enters the spinal cord). Some authors define the peripherally directed axon as a long dendritic process that extends from the

133

Section 2: Nerve Conduction Studies

peripheral sensory receptor to the cell body of the sensory neuron within the DRG. Like motor axons, when a sensory axon is disrupted, Wallerian degeneration occurs from the disruption site, distally (i.e., away from the cell body), whether it involves the peripherally directed axon or the centrally directed axon. Likewise, the segment distal to the disruption site is termed the distal stump. Similar to motor neurons, when the cell body is destroyed, both processes undergo Wallerian degeneration. Regarding the latter situation, because the cell body is lost, nerve fiber regeneration cannot occur. Like the motor NCS, the sensory NCS only assess the larger, more heavily myelinated axons. Thus, they assess the peripheral axons of sensory neurons whose centrally directed axons compose the posterior column. The latter axons synapse in the gracile nucleus (lower extremity) and in the cuneate nucleus (upper extremity). As previously stated, the centrally directed axons are not assessed by the sensory NCS. Thus, sensory NCS testing does not detect preganglionic lesions (e.g., radiculopathies). Indeed, a radiculopathy that involves solely the sensory axons of the root is not detectable by EDX testing at all. Conversely, because sensory NCS assess the cell body and its peripherally directed axon, sensory NCS are able to detect ganglionic lesions (sensory neuronopathies) and postganglionic lesions (plexopathies and neuropathies). Thus, the sensory NCS have localizing utility – the presence of a sensory response abnormality supports a ganglionic or postganglionic lesion and argues against a preganglionic lesion (exceptions occur, such as when the DRG is located within the intraspinal canal or an extruded disk fragment affects the DRG through migration into its intervertebral foramen). Thus, in general, whenever an individual is encountered who has significant loss of large-fiber sensory modalities in a body region that is assessable by sensory NCS and the recorded sensory response is normal, the responsible lesion is preganglionic. With radiculopathies involving the sensory and motor axons of the root, Wallerian degeneration occurs along the distal segments of both. For the motor axons, this is toward the muscle fibers, whereas for the sensory axons, this is toward the spinal cord. Radiculopathies may be detectable on motor NCS (when fairly severe and affecting muscles from which motor responses can be recorded) and on needle EMG (especially when the affected muscles are

134

studied during the 21 to 35-day period when fibrillation potentials are the densest or when the lesion is severe enough to generate appreciable MUAP changes related to reinnervation by collateral sprouting). Like the motor NCS, the sensory NCS assess for focal demyelination between the stimulating electrodes and the recording electrodes. Unlike the motor NCS, which screen for axon loss from the motor neurons, proximally, to the distal ends of the motor axons, distally, the sensory NCS do not assess the sensory axon segments distal to the distalmost pair of surface electrodes (i.e., the recording electrodes with antidromic sensory NCS; the stimulating electrodes with orthodromic sensory NCS). It is important to understand the concept of sensory domains. Although the sensory receptors of the body are constant, the sensory axons undergo rearrangements as they advance peripherally. As a result, lesions at different levels of the PNS affect different groups of sensory axons and, thus, produce different distributions of sensory abnormalities. It is the distribution of the sensory abnormalities that contributes to lesion localization, clinically. The cutaneous domain (sensory domain) of any PNS element (e.g., root, trunk, cord, branch, nerve trunk) is defined as the cutaneous territory innervated by the sensory axons contained within it. The sensory domains of the roots have special names (dermatomes) that reflect the spinal cord segment from which they are derived. However, as the sensory axons advance peripherally and undergo rearrangements, this segmental relationship is lost. Thus, for lesions distal to the root level, the term cutaneous domain or sensory domain is more accurate. These sensory domains can be calculated by adding the sensory domains of PNS elements that join an element or by subtracting the sensory domain of a departing PNS element. For example, the posterior cord is formed by the radial and axillary nerves. Thus, its sensory domain includes the sensory domain of the axillary nerve (the superior lateral brachial cutaneous nerve, which is derived from the axillary nerve), plus the sensory domain of the radial nerve (the sensory domain of the radial nerve includes the sensory domains of its sensory branches, i.e., the inferior lateral brachial cutaneous nerve, posterior brachial cutaneous nerve, posterior antebrachial cutaneous nerve, and the superficial radial nerve).

Chapter 9: The NCS Manifestations of Various Pathologies

Sensory NCS offer a number of advantages, including their ability to identify disorders restricted to sensory PNS (sensory neuronopathies, sensory polyneuropathies, sensory mononeuropathies), their greater sensitivity to axon disruption, their lesser degree of recovery (generates a greater ability to identify remote lesions), their ability to differentiate preganglionic lesions from ganglionic and postganglionic lesions, their ability to localize lesions to different regions of the brachial plexus, and their contribution to defining the onset time of a lesion (Wallerian degeneration appears around day 7 and is complete around day 11). The major disadvantages of sensory NCS testing include their susceptibility to minor injury, their susceptibility to intervening tissue (e.g., adipose; edema), their decrement and occasional loss with normal aging, their inability to assess the sensory nerve fiber segments distal to the electrodes, and the much greater technical demands required in their procurement (i.e., they are unforgiving). Unlike motor responses, which give a semiquantitative estimate of the percentage of motor axons affected by a lesion, because of their small size, sensory responses overestimate the severity of a lesion.

The Timing of NCS Manifestations The NCS manifestations of the various pathologies and pathophysiologies do not occur instantaneously. When sensory and motor axons are disrupted, Wallerian degeneration occurs distal to the site of the disruption. As previously discussed, this is a multistep process that occurs over a number of days. Prior to its completion, the distal stump remains excitable and, thus, is able to generate and propagate APs. In general, the process of motor and sensory response decrement related to Wallerian degeneration occurs over an approximate 4-day period. For motor axons, the motor response (CMAP) begins to decline in amplitude around day 3 and is typically complete by day 7, whereas for sensory axons, the sensory response (SNAP) decrement begins around day 5 and is typically complete by day 10 (Wilbourn, 1983) (see Figure 9.10). The motor response is lost before the sensory response because neuromuscular transmission failure precedes nerve conduction failure (Gilliatt and Hjorth, 1972). Thus, because

100 NCS response amplitude (% of normal)

The Major Advantages and Disadvantages of the Sensory NCS

80 60 CMAP

SNAP

40 20

1 2 3 4 5 6 7 8 9 10 11 12 Days after injury Figure 9.10 The progressive decline of the motor and sensory response amplitudes in the setting of a complete axon loss lesion of a mixed nerve with stimulation distal to the lesion. Initially, these responses are normal because degeneration does not occur instantaneously. Because the neuromuscular junctions degenerate earlier than the axons and because the motor responses reflect neuromuscular junction transmission, the amplitude of the motor response begins to decrease several days before the sensory response amplitude.

the motor responses include NMJ conduction (in addition to nerve and muscle conduction), they are lost prior to the sensory responses, which solely reflect nerve conduction. Thus, except in certain situations, we do not perform motor NCS before day 7 or sensory NCS before day 11. The potential for mislocalization is greatest around day 5–7, when the CMAPs are appreciably affected and the SNAPs are either minimally affected or not yet affected. There are three explanations for CMAP involvement without SNAP involvement when these sensory and motor NCS are assessing the same PNS elements: (1) an intraspinal canal lesion (e.g., anterior horn cell disease; radiculopathy), (2) a distal lesion located beyond the exit site of the cutaneous sensory branch (e.g., a musculocutaneous neuropathy located distal to the exit site of the lateral antebrachial cutaneous nerve branch), and (3) incomplete SNAP degeneration. This pattern also is seen with NMJ disorders and myopathies, but these are more generalized processes and would not be in the differential diagnosis of a focal nerve lesion. When axon loss involves a mixed nerve (e.g., the musculocutaneous nerve), the sensory response from that nerve (e.g., LABC) is much more likely to be affected than is the motor response from that nerve (e.g., musculocutaneous nerve, recording biceps). With mild to moderate axon loss lesions involving a mixed

135

Section 2: Nerve Conduction Studies

nerve, the sensory response is typically abnormal, whereas the motor response is frequently normal. This pattern reflects the smaller size (low amplitude; short negative phase duration) and lower synchrony of the sensory responses, as well as the ability of motor axons to reinnervate denervated muscle fibers through collateral sprouting, thereby normalizing

References Ad Hoc Subcommittee of the American Academy of Neurology AIDS Task Force. Research criteria for the diagnosis of chronic inflammatory demyelinating polyneuropathy. Neurology 1991;41:617–618. Bostok H, Sears TA. The internodal axon membrane excitability and continuous conduction in segmental demyelination. J Physiol 1978;274:385–387. Campbell WW. Essentials of electrodiagnostic medicine. Baltimore: Williams & Wilkins, 1999. Cornblath DR, Sumner AJ, Daube J, Gilliatt RW, Brown WF, Parry GJ, Albers JW, Miller RG, Petajan J. Conduction block in clinical practice. Muscle Nerve 1991;14:869–871.

136

the CMAP. Even if collateral sprouting occurred and reinnervated all of the denervated sensory receptors, it would not normalize the SNAP because the number of sensory nerve fiber APs would not increase. With acute lesions generating a 50% drop in CMAP size, the SNAP is usually very low in amplitude or absent.

Gilliatt RW, Hjorth R. Nerve conduction during Wallerian degeneration in the baboon. J Neurol Neurosurg Psychiatry 1972;35:335–341. Hodes R, Larrabee MG, German W. The human electromyogram in response to nerve stimulation and the conduction velocity of motor axons. Arch Neurology Psychiatry 1948;60:340–365. Lambert EH, Mulder DW. Electromyographic studies in amyotrophic lateral sclerosis. Mayo Clin Proc 1957:441–446. Rhee EK, England JD, Sumner AJ. A computer simulation of conduction block: effects produced by actual block versus interphase cancellation. Ann Neurol 1990;28:146–156. Rotshenker S. Traumatic injury to peripheral nerves. In Tubbs RS, Rizk E, Shoja MM, Loukas M, Barbaro N, Spinner RJ, editors,

Nerves and nerve injuries. Amsterdam: Elsevier; 2015:611–628. Uncini A, DiMuzio M, Sabeatelli M, Magi S, Tonali P, Gambi D. Sensitivity and specificity of diagnostic criteria for conduction block in chronic inflammatory demyelinating polyneuropathy. Electroencephalogr Clin Neurophysiol 1993;89:161–169. Waller A. Experiments on the section of the glossopharyngeal and hypoglossal nerves of the frog, observations of the alterations produced thereby in the structure of their primitive fibers. Philos Trans R Soc Lond B Biol Sci 1850;140:423–429. Wilbourn AJ. How can electromyography help you? Postgrad Med 1983;73:187–195. Wilbourn AJ. Common peroneal neuropathy at the fibular head. Muscle Nerve 1986;9:825–836.

Chapter

10

The Utility of NCS for Lesion Localization and Characterization

Introduction The purpose of this chapter is to review how the NCS are utilized in lesion localization. The NCS assess the lower motor neurons (anterior horn cells) and the sensory neurons (dorsal root ganglion cells) of the peripheral neuromuscular system. The cell bodies of the lower motor neurons are located in the anterior horn of the spinal cord, whereas the cell bodies of the sensory neurons are located in the intervertebral foramina of the spinal column. Regarding the motor NCS, the motor axons are stimulated distal to their cell bodies of origin (e.g., supraclavicular fossa, axillary, above-elbow, elbow, below-elbow, and wrist), and the motor response (compound muscle action potential; CMAP) is recorded distally. Regarding the sensory NCS, the sensory axons are stimulated distal to their cell bodies of origin, and the sensory response (sensory nerve action potential; SNAP) is recorded distal to the stimulation (antidromic technique) or proximal to the stimulation (orthodromic technique).

The Effect of Focal Demyelination and Axon Disruption on the NCS Because focal demyelination does not produce distant pathological changes, its effects are local. As a result, it is not possible to identify focal demyelination by NCS testing without passing electrical current through the lesion. Thus, the surface stimulating and recording electrodes must lie on opposite sides of the lesion. Because motor NCS are performed orthodromically, the stimulating electrodes must lie proximal to the focus of demyelination. Because sensory NCS can be performed using orthodromic or antidromic techniques, the stimulating electrodes can lie proximal to the recording electrodes (antidromic technique) or distal to the recording electrodes (orthodromic technique), as long as the stimulating

and recording electrodes are on opposite sides of the focus of demyelination. Because focal axon disruption generates Wallerian degeneration distal to the site of axon disruption, its pathological effects are present not only at the lesion site but also distal to it. As a result, in addition to the lesion site, the distal segments of the disrupted axons also become nonconducting. With postganglionic sensory axon disruption, the peripherally directed sensory axon degenerates from the lesion site to the sensory receptor. With preganglionic sensory axon disruption, the centrally directed sensory axon degenerates from the lesion to where it synapses with the second-order sensory neuron. With motor axon disruption, the motor axon degenerates from the lesion site to the neuromuscular junction (NMJ); the NMJ also degenerates, and the muscle fibers innervated by the disrupted motor axon become denervated. Muscle fiber denervation results in the spontaneous generation of muscle fiber action potentials (fibrillation potentials), which continues until the denervated muscle fiber is either reinnervated or undergoes degeneration (transforms to fibrofatty tissue) approximately 20–24 months after the onset of the denervated state. As a result of Wallerian degeneration, there is loss of conduction at the lesion site and distal to it. Thus, the focal nonconducting site related to the lesion expands to include the distal nonconducting segment. Consequently, unlike focal demyelination, which has only local effects, axon loss has local and distant effects. As a result, the stimulating and recording electrodes do not need to straddle the lesion for it to be detected. As long as the recording electrodes, the stimulating electrodes, or both electrode pairs overlie the nonconducting segment, the underlying pathology will be identifiable. Once Wallerian degeneration is complete, the focus of the original axon disruption cannot be determined. Prior to complete

137

Section 2: Nerve Conduction Studies

Wallerian degeneration, the lesion site is recognizable when the stimulating and recording electrodes straddle the lesion. Regarding the sensory NCS, regardless of whether Wallerian degeneration has started or is complete, whenever the stimulating and recording electrodes lie proximal to the axon loss disruption site, the lesion and its distant effects are not recognizable. In other words, for every sensory or motor NCS performed, the nerve segment between the stimulating and recording electrodes is being assessed for focal demyelination and axon disruption, whereas the nerve segment proximal to the stimulating and recording electrodes is being assessed for axon disruption. For example, when an antidromic ulnar sensory NCS is performed, the segment of ulnar nerve between the stimulating and recording electrodes is being assessed for both focal demyelination and focal axon disruption, whereas the segment of ulnar nerve proximal to the surface electrodes, along with the medial cord, lower trunk, and C8 anterior primary ramus, is being assessed for axon disruption; the C8 DRG is being assessed for neuron loss. Thus, when performing the sensory and motor NCS, it is important for the EDX provider to be aware of the cell bodies of origin of the sensory and motor axons under study. Because the sensory neurons are located in the DRG within the intervertebral foramina of the spinal column, lesions proximal to the DRG (i.e., those involving the centrally directed sensory axon) are not recognized by the routine sensory NCS. Thus, because the sensory NCS do not assess the preganglionic (centrally directed) axons of the sensory neurons, intraspinal canal lesions (e.g., radiculopathies) that disrupt sensory axons produce sensory symptoms, clinically, but do not produce sensory response abnormalities. Consequently, whenever the motor responses from a specific spinal cord level(s) are affected and the sensory responses from that same level(s) are spared, there are only three possibilities: (1) the lesion is located proximally, within the intraspinal canal (e.g., motor neuron disease; radiculopathy); (2) the lesion is located distally, beyond the takeoff sites of any sensory branches; or (3) the lesion is only about 1 week old. When this pattern of sensory and motor responses is observed in a generalized distribution (i.e., generalized motor response abnormalities in the setting of normal sensory responses),

138

the term generalized low motor normal sensory (GLMNS) is applied. This pattern is observed with motor neuron disease (e.g., amyotrophic lateral sclerosis), NMJ transmission disorders (e.g., LambertEaton myasthenic syndrome), and myopathies.

Determining the PNS Elements Assessed by the Various Sensory and Motor NCS Motor NCS Because the spinal cord is segmental in nature, the term myotome (muscle slice) is used to refer to those muscles innervated by a single spinal cord segment. For example, the C6 myotome refers to those muscles receiving innervation via motor axons derived from the C6 spinal cord segment and would include all of the C5,6 muscles and all of the C6,7 muscles. As the motor axons advance beyond the root level (i.e., peripherally), their segmental grouping is lost. This is especially obvious when they traverse one of the somatic plexuses (e.g., the cervical, brachial, or lumbosacral plexus). For example, as the motor axon groups traverse the brachial plexus, the groups join each other, exchange motor axons, and separate, resulting in motor axon group rearrangements (e.g., trunks, divisions, cords, terminal nerves). As a result, the term myotome no longer makes sense. For this reason, it is easier to term the muscles innervated by the motor axons traversing a PNS element as a domain. For example, the muscle domain of the lateral cord includes all muscles innervated by the motor axons that traverse the lateral cord. These muscle domains can be calculated from the known myotomal charts. For example, the muscle domain of the lower trunk equals the muscle domain of the C8 root plus the T1 root. The muscle domain of the medial cord equals the muscle domain of the lower trunk minus the muscle domain of the posterior division of the lower trunk. The calculated domains are provided in the appendix (see Section 6, Appendix 4). Although the majority of these muscles can be studied during the needle EMG examination, they are not all amenable to study by motor NCS. As a result, the CMAP domains of the PNS differ from its muscle domains. The CMAP domains of the brachial plexus are provided in the appendix as well (see Appendix 4). To determine which brachial plexus elements are being studied by a needle EMG study

Chapter 10: The Utility of NCS for Lesion Localization and Characterization

Figure 10.1 The peripheral nervous system elements assessed by the lateral antebrachial cutaneous nerve conduction study.

Figure 10.2 The peripheral nervous system elements assessed by the median sensory nerve conduction study, recording from the thumb.

Figure 10.3 The peripheral nervous system elements assessed by the median sensory nerve conduction study, recording from the index finger.

Figure 10.4 The peripheral nervous system elements assessed by the median sensory nerve conduction study, recording from the middle finger.

Figure 10.5 The peripheral nervous system elements assessed by the superficial radial sensory nerve conduction study.

Figure 10.6 The peripheral nervous system elements assessed by the ulnar sensory nerve conduction study, recording from the small finger.

Figure 10.7 The peripheral nervous system elements assessed by the medial antebrachial cutaneous nerve conduction study.

or by a motor NCS, one needs to know the root and nerve innervation of the muscle under study. For example, when the biceps is being studied by needle EMG or by motor NCS, because the innervating roots are C5 and C6 and the innervating nerve is the musculocutaneous nerve, the C5 and C6 roots, the upper trunk, the lateral cord, the musculocutaneous nerve, and the motor nerve branch to the biceps muscle are being studied. If evidence of motor axon loss is identified (e.g., acute or chronic changes on needle

139

Section 2: Nerve Conduction Studies

EMG or a low-amplitude musculocutaneous motor response), then the anatomic differential diagnosis includes these structures. The pattern of involved muscles then localizes the lesion.

Sensory NCS Due to the greater sensitivity of the sensory NCS in identifying postganglionic axon loss (i.e., plexopathies and neuropathies), the sensory NCS are especially helpful in the localization of postganglionic lesions. In a similar manner to the motor axons, when the DRG of origin of the sensory axons contained within a particular nerve is known, the structures studied by the sensory NCS being performed can be determined. These pathways have been previously worked out (see Figures 10.1 through 10.7) (Ferrante and Wilbourn, 1995). The SNAP domains of the brachial plexus are provided in the appendix (see Appendix 4). By combining the sensory NCS, motor NCS, and needle EMG studies, the localization of the lesion can be precisely defined. This is much more readily accomplished in the setting of PNS lesions involving the C6 through T1 segments, because there are

References Ferrante MA. Brachial plexopathies. Continuum 2014;20:1323–1342.

140

numerous sensory NCS available. Their availability allows preganglionic lesions to be differentiated from postganglionic and ganglionic lesions. Also, they allow lesion localization to specific elements of the brachial plexus (e.g., to the lower trunk or to the lateral cord) (Ferrante, 2014). In fact, all of the brachial elements are assessable by sensory NCS except for the C5 DRG and APR. Because the routine screening sensory and motor NCS primarily assess the C8- and T1-derived sensory and motor axons, whenever a lesion is suspected to involve C5-, C6-, or C7-derived elements, additional studies are required. Unfortunately, because the number of reliable sensory NCS assessing the PNS elements of the lower extremity are limited to the sural nerve, superficial nerve, and the medial and plantar nerves, differentiating preganglionic and postganglionic lesions and localizing lesions to specific portions of the lumbosacral plexus remains challenging and, thus, relies more on the motor NCS and especially the needle EMG examination. Numerous examples of our approach to lesion localization are provided in the 50 EDX exercises at the end of this textbook (see Section 5).

Ferrante MA, Wilbourn AJ. The utility of various sensory nerve conduction responses in assessing brachial plexopathies. Muscle Nerve 1995;18:1–11.

Chapter

11

Late Responses and Blink Reflexes

Introduction The late responses (H reflexes and F waves) are an additional component of the EDX examination. As their name implies, they represent motor responses (CMAPs) that, in comparison to the M wave, appear much later. Their delayed appearance is related to the much longer distance that the APs contributing to them must travel before reaching the recording electrodes (i.e., from the site of stimulation to the spinal cord and then from the spinal cord to the recording electrodes). Like M waves, H waves and F waves are compound electrical potentials composed of muscle fiber APs. Unlike M waves, they only represent a subpopulation of the motor nerve fibers within the studied nerve rather than the entire population. When APs are generated physiologically, they propagate in only one direction – centrifugally for motor nerve fibers and centripetally for sensory nerve fibers – because the inward Na+ current following membrane depolarization encounters a membrane region in its absolute refractory period on one side of the depolarized segment but not on the other side (see Chapter 3). When APs are generated through transcutaneous stimulation, however, neither side of the depolarized segment is in its absolute refractory period. For this reason, APs propagate bidirectionally when they are stimulator induced. Hence, bidirectionally propagating APs are always generated during the routine sensory and motor NCS. However, because routine NCS are not designed to record them, they are not observed. With late response studies, however, a longer recording period is utilized so that two motor responses are recorded: an early motor response (the M wave) and a later-appearing response (either the H wave or the F wave). Because of the required design modifications of late response studies, most EMG machines employ separate software packages for these studies.

Because late responses are CMAPs, the surface recording electrodes are placed over the muscle of interest using the belly-tendon method. The specific details of the H wave and F wave studies are discussed later here. These late responses (F waves and H waves) allow nerve segments of the peripheral nervous system located proximal to the segments examined by the routine NCS to be assessed. Because there is no way to accurately measure the distance of the nerve segments under study, accurate conduction velocities cannot be calculated for late responses. For this reason, latency values are utilized to estimate the AP propagation speed. Because the time required for the APs to propagate first centrally to the spinal cord and then peripherally to the recording electrodes varies with the length of the limb, these values are limb length–dependent and, consequently, heightdependent (because taller individuals typically have longer limbs). As a result, once collected, the latency values must be compared to the height-based normal control values used by the EMG laboratory. In addition to comparison to normal control values, to identify absolute abnormalities, and like the routine NCS, the recorded values are also compared to the values collected from the contralateral side (assuming the contralateral side is asymptomatic) to assess for relative abnormalities. Whether or not late responses are routinely performed depends on the clinical indication for the EDX study and the preference of the EDX provider. For example, in our EMG laboratories, we consider H reflex studies to be especially useful for identifying generalized polyneuropathies (especially sensory polyneuropathies) and S1 nerve root disease and, consequently, perform them routinely, whereas we consider F waves to provide information superfluous to that already generated by the routine NCS and, for this reason, rarely perform them.

141

Section 2: Nerve Conduction Studies

H Reflexes

Technique

Introduction

H reflex studies can be performed in more than one manner. In our EMG laboratories, the patient is positioned supine, and the E1 electrode is placed on the gastrocnemius-soleus muscle complex, one fingerbreadth medial to the medial edge of the tibia, at the level where the girth of the calf muscle complex is greatest. The E2 electrode is situated at the anterior aspect of the ankle. With the knee slightly flexed, the handheld stimulator is positioned in the popliteal fossa, with the cathode proximal to the anode and overlying the posterior tibial nerve. The patient is relaxed and the gastrocnemius-soleus muscle complex is at rest (muscle contraction enhances the H wave) (Fisher, 1992) (see Figure 11.1). Another technique for performing the H reflex positions the patient prone, with the E1 electrode overlying the soleus muscle, just inferior to the gastrocnemius muscle heads, and the E2 electrode overlying the Achilles tendon. Stimulation is provided in the center of the popliteal fossa, with the cathode proximal to the anode and overlying the posterior tibial nerve. Due to H-wave habituation, the amplitude of the H wave may decrease when stimuli are delivered at rates exceeding 0.5 Hz. Although a needle recording electrode can be utilized in place of the E1 surface recording electrode to better define the H wave latency, it does not permit accurate assessment of the H wave amplitude, which, in the experience of the author, is much more sensitive than the latency in the diagnosis of S1 radiculopathies and mild polyneuropathies. Unlike nerve stimulation during routine motor NCS, tibial nerve stimulation differs with H wave

The H reflex study was named after Hoffman, the individual who first described it in 1918 recording from the gastrocnemius-soleus complex (Hoffman, 1918). This study primarily reflects the S1 nerve root. Because it is a monosynaptic reflex, the term H reflex will be used when referring to the technique itself, whereas the term H wave will be used when referring to the late response elicited by this technique. Simply put, the H reflex study generates two tibial motor responses. These two motor responses arrive at different times – the one with the smaller onset latency value is derived from the distally propagating motor nerve fiber APs and is termed the M wave, whereas the one with the larger onset latency value is derived from the proximally propagating sensory nerve fiber APs and is termed the H wave (discussed in detail further on). Unlike motor NCS, which can be performed on any nerve-muscle combination to which the stimulating and recording electrodes have access, H reflex studies are much more limited. Most commonly, H reflexes are recorded from the gastrocnemius-soleus muscle complex following tibial nerve stimulation. Much less frequently, H reflex testing is performed on the median nerve while recording from homologous forearm flexors, such as the flexor carpi radialis muscle (Deschuytere et al., 1976). They are also frequently obtainable from the quadriceps muscle complex and occasionally from plantar foot muscles (Fisher, 1992). Prior to the age of 2 years, they are much more widely distributed because central nervous system maturation is not yet complete (Fisher, 1992).

Figure 11.1 The H-reflex technique. With the patient supine, the E1 electrode is placed just medial to the tibia at the level where the girth of the calf musculature is greatest. The E2 electrode is placed at the dorsal aspect of the ankle. Stimulation is applied in the center of the popliteal fossa, overlying the tibial nerve, with the cathode oriented proximal to the anode.

142

Chapter 11: Late Responses and Blink Reflexes

studies. First, the stimulator is oriented in reverse, with the cathode proximal to the anode. Second, the initial stimulus current settings differ. Instead of beginning with a short-duration, low-intensity current pulse, the H wave study begins with a longduration (1.0 msec), low-intensity current pulse. The long-duration, low-intensity current pulse combination favors the activation of larger-diameter, more heavily myelinated sensory nerve fibers over motor nerve fibers (Panizza, 1989). Third and most importantly, with H wave studies, low-intensity stimuli are utilized to collect the maximum H wave, whereas supramaximal stimulation is used to collect the maximum M wave. In general, the H wave appears with stimulus strengths below the threshold for the appearance of the M wave and the H wave maximizes far below the threshold for the maximal M wave. Because submaximal stimulus strengths are used to collect the H waves, they are seen on the monitor prior to the appearance of the M waves. Because they are late responses, they appear to the right of where the M wave will eventually appear. When the tibial sensory fibers are initially activated, bidirectionally propagating APs are generated. The distally propagating sensory nerve fiber APs do not contribute to either the H wave or the M wave, because the recording electrodes overlie the gastrocnemius-soleus muscle complex and, hence, do not record them. The proximally propagating sensory nerve fiber APs reach the cell bodies of the sensory neurons of the S1 dorsal root ganglion and then propagate along their central processes into the parenchyma of the S1 segment of the spinal cord, where they synapse on lower motor neurons in that segment and trigger their depolarization, thereby resulting in a distally propagating volley of APs along the motor nerve fibers. These motor nerve fiber APs generate muscle fiber APs in the muscle fibers of the motor units to which they belong. The muscle fiber APs reach the surface recording electrodes and summate to form the H-wave, typically at about 30 msec. Thus, as previously stated, the H wave is a compound electrical potential that is composed of muscle fiber APs. Once the H wave first appears, it is important to increase the stimulus current intensity in extremely small increments until the H wave is maximized. Large incremental adjustments may result in diminution of the H wave amplitude value due to collisions (discussed later here), although other inhibitory

processes likely contribute to H wave diminution (Fisher, 1992). Once a maximal H wave is collected, the stimulus intensity is incrementally increased further. The increases in stimulus current intensity cause some of the tibial motor nerve fibers to also be stimulated. Consequently, at this point, the stimulation current generates bidirectionally propagating APs along both the sensory and the motor nerve fibers of the tibial nerve. The outcomes of the distally and proximally propagating sensory nerve fiber APs have already been discussed. The distally propagating motor nerve fiber APs generate the M wave (i.e., the tibial CMAP), which has a latency of around 5–6 msec. The proximally propagating motor nerve fiber APs collide with the distally propagating motor nerve fiber APs generated by the lower motor neurons (i.e., the ones forming the H wave). These collisions diminish the amplitude of the H wave (Magladery and McDougal, 1950). Thus, at this point, there are two CMAPs on the screen, an M wave and an H wave. As the stimulus intensity is increased further, the H wave decreases in size (due to an increasing number of AP collisions) and eventually disappears, whereas the M wave progressively increases in size and eventually maximizes, at which point the study is complete (see Figure 11.2). Among the EDX studies, the H wave is the only one that assesses the preganglionic processes of the sensory neurons.

Technical Errors Of the various NCS, H wave testing is the most challenging. However, once the anatomy and physiology underlying H wave studies are understood (discussed earlier), technical errors are much less likely. To ensure that a maximal H wave is recorded, it is important to begin with a very-low-stimulus current intensity and increase it in very small increments to avoid collision-related H wave amplitude decrements. The maximal value is the one achieved when stimulation maximizes lower motor neuron depolarization without the occurrence of any collisions to diminish it. It is easy to overshoot this point (i.e., to go from a submaximal H wave [because a submaximal number of lower motor neurons are contributing to the H wave] to a larger, but still submaximal, H wave [because the stimulus current activates the maximum number of lower motor neurons but also activates some of the tibial motor nerve fibers, resulting in collision-related loss]). Thus,

143

Section 2: Nerve Conduction Studies

H waves may also be unelicitable among individuals with a large body habitus or significant lower extremity girth (e.g., edema; obesity). In these settings, there is a large distance between the stimulation site and the tibial nerve fibers, limiting the stimulation current from activating the tibial nerve fibers. When the ability of the stimulator to stimulate the tibial nerve fibers is inadequate, the M wave (and sometimes even the H wave) will be submaximal or even absent. This is a limit of the EMG machine and does not indicate pathology. When this happens, a comment should be included in the report indicating that the submaximal or absent response is not pathological, but instead is related to the limitations of the EMG machine and the intervening tissue. Because fat storage is uncommon along the sole of the foot, a medial plantar response can be added when the lower extremity sensory responses are absent and obesity is felt to be responsible. Its presence argues against a length-dependent polyneuropathy and supports obesity as the underlying cause. Age must also be considered when performing medial plantar responses because they may be unelicitable among normal individuals over the age of 45 years. Figure 11.2 The tibial H reflex. As the stimulus current intensity is increased, an H wave eventually appears. With further stimulus current increments, the H wave maximizes and then begins to decrease in size as collisions occur between the proximally propagating motor nerve fiber action potentials and the sensory nerve fiber action potential-induced distally propagating motor nerve fiber action potentials. Eventually, the H wave is completely annihilated. With still further increases in stimulus current intensity, the M wave is maximized.

to avoid skipping over the true peak value, once a maximum H wave is collected, it is useful to lower the stimulus intensity slightly and determine if the response gets larger. Another mistake occurs when an F wave is mistakenly thought to be an H wave. Differentiation between H waves and F waves is straightforward. With H waves, the latency should be constant, the morphology of the waveform should be uniform as it increases in size, and the amplitude should maximize at submaximal intensities and then diminish and disappear as the stimulus intensity is increased further. It should not display the characteristics of an F wave: variations in the waveform conformation, variations in the onset latency, not increasing in size beyond 5–10% of the size of the M wave, and persistence despite supramaximal stimulation. Not infrequently, the monitor displays both wave types simultaneously. At supramaximal stimulation, however, only the F wave should remain.

144

Amplitude and Latency Measurements Because the H wave is a CMAP, like other CMAPs, its amplitude value is a much better indicator of axon loss than is its latency value because the latency value only reflects the fastest conducting motor nerve fibers activated. For this reason, incomplete lesions do not affect the latency value unless all of the fastest conducting fibers are involved. In addition, because of the great distance traversed by the propagating APs composing the H wave, even when a focal demyelinating lesion causes extreme slowing of every nerve fiber within the nerve, the recorded latency might still be normal, because the conduction time across the lesion is so small in comparison to the conduction time along the much longer normal segments assessed by this study (i.e., shorter distances are more sensitive for identifying focal demyelinating conduction slowing than are longer distances; see Chapter 9). For these reasons, in our EMG laboratories, as for any CMAP, we record the latency and amplitude values of both the M wave and the H wave. In our experience, although the latency value may occasionally be abnormal in isolation, we have found the amplitude value to be the more sensitive measurement. Regarding the H wave amplitude, in our EMG laboratory, the lower limit of normal is 1 mV,

Chapter 11: Late Responses and Blink Reflexes

regardless of age. Because the H wave habituates in response to stimulation frequencies exceeding 0.5 Hz (1 stimulus per 2 seconds) (Olsen and Diamantopoulos, 1967), we wait several seconds between stimuli. However, this value is often much larger in young and middle-aged adults. For this reason, we typically perform H wave studies bilaterally (to look for a relative abnormality), especially when the patient has unilateral symptoms. Although H reflexes are found in many elderly individuals, because H waves diminish with age, they may not be elicitable among some apparently normal individuals over the age of 60 years (Weintraub et al., 1988). Obviously, a unilaterally absent H wave is abnormal at any age. Prior to concluding that an H wave is absent, the patient should be asked to mildly contract the ankle flexors, which will facilitate the response (discussed further on). Facilitation (mild contraction of wrist flexors) is frequently required to elicit a median H wave from the flexor carpi radialis muscle. Although facilitation would not affect the onset latency of the recorded H wave, the amplitude will be increased due to a greater number of activated lower motor neurons. Because it would be challenging to standardize the degree of enhancement applied and because our normal control values were collected using complete relaxation, we do not perform facilitation in our EMG laboratory. Although it has been reported that normal individuals may demonstrate side-to-side H wave amplitude asymmetries of 60% or higher (Fisher, 1992; Jankus, 1994), we nonetheless consider a relative abnormality to be present when the side-to-side amplitude difference exceeds 50%. Regarding the H wave latency, the upper limit of normal is 35 msec (20 msec for the median H wave latency recording flexor carpi radialis), although, as previously stated, this value may be higher than 35 msec in normal individuals who are of above average stature. In this setting, the latency values of the two sides should be compared. In addition to absolute latency value abnormalities, relative abnormalities may also be present. Thus, we perform the study bilaterally to look for side-to-side differences (the upper limit of normal is 1.5 msec) (Fisher, 1992). It is important to realize that the value of the side-to-side difference increases with aging (Falco et al., 1994). Finally, the size of the H wave is a reflection of the excitability of the lower motor neuron pool. The ratio of the H wave amplitude to the M wave amplitude (the H/M ratio) reflects the excitability of the motor

Figure 11.3 Abnormally high-amplitude H wave in a patient with spinal cord compression (8.83/9.75 = 0.9, which is > 0.7 and indicates lower motor neuron hyperexcitability). In this illustration, the M waves and H waves are shown superimposed rather than rastered, as they were in Figure 11.2.

neurons and, in general, is less than 0.7 (Delwaide, 1984). Thus, the H wave can be pathologically high in amplitude in the setting of upper motor neuron disease (e.g., spinal cord compression) (see Figure 11.3). In addition, whenever an H wave can be recorded from skeletal muscles other than the gastrocnemiussoleus complex, the quadriceps, or the flexor carpi radialis (e.g., the APB muscle), an upper motor neuron disorder should be suspected, although H waves are also occasionally recordable from plantar foot muscles (Thomas and Lambert, 1980; Fisher, 1992). Anxiety also affects H wave size, typically in a symmetric manner.

Utility of H Wave Testing The H wave is especially useful in the identification of S1 nerve root disease and generalized polyneuropathies, especially those affecting the sensory nerve fibers (Braddom and Johnson, 1974; D’Amour et al., 1979; Ackil et al., 1981; Aiello et al., 1981). Importantly, once the H wave becomes unelicitable following S1 nerve root disease, it may remain unelicitable despite resolution of the symptoms or successful surgical intervention. Although it is also identifies

145

Section 2: Nerve Conduction Studies

proximal tibial neuropathies, sciatic neuropathies, and plexopathies involving the S1-derived nerve fibers, these disorders are typically recognized during the routine NCS and needle EMG studies, and therefore, H wave testing offers no additional benefit in their recognition. Although the H wave is analogous to the ankle muscle stretch reflex (Katirji et al., 1988; Weintraub et al., 1988), it is not identical. The ankle reflex only assesses the Ia sensory afferent fibers innervating the muscle spindle organs, whereas the H reflex assesses all of the larger-diameter, more heavily myelinated sensory nerve fibers of the tibial nerve, including the Ib Golgi tendon organ afferents. In addition, the degree of synchrony of the propagating sensory nerve fiber APs is greater for the H wave than it is for the ankle reflex, because the distance over which the APs propagate is shorter (knee to spinal cord, as opposed to Achilles tendon to spinal cord). Finally, the H wave studies do not assess the sensory nerve fibers distal to the stimulation site. For these reasons, an H wave may be recordable when the ankle muscle stretch reflex is absent, and conversely, an ankle muscle stretch reflex may be present when the H wave is unelicitable (Weintraub et al., 1988).

F Waves Introduction The F wave is a recurrent motor response that follows the antidromic activation of lower motor neurons that was first described by Magladery and McDougal (1950). They coined the term F wave based on the fact that the technique they described was first performed using foot muscles. Similar to H wave testing, the F wave testing generates two CMAPs, an initial one (the M wave) and a delayed one (the F wave). Unlike the H wave, the F wave is generated with supramaximal stimulation, involves antidromic activation of lower motor neurons, and can be recorded from any muscle on which a motor NCS can be performed. With more proximally situated muscles, however, overlap of the M wave with the F wave may occur. Although collision techniques can be utilized to overcome this problem (Kimura, 1974), they are not employed in routine clinical practice.

Physiology and Technique The technique for generating an F wave is the same as that utilized for the routine motor NCS to generate an

146

M wave, except that the stimulator is oriented with the cathode proximal to the anode, and a longer sweep duration is utilized. As expected for a motor response recording, the surface electrodes are placed on the muscle using the belly-tendon method, and the innervating nerve is stimulated. Because the F wave is fatigable, the stimulation frequency should not exceed 0.5 Hz. Like a routine motor NCS, the stimulus generates bidirectionally propagating action potentials along the sensory and motor nerve fibers of the nerve under study. Because the F wave technique is not designed to record the APs of the stimulated sensory fibers, they do not contribute. Like an H wave, the distally propagating motor nerve fiber APs generate the M wave (the CMAP with the shorter onset latency). The proximally propagating motor nerve fiber APs continue up the nerve, eventually reaching the initial axon segment/axon hillock membrane of their cell bodies of origin. At this point, the internal Na+ current encounters a considerable impedance mismatch because the initial segment of the motor axon is larger in diameter and unmyelinated. For these two reasons, the surface area requiring electrical discharge is much greater. Thus, higher capacitance and greater current leakage, as well as the transmembrane potential at the time of AP arrival, dictate whether the depolarization threshold of the membrane is reached. If it is reached, the incoming AP conducts proximally and reaches the dendritic tree (it cannot propagate distally back down the motor axon because, due to its recent arrival, this segment of membrane is in its absolute refractory period) (Eccles, 1955). From the dendritic tree, an electrotonically propagating AP may be generated, and if it reaches the axon hillock/initial nerve segment region with enough magnitude to depolarize it to its depolarization threshold value, then a distally propagating AP will be generated. Because of these impediments, the overwhelming majority of the lower motor neurons are not reactivated and, thus, do not generate a distally propagating AP. Indeed, only about 1–3 lower motor neurons are reactivated and backfire in this manner (Yates and Brown, 1979; Dumitru, 1995). For this reason, the size of the F wave represents only 1–3 MUAPs and, thus, is typically less than 5% of the size of its associated M wave. The F wave latency value reflects conduction time from the stimulation site to the spinal cord and from the spinal cord to the recording electrodes, plus the backfire time. The backfire time has been stated to be approximately

Chapter 11: Late Responses and Blink Reflexes

1 msec (Eccles, 1955), although this value may be closer to 3 msec (Petajan, 1985). For mathematical reasons, the total conduction time can be considered to be the summation of three separate conduction times: (1) the conduction time from the stimulation site to the spinal cord, (2) the conduction time from the spinal cord back to the stimulation site, and (3) the conduction time from the stimulation site to the muscle, as well as the backfire time. Because the first and second conduction times are identical to each other and because the third conduction time is identical to the conduction time of the M wave, it is possible to estimate the conduction time from the spinal cord to the stimulation site, which is termed the proximal conduction time, as half of the difference between the F wave latency and the M wave latency, minus the backfire time (estimated at 1 msec): Proximal conduction time ¼ ðF wavelatency  M wavelatencyÞ=2  1 Some authors divide this value into the distance to estimate the F wave conduction velocity using the wrist-sternal notch or wrist C7 spinous process as the distance for the upper extremity and the ankle-xiphoid process or ankle T12 spinous process as the distance for the lower extremity. However, similar to the situation with H waves, conduction velocities are best not calculated, because the true distance of the nerve segment is inaccurately depicted by these surface landmarks. The F wave latency, like other latency measurements, only reflects the fastest fibers reaching the recording electrode pair. Because the excitability of the lower motor neuron pool is constantly changing, serial stimulations excite different groups of lower motor neurons. For this reason, the individual F waves differ in their waveform morphologies. In addition, the stimulations may not always excite the larger motor neurons (i.e., those with the fastest conducting axons), although there is some evidence that the larger motor neurons may be favored, possibly through Renshaw cell activation, which inhibit larger motor neurons the least (Fisher, 1992). As a result, the individual F waves vary in their latencies. For unclear reasons, the same group of lower motor neurons will eventually be re-excited, and when this occurs, an F wave with an identical waveform conformation (and hence latency) to that of a previously collected F wave will appear (termed repeater F waves). In our EMG laboratories, we typically collect 8 F waves and report the minimal latency value among the F waves.

Other measurements include mean latency, median latency, the difference between the minimal and maximal latencies (termed, chronodispersion), persistence (the number of F waves obtained divided by the number of stimulations), and the F/M ratio, as well as the amplitude and duration values (Panayatopoulos, 1979; Fisher, 1992; Wilbourn and Ferrante, 1997). Although the mean latency is more accurate than the minimal latency, the added complexity introduced by the required calculations and the increased numbers of F waves required renders this measurement impractical for routine clinical use (Fisher, 1992). This statement is also true for the median latency, which overcomes the non-Gaussian distribution of the minimal latency. Increases in F wave chronodispersion identify disorders producing pathological temporal dispersion (i.e., disorders producing demyelinating conduction slowing) and, because of the longer distances studied by F waves, should be more sensitive than routine NCS. A decrease in F wave persistence occurs when there are less lower motor neurons available for excitation. Normally, the overwhelming majority of stimuli generate an F wave. When an F wave is generated with every stimulus, the lower motor neuron pool may be hyperexcitable (discussed further on). An excellent review discussing the utility of these alternate F wave parameters is available to the interested reader (Fisher, 1992).

Utility Theoretically, F waves should be of considerable utility in the assessment of the nerve fiber segments proximal to the stimulation site (i.e., radiculopathies, plexopathies, and proximal neuropathies). However, based on the underlying pathophysiology of the lesion, there are several issues that render them either insensitive or of limited value. First, if the proximal lesion is axon loss in nature and postganglionic in location (radiculopathies are discussed below), it will have already been identified during the routine NCS because sensory response amplitude measurements are much more sensitive to focal axon loss than are latency measurements. Second, if the proximal lesion is demyelinating conduction block in nature, unless all of the fastest fibers are blocked, the minimal F wave latency will be unaffected. When the lesion is moderate to severe in degree, it is more likely to be identified during the needle EMG study when a neurogenic MUAP firing pattern is noted in a muscle that demonstrated a normal motor response. Third, if the

147

Section 2: Nerve Conduction Studies

proximal lesion is demyelinating conduction slowing in nature, then the involved motor fibers will have delayed latencies. However, if just one of the fastest conducting fibers is spared, the recorded F wave latency will be unaffected. In addition, even when all of the fastest conducting fibers are affected, the resultant delay across the short segment manifesting the pathology will be diluted by the large distance over which the F wave study is being performed and, hence, will likely not be recognized. With regard to their ability to identify radiculopathies, because (1) most skeletal muscles receive innervation via more than one root, unless the lesion involves all of the innervating roots, the F wave latency is unaffected, (2) short segment lesions are diluted by the long distances assessed by F wave studies (discussed above), and (3) F waves are not affected when radiculopathies are restricted to the sensory nerve fibers of the root (the most common kind). In summary, like the onset latencies of other CMAPs, F wave latencies are insensitive to axon loss and demyelinating conduction block pathophysiologies. In addition, due to the dilutional effect described earlier, F wave latencies are insensitive to demyelinating conduction slowing, depending on the degree of slowing and the completeness of the lesion. Finally, because of myotomal overlap, F waves are insensitive to radiculopathies. Even if prolonged, the location of the lesion would not be determinable. Because upper motor neuron disorders result in lower motor neuron hyperexcitability, the recorded F waves may be larger in size, greater in complexity, and demonstrate a higher persistence (i.e., the percentage of stimuli generating an F wave). Reinnervation via collateral sprouting also increases F wave size because the MUAPs composing them are larger. Thus, for all these reasons, we infrequently perform F wave studies in our EMG laboratories. Although studies have shown F waves to be a sensitive measure in polyneuropathies, slowed in amyotrophic lateral sclerosis, abnormal in myotonic dystrophy, and prominently slowed in Guillain-Barre syndrome, these entities are better identified and better characterized by the routine NCS. Thus, we typically only perform F waves when individuals with clinical features suggestive of early Guillain-Barre syndrome are referred. However, as previously stated, even in this setting, the information they provide is unhelpful, because the pathophysiology associated with weakness must be either demyelinating conduction block

148

or axon loss, both of which are much more apparent on the routine NCS. Moreover, demyelinating conduction slowing (i.e., the pathophysiology most likely to produce an F wave conduction delay) does not cause weakness because all of the APs reach their target organs (i.e., the muscle fibers). The number of F waves collected for analysis varies among EMG laboratories. The reason to collect more than one F wave is that it increases the likelihood of activating the fastest conducting fibers (i.e., the ones generating the minimal F wave latency value). Some authors suggest a minimal number of stimuli be delivered or a minimal number of F waves be collected (usually 10–20), whereas others report the minimal latency among the repeater F waves (when a stimulus generates an F wave with a morphology identical to a previous F wave, it is termed a repeater F wave). It has been suggested that 5 F waves suffice for routine clinical purposes (Marra, 1987). In our EMG laboratories, we infrequently perform F wave studies. When we do, we collect 8 F waves and report the minimal latency value among the recorded F waves without the stipulation that they be repeater F waves. In general, the minimal latency should be below 35 msec for upper extremity nerves (with wrist level stimulation) and below 65 msec for lower extremity nerves (with ankle level stimulation). However, it is best to utilize height-based normative data to determine normal versus abnormal and, in the setting of ipsilateral disease, to compare the recorded values to those collected from the contralateral side. In general, a side-to-side difference between F waves collected from homologous nerves exceeding 2 msec for upper extremity or 4 msec for the lower extremity is abnormal. With stimulation in the popliteal fossa, the latencies are just a bit longer than the upper limit of normal for H waves and, hence, may be contaminated by the presence of H waves (typically only at submaximal stimulation levels). Consequently, the opposite issue – H waves contaminated by F waves – is the more frequent problem given that F waves are not typically collected using popliteal fossa stimulation.

A Waves An A wave (axon reflex) is another type of delayed motor response that appears later than the M wave and that can be confused with an F wave. Unlike an F wave, its latency and morphology are constant. The majority of A waves reflect branching of the motor axon at a site more proximal to the stimulus site

Chapter 11: Late Responses and Blink Reflexes

Fullerton and Gilliatt, 1965). Consequently, like an F wave, when stimulation is applied more proximally along the nerve segment, the latency of an A wave decreases. As previously discussed, when motor axons are stimulated, bidirectionally propagating compound electrical potentials are generated along them. The distally propagating potential reaches the muscle under study and generates the M wave, whereas the proximally propagating potential reaches a motor axon branch point. By Kirchoff’s current law, current into the node equals current out of the node. Thus, at the branch point, the current divides into two portions – one portion propagates proximally, toward the spinal cord, and the other portion conducts along the branch, returning to the muscle and generating the A wave. Thus, A waves typically appear earlier than F waves or H waves, their exact latency time being dependent on the level of motor axon branching – the more proximal the branch site, the longer the A wave latency. Other mechanisms that result in A waves include re-excitation of the parent nerve more proximally and activation through ephaptic conduction between two axons (Roth, 1978a, 1978b; Magistratis and Roth, 1985; Falck and Staberg, 1995). As APs propagate along nerve fibers (or muscle fibers), their external currents may interact with the transmembrane potential of an adjacent membrane. When the adjacent fiber is diseased, this interaction might produce ephaptic conduction between the fibers (Granit et al., 1944). It is important to be familiar with A waves so that they are not mistaken for F waves. A waves were not observed among 100 healthy controls, suggesting that their presence might be an early indicator of collateral sprouting related to underlying motor axon pathology, especially in the setting of polyneuropathies (Bischoff et al., 1996).

bilaterally, and their brainstem interconnections. Like the H wave, these three responses reflect a circuit that involves orthodromic sensory conduction and orthodromic motor conduction. Unlike the H wave, neither the R1 response nor the R2 response is monosynaptic. The R1 response involves less synapses (oligosynaptic) than do the R2 responses (polysynaptic). After stimulating the ipsilateral supraorbital nerve, the contralateral supraorbital nerve is stimulated. Although the infraorbital nerve can also be utilized, it less reliably elicits an R1 response. The R2 responses correlate clinically with the bilaterally observed blinks. Similar to late responses, temporally discreet responses are recorded and, consequently, a separate software package is required. Like the F waves, only the latency value is utilized, and thus, only the fastest fibers are assessed. For this reason, blink reflexes have all of the inherent disadvantages associated with latency measurements. In general, facial nerve motor NCS are performed first, to ensure that the facial nerve components of the blink reflexes are normal. With the advent of magnetic resonance imaging, blink reflexes are less frequently performed and, consequently, are less extensively discussed here. The normal response includes three motor responses, an earlier ipsilateral one (termed the R1 response) and two later appearing ones, one from each side (termed the ipsilateral R2 response and the contralateral R2 response). The R1 response has an onset latency of about 10–15 msec, whereas the R2 response has an onset latency of about 30–45 msec, with the ipsilateral R2 response appearing earlier than the contralateral R2 response (see Figure 11.4) (Wilbourn and Ferrante, 1997).

Blink Reflexes Blink reflex studies are typically performed by electrically stimulating the trigeminal nerve ipsilaterally while recording from the facial nerves, bilaterally. More specifically, supraorbital branch of V1 is stimulated at the palpable groove of the supraorbital ridge (located below the medial aspect of the eyebrow) while recording from the orbicularis oculi muscles bilaterally. Consequently, the blink reflex assesses the ipsilateral V1 (ophthalmic division) portion of the fifth cranial nerve, the ipsilateral main sensory nucleus, the facial neurons and facial nerves

Figure 11.4 The blink reflex technique. The cathode of the stimulator (S1) is placed in the supraorbital ridge and the surface recording electrodes (E1 and E2) are placed over the orbicularis oculi as shown). In the illustration, the supraorbital branch of the right trigeminal nerve is stimulated and the ipsilateral and contralateral orbicularis oculi muscles are recorded, after which the left side is stimulated. Each stimulation generates three motor responses: ipsilateral R1, ipsilateral R2, and contralateral R2 (see text for further discussion).

149

Section 2: Nerve Conduction Studies

By performing this study bilaterally, any underlying lesion can be localized to the afferent (trigeminal) or efferent (facial) limb of one of the two sides. For example, with a trigeminal nerve lesion in which the fastest trigeminal nerve fibers are affected, the ipsilateral R1 and both R2 responses are delayed (or absent) with ipsilateral supraorbital nerve stimulation, but with contralateral supraorbital nerve stimulation, the abnormal R2 response becomes normal. In the setting of a facial neuropathy, the R1 response and ipsilateral R2 response are abnormal following stimulation of the ipsilateral supraorbital

References Ackil AA, Shahani BT, Young RR, Rubin NE. Late responses and sural conduction studies. Usefulness in patients with chronic renal failure. Arch Neurol 1981;38:482–485. Aiello I, Rosati G, Serra G, Manca M. The diagnostic value of H-index in S1 root compression. J Neurol Neurosurg Psychiatry 1981;44:171–172. Bischoff C, Stalberg E, Falck B, Puksa L. Significance of A-waves recorded in routine motor nerve conduction studies. Electroencephalogr Clin Neurophysiol 1996;101:528–533. Braddom RI, Johnson EW. Standardization of H reflex and diagnostic use in S1 radiculopathy. Arch Phys Med Rehabil 1974;55:161–166. D’Amour ML, Shahani BT, Young RR, Bird KT. The importance of studying sural nerve conduction and late response in the evaluation of alcoholic subjects. Neurology 1979;29:1600–1604. Delwaide PJ. Contribution of human reflex studies to the understanding of the pyramidal syndrome. In Shahani BT, editor, Electromyography in CNS disorders: central EMG. Boston: Butterworths, 1984:77–109. Deschuytere J, Roselle N, De Keyser C. Monosynaptic reflexes in superficial forearm flexors in man and their clinical significance. J Neurol

150

nerve, and in this setting, the abnormal R2 response remains abnormal with contralateral supraorbital nerve stimulation (i.e., the R1 and R2 responses ipsilateral to the facial nerve lesion are affected regardless of stimulation side). Lesions of the pons and medulla produce site-dependent abnormalities, such as a normal R1 response with bilaterally delayed R2 responses. In addition, the R1 latency can be helpful in the identification of patients with inherited demyelinating polyneuropathies in whom the routine NCS show axon loss related to severity (Wang et al., 2017).

Neurosurg Psychiatry 1976;39:555–565. Dumitru D. Electrodiagnostic medicine. Philadelphia: Hanley & Belfus, 1995.

Katirji MB, Weissman JD. The tibial H reflex and the ankle jerk. Muscle Nerve 1988;11:971–972 (Abstract).

Eccles JC. The central action of antidromic impulses in motor nerve fibres. Pflugers Arch 1955;260:385–415.

Kimura J. F-wave velocity in the central segment of the median and ulnar nerves. A study in normal patients and in patients with Charcot-MarieTooth disease. Neurology 1974;24:539–546.

Falck B, Stalberg E. Motor nerve conduction studies: measurement principles and interpretation of findings. J Clin Neurophysiol 1995;12:254–279.

Magistratis MR, Roth G. Long-lasting conduction block in hereditary neuropathy with liability to pressure palsies. Neurology 1985;35:1639–1641.

Falco FJ, Hennessye WJ, Goldberg G, Braddom RL. H reflex latency in the healthy elderly. Muscle Nerve 1994;17:1350–1351.

Magladery JW, McDougal DB. Electrophysiological studies of nerve and reflex in normal man. I. Identification of certain reflexes in the electromyogram and the conduction velocity of peripheral nerve fibers. Bull Johns Hopkins Hosp 1950;86:265–290.

Fisher MA. AAEM minimonograph #13: H reflexes and F waves: physiology and clinical applications. Muscle Nerve 1992;15:1223–1233. Fullerton PM, Gilliatt RW. Axon reflexes in human motor nerves. J Neurol Neurosurg Psychiatry 1965;28:1–11. Granit R, Leksell L, Skoglund CR. Fibre interaction in injured or compressed region of nerve. Brain 1944;67;125–140.

Marra TR. F wave measurements: a comparison of various recording techniques in health and peripheral nerve disease. Electromyogr Clin Neurophysiol 1987;27:33–37.

Hoffman P. Uber die beziehungen der schenreflexe zur willkurlichen bewegen zum tonus. Z Biol 1918;68:351–370.

Olsen PZ, Diamantopoulos F. Excitability of spinal motor neurons in normal subjects and patients with spasticity, Parkinsonia rigidity, and cerebellar hypotonia. J Neurol Neurosurg Psychiatry 1967;30:325–331.

Jankus WR, Robinson LR, Little JW. Normal limits of side-to-side Hreflex amplitude variability. Arch Phys Med Rehabil 1994;75:3–7.

Panayatopoulos CP. F chronodispersion: a new electrophysiological method. Muscle Nerve 1979;2:69–72.

Chapter 11: Late Responses and Blink Reflexes

Panizza M, Nilsson J, Hallett M. Optimal stimulus for the H reflex. Muscle Nerve 1989;12:576–579. Petajan JH. F-waves in neurogenic atrophy. Muscle Nerve 1985;8:690–696.

1153 nerves. Electromyogr Clin Neurophysiol 1978;18(5):311–351. Thomas JE, Lambert EH. Ulnar nerve conduction velocity and H-reflex in infants and children. J Appl Physiol 1980;15:1–9.

Roth G. Intranervous regeneration of lower motor neuron – I. Study of 1153 nerves. Electromyogr Clin Neurophysiol 1978;18 (3):225–288.

Wang W, Litchy WJ, Mandrekar J, Dyck PJ, Klein CJ. Blink reflex role in algorithmic genetic testing of inherited polyneuropathies. Muscle Nerve 2017;55: 316–322.

Roth G. Intranervous regeneration of lower motor neuron – II. Study of

Weintraub JR, Madalin K, Wong M, Wilbourn AJ, Mahdad M. Achilles

tendon reflex and the H response. Muscle Nerve 1988;11:972 (Abstract). Wilbourn AJ, Ferrante MA. Clinical electromyography. In Joynt RJ, Griggs RC, editors, Clinical neurology, Philadelphia: LippincottRaven, 1997:1–76. Yates SK, Brown WF. Characteristics of the F response: a single motor unit study. J Neurol Neurosurg Psychiatry 1979;42:161–170.

151

Chapter

12

Repetitive Nerve Stimulation Studies and Their Pathological Manifestations

Introduction The anatomy and physiology of the neuromuscular junction (NMJ) was discussed in Chapter 3. This chapter focuses on the EDX manifestations of NMJ disorders. Repetitive nerve stimulation studies (RNSS) are discussed in this chapter and single-fiber EMG (SFEMG) is discussed in a subsequent chapter (see Chapter 15). Dysfunction of the neuromuscular junction (NMJ) is commonly categorized into presynaptic, synaptic, and postsynaptic disorders. Of these, postsynaptic disorders are by far the most commonly encountered. All NMJ disorders result in less acetylcholine (ACh)-ACh receptor (AChR) interactions per nerve terminal depolarization, and hence, all of these disorders reduce the normal safety factor of NMJ transmission (see Chapter 4). The primary problem with presynaptic disorders is a reduction in the number of ACh vesicles released. The ACh vesicles themselves are normal. Because the ACh vesicles that are released are normal, the number of ACh-AChR interactions generated per vesicle is normal. Consequently, miniature endplate potential (MEPP) amplitude is normal. Because fewer vesicles are released, MEPP frequency is reduced. Also, because less ACh vesicles are released per nerve terminal depolarization (i.e., reduced quantal content), the total number of ACh-AChR interactions per depolarization is reduced. As a result, the generated endplate potential (EPP) is lower in amplitude. The most commonly encountered presynaptic disorder is Lambert-Eaton myasthenic syndrome (LEMS). The most commonly encountered postsynaptic disorder is myasthenia gravis. With myasthenia gravis, anti-AChR antibodies result in a reduction in the number of available AChRs. As a result, there are fewer ACh-AChR interactions per vesicle of ACh released, and therefore, the generated MEPPs are smaller in amplitude – typically, about 20% smaller (Lindstrom and Lambert, 1978). Because the

152

spontaneous release of ACh vesicles is normal, the MEPP frequency is normal. Also, because the number of vesicles released per nerve terminal depolarization is normal, quantal content is normal. Because the MEPPs generated with MG are lower in amplitude, MEPP summation generates lower-amplitude EPPs. Whenever the EPP amplitude value is below the depolarization threshold value, neuromuscular junction transmission failure results.

Repetitive Nerve Stimulation Studies Introduction The most common EDX technique utilized to assess the NMJ is termed repetitive nerve stimulation (RNS). With this technique, the nerve is repetitively stimulated and the resulting train of motor responses is recorded from one of the muscles innervated by the nerve while it is being stimulated. As discussed in Chapter 4, two main factors affect ACh release – the quantity of ACh available and the availability of calcium ions in the terminal axon – and both of these factors are affected by RNS. With nerve depolarization, the concentration of calcium ions in the nerve terminal increases (facilitates ACh release), and with repetitive nerve depolarization, the number of available ACh vesicles decreases. The increase in intracellular calcium ion concentration is transient and persists until the calcium ions are sequestered into neighboring mitochondria (100–200 msec). For this reason, calcium ion accumulation occurs when the nerve is depolarized at rates exceeding 5–10 Hz. Thus, when the nerve is repetitively stimulated (or exercised), the collected train is affected by the frequency of the stimulation provided (i.e., low frequency or high frequency) or by exercise (high-frequency stimulation). The RNS protocols vary slightly among EMG laboratories, including the number of stimuli per

Chapter 12: Repetitive Nerve Stimulation Studies and Their Pathological Manifestations

train and the timing and number of trains per study. In addition, the duration of exercise also varies. By adjusting the study parameters – frequency of stimulation, the number of stimuli delivered per train, the number of trains per study, and the duration and timing of exercise – different portions of the neuromuscular junction can be assessed. As a group, the various RNS study protocols are referred to as repetitive nerve stimulation studies (RNSS). Among normal individuals, when RNS is performed at a frequency below 5 Hz (i.e., a stimulation frequency that generates interstimulus intervals > 200 msec), the number of ACh vesicles in the immediately available pool decreases with each stimulation. (Recall from Chapter 4 that stimulation frequencies exceeding 5 Hz permit intracellular calcium ion accumulation, which results in the facilitation of ACh vesicle exocytosis.) As the number of vesicles in the immediately available pool progressively decreases, the quantal content (the number of vesicles released per stimulation) also progressively decreases. As a result, the magnitude of the elicited EPP progressively decreases. This progressive decrement with continuous stimulation is limited by vesicle replenishment of the immediately releasable pool of vesicles (contains about 1,000 vesicles) by the mobilization store (about 10,000 vesicles) and the main reserve (about 250,000 vesicles), as discussed earlier in this textbook (see Chapter 4) (Keesey, 1989). The greatest drop in quantal content occurs between the first and second stimulations, with each subsequent drop in quantal content becoming smaller in magnitude until the trough value is reached (typically the fourth or fifth stimulus), after which the quantal content increases slightly and then plateaus (it does not return to the original value). The plateau value reflects an approximate equilibrium between the rate of ACh vesicle replenishment and the rate of ACh vesicle release. Despite the drop in quantal content with repeating low-frequency depolarizations of the axon terminal, even at the trough value, the generated EPP is suprathreshold. Consequently, each sequence of axon terminal depolarizations generates a train of identically appearing muscle fiber APs. The latter summate to form the motor response train (see Figure 12.1). The overage of the EPP voltage value above that required for depolarization is termed the safety factor of NMJ transmission. NMJ transmission fails whenever the EPP voltage value falls below the threshold value.

Quantal content

Muscle fiber AP

Motor reponse (CMAP)

Figure 12.1 Normal neuromuscular junction transmission. In the top panel, number of vesicles of ACh released per nerve terminal depolarization (i.e., the quantal content) of the first three depolarizations in a series of three is shown (the group of vesicles on the right represents the quantal content of the first depolarization, and the group of vesicles on the left represents the quantal content of the third depolarization; the central group of vesicles represents the second depolarization). Although the number of acetylcholine (ACh) vesicles released with each nerve stimulation (quantal content) decreases with slow repetitive nerve stimulation (i.e., stimulation frequency < 5 Hz), the total number of vesicles remains greater than the number required to depolarize the muscle fiber membrane. As a result, each NMJ generates a normal muscle fiber action potential (AP), and therefore, each nerve stimulation generates a normal compound muscle action potential (CMAP).

Low-Frequency RNSS Introduction Low-frequency RNSS are primarily utilized to identify postsynaptic disorders of NMJ transmission. With low-frequency RNSS, the belly-tendon method (see Chapter 7) is used to determine the stimulator intensity necessary to generate a maximal motor response from the muscle under study. Following this determination, a train of stimuli is delivered at the predetermined intensity and at a frequency of < 5 Hz. The stimuli are delivered at a frequency below 5 Hz to avoid an intracellular calcium ion concentration buildup, which occurs when stimuli are delivered at intervals below 200 msec) (see Chapter 4). As stated earlier, the exact stimulation frequency (typically 2 or 3 Hz), the number of stimuli per train (typically 5–8), the total number of trains collected per study (usually 5 or 6), and the timing of train delivery vary somewhat among different EMG laboratories, as does the duration of the exercise period. Lowfrequency RNSS are used to identify a decremental pattern of motor responses. Harvey and Maslund (1941) were the first to demonstrate this pattern of motor responses.

153

Section 2: Nerve Conduction Studies

5 mV

RECORD

5 ms

Figure 12.2 Slow repetitive spinal accessory nerve stimulation (2 Hz), recording from the trapezius muscle. The baseline train shows no decrement and, thus, the patient is exercised for 1 minute.

Technique In our EMG laboratories, we use a train length of 8 – so that an envelope pattern of motor responses, when present, can be recognized (discussed later here) – and a stimulation frequency of 2 Hz (to lessen the possibility of pseudofacilitation, which may occur at stimulation frequencies of 3 Hz or more). After the baseline train is collected, the muscle is exercised to lower the NMJ transmission safety factor. In our EMG laboratories, the duration of exercise is dictated by whether or not the baseline train shows decrement. When the baseline train shows no decrement, we exercise the muscle under study for 1 minute (see Figure 12.2), whereas when the baseline train shows decrement, we exercise the muscle for 10 seconds (see Figure 12.3). In the setting of a baseline decrement, the exercise period is shortened to lessen the likelihood of missing post-exercise facilitation (discussed further on). On occasion (discussed later), we employ a 2-minute exercise period. Following the exercise period, a series of trains of stimuli are provided at specific intervals. In our EMG laboratory, we deliver a single train of stimuli immediately after the exercise period (t0), then every 30 seconds for a total of two more trains (t30 seconds, t60 seconds), and then every minute for a total of four more trains (t2 minutes, t3 minutes, t4 minutes, and t5 minutes). Thus, we collect a total of up to eight trains:  pre-exercise (the baseline train)

:

to look for response decrement (dictates duration of exercise)

 t0 (immediately following exercise)

:

to look for postexercise facilitation

 t30 seconds, t60 and t5 minutes

:

154

seconds, t2 minutes, t3 minutes, t4 minutes,

to look for postexercise decrement

5 mV

RECORD

5 ms

Figure 12.3 Slow repetitive spinal accessory nerve stimulation (2 Hz), recording from the trapezius muscle. The baseline train shows significant decrement and, thus, this patient is only exercised for 10 seconds.

During the recording of the motor response trains, it is important to limit movement of the body region under study, because movement of the muscle under study may change the configuration of the motor responses collected and, hence, result in erroneous conclusions. This is easiest to accomplish with RNSS of distal extremity muscles (e.g., the ulnar nerve, recording the abductor digit minimi) by using the free hand of the examiner to hold the body region under study in place. However, RNSS of distal muscles is less sensitive than RNSS of proximal muscles for identifying patients with myasthenia gravis. Thus, RNSS using more proximally located muscles are more frequently performed and slightly more challenging to keep from moving. In general, we perform both a distal and a proximal RNSS. Even when the distal RNSS is abnormal, we add a proximal RNSS because it may show greater decrement. The degree of decrement is useful for future comparisons after therapy has begun. We prefer the spinal accessory nerve, recording trapezius, because it is easier to keep the body region under study from moving. For patients referred to the EMG laboratory for suspected ocular myasthenia gravis, we perform RNSS of the facial nerve, typically recording orbicularis oculi, nasalis, or both. In general, we let the clinical phenotype dictate the specific muscles from which the trains are collected (the EDX study is an extension of the clinical examination). Otherwise, of the many proximal RNSS that can be performed, we prefer the spinal accessory nerve (recording from upper trapezius) for two reasons: (1) its stimulation site posterior to the sternocleidomastoid muscle does not result in current spread to adjacent nerves, and (2) the recording electrodes, when placed over the superior aspect of the trapezius muscle, do not record current spread from neighboring muscles. When this study is normal, we follow it with facial nerve RNSS, recording nasalis or

Chapter 12: Repetitive Nerve Stimulation Studies and Their Pathological Manifestations

orbicularis oculi. The latter are more technically challenging due to the potential for co-stimulation of the trigeminal motor axons to the masseter and the challenge, for some patients, of exercising the nasalis muscle. Because cool limb temperatures facilitate NMJ transmission, the studied limb is warmed prior to the performance of RNSS. This lessens the likelihood of a falsely negative study and, additionally, because warm muscles lower the safety factor, may actually increase the sensitivity of the study. Other useful upper extremity RNSS include the axillary nerve, recording deltoid, and the musculocutaneous nerve, recording biceps. In the lower extremity, the femoral nerve, recording rectus femoris, and the common peroneal nerve, recording tibialis anterior, are commonly performed. Again, the ideal choice reflects the clinical features leading to the study.

Postexercise Facilitation and Postexercise Exhaustion As stated in the preceding subsection, when the baseline train shows a decremental response, we exercise 5 mV

the muscle for 10 seconds, whereas when a decremental response is not present on the baseline study, we exercise the muscle for 1 minute. We limit the duration of the exercise period when baseline train decrement is present to avoid missing postexercise facilitation. The period of postexercise facilitation is much shorter than the period of postexercise exhaustion (discussed later here). Postexercise facilitation refers to a motor response increment above the baseline value following exercise. As opposed to pseudofacilitation, which reflects muscle fiber AP synchrony, the phenomenon of postexercise facilitation reflects true facilitation. This period of motor response repair is followed by a period of progressive motor response decline to values below the original decrement. This phenomenon is termed postexercise exhaustion. Over several minutes, the degree of decrement returns to its baseline value (i.e., to the degree of decrement of the original train) (see Figure 12.4). When individuals do not demonstrate a baseline decrement, a longer exercise period (1 minute) is 5 mV

RECORD

RECORD Stimulator impedance high

Stimulator impedance high

05/30/14/ 11:34:35

05/30/14/ 11:39:26

Figure 12.4 Slow repetitive nerve stimulation study showing postexercise facilitation and postexercise exhaustion. The baseline train shows an approximate 50% decrement between the first response and the fourth response. Because there is a baseline decrement, the patient is exercised for only 10 seconds, at which point the second train is recorded. The second train shows a decrement of approximately 5%, which is significantly improved over the baseline train (postexercise facilitation). The third through sixth trains, which were performed 1 minute apart, show decrement values of approximately 15%, 47%, 56%, and 52%, respectively. A shorter exercise period is used when the baseline train shows decrement, to avoid missing the brief period of postexercise facilitation that precedes the postexercise exhaustion.

155

Section 2: Nerve Conduction Studies

employed. This increases the likelihood of precipitating a 10% or greater motor response decrement. In other words, with milder disease, postexercise exhaustion may be the only abnormal finding. When postexercise exhaustion does not follow postexercise facilitation, it is simply referred to as motor response decrement and is quantified.

Pseudofacilitation It is also important to be familiar with the phenomenon of pseudofacilitation. Among normal individuals, at stimulation frequencies of 3 Hz or more, the motor response amplitude increases and the negative phase duration decreases. The negative AUC value is relatively constant because the number of contributing muscle fibers is unchanged. Unlike pseudofacilitation, with true facilitation, there is an actual increase in the number of muscle fiber APs contributing to the recorded motor response, increasing both the amplitude and the negative AUC values of the motor responses. Pseudofacilitation likely reflects an increase in muscle fiber AP synchronization related to increasing conduction velocities among the contributing muscle fiber APs (Stalberg, 1976; Kadrie and Brown, 1978).

Criteria for Abnormal RNSS When the maximum degree of decrement exceeds 10% and the finding is reproducible, the RNSS is considered abnormal. Degrees of decrement below 10% are not labeled as abnormal due to the greater risk of a false positive related to technical error. However, by definition, normal individuals do not show decrement. Thus, this value may be too lenient, increasing the risk of falsely negative studies. In a recent study to determine the ideal cutoff value for decrement with slow RNSS of the facial nerve in patients with myasthenia gravis, it was reported that a cutoff value of 7% (for RNSS of the facial nerve, recording frontalis) or 8% (for RNSS of the facial nerve, recording nasalis) increased study sensitivity without a significant loss of study specificity (Abraham et al., 2017). When the degree of decrement is 5–10%, we repeat the test employing a 2-minute exercise period. When the repeat study shows the same degree of decrement, especially when the motor response trains demonstrate an envelope pattern (i.e., a smooth decrement followed by a smooth increment), we include the following comment:

156

“Although the slow RNSS showed a decrement, the decrement did not meet our criteria for abnormal. However, the reproducibility of the motor response decrement and the envelope pattern of the motor response trains are strongly suggestive of a postsynaptic deficit, such as is observed among individuals with myasthenia gravis.”

Specificity of an Abnormal Study In addition to myasthenia gravis, other disorders associated with a decremental response to slow RNS include other primary NMJ transmission disorders (e.g., Lambert-Eaton myasthenic syndrome, botulism, toxin- and medication-related NMJ transmission defects, and congenital myasthenic syndromes), some neurogenic disorders (rapidly progressive ones), and some primary myopathies (e.g., certain hereditary myotonias and periodic paralyses; myophosphorylase deficiency and other myopathies demonstrating physiological contracture) (Levin, 2000). Rapidly progressive neurogenic disorders, such as amyotrophic lateral sclerosis, are associated with significant reinnervation. As a result, there are a large number of newly formed NMJs (at least during the earlier stages, when reinnervation is attempting to keep pace with denervation). Because newly formed NMJs frequently demonstrate synaptic transmission failure, rapidly progressive neurogenic disorders may show decrement on slow RNSS.

Technical Errors Three technical errors in particular are more of a problem with RNS than with single-nerve stimulation and can result in waveform morphology changes that lead to erroneous conclusions, namely: (1) movement of the stimulating electrodes away from the nerve trunk at the stimulation site, (2) movement of the recording electrodes off of the motor point at the recording site, and (3) performing slow RNSS on a cool limb. With slippage of the stimulating electrodes (due to movement of the stimulator) due to movement of the body region under study or due to the slipperiness of the electrolyte cream, the intensity of the current stimulus received by the nerve decreases, producing a submaximal response. With serial slippage, the motor responses composing the train progressively decrease in size, simulating a decremental response (false positive study). Clues that this may have occurred include an erratically changing train of responses rather than a smoothly changing one (e.g.,

Chapter 12: Repetitive Nerve Stimulation Studies and Their Pathological Manifestations

when one response within the train shows an increment and the subsequent one a decrement, or vice versa). The train also changes in its morphology when the degree of contact between the recording electrodes and the skin changes with serial muscle activation (e.g., serial loss of contact). Because cooling improves the safety factor of NMJ transmission (predominantly by increasing the amount of Na+ influx and, to a lesser extent, by slowing the activity level of acetylcholinesterase), the likelihood of identifying a postsynaptic disorder by slow RNSS is significantly reduced whenever slow RNSS are performed on a cool limb (false negative study).

High-Frequency RNSS High-frequency RNSS are helpful in the identification of presynaptic disorders of NMJ transmission. With highfrequency RNSS, stimulation rates exceeding 20 Hz (usually 40–50 Hz) are utilized. As previously stated, with stimulation frequencies above 5 Hz, intracellular calcium ions accumulate in the nerve terminal and the release of ACh is facilitated, which, in turn, shortens the endplate potential rise time (Rhaminimoff et al., 1978). Among normal individuals, the shorter-duration motor endplate rise times increase the synchrony of the generated muscle fiber APs. Because the muscle fiber APs are more synchronous, the motor response they generate has a shorter negative phase duration and, thus, a higher amplitude. This is termed pseudofacilitation, which, in general, does not result in amplitude increments exceeding 50%. Among patients with presynaptic NMJ transmission disorders, such as Lambert-Eaton myasthenic syndrome, however, a large number of NMJs do not generate muscle fiber APs. In this setting, highfrequency RNSS transforms previously nonfunctioning NMJs into functioning NMJs. Thus, a much larger number of muscle fiber APs are generated, and hence, the recorded motor response is much larger, as evidenced by an increase in its negative area under the curve value. The significant increase in the negative area under the curve value differentiates this phenomenon from pseudofacilitation. This phenomenon also underlies the use of the Lambert test when the first low-amplitude motor response is collected. With the Lambert test, the patient activates the muscle that generated the low-amplitude response for 10 seconds, and the nerve is restimulated. A significant increment in the motor response indicates a presynaptic deficit.

High-Frequency RNSS Technique With high-frequency RNSS, a single train of motor responses is generated and collected. Like slow RNSS, the precise technique varies somewhat among different EMG laboratories. In our EMG laboratory, we apply a 40-Hz stimulation current for 5 seconds (i.e., 200 stimuli), if tolerated. In general, highfrequency RNSS are used to identify presynaptic deficits, such as Lambert-Eaton myasthenic syndrome and botulism. With presynaptic deficits, an incremental response is noted with high-frequency RNS (see Figure 12.5). Because high-frequency RNS can be uncomfortable, an alternative is to ask the patient to voluntarily activate the muscle at maximal capacity. In our EMG laboratories, when the first low-amplitude motor response is encountered, we have the patient voluntarily activate the muscle under study for 10 seconds and then restimulate the nerve looking for an increment in the evoked motor response (this technique is referred to as the Lambert test after its original descriptor). This was done for the patient shown in Figure 12.5 when the median motor response was noted to be low in amplitude despite a normal median sensory response. In this individual, who had been referred by his primary care provider for ataxia suspected to represent the Fisher variant of

Figure 12.5 Incremental response. High-frequency stimulation in a patient with Lambert-Eaton myasthenic syndrome showing an incremental response. In this illustration, although the negative phase duration decreases, the negative area under the curve significantly increases, consistent with a presynaptic NMJ transmission disorder.

157

Section 2: Nerve Conduction Studies

Guillain-Barre syndrome, the Lambert test was positive – the median motor response increased 3.1-fold (210% facilitation) following 10 seconds of exercise. In general, when the response doubles (100% facilitation), a presynaptic defect is suggested, whereas when the response triples (200% facilitation), LEMS is very likely. The positive Fisher test and the incremental response noted on high-frequency RNSS prompted a chest CT (a 2.1  1.5 cm spiculated mass with

References Abraham A, Alabdali M, Alsulaiman A, Breiner A, Barnett C, Katzberg HD, Lovblom LE, Bril V. Repetitive nerve stimulation cutoff values for the diagnosis of myasthenia gravis. Muscle Nerve 2017;55:166–170. Harvey AM, Maslund RL. Method of study for neuromuscular transmission in human subjects. Bull Johns Hopkins Hosp 1941;68:81–93. Kadrie HA, Brown WF. Neuromuscular transmission in human single motor units. J Neurol Neurosurg Psychiatry 1978;41:193–204. Keesey JC. AAEE minimonograph #33: electrodiagnostic approach to defects of neuromuscular transmission. Muscle Nerve 1989;12:613–626.

158

lobulated borders was present in the right upper lobe, along with mediastinal lymphadenopathy) and antibody testing (elevated P/Q type voltage-gated calcium channel antibody); LEMS is more commonly associated with antibodies against the N-type voltage-gated calcium channel, and the patient was diagnosed with paraneoplastic Lambert-Eaton myasthenic syndrome. Thus, the EDX study permitted the lung cancer to be identified while it was still asymptomatic.

Lennon VA, Kryzer TJ, Griesmann GE, O’Suilleabhain PE, Windebank AJ, Woppmann A, Miljanich GP, Lambert EH. Calcium-channel antibodies in the Lambert-Eaton syndrome and other paraneoplastic syndromes. N Engl J Med 1995;332:1467–1474. Levin KH. Disorders of neuromuscular junction transmission. In Levin KH, Luders HO, editors, Comprehensive clinical neurophysiology. Philadelphia: WB Saunders Company, 2000:251–267. Lindstrom JM, Lambert EH. Content of acetylcholine receptor and antibodies bound to receptor in myasthenia gravis, experimental autoimmune myasthenia gravis, and Eaton-Lambert syndrome. Neurology 1978;28:130–138. Rhaminimoff R, Erulkar SD, Lev-Tov A. intracellular and extracellular

calcium ions in transmitter release at the neuromuscular synapse. Ann NY Acad Sci 1978;307:583–598. Sanders DB, Stalberg E. The overlap between myasthenia gravis and Lambert-Eaton myasthenic syndrome. Ann NY Acad Sci 1987;505:864–865. Stalberg E. Electrogenesis in dystrophic human muscle. In Rowland LP, editor, Pathogenesis of human muscular dystrophies. Proceedings of the Fifth International Scientific Conference of the Muscular Dystrophy Association, Durango, Colorado. Amsterdam, Oxford: Exerpta Medica, 1976:570–589. Streib EW. AAEE minimonograph #27: differential diagnosis of myotonic syndromes. Muscle Nerve 1987;10:603–615.

Section

3

The Needle EMG Examination

Chapter

13

The Needle EMG Examination

History In 1825, Sarlandiere, based on the principles of acupuncture, inserted needles into muscles and stimulated muscle contraction through the generation of small sparks (Bonner and DevlescHoward, 1995). This therapeutic technique was painful and often caused tissue necrosis. In 1833, Duchenne demonstrated that muscle could be stimulated percutaneously through the application of skin electrodes, which allowed him to study the normal motions of various skeletal muscles and, later, to apply this technique to the diagnosis and treatment of muscle disorders, thereby giving birth to the field of electrodiagnostic medicine (Bonner and DevlescHoward, 1995). In 1929, with the invention of the coaxial needle electrode, Adrian was able to record the signal of individual motor unit action potentials (MUAPs) and, by incorporating a loudspeaker into the system, added an auditory element to MUAP analysis (Adrian and Bronk, 1929). The auditory characteristics complement the visual characteristics and allow potentials to be recognized by their firing pattern and, for most potentials, their distinct sounds. This is important because the auditory characteristics are often much more distinguishing than the visual features. During this same year, Denny-Brown recorded and further characterized MUAPs (Denny-Brown, 1929).

Introduction All of the electrical activity recorded through a needle electrode represents single or multiple muscle fiber APs, whereas the electrical activity recorded during the motor NCS represents very large groups of muscle fiber APs. Thus, the needle electrode examination and the motor NCS both assess all elements of the motor unit – cell body, axon, and terminal nerve branches, as well as the muscle fibers they innervate and the intervening neuromuscular junctions. Despite this

overlap, the acquired information collected during the needle EMG examination significantly differs, because the needle EMG study provides information about individual muscle fiber APs and individual MUAPs, whereas motor NCS assess the fibers of the entire muscle simultaneously. As a result, needle EMG provides additional information that not only confirms the motor NCS findings (e.g., localization, pathology, and severity) but further refines them and, additionally, provides temporal characteristics of the lesion, such as its chronicity (acute, subacute, or chronic) and rate of progression (remote, slowly progressive, or rapidly progressive). Because of its sensitivity to acute motor axon disruption and the fact that most skeletal limb muscles can be studied by needle EMG, this portion of the EDX study better defines the distribution of the lesion, including subclinical involvement. In fact, most of the information provided by the needle EMG examination is not obtainable by motor NCS. Consequently, the needle EMG examination must never be omitted, although there are a few exceptions to this statement, including: (1) when the lesion is too acute (i.e., not enough time has elapsed for fibrillation potentials to have developed), (2) when the patient is referred for a follow-up study that does not require a needle EMG examination (e.g., follow-up study for suspected carpal tunnel syndrome), and (3) when there is a contraindication to performing the needle EMG study. Because it is invasive, the needle EMG examination is often considered to be slightly more uncomfortable than the NCS. Its disadvantages include the need for patient cooperation, mild discomfort, an inability to identify demyelinating conduction slowing, and an inability to identify mild to mild-moderate degrees of demyelinating conduction block. Like the NCS, the needle EMG study only assesses the larger, more heavily myelinated nerve fibers. As previously stated, because a great number of muscles can be studied during this portion of the

161

Section 3: The Needle EMG Examination

EDX study, the needle EMG examination permits a more widespread assessment than that permissible by the motor NCS. Thus, during a standard needle EMG study, distal, middle, and proximal muscles, including the paraspinal muscles, are sampled. The specific muscles chosen reflect those with differing root, plexus element, and nerve trunk innervation. The sensitivity of the needle EMG for motor axon loss can be extremely high. Due to the high innervation ratio of the skeletal muscles studied, the loss of a single motor neuron or the disruption of a single motor axon results in the denervation of hundreds (sometimes thousands [e.g., gastrocnemius muscle]) of muscle fibers. Consequently, prior to the occurrence of reinnervation by collateral sprouting, the needle EMG examination is by far the most sensitive portion of the EDX study toward identifying motor axon loss. Thus, a lesion undetectable by clinical assessment may be profoundly obvious by needle EMG examination. Unlike the motor NCS, however, the needle EMG examination is fairly insensitive to focal demyelination. It cannot identify focal demyelination that produces conduction slowing, and although it can identify focal demyelination that produces conduction block (i.e., by the presence of a neurogenic MUAP recruitment pattern), this MUAP recruitment pattern is typically not seen with milder lesions. In general, a neurogenic MUAP recruitment pattern is typically only discernible with lesions disrupting 50% or more of the motor axons innervating the muscle. Unlike the NCS portion of the EDX study, the examiner does not use a stimulator to elicit the recorded electrical potentials. Instead, the muscle is either relaxed (seeking spontaneous electrical activity) or active (assessing the MUAPs). Unlike the NCS, in which the monitor displays only a visual output for interpretation, during the needle EMG examination, in addition to the visual output, there is also an auditory output that must be assessed. For example, during muscle relaxation, the electrical activity of the muscle is assessed by listening as the needle is initially inserted and periodically advanced into the muscle. The sound heard, termed insertional activity, is discussed later in this chapter. Listening to the electrical activity is also important between advancements, when the muscle is at rest and the needle is held stationary. The electrical activity heard between advancements is termed spontaneous activity and may be physiologic or pathologic. Finally, during

162

low levels of muscle activation, the voluntary activity – the morphology (duration; amplitude; number of phases; number of turns) and the recruitment pattern of the MUAPs – is characterized. Like insertional activity and spontaneous activity, listening is an important part of MUAP assessment. The characterization of MUAPs through listening is a skill that requires time to develop and, for this reason, can be quite frustrating to residents and fellows eager to learn needle EMG. This chapter focuses on the needle EMG manifestations of normal muscle, including the measurements we make and their meaning. At the end of this chapter, the two basic types of needle recording electrode (concentric and monopolar) are discussed, along with pertinent electrical concepts and instrumentationrelated issues. This is followed by a discussion of the effect that various pathological insults have on these measurements (see Chapter 14). The utility of needle EMG for lesion localization and lesion characterization is also addressed. Other needle EMG techniques (e.g., single-fiber EMG and macro EMG) are discussed in Chapter 15. By far, the most challenging portion of the needle EMG examination is MUAP assessment. For this reason, the EDX provider must possess a detailed understanding of the motor unit. Consequently, this chapter begins with a detailed discussion of motor unit anatomy and physiology, including basic and advanced concepts important to the needle EMG examination. In our EMG laboratories, we assess insertional and spontaneous activity with a vertical resolution of 50 microV/sec and voluntary activity with a vertical resolution of 200 microV/sec. For all of these forms of activity, we set the sweep speed at 10 msec/division. We often increase this value (e.g., to 50 msec/division or more) during the assessment of spontaneous activity (e.g., to better define the firing rates and patterns of myokymic potentials) and during the assessment of voluntary activity (e.g., to better assess MUAP recruitment).

Motor Unit Anatomy and Physiology As previously stated, the motor unit consists of a single anterior horn cell (AHC) and the muscle fibers it innervates, as well as the intervening neuromuscular junctions (NMJs). The AHC, as its name implies, is located in the anterior horn of the spinal cord (laminae VII, VIII, and IX). These cells vary in their

Chapter 13: The Needle EMG Examination

size, with larger AHCs outnumbering smaller AHCs by about 3: 1. Because larger AHCs have largerdiameter axons, it is not surprising that the ratio of larger-diameter to smaller-diameter motor axons within the ventral root is approximately 3:1 (Brown, 1984). AHCs are also referred to as lower motor neurons. There are three types of lower motor neurons – alpha (innervate extrafusal skeletal muscle fibers), gamma (innervate intrafusal skeletal muscle fibers), and beta (innervate both types of skeletal muscle fibers) – the majority of which reside in lamina IX. The cell bodies of the AHCs innervating individual muscles are arranged in vertical columns that span two or more spinal cord segments. In addition to their vertical arrangement, AHCs are also somatotopically arranged, with the truncal and axial motor neurons located ventromedially (with flexor AHCs dorsal to extensor AHCs) and the limb muscles located more laterally. As a result of their vertical arrangement, single skeletal muscles receive innervation from multiple spinal cord segments. This fact – that muscles receive multisegmental innervation – was demonstrated by Thage (1965) by stimulating individual ventral roots while simultaneously recording from multiple skeletal muscles. The central drive to a muscle is distributed to all of the anterior horn cells serving that muscle. Although the descending input to lower motor neurons is primarily via interneurons, a monosynaptic relationship exists between upper motor neurons and lower motor neurons innervating distal upper extremity muscles (Petajan, 1991). When an AHC is activated (at the axon hillock or, more commonly, at the initial axon segment), all of the muscle fibers that it innervates are depolarized (this precipitates their shortening and the generation of contractile force). The amount of force produced by the motor unit reflects the total number of muscle fibers composing it and the activation frequency of the motor unit. In response to a single stimulus, the muscle fiber twitches and then relaxes. When repetitive stimuli are applied at a slow rate, individual twitch contractions are produced, whereas when they are applied at higher frequencies, the twitches produced summate, generating a fused tetanic contraction (a tetanus).

composing a single motor unit. Thus, it is the EDX manifestation of the motor unit, which represents the smallest unit of force generated by muscle tissue. The visual and auditory analysis of the MUAP is the most challenging EDX study component to master, mainly because the MUAPs vary with the muscle under study and with the age of the patient. Thus, in addition to the mastery of the auditory characteristics associated with all of the visually measured needle EMG parameters, the EDX provider must learn to judge the MUAPs based on age and the muscle under study. Because different muscles are composed of muscle fibers of slightly different mean diameters, and because the motor unit innervation ratio varies among different muscles and increases with age, it should not be surprising that the observed MUAPs differ among different muscles and different patient ages. The EMG provider must be familiar with these differences. This requires experience. The MUAP is assessed during the activation phase of the needle EMG study. Thus, the EMG provider must be aware of the function of the muscle under study so that it can be appropriately activated. For example, some of the EMG fellows activate the abductor pollicis brevis by asking the patient to extend the thumb. Although some thenar muscle MUAPs appear on the screen, the muscle is ideally activated by thumb abduction rather than extension. In this manner, less effort is required by the patient, more MUAPs are available for study, and the assessment is much less painful. Both extrinsic (amplitude; duration) and intrinsic (e.g., the number of phases) MUAP measurements are made. These measurements should only be taken when the MUAP is near the recording surface of the needle electrode, as indicated by a rise time below 500 microseconds. The rise time represents the fastest component of the MUAP (the duration of time from the initial positive to negative peak) and generates a sharp clicking sound. The rise time depends on the skew rate (slope) and the amplitude of the MUAP (i.e., the rate of ascent and the degree of ascent) (see Figure 13.1).

The Motor Unit Action Potential

Among the MUAP measurements, MUAP duration (in msec) is the most important one, whereas MUAP amplitude (in mV from the positive peak to the negative peak) is of much lesser utility (unlike with the NCS, where amplitude is typically the most important

Introduction As previously stated, the MUAP is a compound electrical potential composed of the muscle fiber APs

Duration

163

Section 3: The Needle EMG Examination

200 uV

10 ms

Satellite

Amplitude

Duration Figure 13.1 Motor unit action potential measurements. The duration is measured from the initial deflection away from the baseline (the onset point) to the final return to the baseline (the termination point). When a satellite potential is present, it is not included in the duration measurement. For MUAPs, the amplitude is the maximal peak-to-peak value of the amplitude. The rise time (the duration from the initial positive peak to the initial negative peak) is indicated by the two stars. The demonstrated MUAP is triphasic. The number of phases is defined as the number of intersections with the baseline minus one, where baseline intersections include the onset site, the terminal site, and the crossings in between. Alternatively, the number of phases can be defined as the number of baseline crossings plus one. In the figure, the MUAP demonstrates two baseline crossings. The third phase shows three turns (direction changes that do not cross the baseline).

measurement). In general, MUAP duration ranges from 5 to 15 msec and MUAP amplitude ranges from 1 to 3 mV. Most limb muscles have MUAP durations in the 8–10 msec range. When the duration of an MUAP with a satellite potential is measured, the satellite potential is ignored (see Figure 13.1). In addition to the specific muscle being studied and the age of the patient, these measurements are affected by a number of other factors. Thus, it is important to consider the MUAP duration of the muscle under study in relation to the other limb muscles being studied and, when in doubt, to compare it to the contralateral side. Judging the normality of MUAP durations with respect to the specific muscle, the age of the patient, and in relation to the

164

MUAP durations of other muscles of that limb is a skill that requires time to develop. Anterior horn cell senescence related to aging leads to reinnervation via collateral sprouting and the resultant increase in MUAP duration related to age. The MUAP duration of different muscles varies. Regarding the upper extremity muscles that we commonly assess in our EMG laboratory (see Section 6, Appendix 3), the brachioradialis muscle demonstrates MUAPs that are obviously shorter in duration than the MUAPs of the other upper extremity muscles. To the unwary practitioner they may appear myopathic. Conversely, the triceps (we typically study the lateral head) and the deltoid muscles usually exhibit MUAPs that have a somewhat longer duration than the MUAPs of the other upper extremity muscles. In the lower extremity, the shortest duration MUAPs are typically encountered in the hamstring muscles (e.g., the short head of the biceps femoris), whereas the quadriceps muscle group tends to have some of the longest-duration MUAPs. It is important to understand these MUAP relationships because deviation suggests a potential abnormality. When the duration of an MUAP is questionably prolonged, it should immediately be compared to the MUAPs of the homologous muscle on the contralateral side.

Motor Unit Recruitment (Force Generation) The motor units represent the elemental units of muscle force. The quantity of the force generated ranges from minimal to maximal and reflects the number of motor units recruited (spatial recruitment) and their rate of firing (temporal recruitment). In general, the motor units are successively activated based on their size. This principle of motor unit recruitment by size is referred to as the Henneman size principle (Henneman, 1965). There is a correlation between the size of the motor unit (i.e., the diameter of its cell body and its axon) and its force (tension)-generating ability (Petajan, 1991). Based on this relationship, the smaller type I motor units are recruited before the larger type II motor units. More specifically, with low levels of force, the recruited motor units are primarily type I (type S [slow-twitch, fatigue-resistant]). As the amount of force is increased, type IIa motor units (type FR [fast-twitch, fatigue-resistant]) and, later, type IIb motor units (type FF [fast-twitch, fatigue sensitive]) appear. The successive recruitment of motor units based on their

Chapter 13: The Needle EMG Examination

force-generating ability ensures that the increases in the amount of force generated by a muscle are smooth. Motor units with a lower threshold are recruited earlier and fire at lower rates, whereas motor units with a higher threshold are recruited later and fire at higher rates. Because the motor units with lower threshold have longer periods of afterhyperpolarization, their firing frequencies are limited and correlate with one-half of the relaxation time of the muscle fibers of the recruited motor unit (Petajan, 1991). This firing frequency correlation also lessens the tremulousness of the muscle contraction. The slower motor units are the ones studied during the needle EMG examination when the muscle contractile force is limited to mild voluntary contraction (e.g., that required to just overcome gravity [MRC 3]). Because many motor units are of similar size, there must be other factors that contribute to recruitment order. In one study of the first dorsal interosseous muscle, motor unit recruitment order varied based on the task (i.e., whether it was functioning as an agonist in finger abduction or as a synergist in finger flexion) (Desmedt, 1983). In addition to motor unit size and the task being performed, other factors that have been shown to influence motor unit recruitment include axon speed, relative twitch speed, relative tetanic force, relative fatigue, specific membrane resistance, specific membrane conductance, and synaptic density, as well as motor unit location within the muscle. Specific membrane resistance and synaptic density play a greater role in slow motor units, whereas axon speed, relative twitch speed, relative tetanic force, relative fatigue, and specific membrane conductance play a greater role for fast fatigue motor units; fatigue resistant motor units are intermediate (Petajan, 1991). Regarding proprioceptive input, smaller motor neurons generate larger EPSPs in response to Ia afferent stimulation. Also, for a constant amount of current, smaller neurons generate larger depolarizations, which might account for their lower threshold of activation and earlier recruitment (Petajan, 1991). Another important concept regarding motor unit recruitment is that, although there is an increase in the amplitude of the MUAPs of the successively recruited motor units, their durations do not significantly differ (Brown, 1984). The amplitude of the muscle fiber AP is proportional to its diameter. Hence, as spatial recruitment advances from type I

motor units to type II motor units, an increase in MUAP size is typically visible. Because the MUAP amplitude typically reflects only 1–3 neighboring muscle fibers and the individual muscle fiber AP amplitudes are proportional to the diameter of the muscle fiber, an increase in the peak-to-peak amplitude is discernible during strong contraction. However, because the innervation ratio of a muscle is constant and because the duration of the MUAP reflects all of the muscle fibers belonging to the motor unit, the duration of the MUAP does not appreciably increase with greater degrees of effort. In addition, once about 4 MUAPs are present on the screen, MUAP overlap impedes MUAP duration assessment. During the needle EMG examination, because MUAP recruitment is typically assessed during low levels of muscle activation, type I motor units (slowtwitch, fatigue-resistant) are primarily being assessed. As previously stated, the degree of force generated by the muscle is increased in two primary ways: (1) spatial recruitment (an increase in the number of activated motor units) and (2) temporal recruitment (an increase in the firing rate of the already activated motor units). Spatial recruitment and temporal recruitment progress concurrently. With weaker contractions, spatial recruitment plays the predominant role in force augmentation, whereas temporal recruitment provides the basis for fine gradations of muscle contractile force. Obviously, once all or nearly all of the motor units have been recruited, further spatial recruitment is no longer possible. Consequently, further increments in force must be achieved through temporal recruitment (Stein, 1974). During needle EMG examination, with a low level of effort, a single MUAP can be recorded on the monitor. The firing rate at which the first MUAP appears is usually in the 5–8 Hz range and is termed the onset frequency or the basal firing rate (Conwit et al., 1998). The MUAP firing rate increases to 20–40 Hz with increasing effort. At lower levels of activation, lapses in firing may occur (Petajan, 1991). With greater effort, the activated MUAP increases its firing rate (temporal recruitment) and a second MUAP appears (spatial recruitment). Because MUAPs demonstrate minimal variation in their morphology with consecutive firings, different MUAPs are easily distinguished from each other. Through increases in temporal and spatial recruitment, the contractile force is increased. The firing rate of the first MUAP when the second MUAP

165

Section 3: The Needle EMG Examination

appears is referred to as the recruitment frequency (Petajan and Phillip, 1969). This value varies with the muscle under study, but usually ranges from 10 to 12 Hz for most extremity muscles, up to 16 Hz for cranial muscles, and up to 30 Hz for facial muscles (Frascarelli and Rocchi, 1992). The onset and recruitment frequencies are calculated by measuring the interpotential interval (IPI) between successive appearances of an individual MUAP when it first appears (onset frequency) and when the second MUAP appears (recruitment frequency). The IPI represents the duration of its cycle. The IPI dictates the number of cycles (MUAP firings) that occur in 1 second. The cycles per second (CPS) is termed Hertz (Hz) and represents the firing frequency of the MUAP. The firing frequency of a repeating MUAP is easily calculated by dividing 1,000 msec (which is equivalent to 1 second) by its IPI, in msec, which dictates the number of repeats per second: Hz ¼ cycle=sec ¼ cycles per 1,000 msec 1 cycle ¼ the IPI ðin msecÞ Thus, the MUAP firing frequency (in Hz) at any moment = 1,000 /IPI at that moment. It is quite easy to estimate the MUAP firing frequency when the sweep speed is set at 10 msec per division. At this setting, 10 divisions represent 100 msec (1/10 of a second). For example, when the interpotential interval for an MUAP is 100 msec, it is firing at a rate of 10 Hz (1,000/100 = 10 Hz) (see Figure 13.2).

Figure 13.2 Recruitment frequency. The recruitment frequency is the firing frequency of the first motor unit action potential (MUAP) when the second MUAP appears. In this figure, there are three MUAPs on the screen. Of these three, the first and third MUAPs represent the same motor unit. In this example, when the second MUAP appears, the first MUAP is firing at 16.4 Hz, based on its interpotential interval of 61 msec at that moment (1,000 msec/ 61 msec = 16.4). For this reason, the recruitment frequency is 16.4 Hz, which is abnormally high.

166

The firing frequency of an MUAP can also be estimated based on the duration of the sweep. For example, if there are 10 divisions on the screen and each division represents 10 msec, then the screen (and each sweep) represents 100 msec (10 msec/division  10 divisions = 100 msec). For this reason, whenever the MUAP falls on the exact same spot with sequential sweeps, the MUAP is firing at exactly 10 Hz, whereas when it appears earlier on the subsequent sweep, it is firing at a rate exceeding 10 Hz. Similarly, when it is firing at a rate below 10 Hz, then sequential sweeps cause it to appear later with sequential sweeps. In other words, when sequential sweeps show the MUAP to appear at the same site on the screen, the MUAP is firing at 10 Hz, whereas when it appears to be moving across the screen to the left (i.e., appearing earlier), it is firing at a rate above 10 Hz. Finally, when it moves across the screen to the right (i.e., appearing later), it is firing at a rate below 10 Hz (see Figure 13.3). When it moves to the left or the right, its value can be estimated by the number of divisions it is away from the 10-Hz site (see Figure 13.4). Recruitment can also be characterized by the ratio of the fastest firing MUAP on the monitor to the total number of individual MUAPs. This ratio is termed the recruitment ratio. The recruitment ratio is normally in the 5–7 range. In our EMG laboratories, we typically do not consider the recruitment ratio to be abnormal until it exceeds 10 (see Figures 13.5 and 13.6). The rule of fives is based on an expected recruitment ratio of 5 (Daube, 1991). With this supposition, the first MUAP appears on the screen at about 5 Hz (onset frequency), and as the force of contraction is increased, its firing frequency increases (temporal recruitment). When it reaches 10 Hz, a second MUAP appears on the screen (spatial recruitment). At this moment, the recruitment ratio is 5 (10/2 = 5). With even greater degrees of effort, temporal recruitment increases further, and when it reaches approximately 15 Hz, a third MUAP appears on the screen, at which point the recruitment ratio is still 5 (15/3 = 5). The recruitment ratio is not constant. For example, just before the third MUAP appears, when the fastest firing MUAP is firing at a frequency of 14 Hz and there are only two MUAPs on the monitor, the recruitment ratio is 7 (14/2 = 7), and, just after the third MUAP appears on the screen, when the fastest MUAP is firing at a frequency of 16, the recruitment

Chapter 13: The Needle EMG Examination

Figure 13.3 In the upper panel, the MUAP appears at the same position on the screen, indicating that the potential is firing at a frequency of 10 Hz. In the middle panel, the MUAP appears earlier with each sweep (i.e., it is moving to the left). Thus, it is firing at a frequency faster than 10 Hz. In this case, it is firing at 11 Hz (1,000 msec/90 msec). In the lower panel, the MUAP appears later with each sweep (i.e., it is moving to the right). Thus, it is firing at a frequency below than 10 Hz. In this case, it is firing at 9.1 Hz (1,000 msec/110 msec).

ratio is 5.3 (16/3 = 5.3). In other words, the recruitment of the next MUAP causes the recruitment ratio to drop to 5. It then increases (temporal recruitment) until the next MUAP is recruited (spatial recruitment),

at which point it drops again. Values over 10 Hz indicate neurogenic recruitment (a pathological decrement in spatial recruitment). This is discussed in detail in Chapter 14.

167

Section 3: The Needle EMG Examination

200 uV MUAP 1

Figure 13.4 Diagram showing the firing frequency values for leftward and rightward movement across the screen. For example, when a potential appears at the same position with repeated firing, its firing frequency is 10 Hz, whereas when it moves across the screen to the right at 1 box per sweep, its firing frequency is 9.1 Hz. When it moves across the screen to the right at 2 boxes per sweep, its firing frequency is 8.3 Hz. When it moves across the screen to the left, its firing frequency is faster than 10 Hz, as shown in the illustration.

200 uV MUAP 1

10 ms MUAP 1

MUAP 3

MUAP 2

MUAP 2

Figure 13.6 Determining the recruitment ratio with three motor unit action potentials (MUAPs) on the monitor. MUAP 1 is firing at 22.2 Hz (1,000 msec/45 msec), MUAP 2 is firing at 16.9 Hz (1,000 msec/59 msec), and MUAP 3 has just appeared. Thus, at this point, the recruitment ratio is 22.2/3, which equals 7.4 and would be considered normal.

Eventually, as more and more MUAPs firing at faster and faster rates appear on the screen, individual MUAPs can no longer be discerned by visual or auditory means, and hence, a recruitment ratio can no longer be calculated. At this point, a full interference pattern is said to be present (discussed further on). When the calculated recruitment ratio exceeds 10, a reduction in spatial recruitment is indicated. When

168

10 ms MUAP 2

MUAP 1

Figure 13.5 Determining the recruitment ratio with two motor unit action potentials (MUAPs) on the monitor. The recruitment ratio is defined as the firing frequency of the fastest MUAP divided by the total number of MUAPs on the screen. The monitor shows 2 MUAPs near the recording surface of the needle electrode. The fastest one is firing at a frequency of 16.4 Hz (see Figure 13.2). Thus, the recruitment ratio is 8.2 (16.4 Hz/2 MUAPs = 8.2).

spatial recruitment is reduced (e.g., MUAP drop out due to demyelinating conduction block or axon loss), the activated MUAPs can be observed to fire at rates faster than normally appreciable. In other words, there is discordance between the number of MUAPs (spatial recruitment) and their firing frequencies (temporal recruitment). This phenomenon is termed neurogenic recruitment and it reflects AP dropout, either from dysfunction of the lower motor neuron at the cell body or motor axon level (i.e., proximal to the intramuscular arborization point) or to demyelinating conduction block. Its presence (temporal recruitment out of proportion to spatial recruitment) indicates a good effort and genuine disease. Conversely, with decreased effort, spatial recruitment is proportional to temporal recruitment (e.g., a limited number of MUAPs firing at a slow or moderate rate). Neurogenic recruitment is also termed decreased recruitment (we prefer to avoid this term because it is less specific and may be confused with decreased effort). When single MUAPs are firing at rates exceeding 20 Hz, neurogenic recruitment must exist because this rate is approximately 3 standard deviations above the mean rate observed with a 30% maximum isometric contraction (i.e., a level of effort at which MUAP recruitment should fill the screen [termed a full interference pattern] and MUAPs should not be individually discernible) (Dorfman et al., 1988) (see Figure 13.7). This is discussed in detail in Chapter 14. Regarding MUAP recruitment, maximal spatial recruitment precedes maximal temporal recruitment. It is estimated that maximal temporal recruitment increases the level of contractile force threefold

Chapter 13: The Needle EMG Examination

500 uV

10 ms

EMG laboratories, we seldom find it necessary to assess full interference patterns. Moreover, this often puts considerable torque on the needle electrode, which often bends the needle and typically increases the discomfort of the study. In those rare instances when we do assess the interference pattern, we hold the limb in place so that the contraction is isometric, thereby limiting needle torque.

The Measurements We Make and Their Meanings Figure 13.7 Neurogenic recruitment demonstrated by a single motor unit action potential (MUAP). In this illustration, the interpotential interval for the MUAP is 50 msec. This equates to a firing frequency of 20 Hz (1,000 msec/50 msec = 20 Hz). In addition, the MUAP duration is approximately 30 msec, indicative of reinnervation via collateral sprouting.

(or more) higher that the contractile force obtained at maximal spatial recruitment (Ferrante and Wilbourn, 2015). When disease incompletely affects the motor unit distal to the arborization point of the motor axon (i.e., it involves some of its terminal branches [early Guillain-Barre syndrome], NMJs (Lambert-Eaton myasthenic syndrome], or muscle fibers [myopathy]), the MUAP generated is composed of less muscle fiber APs and, consequently, is smaller. This is termed disintegration of the motor unit. Because these motor units generate less contractile force, it takes more of them to generate a given force. As a result, spatial recruitment occurs at a quicker pace than usual (termed early recruitment). Spatial and temporal recruitment are concordant with hysteria/conversion, malingering, and factitious disorder, when pain limits effort, with upper motor neuron disorders, and when stronger muscles are studied (e.g., the gastrocnemius). In these settings, spatial and temporal recruitment are appropriate for the degree of effort (i.e., a reduced number of MUAPs firing at a reduced rate; an intermediate number of MUAPs firing at an intermediate rate). This differs from the discordance seen with neurogenic recruitment – a reduced number of MUAPs firing at an increased rate. These MUAP recruitment patterns – neurogenic recruitment, early recruitment, poor effort, strong muscles, and upper motor neuron disorders – are most appreciable at lower levels of muscle activation (i.e., 1–4 MUAPs on the screen). Therefore, in our

Like the NCS portion of the EDX study, the needle EMG portion is not standardized among practitioners. Indeed, it is even less standardized than the NCS, including the number of muscles studied, the choice of muscles studied, the amount of time spent in each phase of the study, and many other factors. Many practitioners begin the needle EMG study with a core group of muscles that includes distal, intermediate, and proximal muscles that are innervated by different roots, plexus elements, and nerves, as well as the paraspinal muscles. To this initial group of muscles they add muscles based on the specific diagnostic considerations. In our EMG laboratory, we begin with 6 or 7 muscles plus the paraspinal muscles and add muscles to this group based on the diagnostic considerations. We also add muscles based on the findings as the needle EMG study progresses. Our typical screening muscles are provided in Appendix 7, and the 50 cases demonstrate the addition of those muscles that we prefer in specific situations (see Sections 5 Section 6, Appendix 7). To increase the yield of the paraspinal muscle assessment, we study these muscles at two levels because of their significant multiple root innervation and their tendency to reinnervate quickly. If the first assessment is positive, then the second level is not required. Because the paraspinal muscles are innervated by the posterior primary rami, which emanate from the mixed spinal nerve just after it exits from the intervertebral foramina, the presence of fibrillation potentials in the paraspinal muscles most commonly indicates a lesion within the intraspinal canal (less likely the intervertebral foramina or the distal portion of the posterior primary ramus). Paraspinal muscle fibrillation potentials are also seen with generalized processes, such as myopathies. With myopathies, fibrillations potentials tend to be more pronounced in the paraspinal muscles and, indeed, may be restricted to these muscles.

169

Section 3: The Needle EMG Examination

During the needle EMG study of an individual muscle, the needle electrode is inserted into the resting muscle and then intermittently advanced in small increments. At the end of each advancement (insertion), the insertional activity triggered by the advancement is assessed. In addition, other activity precipitated by the needle advancement is awaited. Importantly, the latter activity, which follows the insertional activity, has a different significance. We refer to this activity as provoked activity because it occurs after the insertional activity but is temporally related to the needle insertion. During the pauses between advancements, the needle is held motionless in the resting muscle and spontaneous activity is sought. Intermittently, the patient is asked to lightly activate the muscle under study so that the morphology and recruitment characteristics of the MUAPs, termed voluntary activity, can be assessed. Although the needle EMG study is divided into three phases (insertional phase assessing insertional activity, resting phase seeking spontaneous activity, and activation phase assessing voluntary activity), the three phases are not sequentially performed. Rather, the needle electrode is incrementally advanced into resting muscle and the insertional activity following each advancement is assessed and the presence of spontaneous activity between advancements is sought (i.e., the insertional and spontaneous phases are performed simultaneously). During the voluntary phase, while the voluntary activity is being studied, spontaneous activity (e.g., fibrillation potentials) is still being sought. When the characteristic auditory features of fibrillation potentials are perceived during the voluntary phase of MUAP assessment, the screen sensitivity is adjusted and the fibrillation potentials characterized, after which the voluntary phase is continued. Thus, although these phases will now be discussed individually, it is important to realize that their performance is overlapping.

Insertional Phase Once the needle electrode penetrates the skin and subcutaneous tissue, it lies in the substance of the muscle. When it initially enters the resting muscle and with all subsequent advancements, the needle electrode induces a burst of muscle fiber discharges, termed insertional activity. This activity reflects a combination of mechanical stimulation (excitation) of muscle fibers by the advancing needle electrode

170

and needle-induced muscle fiber injury. Because larger needle electrodes affect a larger number of muscle fibers, they produce a greater amount of insertional activity with needle electrode advancement than do smaller needle electrodes. Nonetheless, even large amounts of insertional activity are brief in duration and do not outlast needle advancement by more than 400 microseconds. The primary significance of insertional activity is that its presence indicates that the muscle tissue under study is viable. Insertional activity is assessed visually and aurally at sensitivity settings of 50–200 microV/division. Thus, it can be assessed throughout the needle EMG study each time the needle electrode is advanced. A form of normal insertional electrical activity that follows the insertional activity described above is snap-crackle-pop (Wilbourn, 1982). This normal type of electrical activity is transient, typically lasting just a few seconds. Unlike the insertional activity that occurs during needle advancement and continues for up to 400 microseconds after needle advancement stops, snap-crackle-pop occurs during the needle advancement pause, just after the insertional activity discussed above. To avoid confusion, this form of electrical activity may be better referred to as provoked activity because it is provoked by needle advancement and because it follows the insertional activity related to the advancement. Wilbourn coined the term “snap-crackle-pop” to signify that the individual electrical potentials composing this benign form of electrical activity differ from each other, not only in their visual characteristics, but also in their auditory characteristics (Wilbourn, 1982). In addition, the number of potentials in each spurt of discharges also differs, as do the interpotential intervals separating them. This type of electrical activity is most frequently encountered during the needle EMG study of the medial head of the gastrocnemius muscles of young, muscular males, but can be seen in other limb muscles, including those of the upper extremity. It is important to be familiar with this benign form of electrical activity so that it is not mistaken for a pathological form of electrical activity. A third form of increased insertional activity consists of runs of insertional positive waves triggered by needle advancement (Weichers and Johnson, 1982). It is termed Weichers-Johnson syndrome (after the two physicians first describing it) or EMG disease. This type of electrical activity has a wide distribution and may have an autosomal dominant inheritance pattern

Chapter 13: The Needle EMG Examination

and represent a forme fruste (an attenuated manifestation of a disease or syndrome suggesting partial presence) of myotonia congenita (Mitchell and Bertorini, 2007).

Resting Phase Between needle electrode advancements, when the muscle is resting and the needle is motionless, any electrical activity generated by the muscle is termed spontaneous activity. Spontaneous activity can be normal or abnormal. At times, it is difficult for the patient to relax the muscle, making it challenging to assess for spontaneous activity. In this situation, the limb can be repositioned (e.g., passively elevate the forearm to relax the extensor indicis), the antagonist can be activated (e.g., neck flexion to relax the paraspinal muscles; arm adduction to relax the deltoid), or the patient can be asked to relax the muscle by using the EMG activity as a form of biofeedback. Spontaneous activity is assessed visually and aurally at a sensitivity setting of 50 microV/division, although it is frequently apparent at higher settings (e.g., 200 microV/division). Unlike insertional activity, which requires needle electrode advancement for its generation, spontaneous activity is independent of needle electrode advancement. In addition, unlike the transient nature of insertional activity, most forms of spontaneous activity are sustained. Like insertional activity, spontaneous activity may be physiologic (when the needle electrode is in the vicinity of the endplate zone) or pathologic. Physiologic endplate activity, of which there are two types, is discussed here, whereas pathologic spontaneous activity is discussed in the subsequent chapter of this textbook (see Chapter 14). Although fasciculation potentials may be benign or pathologic, there is no definite way to differentiate the two types. Rather, they are differentiated by the company they keep (by whether or not they are observed with pathological potentials, such as fibrillation potentials). For this reason, they are discussed with the pathological potentials in the subsequent chapter (see Chapter 14).

Endplate Activity Within the endplate region, spontaneous activity is termed endplate activity or endplate noise. It consists of miniature endplate potentials (MEPPs) and endplate spikes. MEPPs result from the spontaneous release of ACh vesicles and summate to form endplate

potentials (previously discussed in Chapter 4). Endplate spikes are caused by the mechanical irritation of intramuscular terminal branches by the needle electrode, which results in the generation of individual muscle fiber APs (Brown, 1984; Blight and Precht, 1980). Because the needle electrode is causing the potential, an approaching phase is unexpected. As suggested by the term “endplate activity,” this type of activity is only observable when the needle electrode is in the endplate region (innervation zone), which is typically more painful for the patient and should prompt withdrawal of the needle electrode and redirection (see Figure 13.8). Miniature Endplate Potentials Miniature endplate potentials are low in amplitude (usually sensory Axon loss, pure sensory Mixed axon loss and demyelinating, sensory and motor

Neuromuscular Junction Disorders The anatomy and physiology of the neuromuscular junction (NMJ) was discussed in Chapter 4, and repetitive nerve stimulation studies (RNSS) were discussed in Chapter 12. NMJ transmission disorders may be divided into those affecting the nerve terminal (presynaptic), the synapse (synaptic), or the endplate region (postsynaptic). In the EMG laboratory (and clinically), the most common postsynaptic NMJ disorder encountered is myasthenia gravis (MG), and the most common presynaptic NMJ disorder is LambertEaton myasthenic syndrome (LEMS), both of which are autoimmune disorders. With MG, the antibodies are directed against the AChR on the postsynaptic membrane, whereas with LEMS, they are directed against the VGCCs of the presynaptic membrane. Synaptic disorders are much less frequently encountered. Of these, the most common is reduced acetylcholinesterase activity, which may be acquired or congenital. In this setting, the breakdown of ACh is impeded, which may lead to prolonged muscle fiber depolarization. On motor NCS, this phenomenon may be observed as the appearance of two motor responses in response to a single stimulation. With presynaptic disorders, the primary problem is ACh release, whereas with postsynaptic processes, there is a decrease in ACh-AChR interactions, primarily related to AChR loss or block. Both presynaptic and postsynaptic disorders lessen the safety factor of NMJ transmission (see Chapter 4). Myasthenia gravis and Lambert-Eaton myasthenic syndrome are discussed here, but those portions of the discussion

251

Section 4: Other Pertinent EDX Information

related to NMJ anatomy, pathophysiology, and EDX assessment are limited, as they have already been reviewed (see Chapters 4, 12, and 15).

Myasthenia Gravis Background Myasthenia gravis is an autoimmune disorder related to the development of antibodies against the postsynaptic acetylcholine (ACh) receptors. Receptor cross-linking by antibody bindings leads to AChR internalization (i.e., shortened AChR half-life). Binding also activates complement, which, in turn, leads to complement-mediated membrane destruction. The latter results in a reduction of the surface area of the postsynaptic membrane, a widening of the primary synaptic cleft, and a simplification of the postsynaptic membrane (i.e., the postsynaptic folds are shallower in depth and less numerous). The shortened half-life of AChRs results in a decrease in the AChR density of the postsynaptic membrane. The widened gap between the presynaptic and postsynaptic membranes means that more ACh diffuses out of the synapse and that it takes longer for ACh molecules to cross the synapse (increased transit time). The increased transit time increases the chance that the ACh molecule will be hydrolyzed by acetylcholinesterase (AChE) rather than binding to a postsynaptic AChR. All of these changes lessen the likelihood of successful AChAChR interactions, thereby decreasing the safety factor of NMJ transmission. Because this disorder has a predilection for the oculobulbar and proximal upper extremity muscles, the sensitivity of EDX testing is higher in this distribution than outside of it. Because MG is an NMJ transmission disorder, the sensory NCS are normal. Because it is a disorder of pathological fatigue, the motor NCS and late responses are typically normal (i.e., the stimuli associated with the collection of a motor response does not induce fatigue). The needle EMG examination also is typically normal, although moment-to-moment variation of MUAP morphology may be observed (because the subset of failing NMJs varies with each activation, the individual muscle fiber APs contributing to the MUAP also vary). With more advanced disease, short duration, lowamplitude MUAPs are observed (due to disintegration of the motor unit). Nonetheless, routine EDX testing is frequently normal among MG patients.

252

Routine NCS As expected of an NMJ transmission disorder, with myasthenia gravis, the sensory NCS are normal. Except in very severe cases, the motor NCS are also normal, because most of the EPPs, although low in amplitude, are suprathreshold. With very severe disease, however, low-amplitude motor responses may be observed. When the Lambert test (10 seconds of exercise) results in a motor response increment, the low-amplitude motor response may be erroneously attributed to reflect a presynaptic NMJ transmission deficit (Sanders and Stalberg, 1987).

Repetitive Nerve Stimulation Studies The major technique used to identify MG in the EMG laboratory is low-frequency RNSS, which represent the electrical correlate of sustained effort (sustained effort is used clinically to bring out fatigable weakness). The progressive diminution of quantal content resulting from low-frequency RNSS results in progressive NMJ transmission block (increases the number of subthreshold EPPs). The addition of exercise to the slow RNSS further reduces the safety factor of the exercised NMJs and, thus, further increases the sensitivity of the study. Exercise is equivalent to nerve stimulation at 20–40 Hz. The technique used in our EMG laboratories is reviewed in Chapter 12. Because NMJ transmission is enhanced by cooler temperatures, it is mandatory to ensure that the limb under study is warm. Muscle coolness has been touted as the explanation for why distal muscles are less likely to show decrement on slow RNSS than more proximal (i.e., warmer) muscles. As the severity of disease increases, the sensitivity of slow RNSS increases. When disease is mild, slow RNSS may be normal. The application of exercise increases the diagnostic sensitivity of slow RNSS among this patient population. Because of the predilection of oculobulbar and proximal limb muscles, the sensitivity of slow RNS is improved in proximally located muscles. The higher involvement of extraocular muscles may reflect their tonic activity in the awake state. We typically perform slow RNSS on the spinal accessory nerve (recording trapezius) and, when negative, study the facial nerve recording from one of the facial muscles (e.g., orbicularis oculi or nasalis). The ideal muscle to study is one that demonstrates fatigue on clinical examination. Any identified

Chapter 17: The Electrodiagnostic Manifestations of Disorders at Various Levels of the Neuraxis

abnormalities should be reproducible. When slow RNSS are normal and MG remains a strong consideration, single-fiber EMG is performed because of its higher sensitivity.

Needle EMG Study Depending on disease severity, needle EMG study abnormalities may be present. With more severe disease, EDX features of motor unit disintegration (i.e., the early recruitment of short-duration, low-amplitude, polyphasic MUAPs), the presence of fibrillation potentials, and moment-to-moment variation (MMV) in waveform morphology as motor units sequentially fire. The presence of fibrillation potentials reflects chemical denervation of the muscle fibers, whereas the phenomenon of MMV reflects the constantly changing distribution of NMJ transmission failures within the motor unit generating the MUAP (i.e., the muscle fiber APs contributing to the compound MUAP differ with sequential firings of the motor unit). On the monitor, the MUAP demonstrates a constantly changing waveform morphology during sequential activation.

Lambert-Eaton Myasthenic Syndrome Background Lambert-Eaton myasthenic syndrome (LEMS) is the most common presynaptic disorder of NMJ transmission. Like MG, LEMS is an autoimmune disorder. With LEMS, the antibodies bind to voltage-gated calcium channels (VGCCs) of the presynaptic membrane, resulting in their destruction, the reduction of calcium ion entry, and presynaptic membrane disruption. In response to the decreased release of ACh vesicles, there is postsynaptic membrane hypertrophy (which increases the complexity of the folding pattern of the postsynaptic membrane). This, in turn, increases the number of postsynaptic ACh receptors and, hence, the likelihood of an ACh–ACh receptor interaction. The antibodies may be primary (autoimmune LEMS) or secondary (paraneoplastic LEMS, usually small-cell lung cancer) (Lennon et al., 1995). Because of the reduction in calcium ion entry with nerve terminal depolarization, the number of ACh vesicles released (quantal content) is significantly reduced. As a result, the number of postsynaptic muscle fiber APs generated per nerve terminal depolarization is significantly reduced. This pathophysiology also explains the mechanism of action of

3,4-diaminopyridine (DAP), which is used in the treatment of LEMS. This medication antagonizes DAP-sensitive presynaptic K+ channels, thereby delaying the onset of repolarization and, in effect, prolonging depolarization. The prolonged depolarization causes the VGCCs to remain open for a longer period of time, thereby increasing the entry of calcium ions and, hence, quantal content.

Routine NCS Because LEMS is a disorder of the NMJs, the sensory NCS are normal. Because of the significant reduction in ACh vesicle release, the motor responses are very low in amplitude, often less than 10% of the lower limit of normal. The CMAP abnormalities are generalized in their distribution. In our EMG laboratories, as a general rule, whenever the first low-amplitude motor response is encountered during motor NCS, a Lambert test is performed – the muscle demonstrating the low motor response is exercised for 10 seconds and the motor response is collected again. Ideally, restimulation is performed within 3 seconds of exercise cessation (Brown, 1984). When the motor response demonstrates a significant increment, we perform fast RNSS (discussed later here).

Repetitive Nerve Stimulation Studies On slow RNSS, as the quantity of ACh release decreases with each stimulation, the train of motor responses demonstrates a decrement. Because the baseline motor response is so low to begin with, this decrement is less noticeable. With fast RNSS above 5 Hz (e.g., 40 Hz), there is an increase in the intracellular calcium concentration within the axon terminal with repetitive nerve stimulation. As a result, there is a significant increase in the number of ACh vesicles released per stimulation. This causes the train of motor responses to demonstrate an incremental pattern, termed an incremental response. Typically, the increment results in responses several-fold larger than the baseline motor response. Because fast RNSS are somewhat uncomfortable for the patient, voluntary exercise of the muscle under study can be substituted. Although postactivation exhaustion may be observed, it usually is not sought.

Needle EMG Study Similar to MG, the MUAPs observed are short in duration, low in amplitude, and show early recruitment and moment-to-moment variation in their

253

Section 4: Other Pertinent EDX Information

waveform conformation. Fibrillation potentials, related to chemical denervation, are infrequent.

Myopathic Disorders The myopathies constitute a heterogeneous group of disorders with differing EDX manifestations, depending on the underlying etiology, that reflect the pathology, pathophysiology, level of severity, stage of disease, and the effect of treatment. Because myopathies involve muscle fibers, the sensory NCS are normal. In general, unless the studied individual has a distal myopathy, routine motor NCS are also normal. Even when proximal motor NCS are performed, the collected responses are often normal unless the disorder is severe. Consequently, most EDX abnormalities are observed during the needle EMG portion of the EDX study. As previously stated, these abnormalities vary with the underlying disorder, may be limited in their distribution (e.g., to the girdle or paraspinal muscles), and may be quite patchy in their distribution. On needle EMG, the observed MUAPs may be normal, may be short in duration, or may be both short in duration and low in amplitude. The presence of low-amplitude MUAPs with a normal duration is unexpected. Importantly, shortduration, low-amplitude MUAPs are not specific for myopathies and may be seen with other disorders producing motor unit disintegration (i.e., disorders of the terminal axon; disorders of the NMJ). With more advanced disease, MUAP recruitment is early because the force per motor unit is reduced, and thus, more motor units are required to generate a given level of contractile force. Because most myopathies have a proximal distribution, the needle EMG study should be expanded to include a greater number of proximal muscles. Of these, the paraspinal muscles are most likely to show abnormalities, followed by the adjacent limb girdle muscles (e.g., spinati; iliacus; glutei). In general, the brachioradialis and tibialis anterior are helpful midlevel limb muscles (Wilbourn, 1993). Other muscles with a high yield reflect the distribution of the specific myopathy. For example, with inclusion body myopathy, the wrist flexors (e.g., flexor carpi radialis), finger flexors (e.g., flexor digitorum profundus; flexor pollicis longus), and quadriceps muscles (e.g., rectus femoris; vastus lateralis) are more likely to demonstrate abnormalities on needle EMG. Because the abnormalities associated with myopathies may be

254

patchy in their distribution, a more thorough needle EMG examination is often required, including sampling more than one site within the same muscle. The EDX manifestations vary depending on the underlying etiology. Some myopathies are normal on EDX testing (termed bland myopathies), others show only changes in the conformation of the MUAPs (e.g., short duration, low amplitude; also termed bland myopathies), others demonstrate fibrillation potentials (termed necrotic myopathies), and still others are associated with myotonic potentials (myotonic disorders). These EDX manifestation differences help with the initial categorization of the myopathy. Examples of myopathies with no EDX abnormalities or with solely abnormalities of the MUAP waveform include congenital and metabolic myopathies; examples with associated fibrillation potentials include toxic myopathies, muscular dystrophies, and immune-mediated myopathies; and examples with associated myotonic potentials include myotonic dystrophy, congenital myotonias, and acid maltase deficiency. As with neurogenic disorders, fibrillation potentials indicate muscle fiber denervation. They are the most common form of spontaneous activity observed in myopathies. The denervation results from muscle fiber transection or muscle fiber splitting so that one portion of the affected muscle fiber becomes separated from its endplate region (i.e., it is no longer innervated). Fibrillation potentials are observed with a number of myopathies, including inflammatory myopathies (e.g., polymyositis; dermatomyositis), rapidly progressive dystrophies (e.g., Duchenne muscular dystrophy), toxic myopathies, infectious myopathies, rhabdomyolysis, acid maltase deficiency, and certain congenital myopathies (e.g., myotubular myopathy). When present, myotonic potentials also help identify the underlying disorder because they are seen with a limited number of etiologies. Complex repetitive discharges may also be seen. In addition to its diagnostic utility with myopathic disorders, EDX testing may also be helpful in selecting the ideal muscle to biopsy (i.e., one that shows abnormalities of moderate severity). To avoid biopsying muscle demonstrating needle EMG-related inflammatory and structural changes, the muscle identified as ideal for biopsy should be biopsied contralaterally. The needle EMG can also be used to distinguish conditions not easily differentiable on clinical examination. For example, when a polymyositis patient

Chapter 17: The Electrodiagnostic Manifestations of Disorders at Various Levels of the Neuraxis

develops weakness during steroid treatment, the weakness could be related to the use of corticosteroids or could represent an exacerbation of the polymyositis.

References Aguayo A, Nair CPV, Midgley R. Experimental progressive compression neuropathy in the rabbit. Arch Neurol 1971;24:358–364. Bolton CF. AAEM minimonograph #40: clinical neurophysiology of the respiratory system. Muscle Nerve 1993;16:809–818. Bromberg MB. Review of the evolution of electrodiagnostic criteria for chronic inflammatory demyelinating polyradiculoneuropathy. Muscle Nerve 2011;43:780–794. Brooks BR, Miller RG, Swash M, Munsat TL; World Federation of Neurology Research Group on Motor Neuron Diseases. El Escorial revisited: revised criteria for the diagnosis of amyotrophic lateral sclerosis. Amyotroph Lateral Scler Other Motor Neuron Disord 2000;1:293–299. Brown WF. The physiological and technical basis of electromyography. London: Butterworth, 1984:372-375. Deymeer F, Matur Z, Poyraz M, Battaloglu E, Oflazer-Serdaroglu P, Parman Y. Nerve conduction studies in Charcot-Marie-Tooth disease in a cohort from Turkey. Muscle Nerve 2011;43:657–664. Donofrio PD, Albers JW. ASEM minimonograph #34: polyneuropathy: classification by nerve conduction studies and electromyography. Muscle Nerve 1990;13:889–903. Ferrante MA. Brachial plexopathies: classification, causes, and consequences. Muscle Nerve 2004;30:547–568. Ferrante MA. Electrodiagnostic assessment of the brachial plexus. Neurol Clin 2012;30:551–580.

When fibrillation potentials are noted on NEE, polymyositis is favored, whereas their absence favors corticosteroid-induced weakness.

Ferrante MA. The relationship between sustained gripping and the development of carpal tunnel syndrome. Federal Practitioner 2016;33:10–15. Ferrante MA, Wilbourn AJ. The utility of various sensory nerve conduction responses in assessing brachial plexopathies. Muscle Nerve 1995;18:879–889. Ferrante MA, Wilbourn AJ. The characteristic electrodiagnostic features of Kennedy’s disease. Muscle Nerve 1997;20:323–329. Fowler CJ, Danta G, Gilliatt RW. Recovery of nerve conduction after a pneumatic tourniquet. Observation on the hind limb of the baboon. J Neurol Neurosurg, Psychiatry 1972;35:638–647. Friedrich JM, Robinson LR. Prognostic indicators from electrodiagnostic studies for ulnar neuropathy at the elbow. Muscle Nerve 2011;43:596–600. Frontera WR, Grimby L, Larsson L. Firing rate of the lower motoneuron and contractile properties of its muscle fibers after upper motoneuron lesion in man. Muscle Nerve 1997;20:938–947. Harding AE, Thomas PK, Baraitser M, Bradbury PG, Morgan-Hughes JA, Ponsford JR. X-linked recessive bulbospinal neuropathy: a report of ten cases. J Neurol Neurosurg Psychiatry 1982;45:1012–1019. Kennedy WR, Alter M, Sung JH. Progressive proximal spinal and bulbar muscular atrophy of late onset: a sex-linked trait. Neurology 1968;18:671–680. Kugelberg E. Accomodation in human nerves and the significance for symptoms in circulatory disturbances and tetany. Act Physiol Scand 1944;8(suppl)24:1–105.

Lennon VA, Kryzer TJ, Griesmann GE, O’Suilleabhain PE, Windebank AJ, Woppmann A, Miljanich GP, Lambert EH. Calcium-channel antibodies in Lambert-Eaton syndrome and other paraneoplastic syndromes. N Engl J Med 1995;332:1467–1474. Levin KH. L5 radiculopathy with reduced superficial peroneal sensory responses: intraspinal and extraspinal causes. Muscle Nerve 1998;21:3–7. McComas AJ, Sica REP, Campbell AJ, Upton AR. Functional compensation in partially denervated muscles. J Neurol Neurosurg Psychiatry 1971;34:453–460. Narayanaswami P, Geisbush T, Jones L, Weiss M, Mozaffar T, Gronseth G, Rutkove SB. Critically reevaluating a common technique: accuracy, reliability, and confirmation ias of EMG. Neurology 2016;86:218–223. Ochoa J, Fowler TJ, Gilliatt RW. Anatomical changes in peripheral nerves compressed by a pneumatic tourniquet. J Anat 1972;113:433–455. Sanders DB, Stalberg E. The overlap between myasthena gravis and Lambert-Eaton myasthenic syndrome. Annals New York Acad Sci 1987;505:864–865. Sunderland S. Nerves and nerve injuries, 2nd ed. Edinburgh: Churchill Livingstone; 1978. Wilbourn AJ. The electrodiagnostic examination with myopathies. J Clin Neurophysiol 1993;10:132–148. Wilbourn AJ. Diabetic neuropathies. In Brown WF, Bolton CF, editors, Clinical electromyography, 2nd ed. Boston: Butterworth-Heinemann; 1993:477–515.

255

Section 4: Other Pertinent EDX Information

Wilbourn AJ. The electrodiagnostic examination with hysteriaconversion reaction and malingering. Neurol Clin 1995;13:385–404. Wilbourn AJ. Brachial plexopathies. In Katirji B, Kaminski HJ, Preston DC,

256

Ruff RL, Shapiro BE, editors, Neuromuscular disorders in clinical practice. Boston: ButterworthHeinemann; 2002:884–904. Wilbourn AJ, Aminoff MJ. AAEE minimonograph #32. the electrophysiologic

examination in patients with radiculopathies. Muscle Nerve 1988;11:1099–1114. Wilbourn AJ, Furlan AJ, Hulley W, Ruschhaupt W. Ischemic monomelic neuropathy. Neurology (Cleve) 1983;33:447–451.

Chapter

18

Common Pitfalls and Their Resolution

Introduction It is important that EDX providers have an awareness of the large number of potential pitfalls that they might encounter during the performance of the various EDX studies. They should also appreciate the cause of those pitfalls and have a method for overcoming them. Although there is some overlap, the pitfalls encountered during the performance of EDX testing can be divided into those related to the environment, those related to the EMG machine, those related to the patient, and those related to the EDX provider.

Physiological Pitfalls Studying Cool Limbs In the opinion of the author, the biggest error made in the EMG laboratory is the performance of EDX studies on a cool limb. This error adversely affects every part of the EDX study (discussed later in this chapter) and must be avoided. The widespread effect that low temperature has on the EDX studies reflects the temperature sensitivity of many of the elements composing the neuromuscular system, such as voltage-gated ion channel opening and closing, enzymes (e.g., acetylcholine), NMJ transmission, and muscle fiber contraction. Thus, not surprisingly, when cool limbs are studied in the EMG laboratory, all components of the EDX study (i.e., NCS, late responses, RNSS, and needle EMG) are affected. When the unwary EDX provider does not recognize these low temperature alterations, erroneous report conclusions may be generated that, in turn, lead to potential patient mismanagement by the referring provider. For the aforementioned reasons, in our EMG laboratories, we expend a great deal of effort to limit the likelihood of cool limbs being studied. When we mail the patient their appointment letter, in addition to a description of what they can expect to occur

during their EDX study, we also include an instructions sheet. Among the instructions, they are asked to wear closed shoes with socks. We keep the thermostat in the EMG laboratories set at 75 degrees so that the room is slightly warm. In institutions where the room temperature is controlled by the engineering department, we have had to resort to having the engineering department weld the air vent shut so that cool air cannot enter the EMG laboratory. In this setting, the room is periodically cooled by opening the door between patients’ visits. Signs are placed on the EMG laboratory doors instructing the overnight housekeepers to keep the doors shut at all times, as leaving them open overnight leads to a cold room in the morning. Despite these efforts, a large percentage of our patients require limb warming. Although a number of techniques have been described, in the opinion of the author, the fastest and most effective technique is to place a plastic wastebasket liner into a tall plastic sink basin and then incompletely fill the basin with warm water. We do not use a thermometer to guide us, but simply fill the basin with warm (not hot) water. We then place the hands and distal forearms (upper extremity study) or the feet and distal legs (lower extremity study) into the warm water for 2–3 minutes. We dry the limbs well, wrap the contralateral limb in a dry bath towel, and immediately begin to study the ipsilateral limb. We do not use heating lamps for this purpose because they only warm the limb superficially. Thus, there is a discrepancy between the recorded skin surface temperature and the actual near-nerve temperature, at least initially (Halar et al., 1980). Although the deeper tissue elements eventually do warm up with heating lamps, the time required (15–30 minutes) is impractical (Bjorkvist et al., 1977; Geerlings and Mechelse, 1985; Franssen and Wieneke, 1994). We do not use chemical hot packs or reusable microwave packs because, although they may be as effective as a water bath, they take much longer.

257

Section 4: Other Pertinent EDX Information

Maintaining a warm limb can also be challenging. Many patients, especially those with a thinner body habitus, do not remain warm following a water bath and require repeat warming. We do not warm single limbs, but always warm either both upper extremities or both lower extremities (depending on which limbs are being studied). Once the limbs have been warmed, we wrap the contralateral limb in a towel and immediately begin to study the ipsilateral limb. Although heating lamps are likely useful to keep the ambient air surrounding the limb warm, we have not found this to be necessary when the contralateral limb is wrapped. We use thermistor tape to monitor the temperature of the extremity under study (from the beginning of the EDX study to its completion). We place the thermistor tape between the second and third metacarpal bones on the dorsum of the hand for upper extremity studies (at least 34 degrees Celsius) and over the dorsal aspect of the ankle for lower extremity studies (at least 32 degrees Celsius). Another issue is isolated distal extremity cooling. Frequently, in the setting of a cool limb, the distal aspects of the limb (i.e., distal to the thermistor tape) are much cooler than its more proximal portions (at or proximal to the thermistor tape). Thus, although the thermistor tape registers a temperature that meets the temperature requirements listed above, whenever the digits are cool to the touch, we warm the limb. Infrequently, we have had to warm an individual 3–4 times but, in most instances, 1–2 times suffices. Another advantage of warming the limb is that it softens the skin, thereby lessening its resistance and, hence, the intensity of the current required to achieve maximal stimulation. Although time is required for the warming process, that time is not wasted. During the initial warming period, we enter data into the EMG machine and perform other study-related activities. As the seasoned EDX provider is aware, there are certain signs that indicate that the collected NCS responses were recorded from a cool limb. For example, when the median sensory response is delayed, there may be underlying carpal tunnel syndrome. However, when the subsequent ulnar sensory response is also delayed, the patient should be warmed and the study repeated before too many more studies are performed. This is recorded on the worksheet. Another telling sign is in the setting of carpal tunnel syndrome, when the median motor response is delayed out of proportion to the sensory response; a semicolon is required rather than a comma the patient had likely cooled down when

258

the median motor response was performed. When these patterns are noted, we warm the limb and repeat the suspect NCS. Obviously, it is better to lose 3 minutes warming a patient than to lose 30 minutes repeating the NCS portion of the EDX study.

The Effects of Reduced Temperature on the EDX Study Nerve Conduction Studies When a cool limb is studied, the NCS are significantly altered. These changes are due to the gate effects that cold temperature has on the VGNCs, as originally described by Hodgkin and Katz in 1949. First, cooling results in delayed opening of the activation gate of the VGNC, which increases nerve activation time and, hence, the onset latency of the response. For each degree Celsius drop, the onset latency increases by approximately 0.2 msec. The distribution of the cooling (focal or more widespread) also plays a role in the observed conduction effects (Lang and Puusa, 1981). With focal cooling involving the tissue below the recording electrodes, the onset latency is delayed and the conduction velocity is relatively unaffected. However, when cooling is more widespread and includes the segment between the stimulating and recording electrodes, the activation time is prolonged at the stimulating electrodes and all of the nodes of Ranvier from the stimulating electrodes to the recording electrodes. This significantly reduces the conduction velocity through the summation of these sequential delays. In addition to the delayed opening of the activation gate of the VGNC, cooling also results in delayed closure of its inactivation gate. This delay allows more Na+ current to pass through the channel, thereby generating a much larger response (increased in amplitude and duration). Although cooling delays VGNC opening and closing, the delay in closure is more pronounced than the delay in opening (Hodgkin and Katz, 1949). Thus, the cold-induced increment in response amplitude may be more pronounced than the cold-induced increment in onset latency. In addition, delayed closure of the inactivation gate leads to lengthening of the absolute refractory period (Paintal, 1965). Generalized cooling also enhances temporal dispersion. For example, in a nerve in which the fastest axon conducts at 50 m/sec and the slowest axon conducts at 40 m/sec, over a 20-cm distance, the difference in their

Chapter 18: Common Pitfalls and Their Resolution

arrival times is 1 msec (i.e., the negative phase duration of the recorded response is 1 msec): CV ¼ distance=time; thus, time ¼ distance=CV Faster axon time ¼ 200mm=50mm=msec ¼ 4:0 msec Slower axon time ¼ 200mm=40mm=msec ¼ 5:0 msec Difference in arrival times ¼ 1 msec If the drop in CV equals 2 m/sec/degree C and there is a 10-degree Celsius drop (i.e., 20 m/sec), the faster axon now conducts at 30 m/sec (30 mm/msec) and the slower axon at 20 m/sec (20 mm/sec). Thus, at the cooler temperature, the difference in their arrival times increases to 3.3 msec: Faster axon time ¼ 200mm=30mm=msec ¼ 6:7 msec Slower axon time ¼ 200mm=20mm=msec ¼ 10:0 msec Difference in arrival times ¼ 3:3 msec Thus, the duration of the negative phase increases from 1 msec to 3.3 msec. This increase in temporal dispersion increases the amount of negative phase cancellation, thereby lowering the response amplitude. Given the typical temperature gradient of a cold limb (i.e., colder distally than proximally), antidromic sensory responses are more susceptible to cold-induced changes than are orthodromic sensory responses. Consequently, when a cool limb is studied, an unwary EDX practitioner may erroneously conclude that there is underlying demyelination (because of the cold-induced increase in onset latency and decrease in conduction velocity) or may fail to recognize underlying axon loss (because of the coldinduced increase in amplitude). These false positive and false negative conclusions, respectively, can have negative outcomes on clinical management and must be avoided. The sensory responses are more susceptible to these cold-induced effects than are the motor responses (Denys, 1991). At the extremes of temperature, conduction fails. Between these extremes, the relationship between conduction velocity and temperature was initially reported to be fairly linear. These reports led to the use of correction factors to correct for the cold-related conduction velocity reductions and latency prolongations. In 1956, Henriksen reported that over a nearnerve temperature range of 28–38 degrees Celsius, the median and ulnar motor nerve conduction velocity value dropped 2.4 m/sec for each degree Celsius drop (Henriksen, 1956). Subsequent authors calculated lower temperature correction factors ranging from 1.5–2.0 m/sec/degree Celsius drop (Buchthal and

Rosenfalck, 1966; Lowitzsch et al., 1977; Luden and Beyelar, 1977; Rutkove, 2001). A more recent manuscript reported 2.4 m/sec/degree Celsius drop for the median and ulnar nerves, 1.8 m/sec/degree Celsius drop for the peroneal nerve, and 1.1 m/sec/ degree Celsius drop for the tibial nerve (Denys, 1991). Although formulas are available to correct for the conduction velocity obtained from a cool limb (e.g., 2 m/sec loss per degree Celsius drop) (Paintal, 1965), these formulas do not correct for the amplitude increment. In addition to not correcting for amplitude increment, correction factors were calculated from normal nerves and, thus, may not be accurate when abnormal nerves are studied (Ashworth et al., 1998). Consequently, whenever a response suggests that the limb may be cool (e.g., a delayed latency with a larger than expected amplitude; multiple delayed latency values), we warm the limb and repeat the suspicious NCS. The effect of elevating the limb temperature has also been reported (Rutkove et al., 1997). At higher temperatures (44 degrees Celsius [111 degrees Fahrenheit]), the sodium channel inactivation gates close quicker, resulting in less Na+ current entry (reduces response size) and, potentially, heat-induced conduction block (Rogart and Stampfli, 1982). This effect is significant and, like that of cooler temperatures, more pronounced for the sensory responses. This phenomenon explains why patients with focal demyelination get weaker in hot environments (e.g., hot tubs); the already compromised fibers cannot tolerate further current decrements.

Repetitive Nerve Stimulation Studies Cooling affects several steps of neuromuscular junction transmission. For example, it prolongs the duration of influx of Ca++, the binding of ACh vesicles to the presynaptic membrane, and the binding of ACh to the postsynaptic receptor. It delays the generation of the endplate potential and slows the rate of hydrolysis of ACh by acetylcholinesterase (Rutkove, 2001). The net effect of these changes is that cooling improves the safety factor of neuromuscular junction transmission. This primarily reflects the increase in open time of the ACh receptor in the setting of cooling – more Na+ enters the myoplasm, increasing the amplitude of the MEPP and, in turn, the amplitude of the EPP. Thus, the depolarization threshold of the muscle fiber is easier to achieve. Thus, when slow repetitive nerve stimulation studies are performed on

259

Section 4: Other Pertinent EDX Information

the cool limb of a patient with myasthenia gravis, NMJ transmission is improved. This lessens (or eliminates) any CMAP train decrement. To avoid a false negative conclusion, when slow RNSS are performed on a patient with suspected myasthenia gravis, the limb must be warmed.

Needle EMG As with NCS, the resting membrane potential does not change significantly, and therefore, the quantity of insertional activity noted with needle advancement is not appreciably different. Regarding spontaneous activity, although the mechanism is unclear, fibrillation potentials may decrease in abundance or completely disappear when a cool limb is studied (Rutkove, 2001). Regarding voluntary activity, as with NCS, the duration and amplitude values of the voluntarily elicited MUAPs are larger due to increased inward Na+ current (i.e., delayed inactivation of the sodium channels). In general, the duration values are increased to a greater degree than are the amplitude values. This may give the false impression of reinnervation via collateral sprouting. Due to temporal dispersion, the number of polyphasic MUAPs is greater. MUAP recruitment is not affected. Cooling can also be used therapeutically. By increasing the duration of the VGNC open time, cooling can restore AP conduction to a nerve fiber with a demyelinating conduction block (Rasminsky, 1973). This improvement reflects an increase in the axonal safety factor. As previously discussed, the advancing Na+ current has approximately 5 times more current than what is required to advance to the next node. Focal demyelination diminishes this safety factor through increased current leakage (see Chapter 3). In addition, cooling may normalize a postsynaptic NMJ transmission defect. Cooling vests, by reducing the core temperature of the body, may permit some individuals who are just barely nonambulatory to become ambulatory. This may be desirable for short-term use, such as walking a daughter down the aisle.

Age-Related Issues Nerve Conduction Studies Nerve conduction velocities are affected by age. In term babies, the calculated values are approximately 50% of the adult values. These values increase over time and typically reach the adult values during the

260

teenage years (Falck and Stalberg, 1995). From the third decade, these values decrease at a rate of approximately 1 m/sec per decade (range 0.5–1.8 m/sec per decade) (Falck and Stalberg, 1995). Beyond the age of 60 years, the rate of decrement increases to about 3 m/sec per decade (Campbell, 1999). As conduction velocity decreases, latencies increase. Thus, the onset latencies (including F wave latencies) and peak latencies increase as the conduction velocity decreases. Along with these changes, the degree of physiological temporal dispersion, along with its negative effect on amplitude, also increases as the conduction velocity decreases. Although it is well known that sensory response amplitudes decrease with aging, it is unclear at what point, if ever, they eventually become unelicitable among apparently normal individuals. This issue is more of a concern for the lower extremities. Regarding the lower extremities, most EDX providers agree that lower extremity sensory responses are normally recordable under the age of 60 years. Thus, this issue only applies to individuals over the age of 60 years. In this age group, the majority have a recordable sural sensory response (Falco et al., 1994; Esper et al., 2005; Tavee et al., 2014). In a recent study of healthy elderly individuals, 96% (48 of 50) over the age of 60 years had an obtainable sural sensory response (Tavee, 2014). In that study, absent sural sensory responses were not observed until the age of 80 years. The lower limits of normal proposed by the authors of that study were 3 microvolts for those aged 60–70 years and 1 microvolt for those aged 70–74 years. Although such low values may have been challenging to record in the past, this is less of an issue with modern EMG machines and good technique. Thus, in conclusion, bilaterally absent sural sensory responses in individuals over the age of 75 years should not be considered abnormal. Of course, unilaterally absent sural sensory responses are abnormal at all ages. Although H waves have been reported to be absent in some apparently normal individuals over the age of 60 years, a prospective study of healthy elderly individuals did not support this statement (Falco et al., 1994). Among apparently normal individuals over the age of 70 years, mild abnormalities may be noted on NCS. The lower extremity motor NCS and the upper extremity sensory NCS may show mild amplitude reductions in amplitude and mild conduction velocity slowing. The motor NCS may be due to age-related

Chapter 18: Common Pitfalls and Their Resolution

motor neuron attrition. Thus, when we encounter borderline-mild abnormalities in the symptomatic limb, we compare them to the asymptomatic, contralateral side. When the other side shows similar findings, we do not report the changes as pathological. The challenge arises when both sides are symptomatic and both sides show these borderline-mild abnormalities. In these cases, we state the ambiguity of the findings and attempt to correlate them with the clinical examination. For example, the finding of low amplitude or absent sural responses in an individual without distal sensory complaints or sensory loss is less likely to be pathological.

Needle EMG On needle EMG, most likely due to the loss of anterior horn cells with age and the consequent reinnervation via collateral sprouting, the MUAP duration increases with aging. Reinnervation via collateral sprouting also results in an increase in muscle fiber density. Because the increase in muscle fiber density is much less pronounced than is the increase in the innervation ratios of the individual motor units, the MUAP amplitude is not affected to the same degree as the MUAP duration. Because aging involves all muscles equally, the MUAP duration relationship among the different limb muscles is preserved. For example, regarding the muscles of the upper extremity, the mean MUAP duration of the brachioradialis generally has a lower value than that of the other upper extremity muscles, whereas the mean MUAP durations of the triceps and deltoid muscles typically have higher values. The MUAPs of the extensor indicis muscle may also be of longer in duration than the other upper extremity muscles. For the lower extremity, MUAPs with the shortest duration are typically seen in the iliacus and the short head of the biceps femoris (and other hamstring muscles) muscles. The MUAPs of the gluteus maximus and medius muscles are also shorter in duration. At the other end of the spectrum, the MUAPs of the vastus lateralis (and other quadriceps muscles) tend to show the longest durations. When a muscle is studied by needle EMG and the MUAPs seem to be of longer duration than expected, it is helpful to immediately assess the homologous muscle on the contralateral side. When an obvious asymmetry is present, it can be graded and reported. When the units appear similarly on the two sides, they most likely are normal. To avoid false positive

conclusions, subtle side-to-side differences are not considered abnormal.

Body Habitus–Related issues As previously discussed, nerve conduction velocity is affected by axon diameter, myelin thickness, internode distance, and temperature, as well as by age (see discussion later in the chapter). However, even when age and temperature are controlled, variability among apparently normal individuals remains. The effect that body height has on the NCS measurements is unclear but appears to account for at least some of this variation, at least for the lower extremities. Because more proximal segments of a limb demonstrate faster conduction velocities than do more distal segments (Behse and Buchthal, 1971), it is likely that the distal segments of longer limbs conduct slower than the distal segments of shorter limbs. This is supported by the fact that the upper extremities conduct faster than the lower extremities. Many studies have shown an inverse relationship between body height and lower extremity nerve conduction velocity values, with taller individuals demonstrating slower velocities than shorter individuals (Campbell et al., 1981; Stetson et al., 1992). The magnitude of this relationship has been estimated. The conduction velocity of the lower extremity decreases by 2–3 m/sec for every 10-cm increase in height (Falck and Stalberg, 1995). The explanation for this inverse relationship between limb length and conduction speed is unclear. Because nerve conduction velocity values decrease with cooling, temperature was initially assumed to be responsible for this observation. In one study in which temperature was controlled, the inverse relationship between body height and nerve conduction velocity was not noted (Trojaborg et al., 1992). However, a large number of studies have clearly demonstrated that temperature gradients alone cannot fully account for the inverse relationship between limb length and conduction velocity (Arrigo et al., 1952; Spiegel and Johnson, 1962; Campbell et al., 1981; Soudmand et al., 1982; Fong et al., 2016). A number of other explanations have been suggested, including: (1) longer axons are thinner along their entire length (Zwarts, 1995); (2) there is axonal tapering distally (Soudmand et al., 1982); and (3) internodal segment length varies with distance along the nerve (Caruso et al., 1992). Because more distal segments of a limb conduct slower than

261

Section 4: Other Pertinent EDX Information

more proximal segments, this inverse relationship may simply reflect the fact that taller individuals typically have longer lower extremities than shorter persons. An inverse relationship between body height and upper extremity nerve conduction velocity values has not been noted (Soudmand et al., 1982).

Weight An inverse relationship exists between body mass index (BMI) and sensory response amplitude (Buschbacher, 1998; Buschbacher, 2003; Awang et al., 2006; Huang et al., 2009; Fong et al., 2016). The BMI reflects the mass of the body divided by the square of its height (typically expressed in kg/m2). The BMI is the weight normalized to height and permits an individual to be categorized as underweight, normal weight, overweight, or obese. The inverse relationship between BMI and sensory response amplitude likely reflects the thickness of the subcutaneous tissue separating the nerve fibers and the recording electrodes. Differences in digit girth exemplify this relationship. For example, when individuals with thick digits are studied in the EMG laboratory, the amplitude values of the digital sensory responses are lower than the values recorded from individuals with thin digits. On occasion, the sensory responses are mildly abnormal. Lower extremity sensory responses may not be elicitable in the setting of obesity. In our EMG laboratories, when obese individuals under the age of 40 years are referred for EDX assessment of suspected sensory polyneuropathy, and the routine sensory NCS (sural and superficial peroneal) are absent, we add the medial plantar and lateral plantar NCS, because individuals do not store fatty tissue on the soles of their feet. When the plantar responses are present, they indicate that the sural and superficial peroneal sensory responses are absent due to intervening adipose tissue, because a stocking-distribution sensory polyneuropathy would affect the plantar responses before the sural and superficial peroneal responses.

Gender-Related Issues Regardless of age, females have larger-amplitude antidromic digital sensory responses than do males. Presumably this reflects thinner digits among females (i.e., less subcutaneous tissue between the nerve fibers and the recording electrodes), as discussed above (Bolton and Carter, 1980). It is unclear whether a relationship between conduction velocity and gender

262

exists because gender and height are not independent variables (Flack and Stalberg, 1995).

Anomalous Innervations Anastomotic connections between the median and ulnar nerves in the forearm and hand produce changes in the innervation pattern of the hand intrinsic muscles. These variations are not infrequent. Because a number of these anomalous innervations have major effects on the NCS and the needle EMG, it is imperative that EDX providers recognize the NCS manifestations suggesting their presence and be familiar with the necessary steps to verify their suspicion. When these anomalies go unrecognized, erroneous EDX conclusions may result in clinical mismanagement (e.g., unnecessary surgery). This section reviews those anomalies pertinent to the performance of EDX medicine, the effects that they have on the EDX studies, and the techniques necessary to verify their presence. Anomalous innervations in the upper extremity include the Martin-Gruber anastomosis and the Riche–Cannieu anastomosis. In the lower extremity, the presence of an accessory deep peroneal nerve permits the dual innervation of the extensor digitorum brevis muscle.

Martin-Gruber Anastomosis Introductory Comments Among these anatomic anomalies, the most commonly encountered in the EMG laboratory is the Martin-Gruber anastomosis (MGA), named after Martin, a Swedish anatomist who initially reported the anomaly in 1763, and after Gruber, who subsequently reported it in 1870 (Gutmann, 1993). The exact incidence of an MGA, which is also referred to as a median-to-ulnar anastomosis or a median-toulnar crossover, is unclear. Although NCS have suggested an incidence of 15–32% among normal individuals, anatomical studies of aborted fetuses and other studies have suggested an incidence closer to 15% (Mannerfelt, 1966; Srinivasan and Rhodes, 1981). With this anomaly, some ulnar motor axons initially travel within the median nerve before crossing over to the ulnar nerve. Anatomically, the crossover usually occurs in the proximal one-third of the forearm and more frequently exits from the anterior interosseous nerve branch of the median nerve rather than from the main trunk of the median nerve (Srinivasan and Rhodes, 1981; Falck and Stalberg, 1995).

Chapter 18: Common Pitfalls and Their Resolution

This anomaly is frequently bilateral and may have an autosomal dominant mode of inheritance. In one study of 5 patients with an MGA, the anomaly was noted to be present in 62% of family members (Crutchfield and Gutmann, 1980). As expected of an anomaly with an autosomal dominant inheritance pattern, a gender difference has not been reported. As stated earlier, with an MGA, the crossover fibers are initially located within the main trunk of the median nerve, exit the median nerve (typically from the anterior interosseous nerve), travel across the forearm, and enter the main trunk of the ulnar nerve. They continue within the ulnar nerve and ultimately innervate one or more target muscles within the hand, including a hypothenar muscle (type I), the first dorsal interosseous (type II), or a thenar muscle (adductor pollicis or deep head of the flexor pollicis brevis; type III); of these, the most common target is the first dorsal interosseous muscle (Wilbourn and Lambert, 1976; Uchida and Sugioka, 1992). In one study of 22 limbs with an MGA, the FDI was the target in 21, a hypothenar muscle in 9, and a thenar muscle in 3 (Wilbourn and Lambert, 1976). Because this is a median-to-ulnar crossover, the median and ulnar motor nerve conduction studies are affected. The primary concept is that the median nerve contains more motor axons proximal to the crossover and less motor axons distal to the crossover, whereas the ulnar nerve contains less motor axons proximal to the crossover and more motor axons distal to the crossover. The specific NCS findings vary with the specific muscles innervated by the crossover branch and the number of nerve fibers composing the crossover branch. When conceptualizing the motor NCS findings associated with an MGA, it is easier to think in terms of three nerves (median, ulnar, and crossover nerves) – and, hence, three motor responses (median, ulnar, and crossover responses) – and realize that the crossover nerve moves from the median nerve to the ulnar nerve, and thus, the crossover response summates with the median response above the crossover and with the ulnar response below the crossover. Although the target ulnar nerve innervated muscle of the crossover varies, the crossover fibers always emanate from the median nerve. For this reason, its presence may be noted on median motor NCS when the proximal median motor response is larger in amplitude than the distal median response. When this pattern is observed, the presence of the anomaly can be confirmed by stimulating the ulnar

nerve at the wrist and elbow while maintaining the recording electrodes over the thenar eminence.

MGA to the Hypothenar Eminence When the crossover branch innervates one of the hypothenar eminence muscles (e.g., the abductor digiti minimi muscle), the routine ulnar motor nerve conduction study (recording hypothenar eminence) demonstrates a demyelinating conduction block (DMCB) pattern of motor responses with wrist and above-elbow stimulation. The wrist response is larger because it represents both the ulnar fibers and the crossover fibers, whereas the above-elbow response is smaller because it only represents the ulnar fibers. When this pattern is noted, it is important to stimulate the ulnar nerve below the elbow. With a DMCB across the elbow segment of the ulnar nerve, the wrist and below-elbow responses are higher in amplitude than the above-elbow response (see Figure 18.1). With an MGA, the below-elbow and above-elbow responses are both smaller in size than the wrist response, indicating that the “block” lies between the below-elbow and wrist stimulation sites. Because DMCBs in this segment of the ulnar nerve are quite uncommon, this finding is usually indicative of an MGA (see Figure 18.2). To differentiate an MGA from a DMCB across the elbow segment, further testing is required. With the recording electrodes left in place over the hypothenar eminence, the median nerve is stimulated at the wrist and at the elbow. In the setting of an MGA to the hypothenar musculature, because the median nerve does not innervate any of the hypothenar eminence

5 mV

5 ms Lat ms

AMP mV

2.8

9.1

7.5

8.3

11.8

.750

Figure 18.1 Ulnar neuropathy across the elbow segment. The above-elbow response is smaller than the below-elbow and wrist responses, consistent with a demyelinating conduction block involving the ulnar motor nerve fibers somewhere between the below-elbow and above-elbow stimulation sites (i.e., proximal to the below-elbow stimulation site and distal to the above-elbow stimulation site).

263

Section 4: Other Pertinent EDX Information

5 mV 5 mV

5 ms

Lat ms

Amp mV

Stimulus Site

Lat ms

Amp mV

2.9

7.6

1. Wrist

3.7

11.0

8.0

5.8

2. B.Elbow

6.9

4.7

10.8

5.8

3. A.Elbow

8.7

4.4

Figure 18.2 Martin-Gruber anastomosis to the first dorsal interosseous muscle. In this figure, the distal response is larger than both the below-elbow and above-elbow responses, indicating that the “block” lies proximal to the wrist stimulation site and distal to the below-elbow stimulation site. Because DMCB lesions are atypical in this segment of the ulnar nerve, whenever this pattern of ulnar motor responses is identified, a Martin-Gruber anastomosis should be considered and sought (see text).

muscles and because the crossover fibers are not in the median nerve at the wrist level, wrist stimulation generates no motor response. However, with median nerve stimulation at the elbow, a response appears because the crossover fibers are stimulated. The size difference between the proximal median and distal median motor response should roughly match the size difference between the ulnar motor responses obtained with wrist and below-elbow stimulation (see Figure 18.3).

MGA to the First Dorsal Interosseous Muscle Similar to what occurs with an MGA involving the abductor digiti minimi muscle, when the crossover fibers innervate the first dorsal interosseous muscle, the ulnar motor nerve conduction study (recording first dorsal interosseous) suggests a focal DMCB between the below-elbow and wrist stimulation sites because the wrist response is larger (ulnar and crossover nerve fibers are represented) and the below-elbow response is smaller (only ulnar nerve fibers are represented). Again, the above-elbow and below-elbow responses are the same in appearance, indicating that the discrepancy lies between the below-elbow and wrist stimulation sites and not across the elbow. As discussed earlier, to differentiate an MGA from a focal DMCB, the median nerve is stimulated at the wrist and at the elbow with the recording electrodes left in place over the FDI muscle. In the setting of an MGA to the FDI muscle, the median motor response with wrist stimulation is

264

5 ms

Figure 18.3 Martin-Gruber anastomosis to the hypothenar eminence. When a “block” pattern is noted between the belowelbow and wrist ulnar stimulation sites, an MGA is indicated and is sought by stimulating the median nerve at the wrist and at the elbow. The illustration demonstrates the crossover fiber motor response with median nerve stimulation at the elbow (trace 5), but not at the wrist (trace 4), confirming that the ulnar “block” is due to the presence of motor nerve fibers crossing from the median nerve to the ulnar nerve.

smaller than the response elicited with elbow stimulation. Although the median nerve fibers do not innervate the FDI muscle (they innervate muscle fibers of the thenar eminence), because the recording electrodes are positioned over the FDI muscle (i.e., on the lateral side of the hand), median nerve stimulation at the wrist generates a thenar motor response that is picked up by recording electrodes through volume conduction. With elbow stimulation, the recorded motor response reflects stimulation of the median fibers (generates the same volume conducted median motor response) plus stimulation of the crossover fibers (the FDI motor response). Again, the difference in the two median motor responses represents the magnitude of the crossover fibers and should be similar to the difference between the two ulnar motor responses (see Figure 18.4). As previously stated, the possibility of inadvertent stimulus spread must be considered and excluded.

MGA to the Adductor Pollicis Muscle Finally, when the crossover fibers innervate the adductor pollicis muscle (i.e., a muscle located in the thenar eminence area), the median motor nerve

Chapter 18: Common Pitfalls and Their Resolution

5 mV

5 ms

Stimulus Site

Lat ms

Amp mV

1. Wrist

3.7

11.0

2. B.Elbow

6.9

4.7

3. A.Elbow

8.7

4.4

Wrist

4.5

.417

A.Elbow

8.8

6.1

Figure 18.4 Martin-Gruber anastomosis involving the first dorsal interosseous muscle. In this case, median nerve stimulation at the wrist also shows a motor response. The latter reflects volume conduction from the thenar eminence to the surface recording electrodes overlying the first dorsal interosseous muscle. Because the thenar eminence is below the surface recording electrodes, it does not approach the recording electrodes (i.e., it is generated below them) and therefore has a negative polarity. In this case, the proximal median response is a composite of the volume conducted median response and the crossover fiber response. Thus, again, the difference between the two represents the crossover fiber response, which should be roughly equivalent to the difference between the ulnar below-elbow and wrist responses.

conduction study (recording thenar eminence) more typically suggests the crossover. The median motor response with wrist stimulation is smaller (stimulates the median nerve) than with elbow stimulation (stimulates the median nerve and the crossover fibers). In a verification process similar to the two described above, the recording electrodes are left in place over the thenar eminence and the ulnar nerve is stimulated at the wrist and below-elbow sites. With an MGA, the ulnar wrist response is larger (ulnar and crossover nerve fibers are stimulated) than the below-elbow response (only ulnar nerve fibers are stimulated). There will be no difference between the below-elbow and above-elbow ulnar motor responses. In summary, an MGA is suspected when there is a motor response discrepancy between the wrist and elbow stimulation sites for either the ulnar nerve or the median nerve. When an amplitude discrepancy is noted in one of these two nerves, the presence of an

MGA is verified by stimulating the other nerve at the wrist and elbow and looking for an equal but opposite amplitude discrepancy. The identification and characterization is more complicated when the crossover fibers innervate more than one target muscle, as they often do. In our EMG laboratories we calculate the size of the crossover to each muscle as described earlier. An approach to quickly verify the presence of dual targets is to stimulate the median nerve at the elbow, while recording from the hypothenar eminence (a motor response appears with an MGA) and then while recording from the first dorsal interosseous muscle (a motor response appears with an MGA). To avoid concomitant stimulation of the median and ulnar nerves with elbow stimulation, the median nerve is first stimulated at the elbow while recording from the thenar eminence. The minimal current required to achieve a maximal response is determined. This same intensity is then applied with the recording electrodes over the hypothenar eminence and then over the first dorsal interosseous muscle.

MGA with Concomitant Carpal Tunnel Syndrome With concomitant carpal tunnel syndrome, a number of EDX manifestations may be noted. Because the crossover fibers are contained within the median nerve at the elbow level but not at the wrist level, the median motor response with elbow stimulation has a positive dip (i.e., it is triphasic), whereas the median motor response with wrist stimulation has the typical biphasic morphology (Gutmann, 1977). The positive dip represents a volume conducted motor response from the crossover fibers to the recording electrodes (volume conduction has already been discussed; see Chapter 6). This occurs because the crossover fibers bypass the carpal tunnel, and thus are not delayed in the carpal tunnel as are the median nerve fibers. Because they are not delayed, the crossover response is detected by the recording electrodes before the median motor response is detected. Because the crossover response is generated away from the E1 electrode, it generates a positive phase as it approaches the E1 electrode (see Figure 18.5). The positive dip is more pronounced when the crossover fibers innervate a muscle nearer to the E1 recording electrode (e.g., adductor pollicis; first dorsal interosseous). In Figure 18.5, a positive dip is also present in both ulnar motor responses. This is easiest to understand when the involved nerve fibers are considered

265

Section 4: Other Pertinent EDX Information

5 mV

5 ms

Figure 18.5 Martin-Gruber anastomosis in the setting of carpal tunnel syndrome. For the 4 motor responses shown, the E1 and E2 electrodes are located over the thenar eminence using the bellytendon method. The first response was recorded with median nerve stimulation at the wrist and the second response with median nerve stimulation at the elbow. The distal median response is low in amplitude and its onset latency is delayed, indicating both axon loss and demyelination involving the median nerve. Because of the delayed onset latency, the demyelinated fibers permit localization to between the wrist stimulation site and the recording electrodes (i.e., carpal tunnel syndrome). The proximal median motor response is larger and includes a positive dip, suggesting possible inadvertent ulnar nerve stimulation at the elbow. Once this technical error is excluded, the only other possibility is a Martin-Gruber anastomosis (to account for the amplitude discrepancy) coupled with carpal tunnel syndrome (to account for the positive dip limited to just the proximal motor response). By stimulating the ulnar nerve at the wrist (response 3) and at the elbow (response 4), this hypothesis is verified. There is a 1.3 mV difference between the two median motor responses and, in the opposite direction, between the two ulnar motor responses. The two ulnar motor responses both show a positive dip, which represents the approaching phase of the ulnar motor response (see text for discussion of this phenomenon).

as median fibers, crossover fibers, and ulnar fibers. Because the distal median motor response stimulates only median fibers, there is no dip. The proximal median motor response reflects stimulation of median and crossover fibers, so it is larger than the distal median motor response. Also, because the crossover fibers bypass the carpal tunnel, they arrive first. Because they innervate ulnar nerve innervated muscles located away from E1, there is a positive dip

266

at the onset of the response. With distal ulnar nerve stimulation, the crossover fibers and the ulnar fibers innervate muscles away from E1, and hence, a positive dip is present at the onset of the response. Finally, with proximal ulnar nerve stimulation, because the innervated muscles are located away from the E1 electrode, a positive dip is generated. In addition, when this anomaly occurs in the setting of carpal tunnel syndrome, the positive dip with proximal stimulation causes the calculated conduction velocity to be spuriously fast. This occurs because the two waveforms represent different nerve fiber populations (the crossover and median fibers with elbow stimulation and only the median fibers with wrist stimulation). The onset latency of the proximal response is generated by the crossover nerve fibers, which are not delayed in the carpal tunnel, whereas the onset latency of the distal response is generated by the median nerve fibers, which are delayed in the carpal tunnel. Because two different fiber populations are being compared, the Δt is lower, thereby spuriously increasing the calculated nerve conduction velocity value (i.e., CV = Δd/Δt, therefore, when the Δt value decreases, the CV value increases). It is important to be familiar with this concept, because when patients with carpal tunnel syndrome have normal routine NCS (including palmar mixed NCS), the presence of a positive dip with elbow stimulation but not with wrist stimulation indicates its presence (Guttman, 1977). With more advanced carpal tunnel syndrome (in which the median motor response is delayed even further), both the positive phase and the negative phase of the crossover motor response may precede the negative phase of the median motor response, thereby causing the negative phase to have two peaks with elbow stimulation (Lambert, 1962). When the crossover motor response has two peaks, the proximal median response may demonstrate three peaks (see Figure 18.6). After technical error is excluded, the ulnar nerve should be stimulated at the wrist and below-elbow sites seeking an MGA. Otherwise, the presence of the MGA is identifiable during the ulnar motor NCS (see Figure 18.7).

MGA with a Concomitant Ulnar Neuropathy With ulnar neuropathies situated proximal to the crossover (e.g., the elbow), the lesion will appear less severe in the presence of an unrecognized MGA, because some of the ulnar nerve-innervated muscles are spared because they receive their innervation via

Chapter 18: Common Pitfalls and Their Resolution

5 mV

5 ms

Lat ms

Amp mV

7.5

1.7

8.8

2.4

Figure 18.6 The proximal median motor response has three peaks, and the negative area under the curve value of the proximal response is greater than that of the distal response. After a technical error is excluded, an MGA should be sought (see text for explanation and see Figure 18.7 for verification).

the crossover fibers. For example, with a complete ulnar neuropathy at the elbow segment (e.g., complete axon loss or complete DMCB), an above-elbow response is lacking (because no action potentials can traverse the lesion), whereas a wrist response is present (due to the unaffected crossover fibers). Thus, the complete elbow segment lesion mimics a partial DMCB situated between the wrist and above-elbow stimulation sites. The underlying pathology is determined by the response evoked with below-elbow stimulation. When the elbow lesion is complete DMCB, the below-elbow stimulation evokes a response that reflects the ulnar nerve fibers proximal to the crossover. Its size difference between the above-elbow and below-elbow responses reflects the magnitude of the block. When the lesion is complete axon loss, the below-elbow response is also absent. As with any MGA, the median motor response is larger with elbow stimulation than with wrist stimulation.

Atypically Proximal MGA Much less frequently, the fibers of the MGA cross at a more proximal site. When the crossover site lies proximal to the below-elbow stimulation site, the conduction block pattern of the MGA with ulnar nerve stimulation occurs between the below-elbow and above-elbow stimulation sites rather than between the wrist and below-elbow stimulation sites. Consequently, it truly mimics a DMCB across the elbow segment. When the possibility of an MGA is not considered and sought using the techniques described earlier, an inaccurate conclusion and possible mismanagement may result. Conversely, when an MGA is considered and the median is stimulated at the elbow and wrist without relocating the recording electrodes, a median response difference equivalent in magnitude to the

5 ms Lat ms

Amp mV

3.5

15.1

7.3

7.8

10.0

6.4

5.2

.333

9.1

7.3

Figure 18.7 Martin-Gruber anastomosis to FDI muscle in the same patient shown in Figure 18.6.

ulnar response difference across the elbow segment verifies its presence. Consequently, the standard practice of screening for an MGA whenever an amplitude discrepancy is noted between the wrist and below-elbow ulnar nerve stimulation sites should also be applied whenever an amplitude discrepancy exists across the elbow. In one report, the crossover occurred 3 cm distal to the medial epicondyle (Uncine et al., 1988). Thus, the ideal below-elbow stimulation site might be 3 cm below the medial epicondyle, which would still be distal to the cubital tunnel, but would be proximal to these unusually proximal MGAs (Madras and Midroni, 1999). There are two other EDX manifestations that may suggest the presence of an unusually proximal MGA rather than a DMCB across the elbow: (1) the routine median motor NCS response shows an amplitude discrepancy consistent with an MGA (i.e., the median motor response with elbow stimulation is larger than the median motor response with wrist stimulation) and (2) whenever the “block” is large, the lack of a neurogenic MUAP recruitment pattern during needle EMG of the ulnar nerve innervated muscle showing the amplitude discrepancy.

267

Section 4: Other Pertinent EDX Information

Situations in Which MGA Is Unrecognizable In addition, there are three situations in which an MGA anomaly is not discernible by EDX testing: (1) when the crossover innervates the flexor digitorum profundus, (2) when the crossover contains only a small number of axons, and (3) when the crossover occurs proximal to the elbow joint (Lee et al., 2005).

Ulnar-to-Median Anastomosis in the Forearm The opposite crossover (ulnar-to-median anastomosis) has been reported, but is quite rare (Gutmann, 1993). This anomaly is also referred to as the reversed MartinGruber anastomosis and the Marinacci communication (Unver Dogan et al., 2009). As expected, when unrecognized, its presence causes confusing EDX findings. The median and ulnar motor responses with distal and proximal stimulation show discrepancies, but the discrepancies have a relationship opposite to that of the MGA – the median motor response is larger with wrist stimulation than with elbow stimulation, whereas the ulnar motor response is larger proximally (aboveelbow and below-elbow stimulation) and smaller distally (wrist stimulation). Similar to MGA with a concomitant ulnar neuropathy, an ulnar-to-median crossover in the setting of a severe median neuropathy will be underestimated or potentially misinterpreted without stimulating the ulnar nerve proximally and distally while recording from the median nerve. A carpal tunnel syndrome equivalent is not expected because Guyon canal lesions typically produce axon loss or DMCB rather than DMCS.

Riche–Cannieu Anastomosis The hand intrinsic muscles are innervated by the median and ulnar nerves. In general, the median nerve innervates the abductor pollicis brevis, opponens pollicis, and the first and second lumbricals, whereas the ulnar nerve innervates the hypothenar muscles, third and fourth lumbricals, dorsal and palmar interossei, and adductor pollicis. The flexor pollicis brevis is frequently innervated by both nerves, with the superficial head innervated by the median nerve and the deep head innervated by the ulnar nerve. Anomalous communications between the median and ulnar nerves are common and can present significant EDX challenges when not considered (as well as confusing clinical findings).

268

In 1897, an ulnar-to-median nerve crossover at the thenar eminence was independently described by Riche and by Cannieu (Riche, 1897; Cannieu, 1897), and therefore, this anomaly is often referred to as a Riche–Cannieu anastomosis (RCA). With an RCA, there is a neural connection between the deep branch of the ulnar nerve and the recurrent branch of the median nerve that supplies motor axons to the abductor pollicis brevis or opponens pollicis muscles. The prevalence of this anomaly is unclear, ranging from 15% to 77% (Cannieu, 1897; Harness and Sekeles, 1971; Ajmani, 1996). As expected, this can lead to confusing EDX findings when unrecognized. As with the MGA anomaly, this anomaly may also have an autosomal dominant inheritance pattern (Boland et al., 2007). On NCS, an isolated low-amplitude median motor response is observed, suggesting a median neuropathy. The median motor response is low in amplitude with an RCA because some of the thenar eminence musculature is ulnar nerve innervated (i.e., not activated by median nerve stimulation at the wrist). The onset latency of the median motor response is normal, the median sensory responses are unaffected, and the needle EMG study is normal. When a patient with an RCA develops an ulnar neuropathy proximal to the crossover, needle abnormalities in both the ulnar nerve and the median nerve distribution may result in lesion mislocalization to the medial cord, lower plexus, C8 or T1 root, or C8 or T1 spinal cord segments (e.g., motor neuron disease), or the patient may be thought to have concomitant involvement of both nerves (Saperstein and King, 2000). When a median neuropathy is located proximal to the crossover and the RCA is not recognized, it will be underestimated (with an MGA, an unrecognized ulnar neuropathy proximal to the crossover is underestimated). For example, when a patient with an RCA develops a complete median neuropathy, it may be misinterpreted as a partial median neuropathy because some of the traditionally median nerve innervated muscles are being innervated by the ulnar nerve via the crossover branch.

Identifying an RCA When the distal median motor response is reduced in amplitude and an RCA is suspected, its presence can be verified by stimulating the ulnar nerve distally and proximally while recording from the thenar eminence. In the setting of an RCA, a small positive dip

Chapter 18: Common Pitfalls and Their Resolution

may be noted due to volume conduction from a muscle near the E1 electrode (i.e., one of the ulnar nerve innervated thenar eminence muscles). At both stimulation sites, the stimulus intensity should be kept as low as possible to avoid co-stimulation of the median and ulnar nerves.

Other Hand Muscle Innervation Patterns In addition to the RCA, many other hand muscle innervation patterns have been reported. In one study of 226 individuals, a number of variations were identified (Rountree, 1949). Of the 226 individuals, in 33%, the three thenar eminence muscles were innervated by the median nerve; in 32%, both heads of the flexor pollicis brevis muscle were innervated by the ulnar nerve; in 15%, the innervation of the flexor pollicis brevis was shared between the two nerves; in 2%, all three thenar eminence muscles were ulnar nerve innervated (ulnar hand); and in 1%, the median nerve innervated the thenar eminence muscles and the adductor pollicis. Indeed, in this report, the “typical” innervation of the flexor pollicis brevis muscle (median nerve innervation of the superficial head and ulnar nerve innervation of the deep head) was observed in a minority of the individuals. Other authors have also reported hand muscle innervation patterns different from the traditional description provided in anatomy textbooks (Kimura and Ayyar, 1984; Dumitru et al., 1988). For this reason, whenever abnormalities involve muscles in the ulnar nerve and median nerve domain and the abnormalities are limited to the hand, the EDX provider must consider the possibility of a nontraditional hand muscle innervation pattern, especially when only a single involved muscle lies outside of the muscle domain of the involved nerve. The abductor pollicis brevis muscle is only rarely innervated by the ulnar nerve, and the abductor digiti minimi muscle is infrequently innervated by the median nerve.

Berretini Anastomosis With this anomaly of sensory axons, there is a communication between the common digital nerve branches of the ulnar and median nerves (Don Griot JPWD et al., 2000). For example, a communication between the ulnar fourth common digital nerve and the median third common digital nerve result in variations in digital sensory patterns. Because the incidence of this anastomosis exceeds 80%, it is

considered a normal structure rather than an anatomic variation (Unver Dogan et al., 2009).

Accessory Deep Peroneal Nerve The peroneal motor NCS, recording extensor digitorum brevis (EDB), is routinely performed in most EMG laboratories. This muscle extends the second through fifth digits and is innervated by the deep peroneal nerve. On occasion, this muscle receives dual innervation – it is partially innervated by the deep peroneal nerve (the medial aspect of the muscle) and partially innervated by the superficial peroneal nerve (the lateral aspect of the muscle). When the superficial peroneal nerve gives off a branch to the EDB muscle, the branch is termed the accessory deep peroneal nerve. Typically, this branch exits the superficial peroneal nerve at the mid-leg level, descends distally along the leg, and then passes behind the lateral malleolus to reach the EDB muscle. The presence of an accessory deep peroneal nerve is much less frequent than an MGA.

Identifying an Accessory Deep Peroneal Nerve In most individuals, the deep peroneal nerve innervates the extensor digitorum brevis muscle. Infrequently, this muscle is innervated by both the deep peroneal nerve and the superficial peroneal nerve. When the superficial peroneal nerve provides partial innervation to the extensor digitorm brevis muscle, the nerve branch is termed the accessory deep peroneal nerve. In the presence of an accessory deep peroneal nerve, on motor NCS, when the deep peroneal nerve is stimulated at the ankle level, the motor response is smaller than when the common peroneal nerve is stimulated at the knee level. This occurs because ankle stimulation only activates the deep peroneal nerve fibers innervating the medial aspect of the EDB muscle. Because stimulation at the knee level activates the common peroneal nerve (i.e., the deep peroneal and the superficial peroneal nerves), both portions of the EDB muscle are activated, and hence, the recorded response is larger. Thus, when the amplitude of the peroneal-EDB motor response with proximal stimulation in the lateral aspect of the popliteal fossa is larger than that with distal stimulation at the ankle, an accessory deep peroneal nerve should be suspected. Its presence is verified by stimulating the accessory deep peroneal nerve behind the lateral malleolus, which generates a motor response (the lateral

269

Section 4: Other Pertinent EDX Information

5 ms × 5 mV

2 1

was presented in the first two chapters of this textbook (see Chapters 1 and 2) and will only be briefly reiterated in this chapter, where appropriate. The technical pitfalls can be divided into issues related to recording, stimulating, filtering, amplification, and other issues.

Issues Related to Recording Electrode Impedance and Impedance Mismatch

7

Figure 18.8 An accessory deep peroneal nerve. When the routine deep peroneal motor NCS is performed, recording EDB, the proximal motor response is larger than the distal motor response, suggesting the possibility of an accessory deep peroneal nerve. This is verified by stimulating behind the lateral malleolus (i.e., the site through which the accessory deep peroneal nerve passes) and recording a motor response of size roughly equivalent to the difference of the routine proximal and distal peroneal motor responses.

component of the EDB motor response) roughly equivalent to the difference. In other words, summation of the two ankle motor responses (the deep peroneal nerve and the accessory deep peroneal nerve) should approximate the motor response recorded with popliteal fossa stimulation (see Figure 18.8).

Technical Pitfalls The EDX provider must fully understand basic electricity, the electrical concepts pertinent to EDX medicine, and the EMG machine in order to perform quality EDX studies. The majority of this information

270

Some of the issues relevant to this section have already been discussed and, hence, are not repeated in detail here (see Chapters 6, 7, and 8). The primary issue concerns the impedance between the surface recording electrodes and the skin. As discussed in Chapter 1, the recording electrodes are in series with the amplifier. Because voltage drops across the resistor in proportion to its resistance (resistors are voltage dividers), the impedance of the surface recording electrode must be minimized so that the majority of the voltage drops across the amplifier (so that it appears on the monitor). Because the amplifiers of modern EMG machines have extremely high input impedances (often exceeding 1 million ohms), this is less of an issue now than it was in the past. However, the resistance across the two surface recording electrodes must be identical (i.e., there should be no impedance mismatch) so that any voltage differences between them reflect the desired signal (differential) rather than the undesired noise (common signal).

The Presence of a Positive Dip Because the muscle fiber APs are generated in the endplate region (termed the motor point of the muscle), the motor response has a biphasic morphology rather than the triphasic morphology typically observed when monophasic responses propagate within a volume conductor. With motor responses, the initial positive phase (the approaching phase) is absent, because the muscle fiber APs are generated below the E1 electrode and, consequently, do not approach it (propagate toward it). When the E1 electrode is not situated over the motor point of the muscle, the motor response is not generated below it, and as a result, through volume conduction, the muscle fiber APs propagate toward the E1 electrode, resulting in the generation of a triphasic response. For this reason, whenever an initial positive phase is identified, referred to as a positive dip, the E1 electrode should be repositioned

Chapter 18: Common Pitfalls and Their Resolution

2 mV 5 msec

Figure 18.9 The top trace represents the median motor response of the author. The middle response was recorded after the E1 electrode was repositioned 1 cm distal to its original position (i.e., 1 cm closer to the E2 electrode), whereas the lower response was recorded with the E1 electrode positioned 2 cm distally (i.e., 2 cm closer to the E2 electrode). The duration of the positive phase (i.e., the duration of the positive dip) increases with distance from the correct position. In addition, the proximity of the E1 and E2 electrodes results in signal cancellation and significant motor response diminution.

and the stimulus reapplied. The duration of the positive phase gives an indication of the distance that the E1 electrode is from the motor point (see Figure 18.9). When the positive dip persists despite E1 electrode repositioning, other explanations should be considered (e.g., stimulus spread; muscle atrophy; anomalous innervation). When stimulus spread activates nerve fibers of an adjacent nerve and that nerve innervates muscle fibers in the vicinity of the recording electrodes, a positive dip may be observed when the inadvertently activated muscle fibers conduct toward the E1 electrode. Stimulus spread is discussed in detail later here. In the setting of significant muscle atrophy at the recording site, volume conduction from neighboring muscle fibers may produce a positive deflection (see Figure 18.10). When a positive dip precedes the proximal median motor response but not the distal one, the combination of a Martin-Gruber anastomosis with carpal tunnel syndrome should be suspected (this was discussed in detail earlier in this chapter). A positive dip may be observed in an individual with

2 mV

5 ms Lat ms

Amp mV

5.1

2.6

Figure 18.10 The median motor response shows a positive dip that is related to thenar muscle atrophy. (Note that the voltage setting is 2 mV per division rather than the usual 5 mV per division. This was done to make the positive dip more apparent.)

a Martin-Gruber anastomosis who does not have carpal tunnel syndrome when the proportion of crossover fibers is much greater than the number of median nerve fibers. In this scenario, the number of APs advancing toward the E1 electrode exceeds the number of APs that are generated below the E1 electrode.

Issues Related to Stimulation Stimulus (Shock) Artifact Like the compound APs traversing nerve trunks, the stimulator generates an electrical field at the cathode that, through volume conduction, instantaneously spreads throughout the body, particularly across its surface (the skin). The strength of this electrical field, like that of any electrical field, diminishes over distance. More specifically, the force decreases exponentially with distance as the field line density decreases (the field decays at 1/r3). Consequently, locations equidistant from the cathode have similar potential (voltage) values, which can be depicted as isopotential lines (see Figure 18.11). Because the E1 surface recording electrode is closer to the cathode than is the E2 electrode, it is on a stronger isopotential line (voltage) than is the E2 electrode. Because they are on different isopotential lines, there is a voltage difference between them. In other words, the stimulus artifact appears differently at the two electrodes (differential signal), and therefore, their voltage difference is amplified by the differential amplifier. Due to its amplification, this unwanted electrical activity, termed stimulus artifact, is large in magnitude and causes the tracing to deviate away from the baseline (i.e., upward or downward from the baseline). Depending on its duration, it may interfere with the subsequently recorded NCS response. The amplitude of this phenomenon may be reduced through rotation of the anode about the cathode (discussed further on).

271

Section 4: Other Pertinent EDX Information

Figure 18.11 The isopotential lines of the cathode (negative) and anode (positive) are shown, as well as their relationship to the E1 and E2 electrodes when the stimulator is aligned with the E1 and E2 electrodes.

Although the pulse of stimulation current passes through the body nearly instantaneously, its effect on the amplifier is more prolonged, as evidenced by the slow return of the trace to baseline. Consequently, when the distance between the cathode of the stimulator and the E1 electrode is short, the monitor may display the stimulus artifact overlapping with the desired waveform, thereby interfering with the recording (i.e., the waveform begins before the stimulus artifact terminates). As compared to motor responses, because of the much smaller size of sensory and mixed responses, stimulus artifact is an even bigger issue, especially when there is a short distance between the cathode and the E1 electrode. When the impedances of the two recording electrodes are different (termed impedance mismatch), the voltage drop related to the stimulus artifact will differ even more (see Chapter 1), further increasing the amplitude of the stimulus artifact through differential amplification. Impedance mismatch at the two recording electrodes occurs when the resistance of the skin at the two sites differs, when the volume of conductive paste at the two sites differs, or when the degree of attachment of the two electrodes differs (i.e., one is not firmly attached to the skin). Except for indicating the onset of the stimulus pulse, stimulus artifact has no value, and efforts should be made to reduce it.

Types of Stimulators There are two types of stimulators, constant current stimulators (0–100 mA) and constant voltage stimulators (0–400 V). With either stimulator type, the resistance of the patient is constant. Thus, based on Ohm’s law, for a given skin resistance (R = V/I), a constant current stimulator varies the voltage to deliver the current value selected by the EDX provider, whereas a constant voltage stimulator varies the current to achieve the voltage selected by the provider. The electrodes of the stimulator act as capacitors and, thus, build up charge when the stimulus is

272

delivered. This accumulated charge is the greatest contributor to the charge passed to the recording electrodes as stimulus artifact (Hua, 2006). Because the capacitance of a capacitor is proportional to its surface area (see Chapter 1), the amount of charge accumulated is proportional to the surface area of the stimulating electrodes (because larger surface areas hold larger charge quantities). A larger charge requires a greater amount of time to dissipate and, thus, is associated with a stimulus artifact of higher amplitude and longer duration. Thus, as stated earlier, when the stimulating and recording electrodes are closer to each other, the baseline drift may distort the signal. With constant current stimulators, the output resistance (impedance) between the stimulator and the cathode and between the stimulator and the anode is high, whereas it is much lower at the cathode-skin interface and the anode-skin interface. Recall that in a series circuit, the current is constant (I = V/R), and hence, the voltage is proportional to the resistance (i.e., the resistors act as voltage dividers) (see Chapter 1). Thus, because much of the voltage drop occurs within the stimulator rather than at the electrode-tissue interfaces, the amplitude of the stimulus artifact is reduced; this is the main advantage of this type of stimulator. However, because of the high internal impedance of the stimulator, the stimulus charge (artifact) cannot dissipate back into the stimulator. This lengthens the duration of the stimulus artifact, which is the main disadvantage of this type of stimulator. Thus, with constant current stimulators, the stimulus artifact is lower in amplitude but longer in duration. With constant voltage stimulators, the output resistance (impedance) between the stimulator and the anode and between the stimulator and the cathode is low. For this reason, more voltage drops at the electrode-tissue interfaces, increasing the amplitude of the shock artifact, which is the main disadvantage of this type of stimulator. On the other hand, because of the low output resistance of the stimulator, there is a low-resistance pathway available for the stimulus artifact that increases its rate of dissipation, thereby decreasing the duration of the stimulus artifact. This is the main advantage of this type of stimulator. Thus, with constant voltage stimulators, stimulus artifact is shorter in duration but higher in amplitude.

Stimulus Artifact Reduction The ability to reduce stimulus artifact is a required skill for the practitioner of EDX medicine. There are a

Chapter 18: Common Pitfalls and Their Resolution

number of methods available to reduce this unwanted electrical activity. The size of the stimulus artifact is proportional to the size of the stimulus current delivered. Thus, the maximal response should be elicited using the lowest stimulus intensity and duration possible. For the same reason, the stimulating cathode must be situated directly over the nerve so that the distance between it and the nerve trunk under study is as short as possible. When it is not positioned directly over the target nerve fibers, the greater amount of intervening tissue requires a stronger stimulus and, therefore, generates a larger stimulus artifact. One of the most important techniques to reduce stimulus artifact is to minimize the impedance (resistance) at the stimulating and recording electrode interfaces with the skin. This is accomplished by (1) ensuring that the electrodes are clean; (2) skin preparation, such as light skin abrasion (e.g., abrasive tape; skin preparation gel); and (3) massage of electrolyte lotion into the abraded sites. Ultrasound gel should not be substituted for electrolyte lotion because it does not contain electrolytes and because it is a fire hazard. Reducing the resistance (impedance) at the stimulation site reduces the stimulus current required to generate a maximal response, thereby lessening the stimulus artifact (as stated earlier). In addition, it reduces the baseline noise and the stray capacitance between the stimulating electrode leadwires and the ground leadwire, as does the use of a ground with a larger surface area. The impedance of the skin between the cathode and the E1 electrode should be kept high to limit transcutaneous current flow (i.e., shock artifact). (Indeed, the impedance of the skin between all four of the surface electrodes should be high.) Transcutaneous flow between the cathode and the E1 electrode is facilitated when the skin impedance between them is reduced, such as by sweating or intervening conductive lotion. Placing the ground electrode between the stimulating and recording electrodes is somewhat helpful, but not as helpful as it was in the past, because the ground electrode is no longer part of the grounding system. The previous generation of EMG machines used earth grounds, which were part of the grounding system, whereas, for reasons related to patient safety, the current generation of EMG machines use iso-grounds (discussed in Chapter 19). Because iso-grounds have only minimal chargecarrying capacity, they are less effective at reducing

shock artifact (see Chapter 2). In addition, the impedance between the cathode and the skin should be lowered (electrode lotion; light abrasion), which lessens the amount of stimulus current required to activate the nerve and which favors current flow through the skin rather than along it. Minimizing the lengths of all leadwires limits the capacitive coupling that occurs between them. In addition, maintaining a greater distance between the stimulator leadwires and the recording leadwires is also helpful. One way to achieve this separation is to place the stimulator cable behind the neck and shoulder area and to move the headbox as close as possible to the limb under study. In addition, the leadwires can be wadded up. In our EMG laboratory, we use shielded acquisition cables. With the latter, the E1, E2, and ground leadwires are contained within a cable that is designed to lessen the effect of environmental noise on them. Shorter power cords are better than longer ones, and extension cords should be avoided. Rotation of the Anode about the Cathode Although these techniques lessen the shock artifact, they do not eliminate it. The stimulus artifact that remains despite the use of good technique (described above) can be reduced further by rotating the anode around the cathode while maintaining the cathode in its original position (Kornfield, 1985). Recall that the E1 and E2 electrodes are at different distances from the stimulator cathode and, thus, experience different voltages related to the stimulus artifact because the electrical field projects radially from the cathode (i.e., they are on different isopotential lines). By rotating the anode about the cathode, the orientation of the isopotential lines with respect to the E1 and E2 recording electrodes is optimized (i.e., the voltage difference between them is reduced because they occupy isopotential lines that are closer to each other) (see Figure 18.12), thereby lessening their signal difference and, thus, the size of the stimulus artifact resulting from differential amplification. Unfortunately, because the stimulation pulse is not a two-dimensional event, but a three-dimensional one, there are also subsurface equipotential lines with more complicated patterns. In addition, there is a tendency for the equipotential stimulus artifact lines to align themselves perpendicular to the limb, thereby negating the effect of the rotation (Netherton, 2010). For these reasons, the ideal orientation of the cathode and anode is not necessarily perpendicular to the

273

Section 4: Other Pertinent EDX Information

Figure 18.12 Anode rotation reorients the isopotential lines surrounding the cathode with respect to the surface electrodes so that the isopotential lines that they occupy are more equivalent, thereby reducing the amplification of the voltage related to the stimulus itself (i.e., thereby reducing the shock artifact). For convenience, the isopotential lines surrounding the anode are not shown.

nerve. Fortunately, the monitor can be used to identify the ideal degree of rotation. To do this, the anode is initially rotated (either clockwise or counterclockwise) about the cathode and the magnitude of the shock artifact on the monitor is assessed. The anode is then rotated further in the same direction and assessed again. When it increases in size, the anode is rotated in the opposite direction, and when it decreases in size, the anode is rotated in the same direction. This approach is continued until the reduction in stimulus artifact is maximally reduced. If the polarity of the stimulus artifact changes (i.e., when it moves to the opposite side of the baseline), then the rotation went too far (see Figure 18.13). Final Options Other electrical equipment in the room can be unplugged, if necessary, to further reduce the environmental electrical noise. Because incandescent lighting generates less electromagnetic interference than fluorescent lighting does, if necessary, the fluorescent lights can be turned off and an incandescent floor lamp located in the corner of the room can be turned on. With incandescent lighting, a wire filament is heated by an electric current until it glows with visible light (incandescence), whereas with fluorescent lighting, an electric current in the gas excites mercury

274

vapor, which then produces short-wave ultraviolet light throughout the bulb that causes a phosphor coating on the inside of the lamp to glow. The stimulator can be situated further from the recording electrodes, which will provide more time for the trace to return to baseline so that it does not overlap with the desired response. This approach invalidates the normal control value of the latency, which can be adjusted by adding 0.2 msec for every centimeter of added distance. Unfortunately, this does not rectify the negative AUC lost due to greater physiological temporal dispersion. Consequently, when we resort to this method, we perform the same study on the contralateral side for comparison purposes. When all of these measures fail, the low-frequency filter setting can be increased so that the duration of the stimulus artifact decreases, allowing the trace to return to baseline quicker. Because modifying the low-frequency filter affects the morphology of the recorded response, in our EMG laboratories, fortunately, we have never had to resort to this approach to lessen the stimulus artifact. The Future A number of advances in stimulator technology have led to the production of stimulators that generate significantly lower-stimulus artifact (e.g., hybrid stimulators; biphasic stimulators). With hybrid stimulators, the first part of the stimulus uses constant voltage (to shorten the duration of the shock artifact), whereas the second part of the stimulus uses constant current (to lower the amplitude of the shock artifact) (Nilsson et al., 1988; Netherton, 2010). Because the stimulus pulse itself is a monophasic square wave, the stimulating electrode-tissue interfaces fill with charge (because the interfaces behave as capacitors). With biphasic stimulators a biphasic pulse is employed to reduce the amount of charge buildup. The first phase (the monophasic negative square wave) generates the desired action potential volley and charges the stimulating electrode–tissue interfaces, whereas the second phase (a monophasic positive square wave of equal dimensions) discharges the unwanted charge buildup, thereby lessening the stimulus artifact (Netherton, 2010).

Submaximal Stimulation Whenever a nerve is submaximally stimulated, the amplitude and the negative area under the curve values are submaximal because not all of the axons

Chapter 18: Common Pitfalls and Their Resolution

10 mV

are contributing to the response. In addition, when the fastest fibers are not stimulated, the latency is prolonged and the calculated conduction velocity value is reduced. Thus, it is important to ensure that the delivered stimulus maximally activates all of the nerve fibers of the nerve under study. With motor NCS, the nerve under study is stimulated at two sites, first distally (e.g., the wrist) and then proximally (e.g., the elbow). It is important to realize that the stimulus intensity utilized to achieve a maximum response distally may not generate a maximal response proximally due to greater nerve depth at the more proximal stimulation site. Thus, although the final stimulus strength used to collect the maximal distal motor response can be applied proximally to collect an initial proximal motor response, the intensity must then be incrementally increased until the proximal motor response is maximal. When this is not done and the proximal motor response is submaximal, the calculated conduction velocity is erroneously slowed. The EDX provider must deliver the appropriate stimulus strength at each stimulation site to be sure that the recorded response is a maximal response. This is especially true for proximal stimulation sites where the nerve lies deeper. For example, the median nerve is much more superficial at the wrist stimulation site than it is at the elbow stimulation site. Thus, the stimulus strength required to generate a maximal

2 ms

Figure 18.13 The effect of anode rotation on the polarity and amplitude of the shock artifact. Upper trace: With the stimulator held in line with the E1 and E2 electrodes, a large-amplitude upward deflection is present in the top trace. Middle trace: With rotation of the anode about the cathode, the amplitude of the shock is reduced and the orientation is maintained (upward). Lower trace: With further rotation of the anode, the amplitude is markedly reduced and the polarity reverses. Thus, the direction and amplitude of the shock artifact can be used to identify the ideal degree of anode rotation that maximally reduces the shock artifact. Given that shock artifact present in the upper trace did not interfere with the motor response, anode rotation was unnecessary. To make this concept more readily appreciated, the recording was made using 10 mV and 2 msec divisional dimensions.

motor response with wrist stimulation is typically lower than that required for elbow stimulation. Consequently, when a maximal motor response is generated at the wrist level and then the same stimulus strength is applied at the elbow level, the proximal motor response is likely to be submaximal, spuriously suggesting the presence of focal demyelination – a conduction block when the proximal response is lower in amplitude and conduction slowing when the fastest fibers are not completely activated by the submaximal stimulus. This error is more likely to occur with ulnar motor NCS. In general, the stimulus strength required to achieve a maximal motor response is typically greatest for the below-elbow stimulation site, lowest for the wrist stimulation site, and intermediate for the above-elbow stimulation site.

Excessive Stimulation In addition to causing a larger-stimulus artifact and more patient discomfort, the larger electrical fields created by excessive stimulation generate other problems that may generate erroneous conclusions when unrecognized. The latter include stimulus lead (distal migration of the virtual cathode) and stimulus spread (inadvertent stimulation of adjacent nerve fibers results in the generation of additional, unwanted motor responses that, when in the vicinity of the recording electrodes, contaminate the desired motor response).

275

Section 4: Other Pertinent EDX Information

Stimulus Lead Regardless of the dimensions of the cathode, nerve fiber stimulation occurs below its center (Henriksen, 1956). However, when excessive stimulus current is delivered, the nerve trunk may be activated more distally. When this occurs, the measured value of the onset latency is spuriously decreased (Pinelli, 1964). This phenomenon is referred to as stimulus lead, distal migration of the virtual cathode, or distal migration of the effective cathode. It occurs because the strength of the electrical field generated at the cathode increases and spreads out radially, causing activation of the nerve trunk distal to the cathode. The resultant reduced onset latency value may result in a false negative conclusion when it normalizes a mildly delayed response. Thus, in general, the supramaximal stimulus should not exceed 10% to 20% of the intensity required to generate the maximal response. When abnormal nerve fibers are being studied (i.e., those with higher depolarization thresholds), a 20% increment may not be high enough. For this reason, it is important to monitor the waveform as the stimulus intensity is increased. A sudden decrease in the onset latency suggests stimulus lead. A sudden change in the morphology of the waveform indicates inadvertent activation of an adjacent nerve (stimulus spread). Stimulus Spread It is important to position the stimulating cathode directly over the nerve so that the distance between the cathode and the target axons is as short as possible. When the cathode is positioned lateral to the nerve, in addition to delaying the latency somewhat (i.e., delayed activation time), more stimulus current is required to maximally stimulate the nerve under study. Not only does this increase stimulus artifact (discussed earlier), but the excessive radial directed stimulus current may inadvertently excite a nearby nerve. When this occurs and the unintentionally excited nerve innervates a muscle in the vicinity of the recording electrodes, an undesired motor response results that affects the desired motor response (e.g., preceding it, summating with it, or canceling a portion of it). If the undesired motor response reaches the E1 electrode before the desired motor response, a positive dip will appear. When the positive dip reflects stimulus spread, relocation of the E1 electrode does not eliminate it. When the undesired motor response reaches the E1 electrode at the same time as the desired signal does, there is

276

signal cancellation between the volume conducted positivity of the unwanted motor response and the negative phase of the desired motor response, causing the desired motor response to be smaller in size. Stimulus spread is more likely where nerves lies close to each other, such as at the palm, wrist, and antecubital fossa stimulation sites (where the median and ulnar nerves are in close proximity), at the axilla (where the median, radial, and ulnar nerves are near each other), and at the popliteal fossa (where the tibial and common peroneal nerves are in close proximity).

Machine Maximum Stimulation Sometimes the machine is limited. For example, when the H reflex is performed on individuals with larger leg girths, although the H wave may be normal in size, the M wave may be reduced in amplitude. As previously stated (see Chapter 11), lower levels of stimulation generate the H wave, which then decreases in size as the stimulus current is increased further. Still further increases lead to the appearance of the M wave, which increases further as the stimulus is increased further, until a maximal M wave is collected. If the M wave is increasing in size when the maximal stimulus strength available from the EMG machine is reached, then it is not maximal. Consequently, a low-amplitude M wave in this setting is not abnormal. This is a limitation of the EMG machine to fully excite the tibial motor axons. In our EMG laboratory, we identify this as a submaximal stimulus on the worksheet (rather than a submaximal response) and include this information in the EDX report so that non-EDX providers do not misinterpret the low amplitude as being pathological in nature.

60 Hz Power Artifact In addition to stimulus artifact, the other major electrical issue with which the EDX provider must contend is the 60 Hz AC signal related to environmental sources (e.g., wiring in the walls and ceiling of the EMG laboratory), the EMG machine itself (e.g., capacitive and inductive current; leakage current), and the leadwires connecting the EMG machine to the patient (ground loops; stray capacitance; stray inductance). Despite the magnitude of 60 Hz artifact, good technique (e.g., avoidance of impedance mismatch) and quality instrumentation (differential amplification) are typically able to prevent it from interfering with the EDX study (see Chapter 2). Briefly, the 60 Hz

Chapter 18: Common Pitfalls and Their Resolution

power artifact reflects the generation of expanding and collapsing electrical and magnetic fields (stray capacitance and inductance) related to the continuous direction changes of the AC signal. The latter induce currents in the EMG machine circuitry, and because resistances exist throughout these circuits, voltage is also generated. These effects are minimized by using a differential amplifier with a high input impedance and a high common mode rejection ratio, by ensuring that there is no impedance mismatch between the two surface recording electrodes (e.g., the electrodes and their leads are of identical composition and properly affixed). With the needle electrode recording surface exposed to the environment, the 60 Hz AC signal in the room can be recorded on the screen. The 60 Hz AC signal may appear as a higher-frequency signal when its individual components are not recognized. The time period of a cycle of 60 Hz AC signal is 16.7 msec (1 cycle  1,000 msec/60 cycles = 16.7 msec). Thus, the waveform on the monitor should be assessed for a repeating pattern with a 16.7 msec interval (see Figure 18.14). When the needle electrode is held exposed to the environment, the 60 Hz AC signal is more pronounced. Once the examiner touches the patient with the other hand, the amplitude of the 60 Hz AC signal immediately decreases. This occurs through a shunting effect (see Figure 18.15). As the concentric needle electrode is advanced into the patient and the E1 and E2 recording surfaces are removed further from the environment, the 60 Hz AC signal diminishes in amplitude. With a monopolar needle electrode, however, because the E2 component is a surface electrode, it remains exposed to the environment, and thus, significant 60 Hz AC signal may persist and interfere with the recording. In this scenario, contact by the examiner with the E2 electrode can be used to shunt the AC signal away from the E2 electrode and improve the recording. Averaging Historically, because the sensory responses were smaller than the noise generated by the amplifier, techniques to enhance the signal-to-noise ratio (SNR) of the sensory response were developed. This topic was previously discussed (see Chapter 8). When the desired signal is repeatedly elicited and averaged, the SNR is enhanced, assuming the noise is not synchronous to the desired signal. In other words,

200 uV

10 ms

Figure 18.14 The 60 Hz AC signal is not always obvious when the waveform is complicated. In this case, black vertical lines are positioned at each cycle. The interpotential interval is just under 17 msec, which is consistent with 60-Hz artifact.

because the major unwanted signal is the 60 Hz activity in the environment, the nerve should not be stimulated at a frequency of 60 Hz (or a rate that would repeatedly place the signal on the same portion of the 60 Hz sine wave). The mathematical relationship between the SNR and the number of stimuli is: SNR α 1=ðstimulation#Þ1=2 This relationship shows that the SNR is proportional to the inverse of the square root of the number of stimulations. For example, collecting 64 stimulations enhances the SNR 8-fold. Although averaging can be quite helpful for eliminating environmental noise, especially when the desired responses are small (e.g., sensory responses), we employ this technique infrequently.

Measurement Errors The calculated motor conduction velocities are based on the change in distance (Δ distance) between the proximal and distal stimulation sites, divided by the change in time (Δ time) between the onset latencies of the proximal and distal motor responses. The EMG machine determines the time difference but requires that the distance difference be entered by the EDX provider. It then uses these two pieces of information to calculate the conduction velocity. (Because velocities have direction, nerve conduction velocities are more accurately termed nerve conduction speeds. To avoid confusion, the inappropriate term, velocity, is used in this textbook.)

277

Section 4: Other Pertinent EDX Information

200 uV

200 uV

10 ms

10 ms

Figure 18.15 Upper panel: 60 Hz AC signal recorded with the needle electrode held motionless within the air adjacent to the patient. Note that the complex waveform repeats every 16.7 msec (60 Hz frequency). The subcomponent with a bifid peak is marked so that the frequency of the signal is more apparent. Lower panel: The amplitude of the 60 Hz AC signal shown in the upper panel is significantly reduced when the examiner touches the patient with the other hand, allowing some of the 60 Hz AC signal to be shunted (to ground). This phenomenon is analogous to the shunting of chassis current along an individual when that individual contacts an electrical device with an accumulated charge, such as a lamp chassis with a two-pronged plug. This concept is discussed in detail in Chapter 19, which deals with electrical safety.

Regarding the time measurement, although the EMG machine measures the time difference, when the cursors are improperly positioned, the measurement is inaccurate and affects the calculated conduction velocity. Most cursor placement errors result from forgetting to position the cursors rather than from placing them in the wrong position on the trace. Importantly, when the cursors are positioned on the tracing, the screen sensitivity setting should be identical to the one used to collect the normal control values of the NCS techniques used by the EMG laboratory. Many EDX practitioners change the sensitivity to magnify the response during cursor placement. When this approach is used, it causes the onset site to appear earlier and, therefore, may result in a false negative conclusion (e.g., a mildly prolonged latency may be reported as normal). Regarding the distance measurement, when the surface distance between the stimulation sites is inaccurately measured, the calculated velocity becomes inaccurate. The effect that the error has is inversely proportional to the distance. Thus, for a given measurement error, the effect is larger for a shorter distances than for a longer distance. For example, for a

278

20 mm distance (2 cm), a 2 mm measurement error represents a 10% measurement error (2/20 = 0.1), whereas for a 200 mm distance (20 cm), the same 2 mm measurement error represents only a 1% measurement error (2/200 = 0.01). As previously stated in this textbook, the measured surface distance between the two stimulation sites underestimates the actual nerve fiber length, because the nerve fibers composing the nerve are not taut (see Chapter 16). However, for the most part, the surface measurement between successive stimulation sites correlates well enough for the calculation of conduction velocity. However, because nerve fibers are especially redundant where they cross joints, the degree of conduction velocity underestimation is much greater. This is why, for example, when the ulnar motor NCS shows an isolated drop of CV across the elbow segment, it is most likely technical. Although conduction velocity drops across the elbow occur with ulnar neuropathies, they do not occur in isolation. With axon loss or demyelinating conduction block involving the fastest conducting fibers, the amplitude and negative area under the curve values are reduced. With focal demyelinating conduction slowing involving the fastest fibers, there is a change in waveform morphology, because the affected fibers are differentially affected (nonuniform slowing). When we measure the distance across a joint, we use the same angle used to collect our normal control values. For the ulnar nerve, for example, we use 135 degrees. It is important that the extremity be stimulated in this position as well so that the stimulation sites and the measurement sites correlate (i.e., do not stimulate the limb in the extended position and then flex it for the measurements). The relationship between the surface distance and the actual nerve distance is more challenging in the setting of excess tissue (e.g., obesity; edema) or at sites where the length of the nerve under study is significantly longer than the skin surface distance, such as across joints (e.g., the ulnar nerve segment across the elbow). We measure the distance using a tape measure with millimeter gradations (e.g., 207 mm). We round the amplitude and latency values generated by the EMG machine to the nearest tenth (e.g., 10.7 mV and 10.7 msec) and the calculated conduction velocities to the nearest whole number (e.g., 52 m/sec). Finally, although the surface measurement between successive stimulation sites correlates fairly well with the length of the segment of

Chapter 18: Common Pitfalls and Their Resolution

nerve being stimulated along the extremity, this is not the case across the brachial plexus (i.e., the distance between the supraclavicular fossa stimulation site and the axillary stimulation site does not reflect the distance of the nerve segment studied). The use of calipers to perform the measurement does not overcome this issue.

Stimulator Reversal When the stimulator is placed over the nerve so that the anode is closer to the E1 electrode (rather than the cathode) and nerve activation occurs below the cathode, the latency value will increase by the amount of time required to conduct across the added distance (i.e., the distance between the cathode and the anode). For example, if the cathode and anode are 2.5 cm apart and the conduction velocity of the nerve is 50 m/sec, then the latency will be increased by 0.5 msec (time = distance/velocity = 25 mm/50 mm per msec = 0.5 msec). This will not affect the calculated conduction velocity when the stimulator is oriented in reverse at both sites (because the time difference between the onset latencies of the two responses does not change). When the stimulator is properly positioned at the proximal stimulation site and reversed at the distal stimulation site, the Δt value is erroneously lower in value. This causes the calculated conduction velocity to be an overestimate (CV = Δd/Δt). When the stimulator is properly positioned at the distal stimulation site and reversed at the proximal stimulation site, the Δt time is erroneously greater in value. This causes the calculated conduction velocity to be an underestimate (CV = Δd/Δt). Because of the negative charge of the cathode and the positive charge of the anode, the cathode represents the depolarizing pole of the stimulator (because it causes the outside of the axolemma to become more negative than the inside of the axolemma, thereby depolarizing the membrane), whereas the anode is the hyperpolarizing pole of the stimulator (because it causes the outside of the axolemma to become even more positive than it is at rest, thereby hyperpolarizing the membrane). Theoretically, the hyperpolarized segment would block APs from traversing it, termed anodal block. Although this electrical phenomenon has been demonstrated (Ranck, 1975), it requires a stimulus current strength in excess of that deliverable by standard EMG machines and, hence, does not occur in clinical practice (Dreyer et al., 1993). Instead, and by an unclear mechanism, depolarization may

occur below the anode (Dreyer et al., 1993). When this happens, the calculated conduction velocities would not be affected by the stimulator reversal. Although the stimulus strength required for anodal depolarization is higher than that required for cathodal depolarization, it is not outside the limits of the EMG machine. Consequently, when the stimulator is reversed, low stimulus intensities generate cathodal depolarization and an increased onset latency value. At stronger stimulus intensities, the anode causes nerve fiber depolarization, and the onset latency is unaffected. However, the recorded response will have two peaks, an earlier one generated by anodal depolarization (because the anode is nearer to the E1 electrode) and a later one generated by cathodal depolarization (because the cathode is further from the E1 electrode). As the stimulus current intensity is increased further, the anodal peak increases in size and the cathodal peak decreases in size. The cathodal peak diminishes in size because the APs generated at the anode propagate bidirectionally. Those propagating proximally collide with the APs generated at the cathode that are propagating toward the anode (similar to the AP collisions that reduce the size of the H wave) (Dreyer et al., 1993).

Submaximal Stimulation When a nerve is submaximally stimulated, the amplitude and the negative area under the curve values will be reduced because not all of the axons are contributing to the response. Thus, the latency value of the response will be delayed. It is important to realize that the stimulus intensity utilized to achieve a maximum response distally may not generate a maximal response proximally due to greater nerve depth more proximally. Thus, after recording the distal response with the ideal stimulus intensity, the same stimulus intensity may be applied proximally (i.e., it is not necessary to lower the intensity and slowly build up), but after the proximal response is recorded, the intensity must be increased to ensure that it is a maximal response. Otherwise, the calculated conduction velocity will be erroneously decreased and may lead to unnecessary interventions.

Filtering Issues When we filter out particular frequency ranges from a biologic signal composed of a wide range of frequencies, the recorded response is altered – the

279

Section 4: Other Pertinent EDX Information

characteristics of the remaining frequencies become more pronounced, whereas those of the filtered frequencies become less pronounced. Because these filters are removing signal, the amplitude and the negative AUC values of the recorded responses are reduced in magnitude. With EDX medicine, low-frequency filters, high-frequency filters, bandpass filters, and notch filters are employed. Because each region of the recorded responses is composed of varying proportions of high- and low-frequency elements, filtering has different effects on different waveform regions, depending on the primary frequency range of that region. For these reasons and others (discussed further on), the EDX provider must understand filtering and how it affects the various waveform measurements. The low- and high-frequency filter settings employed should be the same ones used to collect the normal control values for that particular study. As stated earlier, with filtering, because signal is being removed, the amplitude and negative area under the curve values are reduced. With lowfrequency signal filtering, the higher signals dominate the recorded response. Because the onset latency reflects high-frequency components, it is not affected by low-frequency filtering. Because the rise time is predominantly composed of higher frequencies, lowfrequency filtering does not significantly affect it. Because the sensory and motor NCS responses contain more low-frequency components than highfrequency components, the effect of low-frequency filtering on amplitude and negative AUC is more pronounced than its effect on latency. Because there is significant loss of total signal, the compound response peaks earlier and terminates earlier. Therefore, the peak latency, negative phase duration, and total duration values of the response are lower. As a result, whenever the lower frequencies are overfiltered, the degree of signal loss may falsely suggest an axon loss process. Also, the number of phases or turns may be increased by low-frequency filtering due to the unmasking of the higher frequencies. These are the general changes that are observed with lowfrequency filtering. When the response being filtered has predominantly higher frequencies, these changes are less pronounced, whereas when the response is composed predominantly of lower frequencies, these changes are more pronounced. Because motor responses contain relatively more low-frequency components than do sensory responses, low-frequency

280

5 mV

5 ms

Lat ms

Amp mV

3.5

11.6

3.5

9.9

3.5 3.5

8.6 8.6

3.5

4.1

Figure 18.16 The effect of progressively increasing the lowfrequency filter setting. In this illustration, the degree of lowfrequency filtering is progressively increased from the top trace to the bottom trace (5 Hz; 20 Hz; 100 Hz; 500 Hz). As the low-frequency filter setting is raised, the amplitude and negative AUC values markedly decrease. Although this causes the traces to peak earlier and to terminate sooner, the onset latency, which reflects higherfrequency components, is not significantly affected (it is unaffected in this recording).

filtering has a more pronounced effect on motor responses than on sensory responses. The figure shows the effect of progressively increasing the degree of low-frequency filtering on the recorded motor response (see Figure 18.16). With high-frequency filtering, the higher frequencies are removed and the lower frequencies are more apparent. Again, because signal is being removed, the response will be smaller in size (i.e., decreased amplitude value). Because high frequencies are removed, those regions of the response composed solely or predominantly of higher frequencies are the most affected. Because the onset of the curve is composed of higher frequencies, the onset of the response is delayed in its appearance, and hence, the onset latency value increases. Because the rise time contains a larger proportion of higher frequencies, it is prolonged, and therefore, the response peaks later. Thus, the value of the peak latency increases. Because the lower frequencies predominate, the duration of the response increases. The response appears smoother because

Chapter 18: Common Pitfalls and Their Resolution

5 mV

laboratory. The effects of filtering on each EDX study component are discussed next.

5 ms

Nerve Conduction Studies

Lat ms

Amp mV

3.3

8.3

3.6

8.3

4.0

8.3

Figure 18.17 The effect of progressively increasing the highfrequency filter setting. In this illustration, the degree of highfrequency filtering is progressively increased from the top trace to the bottom trace (10,000 Hz; 1,500 Hz; 500 Hz). As more and more higher frequencies are removed, the onset latency value progressively increases and the response peaks later and later. The amplitude and negative area under the curve values of the recorded responses are not significantly affected.

high-frequency filtering reduces the sharper contours of the waveform. The figure shows the effect of high-frequency filtering on the median motor response (see Figure 18.17). As the degree of the high-frequency filtering is progressively increased, the onset latency value increases and the curve peaks later and later (due to the prolonged rise time), whereas the amplitude and negative AUC values are much less affected. Because sensory responses contain relatively more high-frequency components than do motor responses, high-frequency filtering has a more pronounced effect on the sensory responses, including obvious decrement of amplitude and negative AUC. As with motor responses, with high-frequency filtering, the onset latency and the peak latency are affected to a greater extent than are the amplitude and the negative AUC. Because the filter settings have a significant effect on the collected responses, the filter settings used should be the same ones used to collect the normal control values that are being used by the EMG

Sensory responses are small in size, and therefore, the display sensitivity is usually set in the 10–20 microV/division range. At this sensitivity setting, the sensory response is well amplified and easily assessed. However, the undesired signal, including the noise within the baseline, is also amplified. To limit this problem, the cutoff frequency setting of the highfrequency filter is reduced (e.g., to 2,000 Hz). Because the motor responses are so much larger than the sensory responses, they do not require as much amplification, and consequently, the resultant baseline noise amplification is less of a problem. For this reason, the cutoff frequency setting of the high-frequency filter can be set much higher (e.g., 10,000 Hz). The cutoff frequency setting of the low-frequency filter for motor NCS is usually around 20 Hz. Although this sacrifices signal (because low-frequency signal constitutes a significant proportion of the electrical signal of the motor response), the use of lower settings (e.g., 2 Hz) results in significant baseline instability, making lower settings impractical (Pease and Pitzer, 1990).

Needle EMG In 1954, Buchthal and colleagues reported the effect of low-frequency filtering on MUAPs (Buchthal et al., 1954). In that study, they collected MUAPs using lowfrequency filter settings ranging from 1 Hz to 100 Hz. Distortions of the MUAP duration were noted when the low-frequency filter setting was raised to 7 Hz. They concluded that a low-frequency filter setting of 2 Hz was optimal for recording the duration of the MUAP. Because the range of frequencies composing MUAPs and fibrillation potentials extends to a much higher value, the high-frequency filter cutoff should not be reduced below 10,000 Hz without the risk of significant distortion. In 1954, Buchthal and colleagues also showed that reducing the high-frequency filter setting from 25,000 Hz to 10,000 Hz resulted in a 5% loss in MUAP amplitude. However, reductions below 10,000 Hz led to exponential amplitude losses. Regarding MUAP duration, the high-frequency filter setting could be reduced to 2,000 Hz without a significant effect on MUAP duration. In summary, to

281

Section 4: Other Pertinent EDX Information

avoid MUAP duration distortions, the low-frequency filter setting should not exceed 2–3 Hz, and to avoid significant distortions of the MUAP amplitude, the high-frequency filter setting should not be reduced below 10,000 Hz. This frequency bandpass is still recommended today. With monopolar needle electrodes, however, the manufacturers recommend that the low-frequency filter be set at 20 Hz (Campbell, 1999).

Conceptual Pitfalls Not Using the Technique Used to Collect the Normal Control Values When EMG laboratories are shared by multiple providers, some of the providers may have their own techniques, their own normal values, and, thus, their own opinions regarding normal and abnormal. More than one technique for the same NCS poses a challenge for the EMG technicians. In addition, when a patient returns to the same EMG laboratory for a follow-up study, and the criteria applied by the second practitioner differ from those of the first practitioner, the diagnosis may be changed. This reflects negatively on the laboratory. Thus, if possible, it is best to generate a compromise regarding the techniques used and the criteria utilized for their interpretation. It is also important to realize that the normal control values associated with a particular NCS are specific to the study design used to collect them. In other words, they were derived using standardized recording parameters (e.g., the location of the stimulating and recording electrodes, the distance between the cathode and the E1 surface electrode, the distance between the E1 and E2 surface recording electrodes, the type of surface recording electrodes used, and the machine settings, including the filter settings, the screen sensitivity, and the sweep speed). As previously stated, changing the filter setting affects the morphology of the response (e.g., amplitude, duration, latency, number of phases), changing the screen sensitivity to magnify the response causes the measured onset latency value to decrease, and changing the sweep speed causes the recorded response to be cramped (lowering it) or spread out (raising it). Thus, when one EMG laboratory opts to use the normal control values of another EMG laboratory, the study design must be precisely replicated, including the machine settings.

282

The normal control values used in our EMG laboratories are age-dependent. Although the normal latency and conduction velocity values reported among different laboratories are fairly comparable, this is not true for the lower limit of normal for amplitude, because the amplitude is much more affected by subtle differences in technique. Therefore, because the most important measurement made is response amplitude, it is imperative that the technique generating the normal control values be precisely duplicated in every manner. Otherwise, normality cannot be determined by comparing the collected response measurements to the normal control values. Whenever the original technique is modified, new normal control values must be acquired.

Changing the Sensitivity to Better Identify the Onset Latency Historically, EDX recordings represented electron beam deviations on the oscilloscope. Through the amplifier, low-voltage events generate visible electron beam deviations. The input is the voltage value entering the EMG machine (in volts) and the output is the vertical deflection of the electron beam from the baseline. The gain is defined as the ratio of the vertical deflection to the voltage input, whereas the sensitivity is the opposite (the ratio of the voltage input to the vertical deflection). By adjusting the display sensitivity or the display gain, the recorded response becomes higher or lower in its amplitude (i.e., its y-axis value). This has no effect on the actual y-axis value. However, this adjustment has an undesirable effect: the onset latency (initial deviation of the trace from the baseline) is visualized earlier, and therefore, the measured onset latency value is lower. Although some EDX providers adjust the sensitivity or gain to better define the onset latency, this practice must be avoided because it can cause an abnormal value to become a normal value (i.e., false negative). Instead, the sensitivity or gain setting used during the collection of the normal control values must be used. When the sensitivity or gain setting is inappropriately manipulated in this manner to identify the onset latency for the proximal response and the distal response, the calculated nerve conduction velocity is not affected (because both values are inappropriately shifted leftward to the same degree). Nonetheless, because this approach affects the onset latency, it is best avoided.

Chapter 18: Common Pitfalls and Their Resolution

An alternate approach that can be used when the takeoff site is not obvious is to place the cursor to the left of the takeoff site and then slowly advance it rightward while watching for a vertical deviation of the cursor, which indicates the onset. It is usually easier to identify a vertical deflection of the cursor because its vertical dimension is much larger in magnitude than the vertical thickness of the trace.

Sweep Speed and Sensitivity As was pointed out in Chapter 2, an analog-to-digital convertor has both horizontal and vertical resolution. The horizontal resolution is dictated by the number of measurements made per unit time (termed the inter-sample time or dwell time), and the vertical resolution is dictated by the number of points available on the vertical axis (voltage resolution). The vertical resolution is described in bits (binary elements), such that a vertical resolution of two bits equates to 4 vertical cuts of resolution (2x, where x = 2; thus, 22 = 4), whereas a vertical resolution of 8 bits equates to 256 vertical cuts of resolution (28 = 256) and a vertical resolution of 12 bits equates to 4,096 vertical cuts of resolution (212 = 4,096). The vertical resolution of the EMG machine is determined when it is built and is not adjustable. EMG machines with greater horizontal and vertical resolving capacity generate more accurate digitized reproductions of the analog signal. On most EMG machines, during NCS, the monitor screen is divided along the horizontal axis into 10 divisions and along the vertical axis into 10 divisions. Voltage is measured on the vertical axis and time is measured on the horizontal axis. The amount of voltage allotted to each division along the vertical axis is termed the gain (e.g., mV/division). Its inverse, the sensitivity, is expressed as divisions per unit voltage (e.g., divisions/mV). Thus, as the sensitivity value is increased, the recorded response is magnified, whereas as the gain value is increased, the displayed response decreases in size. For example, with a sensitivity setting of 1 division per mV, a 2 mV response is displayed over two vertical divisions. When the sensitivity setting is increased to two divisions per mV, the response is displayed over four vertical divisions (i.e., it is twice as tall). Increasing the gain has the opposite effect. Most EMG machines use sensitivity units rather than gain units. The amount of time allotted to each horizontal division is termed the sweep speed (i.e., msec/division). Thus, a sweep speed setting of 1 msec/division

generates 10 msec of recording time across the screen (1 msec/division  10 divisions = 10 msec), whereas a sweep speed setting of 5 msec/division generates a total screen time of 50 msec (5 msec/division  10 divisions = 50 msec). The goal is to collect the responses using the majority of the screen rather than cramping the responses into small regions of the monitor. At lower sweep speeds, the response is spread out over a larger number of screen divisions, whereas with higher sweep speeds, the response is crammed into fewer screen divisions. Spreading the response out over a larger number of horizontal divisions means that a larger number of samplings of the response are taken and the displayed response is more accurately depicted. For example, suppose an EMG machine is capable of making 1,000 samplings per screen (i.e., 100 samplings per division). If the response is crammed into two divisions, it will be sampled at 200 points (2 divisions  100 samplings per division = 200 samplings), whereas if it is spread out over four divisions, it will be sampled at 400 points (4 divisions  100 samplings per division = 400 samplings). Waveforms containing higher frequencies require a greater number of samplings for accurate waveform reproduction. The ideal sampling frequency is dictated by the frequency composition of the response being collected. The ideal sampling frequency must be at least twice the fastest frequency contained within the response, but ideally it should be closer to four times the fastest frequency. As stated previously, the sensory responses contain more high-frequency components and are shorter in duration than the motor responses. Taking these two variables into consideration, they are best recorded at a sweep speed of 1–2 msec/ division (in our EMG laboratories we use 2 msec/ division). The ideal sweep speed for the motor responses based on their frequency composition and their longer duration is 2–5 msec/division (in our EMG laboratories we use 5 msec/division). During the needle EMG study, we use a sweep speed of 10 msec/division. Regarding the needle EMG, because the rise time of a fibrillation potential is approximately 0.5 msec, which equates to 2,000 Hz (1 cycle per 0.5 msec = 2,000 cycles/sec = 2,000 Hz), the sampling frequency should not be below 4,000 Hz (twice the highest frequency) and ideally should be 8,000 Hz (four times the highest frequency). The minimal sweep speed (1 cycle = 2 samplings, where 1 cycle = 0.5 msec), as

283

Section 4: Other Pertinent EDX Information

calculated using the technique of dimensional analysis, is 25 msec/division: 0:5 msec=2 samples  1,000 samples=screen 1 screen=10 divisions ¼ 25 msec=division Because ideal resolution requires a sampling frequency of 4 times the fastest frequency, 12.5 msec/ division is preferred. Thus, a sweep speed of 10 msec/ division more than suffices for fibrillation potential resolution. In our EMG laboratory, during the needle EMG study, because we are primarily focusing on MUAP duration, we use a sweep speed of 5 msec/ division for the majority of the study. We use this setting because it better defines the onset and termination sites of the recorded MUAPs, thereby better defining the duration of the MUAP. As previously stated (see Chapter 14), MUAP duration is by far the most sensitive parameter for identifying reinnervation via collateral sprouting. Although we also assess MUAP amplitude, this parameter is much less often affected, even when the degree of reinnervation via collateral sprouting is quite pronounced.

Time Constraints Time constraints are always an issue. Although warming the limbs prior to study and performing contralateral comparison studies, when indicated, are time-consuming, they represent the two most useful acts performed in the EDX laboratory. Often colleagues ask my opinion of their collected responses. Frequently the difficulty is related to a single response with an unexpected value. In that setting I typically ask them for the value of the contralateral response. A common response is, “In the real world, there is not enough time to assess the contralateral side.” Our private-practice EMG laboratories use the same approach as our university-based EMG laboratories, and all of them are in the real world. The point is that when an individual with symptoms limited to one side of the body has an unexpected finding (whether it is an NCS response or an MUAP), the contralateral side provides the opportunity to identify relative abnormalities. The normal control values of the EMG laboratory only permit absolute abnormalities to be identified. The concept of relative abnormal is discussed further in the following subsection.

Missing Relative Abnormalities The normal control values of the EMG laboratory define absolute abnormal. In addition, most

284

practitioners of EDX medicine consider an amplitude value that is more than 50% smaller than the value recorded on the contralateral, asymptomatic side to be relatively abnormal. Consequently, in our EMG laboratories, whenever the response amplitude of a study is in the low-normal range, we always compare it to the contralateral side to avoid missing a relative abnormality. In addition to side-to-side comparisons, we also make ipsilateral comparisons among the response amplitudes. This helps determine when the contralateral side should be studied. This concept is best illustrated with a case.

Case Example A 45-year-old female referred for pinky finger tingling demonstrated the followed sensory response amplitude values on the symptomatic side: Median-D2, 51 (normal >20); Superficial Radial, 61 (normal >17); and Ulnar-D5, 16 (normal >12). All of these amplitude values are above the lower limit of normal (LLN) based on our normal control values. However, when these values are considered in relation to their LLN cutoff value, the Ulnar-D5 response is suspicious. Note that the Median-D2 response is approximately 2.5 times its LLN value and the Superficial Radial response is slight greater than 3 times its LLN, whereas the Ulnar-D5 response is only 1.3 times its LLN. This is quite a difference, and for this reason, the response was repeated. The suspicious value was verified. Following verification, the Ulnar-D5 response was collected from the contralateral side. The contralateral value was 38 (just over 3 times its LLN), indicating that the suspicious response was indeed abnormal (by relative criteria). It is very helpful to consider the recorded response amplitudes in relation to their LLN values and to compare them to each other.

Identifying Demyelination Based on the Latency of a Very Low Amplitude Response Typically, mild to moderate degrees of focal axon loss do not affect measurements that reflect AP propagation speed (e.g., onset latency, peak latency, conduction velocity, and late response onset latencies), because in most situations, as long as a single large-diameter, heavily myelinated fiber is unaffected, these values remain normal. This is true because they are determined by the fastest conducting nerve fiber. However, when the response is very low in amplitude, the

Chapter 18: Common Pitfalls and Their Resolution

majority of the nerve fibers are affected. Once all of the larger-diameter, more heavily myelinated axons are affected, these measurements are determined by axons with intermediate or slower AP propagation speeds. Reinnervational axons are another source of slowed conduction (due to their shorter and thinner segments of myelin and their narrower axon diameter).

Clinical Bias Another potential pitfall is artifact misinterpretation. I once overheard a colleague comment that he is a clinician first, and therefore, when an electrical signal appears on the monitor that is of unclear origin – fibrillation potential versus artifact – when the clinician suspects that the patient has a radiculopathy, the unknown signal is a fibrillation; otherwise it is an artifact. This approach leads to false positive and false negative conclusions and should be avoided. This is an example of the misapplication of the statement that “the EMG examination is an extension of the clinical examination.” Although the statement is true, it does not mean that clinical examination findings can substitute for performing the corroborative EDX studies or that clinical insight can discriminate between different electrical potentials. Because it would be prohibitively expensive and time consuming (as well as uncomfortable for the patient) to test every major nerve and muscle in the body, EDX testing is not useful as a screening procedure. Instead, it relies on the clinical examination to dictate the phenotype or at least the distribution of the symptoms. In other words, the EDX study should always follow the clinical examination. However, it remains an independent procedure capable of objectively localizing and characterizing peripheral neuromuscular lesions. Thus, it confirms the clinical impression when it is accurate and it redirects the referring provider when it is not.

Low Amplitude Peroneal-EDB Motor Responses among Normal Individuals Not infrequently, individuals studied in the EMG laboratory are found to have bilateral low amplitude peroneal motor responses, recording extensor digitorum brevis (EDB). This is an isolated finding on both sides: A dash would also suffice the superficial peroneal sensory responses and the peroneal motor responses (recording tibialis anterior, TA) are normal. On needle EMG, the abnormalities are limited to the EDB muscle. Thus, following the routine sensory and motor NCS, when an individual has an isolated low amplitude

peroneal-EDB motor response, we add the peronealTA motor NCS ipsilaterally and then perform the superficial peroneal sensory and both peroneal motor NCS on the contralateral, asymptomatic side. We also expand the needle EMG to include more muscles with L5-peroneal and L5-tibial innervation, with comparison to at least one of each on the contralateral side. When the needle EMG also demonstrates isolated bilateral EDB muscle abnormalities, we diagnose “isolated bilateral low amplitude peroneal-EDB motor responses” and suggest that the finding is likely related to trauma related to the wearing of shoes. When the findings are one-sided or clearly asymmetric, then this would not be the case. Again, it is imperative that the abnormal findings be surrounded by normal ones, just as if the clinical examination were being performed. On clinical examination, isolated bilateral EDB muscle atrophy is a frequent finding. The EDB muscle atrophy is not accompanied by sensory loss or weakness in the L5-peroneal distribution. This EDX finding has not only a clinical correlate but also a pathological one. The EDB muscles of normal individuals show muscle fiber type grouping early in life (Jennekins et al., 1972). Presumably, this reflects trauma associated with the wearing of shoes.

Measurement Errors Errors in time measurement parameters (e.g., latency; conduction velocity) are less common now that analog instruments have been replaced by digital ones. Although the EMG machine automatically places the cursors (based on a slope criterion that is not dependent on the sensitivity), it frequently places them incorrectly. Thus, most cursor placement errors result from forgetting to properly position the cursors rather than placing them in the wrong position. Also, the cursors should be placed with the screen sensitivity set to the value used to collect the normal control values of the EMG laboratory. Many EDX practitioners change the sensitivity to magnify the response during cursor placement. However, this causes the onset site to appear earlier and, therefore, may result in a false negative conclusion (e.g., a mildly prolonged latency may be reported as normal). The time cursors on our EMG machines are shaped like a Roman numeral one. Our approach is to begin with the cursor located to the left of the onset site and then slowly slide the cursor rightward, along the baseline, until a vertical deflection of the cursor is noted. This is much quicker than adjusting

285

Section 4: Other Pertinent EDX Information

the screen sensitivity and does not nullify the normal control value for that response. Errors in distance measurement (i.e., between the stimulating and recording sites and between successive stimulation sites) can significantly affect the calculated latency and conduction velocity values, respectively. Although the surface measurement between successive stimulation sites correlates fairly well with the length of the segment of nerve being stimulated along the extremity, this is not the case across the brachial plexus (i.e., the distance between supraclavicular fossa stimulation site and the axillary stimulation site does not reflect the distance of the nerve segment studied). The use of calipers to perform the measurement does not overcome this issue. Another problem area is where nerve segments cross joints. In this setting, the position of the joint strongly affects the measurement. For example, with ulnar motor NCS, with the forearm extended, marking the skin 4 cm above the elbow and 4 cm below the elbow generates a distance of 8 cm between the above-elbow and below-elbow stimulation sites. Because the nerve is relaxed, the true distance between the ulnar nerve stimulation sites is greater than the surface distance of 8 cm. When the patient flexes the forearm (e.g., to 90 degrees), the distance between the two skin marks increases in proportion to the degree of flexion. Thus, in the extended position, the calculated nerve conduction velocity will underestimate the actual nerve conduction velocity. This is why, when the ulnar motor NCS shows an isolated drop of CV across the elbow segment, it is almost always technical. Although this could be seen with the dropout of the APs of the fastest conducting motor axons (i.e., focal demyelinating conduction block or axon disruption selectively involving the larger, more heavily myelinated fibers), there should be other EDX features (e.g., amplitude value and negative areaunder-the-curve value decrements; neurogenic recruitment). It could be seen with focal demyelinating conduction slowing involving the fastest fibers, but then there should be a change in waveform morphology as the fibers are differentially affected (nonuniform slowing). Thus, in isolation, a drop in

References Ajmani ML. Variations in the motor supply of the thenar and hypothenar muscles of the hand. J Anat 1996;189:145–150.

286

the conduction velocity across the elbow is much more likely to be technical, even when it is not observed on the contralateral side. In the latter setting, when the motor studies of the two sides are repeated, the conduction velocity asymmetry usually resolves. If it is present on both sides, a technical etiology is even more likely. Another error is to mark the skin, stimulate the two marks in the extended position, and then measure the distance in the flexed position. The measurements and stimulations must be made in the same limb position so that the ulnar nerve is stimulated below the skin mark. Again, the position of the limb should be identical to the position used to collect the normal control values. In addition to the above, the muscle under study should be relaxed. When the muscle is flexed, the muscle belly shortens. This decreases the duration of the motor response (less time is required for the bidirectionally propagating muscle fiber APs to reach the tendinous ends of the muscle fibers), and the synchrony of the individual muscle fiber APs is increased.

Miscellaneous Pitfalls Impedance between the Signal Source and the Surface Recording Electrode Tissue impedance is dependent on the underlying tissue. As would be expected, it is greatest for bone and fat and lowest for ionic fluids (e.g., blood; interstitial fluid), with other tissues lying somewhere between these extremes. Thus, depending on the underlying tissue, the impedance ranges from 1,000 ohms to, in the setting of thick and calloused skin, values exceeding 100,000 ohms (Koumbourlis, 2002). This is why rough patches of skin should be avoided and, if utilized, should undergo appropriate abrasion prior to the application of the surface electrodes. With sensory NCS, it is important that the surface recording electrodes be properly positioned over the nerve under study. Should one of the two electrodes be positioned slightly away from the nerve (i.e., laterally displaced from the nerve), the other electrode will have a larger contribution to the recorded response.

Amoiridis G. Median-ulnar nerve communications and anomalous innervation of the intrinsic hand muscles: an electrophysiological study. Muscle Nerve 1992;15:576–579.

Arrigo A, Cosi V, Savoldi F. The conduction velocity of the human sciatic nerve. Electroencephalogr Clin Neurophysiol (Suppl) 1952;22: 23–25.

Chapter 18: Common Pitfalls and Their Resolution

Ashworth NL, Marshall SC, Satkunam LE. The effect of temperature on nerve conduction parameters in carpal tunnel syndrome. Muscle Nerve 1998;21:1089–1091. Awang MS, Abdullah JM, Abdullah MR, Tharakan J, Prasad A, Husin ZA, Hussin AM, Tahir A, Razak SA. Nerve conduction study among healthy Malays. The influence of age, height and body mass index on median, ulnar, common peroneal and sural nerves. Malays J Med Sci 2006;13:19–23. Behse F, Buchthal F. Normal sensory conduction in the nerves of the leg in man. J Neurol Neurosurg Psychiatry 1971;34:404–414. Bjorqvist SE, Lang AH, Falck B, Vuorenniemi R. Variability of nerve conduction velocity. Electromyogr Clin Neurophysiol 1977;17:21–28. Boland RA, Krishnan AV, Kiernan MC. Riche-Cannieu anastomosis as an inherited trait. Clin Neurophysiol 2007;118:770–775. Bolton CF. AAEM minimonograph #17. Rochester, NY: American Association of Electromyography and Electrodiagnosis, 1981. Bolton CF, Carter KM. Human sensory nerve compound nerve action potential amplitude: variation with sex and finger circumference. J Neurol Neurosurg Psychiatry 1980;43:925–928. Buchthal F, Guld C, Rosenfalck P. Action potential parameters in normal muscle and their dependence on physical variables. Acta Physiol Scand 1954;32:200–218. Buchthal F, Rosenfalck A. Evoked action potentials and conduction velocities in human sensory nerves. Brain Res 1966;3:1–122. Buschbacher RM. Body mass index effect on common nerve conduction study measurements. Muscle Nerve 1998;21:1398–1404. Buschbacher RM. Sural and saphenous 14-cm antidromic sensory nerve conduction studies. Am J Phys Med Rehabil 2003;82:421–426.

Campbell WW. Essentials of electrodiagnostic medicine. Baltimore: Williams and Wilkins; 1999:94. Campbell WW, Ward LC, Swift TR. Nerve conduction velocity varies inversely with height. Muscle Nerve 1981;4:520–523. Cannieu A. Note sur une anastomose entre le branch profunde du cubital et le median. Bull Soc Anat Physiol Bordeaux 1897;17:339–342. Caruso G, Massini R, Crisci C, Nilsson J, Catalano A, Santoro L, Battaglia F, Crispi F, Nolano M. The relationship between electrophysiologic findings, upper limb growth and histological features of median and ulnar nerves in man. Brain 1992;115:1925–1945. Crutchfield CA, Gutmann L. Hereditary aspects of median-ulnar nerve communications. J Neurol Neurosurg Psychiatry 1980;43:53–55. Denys EH. AAEM Minimonograph #14: The influence of temperature in clinical neurophysiology. Muscle Nerve 1991;14:795–811. Don Griot JPWD, Zuidam JM, Kooten EOV, Prose LP, Hage JJ. Anatomic study of the ramus communicans between the ulnar and median nerves. J Hand Surg 2000;25A: 948–954. Dreyer SJ, Dumitru D, King JC. Anodal block vs anodal stimulation. Fact or fiction. Am J Phys Med Rehabil 1993;72:10–18. Dumitru D, Walsh NE. Practical instrumentation and common sources of error. Am J Phys Med Rehabil 1988;67:55–65. Dumitru D, Walsh NE, Weber CF. Electrophysiological study of the riche-Cannieu anomaly. Electromyogr Clin Neurophysiol 1988;28:27–31. Esper GJ, Nardin RA, Benatar M, Sax TW, Acosta JA, Raynor EM. Sural and radial sensory responses in healthy adults: diagnostic implications for

polyneuropathy. Muscle Nerve 2005;31:628–632. Falck B, Stalberg E. Motor nerve conduction studies: measurement principles and interpretation of findings. J Clin Neurophysiol 1995;12:254–279. Falco FJE, Hennessey WJ, Goldberg G, Braddom RL. H reflex latency in the healthy elderly. Muscle Nerve 1994a;17:161–167. Falco FJE, Hennessey WJ, Goldberg G, Braddom RL. Standardized nerve conduction studies in the lower limb of the healthy elderly. Am J Phys Med Rehabil 1994b;73:168–174. Fong SY, Goh KJ, Shahrizaila N, Wong KT, Tan CT. Effects of demographic and physical factors on nerve conduction study values of healthy subjects in a multi-ethnic Asian population. Muscle Nerve 2016;54:244–248. Franssen H, Wieneke GH. Nerve conduction and temperature: necessary warming time. Muscle Nerve 1994;17:336–344. Geerlings AHC, Mechelse K. Temperature and nerve conduction velocity, some practical aspects. Electromyogr Clin Neurophysiol 1985;25:253–260. Gutmann L. Median-ulnar nerve communications and carpal tunnel syndrome. J Neurol Neurosurg Psychiatry 1977;40:982–986. Gutmann L. AAEM minimonography #2: important anomalous innervations of the extremities. Muscle Nerve 1993;16:339–347. Halar EM, DeLisa JA, Brozovich FV. Nerve conduction velocity: relationship of skin, subcutaneous, and intramuscular temperatures. Arch Phys Med Rehabil 1980;61:199–203. Harness D, Sekeles E. The double anastomotic innervation of thenar muscles. J Anat 1971;109:461–466. Henriksen JD. Conduction velocity of motor nerves in normal subjects and patients with neuromuscular

287

Section 4: Other Pertinent EDX Information

disorders. Minneapolis: University of Minnesota (Thesis); 1956. Hodgkin AL, Katz B. The effect of temperature on the electrical activity of the giant axon of the squid. J Physiol 1943;109:240–249. Hodgkin AL, Katz B. The effect of sodium ions on the electrical activity of the giant axon of the squid. J Physiol 1949;108:37–77. Hua Y, Lovely DF, Doraiswami R. Factors affecting the stimulus artifact tail in surface-recorded somatosensory-evoked potentials. Med Biol Eng Comput 2006;44:226–241. Huang CR, Chang WN, Chang HW, Tsai NW, Lu CH. Effects of age, gender, height, and weight on late responses and nerv e conduction study parameters. Acta Neurol Taiwan 2009;18:242–249. Jennekins FGI, Tomlinson BE, Walton JN. The extensor digitorum brevis: histological and histochemical aspects. J Neurol Neurosurg Psychiatry 1972;35:124–132. Johnson EW. Practical electromyography. Baltimore: Williams and Wilkins Publishers; 1980.

Lee KS, Oh CS, Chung IH, Sunwoo IN. An anatomic study of the MartinGruber anastomosis: electrodiagnostic implications. Muscle Nerve 2005;31:95–97. Lowitzsch K, Hopf HC, Galland J. Changes of sensory conduction velocity and refractory periods with decreasing tissue temperature in man. J Neurol 1977;216:181–188. Ludin HP, Beyelar F. Temperature dependence of normal sensory nerve action potentials. J Neurol 1977;216:173–180. Madras C, Midroni G. Proximal Martin-Gruber anastomosis mimicking ulnar neuropathy at the elbow. Muscle Nerve 1999;22:1132–1135. Mannerfelt L. Studies on the hand in ulnar nerve paralysis. A clinical experimental investigation in normal and anomalous innervations. Acta Orthop Scand 1966;87:23–142. Netherton BL. Electrodiagnostic instrumentation: Understand and manage it. In, Neurophysiology and Instrumentation Course. 57th Annual Meeting of the AANEM, Quebec City, Quebec, Canada, 2010:21–28.

mammalian central nervous system: a review. Brain Res 1975;98:417–440. Rasminsky M. The effects of temperature on conduction in demyelinated single nerve fibers. Arch Neurol 1973;28:287–292. Riche P. Le nerf cubital et les muscles de l’eminence thenar. Bull Mem Soc Anat Paris 1897;5:251–252. Rogart RB, Staempfli R. Voltage-clamp studies of mammalian myelinated nerve. In Culp WJ, Ochoa H, editors, Abnormal nerves and muscles as impulse generators. Oxford: Oxford University Press; 1982:193–210. Rountree T. Anomalous innervation of the hand muscles. J Bone Joint Surg 1949;31B:505–510. Rutkove SB. Effects of temperature on neuromuscular electrophysiology. Muscle Nerve 2001;24:867–882. Rutkove SB, Kothari MJ, Shefner JM. Nerve, muscle, and neuromuscular junction electrophysiology at high temperature. Muscle Nerve 1997;20:431–436. Rutkove SB. AAEM minimonograph 14: effects of temperature on neuromuscular electrophysiology. Muscle Nerve 2001;24:867–882.

Kimura I, Ayyar DR. The hand neural communication between the ulnar and median nerves: electrophysiologic detection. Electromyogr Clin Neurophysiol 1984;24:409–414.

Nilsson J, Ravits J, Hallett M, Stimulus artifact compensation using biphasic stimulation. Muscle Nerve 1988;11:597–602.

Saperstein DS, King RB. Motor neuron presentation of an ulnar neuropathy and Riche-Cannieu anastomosis. Electromyogr Clin Neurophysiol 2000;40:119–122.

Kornfield MJ, Cerra J, Simons DG. Stimulus artifact reduction in nerve conduction. Arch Phys Med Rehabil 1985;66:232–235.

Paintal AS. Block of conduction in mammalian myelinated nerve fibers by low temperatures. J Physiol 1965;180:20–49.

Soudmand R, Ward LC, Swift TR. Effect of height on nerve conduction velocity. Neurology 1982;32:407–410.

Koumbourlis AC. Electrical injuries. Crit Care Med 2002;30[suppl.]: S424–S430.

Pease WS, Pitzer NL. Electronic filter effects on motor and sensory nerve conduction tests. Am J Phys Med Rehab 1990;69:28–31.

Spiegel MH, Johnson DW. Conduction velocity in the proximal and distal segments of the motor fibers of the ulnar nerve of human beings. Arch Phys Med Rehabil 1962;43:57–61.

Lambert EH. Diagnostic value of electrical stimulation of motor nerves. Electroencephalogr Clin Neurophys 1962;22:9–16. Lang AH, Puusa A. Dual influence of temperature on the compound nerve action potential. J Neurol Sci 1981;51:81–88.

288

Pinelli P. Physical, anatomical and physiological factors in the latency measurement of the M response. Electroenceph Clin Neurophysiol 1964;17:86. Ranck JB. Which elements are excited in electrical stimulation of

Srinivasan R, Rhodes J. The medianulnar anastomosis (martin-Gruber) in normal and congenitally abnormal fetuses. Arch Neurol 1981;38:418–419.

Chapter 18: Common Pitfalls and Their Resolution

Stetson DS, Albers JW, Silverstein BA, Wolfe RA. Effects of age, sex, and anthropometric factors on nerve conduction measures. Muscle Nerve 1992;15:1095–1104.

conduction parameters in normal subjects related to age, gender, temperature, and height: a reappraisal. Muscle Nerve 1992;15:666–671.

Tavee JO, Polston D, Zhou L, Shields RW, Butler RS, Levin KH. Sural sensory nerve action potential, epidermal nerve fiber densitiy, and quantitative sudomotor axon reflex in the healthy elderly. Muscle Nerve 2014;49:564–569.

Uchida Y, Sugioka Y. Electrodiagnosis of Martin-Gruber connection and its clinical importance in peripheral nerve surgery. J Hand Surg 1992;17A:54–59.

Trojaborg WR, Moon A, Andersen BB, Trojaborg NS. Sural nerve

Uncini A, Lange DJ, Lovelace RE. Anomalous intrinsic hand muscle innervation in median and ulnar nerve lesions: an

electrophysiological study. Ital J Neurol Sci 1988;9:497–503. Unver Dogan N, Uysal II, Seker M. The communications between the ulnar and median nerves in upper limb. Neuroanatomy 2009;8:15–19. Wilbourn AJ, Lambert EH. The forearm median-to-ulnar nerve communication; electrodiagnostic aspects. Neurology 1976;26:368. Zwarts MJ, Guecher A. The relation between conduction velocity and nerve fiber length. MN 1995;18:1244–1249.

289

Chapter

19

Safety Issues in the EDX Laboratory

EDX providers must understand the risks and potential complications associated with the performance of NCS and needle EMG and how to avoid them. A number of publications exist addressing the safety issues pertinent to an EDX laboratory, including a position statement by the American Association of Neuromuscular and Electrodiagnostic Medicine (AANEM) that can be obtained from their website (AANEM Position Statement). In addition, the Occupational Safety and Health Administration (OSHA) provides guidance in this area, and in regard to the risk of disease transmission, the Centers for Disease Control and Prevention (CDC) has generated a number of publications (discussed later in this chapter). The American National Standards Institute (ANSI) approves methods of testing and sets current flow limits for medical devices. The US Food and Drug Administration (FDA) also sets standards for safety, including current leakage limits. The major potential complications encountered in the EMG laboratory can be divided into four categories: (1) electrical injury, (2) bleeding, (3) transmission of infection, and (4) iatrogenic pneumothorax or pneumoperitoneum. The following discussions are approached from an educational point of view and are not meant to apply to every possible situation or management option. The particular circumstances surrounding each patient must be known to the EDX provider so that the potential benefits and risks of the EDX study can be determined and discussed with the patient as part of the decision-making process. In general, NCS and needle EMG studies are safe, with few contraindications. Even then, in most cases, minor modifications of the EDX study permit its safe application.

Electrical Injury A complete circuit is required for electricity to flow. Thus, the potential for electrical injury exists only

290

when a grounded individual comes into contact with a source current and the body resistance is low enough for current to pass through it. The source current may be natural (lightning) or man-made (power line). Regarding lightning-related injuries, the majority of injuries occur in the South and Midwest between May and September, resulting in approximately 374 deaths annually (Adekoya and Nolte, 2005). The injuries are typically secondary (e.g., from the associated fall, from a falling object, from the blast injury) rather than primary (i.e., due to the lightning strike itself ). Keraunoparalysis, transient lower extremity paralysis following a lightning strike, is an example of a primary injury that, in general, improves within hours (Savica, 2017). Regarding man-made electrical injuries, although high current faults trip the circuit breaker, there is a 100 msec delay that can be lethal during the vulnerable period of the cardiac cycle. The severity of the resultant electrical injury reflects the magnitude of the current, the duration of the contact, and the pathway through the body that the current travels. Regarding current magnitude, from Ohm’s law, we know that the magnitude of the current is equal to the voltage divided by the resistance (I = V/R) (see Chapter 1). Therefore, the severity of the injury is proportional to the voltage of the source and inversely proportional to the resistance of the victim. The duration of the contact also contributes to the outcome. For this reason, AC signal is more dangerous than DC signal, because tetany (continuous muscle contraction) occurs when adult muscle fibers are stimulated at frequencies ranging from 40 Hz to 110 Hz by currents exceeding 6–9 mA (Cooper, 1995). Consequently, because AC signal is 60 Hz, even very low current intensities generate tetany. Although all of the upper extremity muscles may be tetanically stimulated, because the hand and forearm flexors are stronger than the hand and forearm

Chapter 19: Safety Issues in the EDX Laboratory

extensors, continuous hand muscle flexion may render it impossible for the victim to release the current source. Currents producing this phenomenon are referred to as let-go currents, which is a misleading term because it is the opposite of what is actually occurring and would be more fittingly termed can’t let-go currents. Because the let-go current threshold is 6–9 mA for adults, the range of safer current intensities is quite narrow (0.3 msec is abnormal). The left median palmar response is collected to screen for carpal tunnel syndrome and is normal. The left Median-D2 response was collected for comparison purposes and is normal. The peak latency difference between the left and right sides is abnormal (>0.4 msec). At this point, the motor NCS are performed. Given the degree of abnormality of the median sensory and median palmar responses, the median motor response is expected to be normal, but this is not always the case. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET EXERCISE 2 NCS PERFORMED

LEFT DRG

LAT

AMP

Median-D2

C6,7

3.1

46.2

Ulnar-D5 Superficial Radial

RIGHT CV

nAUC

LAT

AMP

3.7

40.7

C8

3.0

24.7

C6,7

2.4

34.3

2.5

35.3

2.0

23.7

3.3

11.5

CV

SENSORY

Median Palmar Ulnar Palmar

2.1

53.7

MOTOR Median-APB

10.9

324

52.3

nAUC

Case 1 through Case 50

(cont.)

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET EXERCISE 2

LEFT

NCS PERFORMED

DRG

LAT

AMP

RIGHT CV

nAUC

Ulnar-ADM

LAT

AMP

2.9

8.2

CV

7.9

nAUC

54.2

As expected, the routine motor NCS are normal, and the onset latency of the median motor response is well within normal limits. Thus, a left Median-APB motor NCS is not necessary. A limited needle EMG is performed of the right APB (compared to the left APB), the FDI, and the pronator teres. UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 2

Insertional Activity Normal

IPSWs

SCP

Spontaneous Activity Other

None

Fibs

Fascs

MUAP Analysis

Other

MUAP Recruitment

MUAP Morphology

RIGHT APB

X

X

Normal

Normal

FDI

X

X

Normal

Normal

Pron teres

X

X

Normal

Normal

X

X

Normal

Normal

LEFT APB

The needle EMG study is normal. It is recommended that the patient return in one year for a repeat study of the two median nerves to assess the rate of progression, if any. In the interim, the patient is treated conservatively by her primary care provider. Over time, her right upper extremity symptoms worsen. She did not return for follow-up assessment 12 months later. Her symptoms continue to worsen. Subsequently, she is referred back to the EMG laboratory (27 months later), where she reports that her symptoms are much more pronounced, much more frequent, and that, in addition, she now has right upper extremity aching pain that is awakening her from sleep. She also reports occasional symptoms on the left side. Follow-Up Nerve Conduction Studies

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET EXERCISE 2 NCS PERFORMED

LEFT Stim Site

LAT

AMP

3.5

34.7

RIGHT CV

nAUC

LAT

AMP

4.2

29.7

Ulnar-D5

3.0

21.7

Superficial Radial

2.2

42.0

CV

nAUC

SENSORY Median-D2

325

Section 5: Case Studies in Electrodiagnostic Medicine

(cont.)

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET EXERCISE 2 NCS PERFORMED

LEFT Stim Site

Median Palmar

LAT

AMP

2.5

57.7

RIGHT CV

nAUC

Ulnar Palmar

LAT

AMP

2.9

27.7

1.9

28.5

4.2

9.0

CV

nAUC

MOTOR Median-APB

3.5

9.3

8.9 Ulnar-ADM

2.9

52.8

7.5 7.5

53.3

The sensory NCS show delayed right Median-D2, bilateral median palmar, and right Median-APB responses, as well as a mild delay of the left median palmar response. Although the amplitudes of the sensory responses are normal, when compared to the EDX study performed 27 months prior, they are lower. Because the ulnar and radial sensory responses are not that different, it is possible that there is also concomitant axon loss. Evidence of motor axon loss can be sought on the needle EMG study. Needle EMG Study

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 2

Insertional Activity Normal

IPSWs

SCP

Spontaneous Activity Other

None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

RIGHT APB

X

X

Normal

Mild CMAL

FDI

X

X

Normal

Normal

Pron teres

X

X

Normal

Normal

X

X

Normal

Normal

LEFT APB

The needle EMG study of the right APB muscle is abnormal. The durations of many of the MUAPs are mildly increased, consistent with reinnervation via collateral sprouting. When motor axon disruption progresses slowly and reinnervation via collateral sprouting is able to keep pace, the motor responses (CMAPs) remain normal despite the loss of motor axons. This is apparent on the needle EMG study and is one reason to study the APB (and to compare it to the contralateral side). Because the APB muscle is often more painful than other upper extremity muscles, we begin with the FDI because it is one of the least painful upper extremity muscles. Also, because the APB muscle infrequently shows fibrillation potentials in the setting of mild to moderate carpal tunnel syndrome and because the APB muscle is typically more painful than the other extremity muscles, we typically limit the needle EMG study of this muscle to an MUAP assessment and spend little time seeking fibrillation potentials.

326

Case 1 through Case 50

EDX Study Conclusion 1. Bilateral Median Neuropathies (e.g., carpal tunnel syndrome) The above are demyelinating and axon loss in nature on the right and demyelinating in nature on the left, affecting the sensory and motor nerve fibers on the right and the sensory nerve fibers on the left, and are located at or distal to the wrist. Since the initial EDX study performed 27 months prior, the right median neuropathy has worsened and there is now also a left median neuropathy. In addition to the progression, the symptoms have intensified and are interfering with sleep. In general, when we identify carpal tunnel syndrome, we do not dictate management. We typically suggest that if the patient is treated conservatively, that the median nerve be restudied in one year to assess the efficacy of conservative treatment. This is important because the clinical features associated with progression may be unreliable. Not infrequently, as the disorder progresses and the pathology changes from demyelination to axon loss, the symptoms become less apparent. Frequently, the episodic tingling (a positive and readily noticeable symptom) is replaced with slowly progressive numbness. Because the motor axon loss is also slowly progressive, reinnervation keeps pace with denervation, and as a result, muscle atrophy is not initially apparent. When reinnervation begins to lag behind denervation, muscle atrophy appears. At this point, there may be significant motor axon loss, limiting the degree of recovery from surgical intervention.

Exercise 3 A 42-year-old male presents with numbness involving the lateral aspects of the left shoulder and superior portion of the arm, along with difficulty performing overhead activities. These symptoms followed a fracture-dislocation of the shoulder 10 weeks earlier. (The shoulder fracture-dislocation suggests an infraclavicular process, the overhead limitations of the upper extremity suggest C5,6 nerve fiber involvement, and the distribution of the sensory symptoms suggests an axillary neuropathy.) Because the screening sensory and motor NCS do not assess the C5 sensory axons or the C5,6 motor axons, additional NCS are required with each component. With the screening sensory NCS, the LABC and Median-D1 sensory NCS are added. Nerve Conduction Studies

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET EXERCISE 3 NCS PERFORMED

LEFT DRG

LAT

AMP

Median-D2

C6,7

3.2

24.5

Ulnar-D5

C8

2.9

16.9

Superficial Radial

C6,7

2.6

21.8

LABC

C6

2.6

14.1

Median-D1

C6

3.3

20.1

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY

A reliable sensory NCS to assess the cutaneous branch of the axillary nerve (superior lateral brachial cutaneous nerve) is not available. Thus, because the screening sensory NCS do not assess the C5 sensory axons, they are normal. The added LABC and Median-D1 sensory NCS, which assess the C6 sensory axons 100% of the time, are also normal. Because the recorded responses are high in amplitude, contralateral sensory NCS were not performed.

327

Section 5: Case Studies in Electrodiagnostic Medicine

Because the distribution of the motor complaints suggests C5,6 motor axon involvement, the Axillary-Deltoid and Musculocutan-BC motor NCS are added to the screening studies. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET EXERCISE 3 NCS PERFORMED

LEFT DRG

LAT

AMP

Median-D2

C6,7

3.2

24.5

Ulnar-D5

C8

2.9

16.9

Superficial Radial

C6,7

2.6

21.8

LABC

C6

2.6

14.1

Median-D1

C6

3.3

20.1

3.6

10.9

RIGHT CV

nAUC

LAT

AMP

2.9

12.3

3.2

6.0

4.1

6.2

CV

nAUC

SENSORY

MOTOR Median-APB

10.9 Ulnar-ADM

2.9

56

9.3 9.3

Radial-ED

2.8

54

11.0 11.0

Musculocutan-BC

3.2

59

5.5 X

Axillary-Deltoid

4.5

1.4

The screening motor NCS are normal, as is the Musculocutan-BC response. The Axillary-Deltoid response is severely reduced in amplitude, indicating an axon loss process. The Radial-ED was added because this response assesses C6 motor axons (in addition to C7 and C8). It was also normal. When normal, latency and CV values are replaced with an X. The low-amplitude axillary motor response generates a lesion localization list that includes the axillary nerve, the posterior cord, and the upper plexus. However, the LABC and Median-D1 sensory responses are normal, thereby excluding an upper plexus from consideration. The superficial radial sensory response was normal and makes a posterior cord localization extremely unlikely (with focal PNS lesions, the sensory axons within the element are typically more affected than the motor axons). Thus, at this point, a severe, axon loss axillary neuropathy is by far the most likely diagnosis. The needle EMG is expanded to verify this suspicion and further characterize the lesion. Needle EMG Examination

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 3

Insertional Activity Normal

IPSWs

SCP

Spontaneous Activity Other

None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

Normal

Normal

LEFT FDI

328

X

X

Case 1 through Case 50

(cont.)

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 3

Insertional Activity Normal

IPSWs

SCP

Spontaneous Activity Other

None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

EI

X

X

Normal

Normal

FPL

X

X

Normal

Normal

Pron teres

X

X

Normal

Normal

BC, MH

X

X

Normal

Normal

TC, LH

X

X

Normal

Normal

Deltoid, MH

X

2+

Normal

Mild CMAL

Deltoid, PH

X

3+

Normal

Mild CMAL

Deltoid, AH

X

2+

Normal

Mild CMAL

Teres minor

X

1+

Normal

Normal

Infraspinatus

X

X

Normal

Normal

Brachioradialis

X

X

Normal

Normal

Low cerv psp

X

X

–

–

High thor psp

X

X

–

–

X

X

Normal

Normal

RIGHT

The abnormal muscles are restricted to the muscle domain of the axillary nerve (deltoid and teres minor muscle). All three heads of the deltoid were assessed. In general, with axillary neuropathies, the posterior head reinnervates first. Thus, we typically study at least the middle and posterior heads. In this case, we also studied the anterior head. Other upper plexus and radial nerve innervated muscles are normal, supporting an axillary nerve localization. EDX Conclusion

The EDX study localizes the lesion to the axillary nerve and characterizes it as an axon loss process that is severe in degree. There is evidence of early reinnervation via collateral sprouting (the increased duration MUAPs). The presence of fibrillation potentials suggests that further reinnervation is possible.

Exercise 4 A 67-year-old right hand–dominant male is referred for EDX assessment of a right foot drop that was noted in the recovery room following a right total hip replacement. The referring provider suspected compression of the right common peroneal nerve and had informed the patient that it would resolve. Six months later, due to minimal improvement, the patient was referred for EDX testing. On focused neurological examination, the patient could not dorsiflex the ankle, plantar flex the ankle, or evert the foot; he could minimally invert the foot. In addition, he has loss of sensation along the anterior, lateral, and posterior aspects of the leg and all aspects of the foot. Although the inability to dorsiflex or evert the ankle is consistent with a common peroneal neuropathy, the inability to plantar flex

329

Section 5: Case Studies in Electrodiagnostic Medicine

and the limited foot inversion indicate concomitant tibial nerve involvement. This suggests a sciatic neuropathy related to the surgery itself. In support of this statement, the sensory complaints are in the cutaneous distribution of the sciatic nerve. Thus, clinically, the lesion must be at or proximal to the sciatic nerve (e.g., lumbosacral plexus or roots). The screening sensory NCS are performed first. Nerve Conduction Studies

LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET EXERCISE 4 NCS PERFORMED

LEFT Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Sural

NR

Supfcl Peroneal

2.9

15.7

NR

The screening sensory NCS show absent sural and superficial peroneal sensory responses, consistent with a ganglionic or postganglionic axon loss process located at or proximal to the sciatic nerve. The sensory response involvement excludes lumbosacral radiculopathies. To demonstrate that this is a unilateral problem (surround abnormal with normal), a contralateral sensory NCS is added. Because the referring provider suspected a common peroneal neuropathy, we chose to add the contralateral superficial peroneal nerve, which was normal. The proximal peroneal motor NCS is included in the initial motor NCS to better quantify the severity of the peroneal motor axon loss. The M wave component of the H reflex is useful for quantifying the severity of the tibial motor axon loss. For lesion severity estimation, for each motor response abnormality, the homologous contralateral motor NCS is performed.

LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET EXERCISE 4 NCS PERFORMED

LEFT Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Sural Supfcl Peroneal

NR 2.9

15.7

4.4

8.3

NR

MOTOR Tibial-AH

18.3

4.8

1.2 0.5

Peroneal-EDB

Peroneal-TA

40.5

NR

3.2

5.1

40.5

4.4

0.4 0.3

330

2.4

M wave

5.8

10.2

H wave

34.5

1.6

8.7

0.7 NR

3.1 44.2

Case 1 through Case 50

The initial motor NCS show an absent distal peroneal motor response. The tibial motor response, the proximal peroneal motor response, and the M wave of the H reflex are extremely low in amplitude. The right H wave is absent. To quantify the degree of motor axon loss, the contralateral distal motor responses are added. Because the distal peroneal motor response is absent, a contralateral comparison study is not necessary (i.e., it is 100% involved). At this point, we know that this is a postganglionic axon loss process that involves the sciatic nerve or the lumbosacral plexus. Based on the degree of asymmetry between the motor response amplitude values recorded on the two sides, it is apparent that the lesion is very severe in degree (calculations are performed below). Because the sensory responses were abnormal, the lesion must be ganglionic or postganglionic. However, because the motor responses are abnormal, it cannot be ganglionic. The right H reflex shows an absent H wave. The M wave is mildly delayed and extremely reduced in amplitude. The left H reflex is normal. The muscles innervated by the superior (gluteus medius, gluteus minimus, and tensor fascia lata) and inferior (gluteus maximus) gluteal nerves are helpful in differentiating a sciatic neuropathy from a lumbosacral plexopathy. They are spared with sciatic neuropathies.

LOWER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 4

Insertional Activity

Spontaneous Activity

Normal IPSWs SCP Other None Fibs Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

RIGHT FHB

3+

V severe neurogenic

Severe CMAL

FDL

2+

Severe neurogenic

Moderate CMAL

TA

2+

V severe neurogenic

Severe CMAL

Gastroc, MH

3+

V severe neurogenic

Moderate CMAL

Normal

Normal

No MUAPs

n/a

Vast lateralis

X

X

BF, SH

3+

Glut medius

X

X

Normal

Normal

Glut maximus

X

X

Normal

Normal

Semitendinosus

X

Mod neurogenic

Moderate CMAL

Low L psp

X

X

–

–

High S psp

X

X

–

–

TA

X

X

Normal

Normal

Gastroc, MH

X

X

Normal

Normal

2+

LEFT

331

Section 5: Case Studies in Electrodiagnostic Medicine

Needle EMG Study

The needle EMG abnormalities are confined to the muscle domain of the sciatic nerve. The gluteus medius and gluteus maximus muscles are normal, arguing against a lumbosacral plexopathy. The contralateral TA and gastrocnemius muscles serve to determine the normal MUAPs of these two muscles (i.e., to determine the normal MUAP duration so that the degree of CMAL can be better approximated). In this patient, because the lesion was 6 months old and there was significant CMAL present on the needle EMG study, we used the negative AUC values to approximate lesion severity. These calculations identified that 100% of the motor axons innervating the EDB muscle are involved, at least 87% of the motor axons to the abductor hallucis are involved (2.4/18.3
100% = 13%), at least 92% of the motor axons to the tibialis anterior are involved (3.1/40.5
100% = 8%), and at least 93% of the motor axons to the medial head of the gastrocnemius muscle are involved (0.7/10.2
100% = 7%). The term at least is used because, due to reinnervation via collateral sprouting (based on the significant CMAL noted on needle EMG), the values underestimate the number of disrupted motor axons. These values indicate that the lesion is very severe in degree. EDX Conclusion

The EDX study identifies a right sciatic neuropathy that is axon loss in nature and extremely severe in degree. Despite the presence of fibrillation potentials and, hence, the potential for reinnervation, the severity of the lesion suggests that any subsequent reinnervation is unlikely to result in significant functional motor improvement for the muscles located below the knee. Assuming no concomitant connective tissue proliferation, reinnervation via proximodistal axon advancement can, in theory, reach the more proximal sciatic nerve innervated muscles (e.g., hamstrings). The likelihood of no concomitant connective tissue proliferation is low given that these lesions are usually traction injuries (traction injuries are often associated with lengthy neuromas).

Exercise 5 A 73-year-old left hand–dominant male is referred for EDX assessment of bilateral hand numbness and tingling. The symptoms started on the right side just over 10 years ago and on the left side 1–2 years ago. Currently the symptoms are present upon awakening, precipitated by driving (relieved with limb lowering), and occur spontaneously while seated at rest. Recently, he developed significant right upper extremity aching pain that awakens him from sleep, and the sensory symptoms have evolved from episodic to nearly continuous on that side. He denies neck pain. The initial sensory NCS include screening studies on the right (the limb with the longer duration and more pronounced symptoms) and a contralateral Median-D2. Palmar NCS may or may not be required. Nerve Conduction Studies

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 5 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Median-D2 Ulnar-D5 Superficial Radial

NR 2.7

11.9

NR 2.6

11.1

2.2

15.7

The screen sensory NCS show absent Median-D2 responses bilaterally, consistent with an axon loss process involving the median nerve, lateral cord, upper or middle plexus, or C6 or C7 DRG on both sides. Although this likely represents advanced carpal tunnel syndrome, the lack of a demyelinating focus does not permit localization.

332

Case 1 through Case 50

Because the sensory NCS are typically more affected than the motor NCS, evidence of focal demyelination can be sought on the median motor NCS subsequently. The palmar NCS were not performed, although they might have identified a demyelinating focus. To surround the abnormal with normal, the left Ulnar-D5 sensory NCS was also performed. At this point, routine motor NCS can be performed on the right with median motor NCS added on the left. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 5 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

2.6

11.1

2.2

15.7

CV

nAUC

SENSORY Median-D2

NR

Ulnar-D5

2.7

NR

11.9

Superficial Radial

MOTOR Median-APB

4.4

3.6 3.2

NR 51.7

Ulnar-ADM

2.3

9.2 7.8

54.9

The right Median-APB motor response is absent. When coupled with the sensory NCS, the lesion is localizable to the right median nerve. Thus, this is a median neuropathy that involves the sensory and motor fibers and is axon loss in nature. However, because there is no concomitant demyelination, the lesion cannot be more specifically localized. For this reason, the Median-L2 motor NCS is added because this response may be present despite the Median-APB response being absent. When it is present, it may show concomitant focal demyelination, allowing the median neuropathy to be localized distal to the wrist stimulation site (i.e., to the carpal tunnel). In our EMG laboratories, this is the main use of this motor NCS. Regarding the left side, the left Median-APB response is low in amplitude (consistent with axon loss) and delayed in onset (consistent with focal demyelination distal to the stimulation site). Thus, the left side is localizable and consistent with carpal tunnel syndrome. Based on the left side, it is reasonable to infer that, because the left hand demonstrates features of carpal tunnel syndrome and because the hand symptoms are the same on both sides, that the right side reflects carpal tunnel syndrome as well. Nonetheless, in our EMG laboratories, we add the Median-L2, and make inferences only as a last resort.

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 5 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Median-D2 Ulnar-D5 Superficial Radial

NR 2.7

11.9

NR 2.6

11.1

2.2

15.7

333

Section 5: Case Studies in Electrodiagnostic Medicine

(cont.)

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 5 NCS PERFORMED

Stim Site

LAT

AMP

4.4

3.6

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

MOTOR Median-APB

NR

3.2

51.7

Ulnar-ADM

2.3

9.2 7.8

Median-L2

3.4

1.9

5.7

54.9

0.7

The right Median-L2 motor response is low in amplitude (approximately one-third the size of the left side) and delayed in onset (compared to the left side). Its delayed onset latency allows it to be localized to the median nerve between the stimulating and recording electrodes, which is consistent with carpal tunnel syndrome. In our EMG laboratories, when a patient has the clinical features of carpal tunnel syndrome and does not have features of another disorder and the referring provider has not suggested another disorder, we typically perform an abbreviated needle EMG study including the APB muscle (C8,T1-median nerve innervated), pronator teres (C6,7median nerve innervated), and the FDI (C8,T1-ulnar nerve innervated). The Needle EMG Study

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 5

Insertional Activity Normal

IPSWs SCP Other

Spontaneous Activity None

Fibs

Fascs Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

RIGHT APB

2+

Severe neurogenic Severe CMAL

FDI

X

X

Normal

Normal

Pron teres

X

X

Normal

Normal

Mod neurogenic

Severe CMAL

LEFT APB

1+

2+

FDI

X

X

Normal

Normal

Pron teres

X

X

Normal

Normal

The needle EMG study shows severe chronic changes in the right and left APB muscles (right worse than left regarding recruitment) and fibrillation potentials. The presence of insertional positive sharp waves in the left APB muscle is consistent with a progressive process.

334

Case 1 through Case 50

EMG Study Conclusion

1. Bilateral Median Neuropathies (carpal tunnel syndrome) –

The above are demyelinating and axon loss in nature, involve the sensory and motor nerve fibers, and are located distal to the wrist stimulation sites. Electrically, the abnormalities are extremely severe on the right and very severe on the left.

–

Given the severity of the right upper extremity pain, the patient underwent a right carpal tunnel release procedure knowing that there would be little chance of any meaningful functional motor recovery. The right upper extremity aching pain resolved. He subsequently had a left carpal tunnel release procedure to avoid the development of aching pain on that side. Some improvement in motor function is anticipated based on the degree of spatial recruitment (the motor response amplitude is misleading in the chronic setting).

Exercise 6 A 64-year-old male is referred for EDX assessment of radiating left hip pain. According to the patient, 6 weeks before the study, he developed sudden-onset left hip and buttock pain. The pain radiates down the left lower extremity to the left foot and is associated with numbness and tingling of the sole of the left foot. He denies right lower extremity symptoms. Clinically, the radiating nature of the pain suggests nerve fiber involvement. Because it radiates below the knee, it suggests nerve root involvement. The distribution of the associated tingling (the sole of the foot) suggests S1 nerve root involvement. Based on this presentation, the sensory NCS are expected to be normal. In general, we usually perform one sensory NCS on the contralateral side. Given that his sensory symptoms are in the S1 distribution, we added the contralateral sural sensory NCS rather than the superficial peroneal sensory NCS. Nerve Conduction Studies

LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 6 NCS PERFORMED

Stim Site

LAT

AMP

Sural

4.2

5.1

Supfcl Peroneal

2.9

7.8

RIGHT CV

nAUC

LAT

AMP

4.3

5.7

CV

nAUC

SENSORY

As expected with a radiculopathy, the screening sensory NCS are normal. Had either been abnormal, it would have indicated a ganglionic/postganglionic localization and prompted performance of both studies on the contralateral side. At this point, the routine motor NCS and H reflex are performed. The H reflex is especially important in this case because it is very sensitive to S1 radiculopathies. LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 6 NCS PERFORMED

Stim Site

LAT

AMP

Sural

4.2

5.1

Supfcl Peroneal

2.9

7.8

RIGHT CV

nAUC

LAT

AMP

4.3

5.7

CV

nAUC

SENSORY

335

Section 5: Case Studies in Electrodiagnostic Medicine

(cont.)

LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 6 NCS PERFORMED

Stim Site

LAT

AMP

5.5

3.0

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

MOTOR Tibial-AH

1.7 Peroneal-EDB

47.1

NR

M wave

5.8

H wave

8.2 NR

The initial motor NCS show an absent left peroneal motor response (recording EDB), a low-amplitude tibial motor response (recording AH), and an absent left H wave. The combination of low-amplitude motor responses with normal sensory responses indicates an axon loss process localized to the intraspinal canal. The EDB muscle is an L5, S1 nerve root innervated muscle (especially L5) and, because this response is absent, suggests involvement of both L5 and S1. However, shoe wear can cause this response to be absent. The low-amplitude tibial motor response supports axon loss involving the S1 or S2 nerve roots (clinically, an S2 radiculopathy is infrequent). The absent H wave supports S1 nerve root involvement (it would not be affected by an S2 radiculopathy), but should be compared to the other side (bilateral absence in a 64-year-old may not be pathological). At this point, it appears that there is motor axon loss involving the left L5 and S1 nerve roots. Because the Peroneal-EDB response is absent, the Peroneal-TA is added to the ipsilateral motor NCS. On the contralateral side, the three motor responses and the H reflex are needed. The contralateral tibial motor response will permit grading of the ipsilateral low-amplitude tibial response, the peroneal-EDB response will be useful to address the possibility of show wear–related decrement (should be bilateral and roughly symmetric), the peroneal-TA response is for comparison to the ipsilateral response after it is done, and the H reflex is to screen the contralateral S1 nerve root.

LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 6 NCS PERFORMED

Stim Site

LAT

AMP

Sural

4.2

5.1

Supfcl Peroneal

2.9

7.8

5.5

3.0

RIGHT CV

nAUC

LAT

AMP

4.3

5.7

5.3

4.8

NR

4.3

5.7

4.4

3.1

5.3

SENSORY

MOTOR Tibial-AH

1.7 Peroneal-EDB

Peroneal-TA

3.1

4.3

336

47.1

52.9

CV

nAUC

Case 1 through Case 50

(cont.)

LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 6 NCS PERFORMED

Stim Site

M wave

LAT

AMP

5.8

8.2

H wave

RIGHT CV

nAUC

LAT

AMP

5.6

9.7

NR

CV

nAUC

NR

The right-sided motor responses are normal and the right H wave is absent. This suggests that the right S1 nerve root might also be involved (versus age-related H-wave loss). This can be determined on the needle EMG study. Needle EMG Study

LOWER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 6

Insertional activity Normal IPSWs

SCP Other

Spontaneous Activity None

Fibs

Fascs Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

Normal

Moderate CMAL

LEFT FHB

X

FDL

X

X

Normal

Mild CMAL

TA

X

X

Normal

Mild CMAL

Normal

Mild CMAL

Gastroc, MH

3+

1+

2+

Vast lateralis X

X

Normal

Normal

BF, SH

X

X

Normal

Mild CMAL

Glut medius X

X

Normal

Mild CMAL

–

–

Low L psp

X

1+

High S psp

X

X

–

–

FHB

X

X

Normal

Normal

FDL

X

X

Normal

Normal

TA

X

X

Normal

Normal

Gastroc, MH

X

X

Normal

Normal

RIGHT

337

Section 5: Case Studies in Electrodiagnostic Medicine

The needle EMG study shows fibrillation potentials in the left lower lumbar paraspinal muscles, indicating an intraspinal canal localization. Because each segment of paraspinal muscles receives its innervation from multiple nerve roots, the presence of paraspinal muscle fibrillation potentials does not localize the intraspinal canal lesion to a specific nerve root level. The limb muscles are more helpful in this regard. In this case, the fibrillation potentials are in the muscle domain of the left S1 nerve root – the FHB and gastrocnemius (medial head) muscles. The insertional positive sharp waves noted in the medial head of the left gastrocnemius muscle are indicative of recent denervation (typically within 14 days), suggesting that this is a progressive process rather than a monophasic one. Also (not noted in the table), some of the fibrillation potentials were high in amplitude, suggesting acuteness, whereas others were very low in amplitude, suggesting chronicity. The chronic changes indicate reinnervation via collateral sprouting. Performing contralateral needle EMG assessments is the most helpful way to grade the degree of CMAL. With this approach, the changes ranged from mild to moderate and were present in the muscle domains of the left L5 and S1 nerve roots. There were no acute or chronic changes on the right side, making the cause for the absent right H wave unclear (age versus early S1 nerve root involvement). EMG Conclusion

1. Intraspinal Canal Lesion (e.g., radiculopathies) –

The above is axon loss in nature and involves the left L5 and S1 nerve roots. The presence of chronic motor axon loss indicates reinnervation through collateral sprouting and, additionally, indicates that these abnormalities preceded the left hip and buttock pain that started 6 weeks ago. Thus, this appears to be an acute process superimposed on a chronic process. The presence of insertional positive sharp waves suggests that this is a progressive process (because insertional positive sharp waves typically appear between day 14 and day 21; thus, their presence suggests very recent worsening). The restriction of fibrillation potentials to the muscle domain of the left S1 nerve root suggests that this is an acute left S1 radiculopathy superimposed on chronic left L5 and S1 radiculopathies. The absent right H wave most likely is related to right S1 nerve root involvement not recognized on needle EMG study.

–

This interpretation is concordant with the clinical features suggesting the onset of a left S1 radiculopathy 6 weeks earlier. Also, the relationship between the acute and chronic abnormalities suggests a slowly progressive process, such as spondylosis.

Exercise 7 A 51-year-old right hand–dominant male is referred for EDX assessment of numbness and tingling of the left upper extremity. These symptoms began 5 months ago and along the medial aspect of the left hand. He denies loss of grip strength. On examination there is sensory loss involving the medial 1.5 digits (i.e., he splits the 4th digit) and the medial aspect of the left hand, both dorsally and ventrally. He also has moderate weakness of finger abduction and the long flexors of the 4th and 5th digits. These clinical features suggest an ulnar neuropathy, primarily because the patient splits the 4th digit (almost always related to an ulnar neuropathy). Features localizing lesion along the ulnar nerve include sensory involvement of the dorsomedial aspect of the hand (i.e., the lesion involves the dorsal ulnar cutaneous sensory axons) and weakness of the flexor digitorum profundus (FDP) muscles of the 4th and 5th digits (i.e., the lesion involves motor axons to the FDP). Thus, based on the clinical features, an elbow segment lesion is suspected. The initial sensory NCS include routine studies plus the DUC and MABC sensory NCS.

338

Case 1 through Case 50

Nerve Conduction Studies

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET EXERCISE 7 NCS PERFORMED

LEFT DRG

LAT

AMP

Median-D2

C6,7

3.3

30.0

Ulnar-D5

C8

2.9

15.0

Superficial Radial

C6,7

2.5

39.3

DUC

C8

2.2

10.3

MABC

T1

2.1

14.6

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY

The initial sensory NCS are normal. However, the DUC sensory NCS is near the lower limit of normal (10.0) for this patient. Thus, it should be performed on the contralateral side to exclude a relative abnormality. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET EXERCISE 7 NCS PERFORMED

LEFT DRG

LAT

AMP

Median-D2

C6,7

3.3

30.0

Ulnar-D5

C8

2.9

15.0

Superficial Radial

C6,7

2.5

39.3

DUC

C8

2.2

10.3

MABC

T1

2.1

14.6

RIGHT CV

nAUC

LAT

AMP

2.9

19.0

2.0

26.2

CV

nAUC

SENSORY

The contralateral DUC response is more than twice the size of the ipsilateral response, indicating a relative abnormality. For this reason, the contralateral Ulnar-D5 was added and was normal. The DUC abnormality indicates an axon loss process located at or proximal to the departure site of the DUC branch of the ulnar nerve. The distribution of sensory NCS involvement is less than the distribution on clinical examination, suggesting possible concomitant demyelinating conduction block. This is also suggested by the moderate degree of weakness on clinical examination and the sensory NCS findings. In general, with mononeuropathies demonstrating significant weakness, the sensory responses are more affected than the motor responses. On the motor NCS, the Ulnar-FDI motor response is added to the ipsilateral routine NCS, and both ulnar motor responses are performed contralaterally (for severity assessment). UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET EXERCISE 7 NCS PERFORMED

LEFT Stim Site

LAT

AMP

Median-D2

3.3

30.0

Ulnar-D5

2.9

15.0

RIGHT CV

nAUC

LAT

AMP

2.9

19.0

CV

nAUC

SENSORY

339

Section 5: Case Studies in Electrodiagnostic Medicine

(cont.)

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET EXERCISE 7 NCS PERFORMED

LEFT Stim Site

LAT

AMP

Superficial Radial

2.5

39.3

DUC

2.2

10.3

MABC

2.1

14.6

3.1

10.0

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

2.0

26.2

2.5

12.4

27.9

2.9

20.3

35.2

MOTOR Median-APB

9.9 Ulnar-ADM

Ulnar-FDI

2.6

3.0

50.8

9.7

21.3

9.7

52.5

20.1

2.7

32.0

5.7

19.8

37.1

18.8

52.5

37.0

8.7

40.5

19.7

The routine motor NCS demonstrate a demyelinating conduction block between the below-elbow and aboveelbow stimulation sites, localizing the lesion to the elbow segment. Because the DUC sensory response was reduced in amplitude (axon loss), this is a mixed lesion, with features of both axon loss and demyelination. In our EMG laboratories, in the setting of demyelination, we typically calculate severity using the negative AUC values, as they are less susceptible to pathological temporal dispersion. Based on the negative AUC values of the distal Ulnar-ADM responses of the two sides, the approximate percentage of axon loss is 24% (21.3/27.9 = 76%; 100% – 76% = 24%). Based on the negative AUC values of the above-elbow and below-elbow responses, the approximate percentage of demyelination is 72% (5.7/ 20.1 = 28%; 100% – 28% = 72%). However, this reflects 72% of the remaining motor axons, not 72% of all of the motor axons. Thus, the approximate percentage of motor axons affected by demyelinating conduction block is 72% of the remaining 76%, which is 55% (0.72
76% = 54.7%). Consequently, regarding the motor axons to the hypothenar eminence, 55% are affected by demyelinating conduction block, 24% are affected by axon loss, and 21% are normal. Importantly, the side-to-side difference between the Ulnar-ADM motor responses is not abnormal by absolute or relative criteria. Thus, it would be reasonable to conclude that there is no motor axon loss. In our EMG laboratories, we decide which approach to take based on the needle EMG study of the ADM muscle. When axon loss is present, as in this case (see further on), we use the distal motor responses to calculate the percentage of axon loss first and then determine the percentage of demyelinating conduction block. Regarding the motor axons to the FDI muscle, the degree of axon loss (if any) cannot be determined because the negative AUC value is larger ipsilaterally. On needle EMG of the FDI (see further on), there was no evidence of axon loss (no fibrillation potentials and no increase in the MUAP duration). Thus, the neurogenic recruitment appears to reflect solely the demyelinating conduction block. The above-elbow and below-elbow Ulnar-FDI responses indicate a demyelinating conduction block involving about 47% of the motor axons (19.7/ 37.0
100% = 53%; 100% – 53% = 47%). The ADM and FDP muscles are added to the screening muscles ipsilaterally, and contralateral ulnar nerve innervated muscles are added to better assess the degree of increase in the MUAP duration.

340

Case 1 through Case 50

The Needle EMG Study

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 7

Insertional Activity Normal

IPSWs SCP

Spontaneous Activity Other

None

Fibs

Fascs Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

LEFT FDI

X

X

Mild neurogenic

Normal

EI

X

X

Normal

Normal

FPL

X

X

Normal

Normal

Pron teres

X

X

Normal

Normal

BC, MH

X

X

Normal

Normal

TC, LH

X

X

Normal

Normal

FDP-3,4

X

X

Normal

Mild CMAL

ADM

X

Severe neurogenic

Moderate CMAL

Low cerv psp

X

X

–

–

High thor psp

X

X

–

–

FDI

X

X

Normal

Normal

ADM

X

X

Normal

Normal

FDP-3,4

X

X

Normal

Normal

3+

RIGHT

The needle EMG study showed fibrillation potentials in the left ADM muscle, indicating motor axon loss, as well as long duration MUAPs in the left ADM and FDP muscles, also indicating axon loss. A neurogenic MUAP recruitment pattern was observed in the left ADM (severe in degree) and FDI (mild in degree), indicating significant MUAP dropout for these two muscles. EDX Study Conclusion

1. Left Ulnar Neuropathy The above is both demyelinating conduction block and axon loss in nature, involves the sensory and motor nerve fibers, and is located at the elbow segment. The primary pathophysiology is demyelination conduction block, which involves 55% of the motor axons to the ADM muscle and 47% of the motor axons to the FDI muscle. Axon loss involves 24% of the motor axons to the ADM muscle. When we identify ulnar neuropathies that are solely or predominantly demyelinating conduction block in nature and located across the elbow segment, we ask the patient to avoid sustained forearm flexion (e.g., sleeping with elbows flexed), elbow leaning, activities requiring repetitive forearm flexion-extension movements, and placement of the symptomatic hand behind the head (this significantly increases the pressure on the ulnar nerve) pending

341

Section 5: Case Studies in Electrodiagnostic Medicine

follow-up with his referring physician. In addition, we suggest that should the patient be treated conservatively, that the ulnar nerve be restudied in 3–4 months. When the follow-up study shows remyelination failure or lesion progression, we suggest to the referring physician that surgical intervention be considered. Of course, these are standard suggestions. Each case must be individualized based on the underlying etiology, the severity of demyelinating conduction block (if it transforms to axon loss, the prognosis is worse), the effect the process has on ADLs (including occupation), the wishes of the patient, and other features.

Exercise 8 A 48-year-old right hand–dominant female is referred for EDX assessment of distal left leg and foot burning, tingling, and numbness following a total abdominal hysterectomy 23 days ago. According to the patient, on the night of the procedure, she developed burning, tingling, and numbness along the top of the left foot. Because the symptoms followed the total abdominal hysterectomy, a sciatic neuropathy is suggested. In our EMG laboratories, we typically perform more extensive EDX studies whenever an iatrogenic injury is possible, because these cases often go to litigation, and extensive studies are more helpful to all parties involved when done early. Thus, contralateral sensory NCS and distal motor responses are included in the initial sensory and motor NCS. Nerve Conduction Studies

LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 8 NCS PERFORMED

Stim Site

LAT

AMP

4.1

RIGHT CV

nAUC

LAT

AMP

4.3

3.8

8.7

NR

2.5

11.3

CV

nAUC

SENSORY Sural Supfcl Peroneal

The screening sensory NCS are remarkable for an absent left superficial peroneal response and a mildly reduced left sural response amplitude. The contralateral sensory NCS are normal. The absent left superficial peroneal sensory response indicates an axon loss process located at the superficial peroneal nerve, common peroneal nerve, sciatic nerve, lumbosacral plexus, or DRG. The low-amplitude sural response excludes a superficial peroneal or common peroneal localization. Unlike the sensory axons assessed during the upper extremity sensory NCS, the specific DRG assessed by the various lower extremity sensory NCS are unclear. Clinically, based on the distribution of numbness associated with superficial peroneal and sural neuropathies, it seems like the sensory axons assessed by the superficial peroneal sensory NCS primarily emanate from the L5 DRG, and that the sensory axons assessed by the sural sensory NCS primarily emanate from the S1 DRG. However, whether they also assess sensory axons emanating from adjacent DRG and what the individual frequencies are is unclear. In summary, the sensory NCS indicate an axon loss process involving the peripheral nervous system at or proximal to the sciatic nerve. The initial motor NCS are expanded to include contralateral distal motor responses and a contralateral H reflex.

342

Case 1 through Case 50

LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 8 NCS PERFORMED

Stim Site

LAT

AMP

4.1

RIGHT CV

nAUC

LAT

AMP

4.3

3.8

8.7

NR

2.5

11.3

6.3

4.2

7.8

3.3

4.8

CV

nAUC

SENSORY Sural Supfcl Peroneal

MOTOR Tibial-AH

4.2

5.9 Peroneal-EDB

3.4

48.1

5.8 5.2

47.6

Peroneal-TA

M wave

4.7

8.6

4.8

8.4

H wave

32.1

0.8

31.9

2.7

The initial motor NCS identify a low-amplitude left H wave. This indicates an axon loss process involving the afferent S1 or efferent S1 axons, including the neurons from which they are derived (i.e., the S1 loop). Because the sensory NCS indicate a ganglionic or postganglionic lesion, at this point, the lesion can be localized to the sciatic nerve or the lumbosacral plexus. On needle EMG, the glutei muscles are helpful in differentiating between these two possibilities (i.e., spared with sciatic neuropathies; involved with plexopathies). Needle EMG Study

LOWER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 8

Insertional Activity Normal

IPSWs SCP

Spontaneous Activity Other None Fibs

Fascs Other

MUAP Analysis MUAP MUAP Recruitment Morphology

LEFT FHB

X

FDL

1+

1+

Normal

Normal

1+

Normal

Normal

TA

X

1+

Normal

Normal

Gastroc, MH

X

1+

Normal

Normal

Vast lateralis

X

Normal

Normal

Normal

Normal

BF, SH

X 1+

2+

Glut medius

X

X

Normal

Normal

Glut maximus

X

X

Normal

Normal

EHL

X

X

Normal

Normal

343

Section 5: Case Studies in Electrodiagnostic Medicine

(cont.)

LOWER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 8

Insertional Activity Normal

IPSWs SCP

Spontaneous Activity Other None Fibs

Fascs Other

MUAP Analysis MUAP MUAP Recruitment Morphology

Peron longus

X

1+

Normal

Normal

Semitendinosus

X

1+

Normal

Normal

Low L psp

X

X

–

–

High S psp

X

X

–

–

FHB

X

X

Normal

Normal

FDL

X

X

Normal

Normal

TA

X

X

Normal

Normal

Gastroc, MH

X

X

Normal

Normal

BF, SH

X

X

Normal

Normal

RIGHT

The needle EMG study shows a small number of insertional positive sharp waves and a larger number of fibrillation potentials in the muscle domain of the left sciatic nerve. The MUAPs are normal in appearance, consistent with the onset of symptoms 23 days ago, and the MUAP recruitment pattern is normal. The gluteus medius and gluteus maximus muscles are normal, arguing against a plexus level lesion. Several contralateral muscles are added, due to the potential medicolegal implications. There is no evidence of relative MUAP abnormalities. EDX Study Conclusion

1. Left Sciatic Neuropathy The above is axon loss in nature, involves the sensory and motor nerve fibers, and is mild in degree (based on the motor response symmetries) except for the sensory axons destined for the superficial peroneal nerve, which are more severely affected. Approximately 3 months later, the patient is referred back to the EMG laboratory for worsening weakness. She is now unable to move her toes, foot, leg, or thigh. The sensory disturbances are unchanged in their distribution (lateral aspect of the left leg, distally, and the top of the left foot). The left lower extremity has “slowly become paralyzed.” Repeat NCS

LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 8 NCS PERFORMED

Stim Site

LAT

AMP

Sural

4.1

Supfcl Peroneal

3.1

RIGHT CV

nAUC

LAT

AMP

4.6

3.8

8.9

2.4

2.5

12.6

SENSORY

344

CV

nAUC

Case 1 through Case 50

(cont.)

LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 8 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

MOTOR Tibial-AH

4.4

6.0 5.7

Peroneal-EDB

3.5

5.9

3.4

8.2

3.3

5.1

3.3

4.8

48.5

5.4 Peroneal-TA

4.3

48.4

5.1 5.0

54.7

M wave

4.7

7.7

4.6

7.8

H wave

32.3

0.6

30.4

2.4

The sensory NCS show the same pattern of axon loss, with the superficial peroneal response much more affected than the sural response. The motor NCS are normal and similar to the responses recorded from the contralateral side. The needle EMG study is expanded to better define the lesion. Repeat Needle EMG Study

LOWER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 8

Insertional Activity Normal IPSWs

SCP

Spontaneous Activity

Other None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

LEFT FHB

X

X

Rare fire

Normal

FDL

X

X

Rare fire

Normal

TA

X

X

Variable rate

Normal

Gastroc, MH

X

X

Variable rate

Normal

Vast lateralis

X

X

Variable rate

Normal

None fire

Mild CMAL

BF, SH Glut medius

X

X

Variable rate

Normal

Glut maximus

X

X

Variable rate

Normal

EHL

X

X

Rare fire

Normal

Peron longus

X

X

Rare fire

Normal

Semitendinosus

X

X

Variable rate

Normal

Low L psp

X

X

–

–

High S psp

X

X

–

–

345

Section 5: Case Studies in Electrodiagnostic Medicine

(cont.)

LOWER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 8

Insertional Activity Normal

IPSWs

SCP Other

Spontaneous Activity None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

RIGHT FHB

X

X

Normal

Normal

FDL

X

X

Normal

Normal

TA

X

X

Normal

Normal

Gastroc, MH

X

X

Normal

Normal

BF, SH

X

X

Normal

Normal

The needle EMG study shows normal insertional activity (indicates viability of the muscle fibers) and a lack of fibrillation potentials (consistent with a lack of denervated muscle fibers). The MUAPs have a normal appearance (i.e., no evidence of reinnervation via collateral sprouting). The recruitment pattern of the intermittently activated motor units is normal (not neurogenic or early). The firing rate is variable, in the slow to medium range, consistent with variation in effort. During each needle EMG study, the patient denied pain as the reason for the inability to activate the muscles. 1. Left Sciatic Neuropathy –

The above is axon loss in nature, involves the sensory and motor axons, and is mild in degree. Since the previous study, there has been mild improvement of the left superficial peroneal sensory response, the fibrillation potentials have resolved, and a single muscle shows evidence of mild chronic motor axon loss when compared to the contralateral side.

The reason for the “paralysis” is unclear. The motor NCS stimulations are able to activate the “paralyzed” muscles. The insertional activity is normal for all studied muscles, there are no fibrillation potentials, and the MUAP recruitment is normal. A single muscle (BF, SH) shows mild chronic changes consistent with successful reinnervation of the previously denervated muscle fibers. The other muscles that previously showed fibrillation potentials no longer show them, indicating successful reinnervation. Moreover, because the MUAPs do not show duration increases, the chronic motor axon loss is below the resolution of the needle EMG study. Comparison to the contralateral muscles shows no difference. In general, surgically induced neuropathies are monophasic events that do not subsequently progress. An exception to this statement is when there is concomitant vessel damage that results in a slow leak and a subsequent compartment syndrome, but this is an ischemic phenomenon that produces axon loss. Thus, at this point, conversion/hysteria or malingering is suspected. In general, when a muscle cannot be voluntarily activated, the list of possibilities includes an upper motor neuron or lower motor neuron process, as well as malingering, factitious disorder, hysteria/conversion, and painlimited movement. The EDX examination is not a sensitive test for identifying upper motor neuron disorders. However, the clinician can perform a thorough neurological examination and appropriate imaging studies to exclude this possibility. In this case, the whole limb is variably paralyzed and the “weakness” does not demonstrate an upper motor neuron pattern. The EDX study excludes a lower motor neuron process (i.e., radiculopathy, plexopathy, neuropathy). Once upper and motor neurons disease is excluded, the remaining possibilities are malingering, factitious disorder, hysteria/ conversion, and pain-limited movement. However, the patient denied pain-limitations during the needle EMG study (it is important to document a lack of pain during the needle EMG when malingering or conversion is suspected). Thus, only malingering, factitious disorder, and hysteria/conversion are possibilities. The latter can be addressed by consultation with mental health.

346

Case 1 through Case 50

Again, in the setting of iatrogenic injury, it is important that the EDX study be complete. In this way, should the patient “worsen” and have a subsequent EMG study performed by a less experienced electromyographer or by an electromyographer who consistently works for the plaintiff attorney, the defense is better prepared to address any “new” findings.

Exercise 9 A 49-year-old right hand–dominant male is referred for EDX assessment of episodic left > right hand numbness and tingling. The hand symptoms are present upon awakening, are precipitated by driving, and occur spontaneously while seated at rest. The symptoms are much worse on the left. He denies neck pain. The clinical features are strongly suggestive of carpal tunnel syndrome. Given that the patient is right hand dominant and the left hand is the more symptomatic limb (symptoms are usually most pronounced on the side of the dominant limb), two possibilities exist: (1) the right side is less symptomatic because it has more advanced disease, or (2) the patient has nondominant limb predominant carpal tunnel syndrome. The latter is frequently seen when occupations or hobbies require the use of both upper extremities in which one limb performs intricate activities and the other performs sustained gripping. In this situation, most individuals use the dominant limb for the more intricate activity and the nondominant limb for the sustained gripping. Sustained gripping is a risk factor for carpal tunnel syndrome (Ferrante, 2016). Because the symptoms are bilateral and worse on the left, the initial sensory NCS include screening studies on the left and contralateral median studies. In our EMG laboratories, when the Median-D2 responses are normal or only mildly delayed, we add the palmar NCS. Nerve Conduction Studies

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 9 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

4.2

10.0

CV

nAUC

SENSORY Median-D2

5.6

8.3

Ulnar-D5

3.2

7.1

Superficial Radial

2.6

13.3

The initial sensory NCS demonstrate Median-D2 responses with delayed peak latencies (demyelination) and low amplitudes (axon loss), left more pronounced than right. Thus, the lesion is mixed (axon loss and demyelination), and it is the presence of the demyelination that permits localization to somewhere between the stimulating and recording electrodes. These EDX features are consistent with bilateral carpal tunnel syndrome, nondominant limb more pronounced than dominant limb. Because the nondominant limb abnormalities are more pronounced than the dominant limb abnormalities, an occupation history is performed. The patient works as a pipe fitter, cutting pipes 6 hours of his 8-hour shift. He uses his left hand to firmly hold the pipe in place while using his right hand to bring the saw blade down to cut the pipe. Because sustained gripping is a risk factor for carpal tunnel syndrome and because this patient uses his left hand to grip the pipe while directing the saw through it with his right hand, his condition is very likely occupation related (Ferrante, 2016). The initial motor NCS include screening studies on the left side and a contralateral Median-APB motor NCS.

347

Section 5: Case Studies in Electrodiagnostic Medicine

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 9 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

4.2

10.0

4.1

6.9

CV

nAUC

SENSORY Median-D2

5.6

8.3

Ulnar-D5

3.2

7.1

Superficial Radial

2.6

13.3

5.4

5.2

MOTOR Median-APB

5.1 Ulnar-ADM

2.8

53.8

6.5

52.7

8.7 8.1

52.5

The motor NCS show increased median motor response onset latencies (left worse than right) and a lowamplitude left median motor response. As indicated by the sensory NCS, the motor NCS indicate bilateral carpal tunnel syndrome, left worse than right. In our EMG laboratories, when the clinical features suggest carpal tunnel syndrome and the NCS identify that disorder, as long as there are no clinical features to suggest a concomitant disorder and as long as the referring provider did not suggest an alternative diagnosis (e.g., radiculopathy; ulnar neuropathy), we perform an abbreviated needle EMG study. The Needle EMG Study

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 9

Insertional Activity Normal

IPSWs

SCP

Spontaneous Activity Other

None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

LEFT APB

X

X

Normal

Moderate CMAL

FDI

X

X

Normal

Normal

Pron teres

X

X

Normal

Normal

X

X

Normal

Mild CMAL

RIGHT APB

The needle EMG study shows chronic changes in the left APB muscle and less severe chronic changes in the right APB muscle, again consistent with bilateral carpal tunnel syndrome, left worse than right. On the right side, the needle EMG study was limited to the APB muscle, given the similarity of the symptoms on the two sides and the normal FDI and pronator teres on the more symptomatic side.

348

Case 1 through Case 50

EDX Study Conclusion

1. Bilateral Median Neuropathies (carpal tunnel syndrome) –

The above are demyelinating and axon loss in nature, involve the sensory and motor nerve fibers, and are located at or distal to the wrist. The abnormalities are more severe on the left side.

–

This pattern of nondominant limb predominant carpal tunnel syndrome is frequently seen among individuals using their nondominant limb for sustained gripping, such as pipefitters (Ferrante, 2016).

Exercise 10 A 54-year-old right hand–dominant male is referred for EDX assessment of right-sided neck pain that began approximately 5 weeks prior to the study. The pain extends into the scapular and shoulder regions but does not radiate down the arm. Concomitant with the onset of the neck pain, he noted right thumb tingling. Sustained neck extension (20 seconds) precipitates tingling of the tip of the right third digit but not radiating neck pain. He denies symptoms on the left side. Clinically, the simultaneous onset of the right thumb tingling and the neck pain suggests C6 nerve root involvement, whereas the precipitation of right third digit tingling with sustained neck extension suggests right C7 nerve root involvement. Thumb tingling could also represent carpal tunnel syndrome, although the clinical features do not support this possibility. Thus, the initial sensory NCS are expanded to include routine sensory NCS on the right, bilateral Median-D1 NCS, and palmar NCS. Nerve Conduction Studies

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 10 NCS PERFORMED

Stim Site

LAT

AMP

3.2

21.4

RIGHT CV

nAUC

LAT

AMP

3.1

22.6

Median-D2

2.9

27.7

Ulnar-D5

2.8

18.3

Superficial Radial

2.3

45.0

Median Palmar

2.1

43.7

Ulnar Palmar

2.0

15.3

CV

nAUC

SENSORY Median-D1

The initial sensory NCS are normal. There is no EDX evidence of carpal tunnel syndrome, although normal NCS do not exclude early carpal tunnel syndrome. The Median-D1 responses are normal and symmetric. The constant right thumb tingling when considered in conjunction with a normal Median-D1 sensory response suggests a preganglionic process. There is no indication for additional motor NCS, so only the routine screening motor NCS are performed.

349

Section 5: Case Studies in Electrodiagnostic Medicine

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 10 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

Median-D2

2.9

26.7

Ulnar-D5

2.8

21.3

Superficial Radial

2.3

45.0

CV

nAUC

SENSORY

Median Palmar

2.1

43.7

Ulnar Palmar

2.0

15.3

3.0

10.8

MOTOR Median-APB

10.6 Ulnar-ADM

2.6

57.3

12.5 9.9

32.1 56.3

29.8

The routine motor NCS were normal. Thus, at this point, there are no EDX abnormalities. The needle EMG study is expanded to include additional C6 and C7 nerve root innervated muscles. UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 10

Insertional Activity Normal

IPSWs

SCP Other

Spontaneous Activity None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

LEFT FDI

X

X

Normal

Normal

EI

X

X

Normal

Normal

FPL

X

X

Normal

Normal

Pron teres

X

2+

Normal

Severe CMAL

BC, MH

X

1+

Normal

Mild CMAL

TC, LH

X

2+

Mild neurogenic

Moderate CMAL

Deltoid, MH

X

Normal

Normal

Normal

Mild CMAL

Brachioradialis

350

X 1+

1+

Middle cerv psp

X

X

–

–

Low cerv psp

X

X

–

–

High thor psp

X

–

–

2+

Case 1 through Case 50

(cont.)

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 10

Insertional Activity Normal IPSWs

SCP

Spontaneous Activity

Other None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

RIGHT Brachioradialis

X

X

Normal

Normal

Pron teres

X

X

Normal

Normal

BC, MH

X

X

Normal

Normal

The needle EMG study shows fibrillation potentials in the paraspinal muscles, indicative of an intraspinal canal lesion (e.g., radiculopathy). There are insertional positive waves in the right brachioradialis muscle, indicative of recent muscle fiber denervation. This suggests that the lesion is progressive because the symptoms started about 5 weeks before the study, but the insertional positive sharp waves could not be more than 2–3 weeks of age. There are fibrillation potentials in the muscle domains of the right C6 and C7 nerve roots. There are features of chronic motor axon loss (e.g., longduration MUAPs, indicative of reinnervation via collateral sprouting) in the right C6 and C7 nerve root distributions. The latter are most pronounced in muscles receiving innervation from both the C6 and C7 nerve roots (i.e., the pronator teres and triceps muscles), with lesser involvement of the C5,6 muscles. Thus, although the smallest lesion that could account for the needle EMG abnormalities is the right C6 nerve root, the pattern of fibrillations is more consistent with concomitant right C6 and C7 nerve root involvement. This also is consistent with the history. EMG Conclusion

1. Right C6 Radiculopathy and probable C7 Radiculopathy –

Although the needle EMG abnormalities could be explained by isolated right C6 nerve root involvement, the fact that the C6,7 muscles are more profoundly involved than the C5,6 muscles suggests concomitant C7 nerve root involvement. Because the chronic changes are so much more pronounced than the acute ones, the process appears to be acute on chronic and slowly progressive.

Exercise 11 A 60-year-old right hand–dominant female is referred for EDX assessment of episodic left hand tingling. The symptoms are most pronounced when she is lying down, are present upon awakening, are precipitated by activities requiring sustained limb elevation (e.g., driving and applying cosmetics) and relieved with limb lowering, and also occur spontaneously while seated at rest. She had similar symptoms on the right side that resolved following a carpal tunnel release procedure 18 years ago. She denied neck pain. Clinically, the features suggest carpal tunnel syndrome, which, in this case, is even more likely because she had similar symptoms on the contralateral side that resolved following a carpal tunnel release procedure. The initial sensory NCS are expanded to include the palmar NCS. Nerve Conduction Studies

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 11 NCS PERFORMED

DRG

LAT

AMP

C6,7

5.6

10.7

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Median-D2

351

Section 5: Case Studies in Electrodiagnostic Medicine

(cont.)

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 11 NCS PERFORMED

DRG

LAT

AMP

Ulnar-D5

C8

2.9

28.3

Superficial Radial

C6,7

2.4

30.9

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

The initial sensory NCS performed on the left upper extremity identify a very delayed Median-D2 response that also is relatively low in amplitude. Regarding the delay, given its severity, palmar NCS are unnecessary. Regarding the amplitude of this response, although it is normal for age (i.e., greater than 10 microvolts), it is only about one-third the size of the ulnar response amplitude (the lower limit of normal for the ulnar response is 5 microvolts). To verify the suspected amplitude decrement, the contralateral Median-D2 response is added. Of course, depending on how severe the original carpal tunnel syndrome was, this comparison may not be beneficial. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 11 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

5.6

10.7

Ulnar-D5

C8

2.9

28.3

Superficial Radial

C6,7

2.4

30.9

RIGHT CV

nAUC

LAT

AMP

3.3

24.0

CV

nAUC

SENSORY

The right Median-D2 response amplitude is 24 microvolts, and thus, the left Median-D2 response is reduced. Thus, there is concomitant axon loss. The right Median-D2 response is relatively low in amplitude compared to the left Ulnar-D5 response (usually the Median-D2 response amplitude is larger than the Ulnar-D5 response amplitude). This suggests that the previous right median neuropathy was severe enough to produce axon loss. This can be assessed by needle EMG of the right APB muscle. At this point, the motor NCS are performed. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 11 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

5.6

7.7

Ulnar-D5

C8

2.9

28.3

Superficial Radial

C6,7

2.4

30.9

5.8

10.4

RIGHT CV

nAUC

LAT

AMP

3.3

24.0

SENSORY

MOTOR Median-APB

10.1 Ulnar-ADM

2.9

8.9 8.7

352

56.0

58.3

CV

nAUC

Case 1 through Case 50

The routine motor NCS are normal. Given the high amplitude of the left Median-APB response, the right side was not studied (the possibility of a relative abnormal is essentially zero). Thus, the possibility of left median motor axon loss will be assessed during the needle EMG study. In our EMG laboratories, when a patient presents with classical features of carpal tunnel syndrome and denies neck pain or any other symptoms that would suggest an alternative or concomitant diagnosis, whenever the NCS identify carpal tunnel syndrome, we limit the needle EMG study. Typically, we assess the APB muscle, a C8,T1 muscle innervated by another nerve (e.g., the FDI), and another median nerve innervated by roots other than C8 and T1, such as the pronator teres, which is a C6,7 muscle. The right APB is added to determine how severe the previous left carpal tunnel syndrome was (given that right Median-D2 response was relatively low in amplitude). The Needle EMG Study

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 11

Insertional Activity Normal

IPSWs

SCP

Spontaneous Activity Other

None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

LEFT APB

X

X

Normal

Mild CMAL

FDI

X

X

Normal

Normal

Pron teres

X

X

Normal

Normal

X

X

Normal

Mild CMAL

RIGHT APB

The needle EMG study shows long-duration MUAPs in the left APB muscle, mild in degree. The latter would typically be compared to the right APB muscle, but because this patient had a previous median neuropathy on that side, it is not a useful comparison. Thus, it was compared to the ipsilateral FDI muscle, and an obvious MUAP duration difference was apparent. The mild CMAL noted in the right APB muscle indicates that prior to the right carpal tunnel release procedure, there was motor axon loss improved via reinnervation related to collateral sprouting. As previously pointed out, the motor response is normalized by successful collateral sprouting. The presence of long-duration MUAPs on needle EMG study is often the only remaining evidence. EDX Impression

1. Left Median Neuropathy (carpal tunnel syndrome) –

The above is demyelinating and axon loss in nature, involves the sensory and motor nerve fibers, and is located at or distal to the wrist.

2. Right Median Neuropathy (c/w successful carpal tunnel release) –

The pattern of abnormalities observed on this study can be observed in two settings: (1) with carpal tunnel syndrome and (2) following a successful carpal tunnel release procedure. Following successful surgery, the myelin formed during remyelination are shorter in length and thinner in caliber and, consequently, slow action potential propagation speed. These two possibilities are differentiated by comparing the latencies of the current EDX study with those of the preoperative EDX study. Improvement indicates a successful procedure.

353

Section 5: Case Studies in Electrodiagnostic Medicine

The presence of chronic motor axon loss on needle EMG of the right APB muscle without current EDX features of focal demyelination suggests that there was axon loss at the time of the previous carpal tunnel release procedure and that the procedure was successful. Preoperative versus Postoperative EDX Studies

In general, following a successful carpal tunnel release procedure, remyelination occurs and the latency values improve or normalize, depending on the degree of demyelination present prior to the procedure. Because the new myelin is shorter in length and thinner in caliber, the speed of action potential propagation through the remyelinated segment may not return to baseline. However, when compared to the preoperative latency values, improvement is appreciable. When the preoperative study used is more remote, the improvement related to surgical intervention may not be appreciable by EDX testing. For example, when a patient with a mild delay on NCS is treated conservatively, worsens over a 3-year period, and then undergoes surgery without repeat EDX testing, the improvement from the surgery may not be apparent, and in fact, the postoperative NCS may show worsening when compared to the original NCS (e.g., if the patient went from a mild delay to a severe delay over the 3-year period and then had successful surgery resulting in improvement to a moderate degree of delay, the repeat EDX testing would seem worse. An orthopedic colleague performed a carpal tunnel release procedure on a patient who reported a history of episodic hand tingling for many years, which progressed to continuous numbness and extremity aching. Preoperatively, the delays were severe. Postoperatively, she developed tingling in the median nerve distribution, which prompted another EDX study 4 months later. The postoperative study showed significant improvement, but the latency delays did not normalize. The electromyographer suggested the surgeon “did not get it all.” The patient initiated a lawsuit that resulted in a $25,000 settlement because the electromyographer did not realize that improvement indicates success and that the values may not normalize because the remyelinated segments conduct slower (i.e., successful surgery results in improvement, but not necessarily normalization).

Exercise 12 A 70-year-old right hand–dominant male inpatient is referred for EDX assessment of left upper extremity sensorimotor dysfunction. The patient reports that 4 days ago, at 0100, he awoke from sleep with left hand tingling. He returned to sleep and when he awoke at 0800, the tingling was accompanied by an inability to extend his fingers or wrist. He reported this to the medicine team, who consulted the neurology service, and subsequently he was brought to the EMG laboratory for further assessment. Because he was at day 4, it would not be possible to differentiate motor axon loss from motor demyelinating conduction block, which is best determined after day 6. It is also too soon for sensory NCS. On the other hand, if it is axon loss, after Wallerian degeneration occurs, it will no longer be localizable. Clinically, the sensory and motor deficits are in the distribution of the left radial nerve. Thus, radial motor NCS are performed on the left upper extremity. To verify the timing reported by the patient, bilateral superficial radial sensory NCS are included. Motor NCS (Day 4)

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 12 NCS PERFORMED

Stim Site

LAT

AMP

2.7

19.7

RIGHT CV

nAUC

LAT

AMP

2.6

24.3

SENSORY Superficial Radial

354

CV

nAUC

Case 1 through Case 50

(cont.)

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 12 NCS PERFORMED

Stim Site

LAT

AMP

Elbow

2.2

4.1

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

MOTOR Radial-EI

Ulnar-ED

28.9

Below SG

3.7

Above SG

1.0

5.1

5.0

33.2

Elbow

3.4

50.0

26.7

Below SG

5.0

52.0

33.0

Above SG

0.4

70.0

1.0

The initial motor NCS show a large conduction block between the below-spiral groove and above-spiral groove stimulation sites for both radial motor NCS. Thus, the lesion lies between these two stimulation sites (i.e., it localizes to the spiral groove segment). On day 4, in addition to a focal demyelinating conduction block, early axon disruption could also be responsible because enough time for Wallerian degeneration has not yet elapsed. To differentiate between these two pathophysiologies, the motor NCS will need to be repeated after day 6. The advantage of performing the motor NCS prior to day 7 is that axon loss lesions are localizable during that period (incomplete Wallerian degeneration). The left superficial radial sensory response is normal and comparable to the contralateral superficial radial response. The superficial radial sensory NCS will serve as baseline studies for the subsequent studies. Because the sensory NCS are best performed after day 10 and needle EMG is best performed in the 21-day to 35day window when fibrillation potentials are maximum in density, the patient is scheduled for a follow-up EDX study on day 25. Complete NCS Studies (Day 25)

The ipsilateral sensory NCS are performed along with a contralateral superficial radial sensory NCS.

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 12 NCS PERFORMED

Stim Site

LAT

AMP

Median-D2

3.3

18.1

Ulnar-D5

2.7

10.8

Superficial Radial

2.5

21.6

RIGHT CV

nAUC

LAT

AMP

2.6

28.7

CV

nAUC

SENSORY

The initial sensory NCS are normal, indicating no significant axon loss. There is a slight asymmetry, but it does not meet our criteria for a relative abnormality. In addition to the routine motor NCS, the initial motor NCS also include bilateral Radial-ED and Radial-EI motor NCS.

355

Section 5: Case Studies in Electrodiagnostic Medicine

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 12 NCS PERFORMED

Stim Site

LAT

AMP

Median-D2

3.3

18.1

Ulnar-D5

2.7

10.8

Superficial Radial

2.5

21.6

3.6

8.7

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

2.6

28.7

2.2

6.4

32.9

2.6

7.1

45.6

SENSORY

MOTOR Median-APB

8.6 Ulnar-ADM

2.8

54.2

7.2 7.1

Radial-EI

Mid-FA

Radial-ED

4.0

28.9

Elbow

3.8

28.2

Below SG

3.8

27.7

Above SG

0.8

3.8

6.2

40.0

Below SG

5.7

38.5

Above SG

1.1

8.4

Elbow

2.1

55.3

2.7

The median and ulnar motor NCS are normal. The radial motor NCS are abnormal, both of which indicate a large demyelinating conduction block across the spiral groove. The distal radial motor responses show side-to-side negative AUC and amplitude differences that do not meet our EDX criteria for relative abnormal. Thus, based on the sensory and motor NCS, this is a demyelinating conduction block lesion located across the spiral groove. If there is concomitant axon loss (as suggested by the response asymmetries), it should be apparent on the needle EMG study, which is much more sensitive for motor axon loss in the 21–35-day window. Regarding lesion severity, approximately 86% of the motor axons to the extensor indicis muscle are affected by the demyelinating conduction block (3.8/27.7
100% = 14%; 100% – 14% = 86%), and approximately 78% of the motor axons to the extensor digitorum communis muscle are affected by the demyelinating conduction block (8.4/38.5
100% = 22%; 100% – 22% = 78%). The needle EMG study is expanded to include additional radial nerve innervated muscles and contralateral radial nerve innervated muscles for comparison purposes. The Needle EMG Study

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 12

Insertional Activity Normal

IPSWs

SCP Other

Spontaneous Activity None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

Normal

Normal

LEFT FDI

356

X

X

Case 1 through Case 50

(cont.)

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 12

Insertional Activity Normal IPSWs

SCP

Spontaneous Activity

Other None

Fibs Fascs 3+

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

Severe neurogenic

Normal

EI

X

FPL

X

X

Normal

Normal

Pron teres

X

X

Normal

Normal

BC, MH

X

X

Normal

Normal

TC, LH

X

X

Normal

Normal

Deltoid, MH

X

X

Normal

Normal

3+

Severe neurogenic

Normal

Brachioradialis ECR-longus

X

3+

Severe neurogenic

Normal

ED

X

2+

Severe neurogenic

Normal

Low cerv psp

X

X

–

–

High thor psp

X

X

–

–

EI

X

X

Normal

Normal

Brachioradialis

X

X

Normal

Normal

RIGHT

The needle EMG study shows a large number of fibrillation potentials and severe neurogenic MUAP recruitment in the muscle domain of the radial nerve. Muscles innervated by the radial nerve prior to its arrival at the spiral groove (e.g., triceps; anconeus) are spared, whereas those innervated distal to the spiral groove are involved, consistent with the spiral groove localization indicated by the motor NCS. The distal segment of the radial nerve innervates the brachioradialis and extensor carpi radialis longus muscles before dividing into the posterior interosseous and superficial radial nerves. Given the severity of the lesion, it is not unexpected to see a large number of fibrillation potentials. The number reflects the timing of the study and not its severity. Recall that hundreds of fibrillation potentials are generated per motor axon disrupted. Given that this lesion was severe enough to generate such a large demyelinating lesion, concomitant disruption of at least a few motor axons is expected. EDX Study Conclusion

1. Left Radial Neuropathy –

The above is demyelinating conduction block  axon loss and is located in the spiral groove segment of the nerve. Assuming that further injury does not occur, significant improvement is expected in 3–4 months. The patient is scheduled for a follow-up assessment at that time.

357

Section 5: Case Studies in Electrodiagnostic Medicine

Follow-Up NCS

The ipsilateral superficial radial sensory NCS and both radial motor NCS are performed. The contralateral superficial radial sensory NCS is also performed. The patient requested that the asymptomatic limb not be studied for comparison purposes. Consequently, the values from the previous study were used for this purpose. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 12 NCS PERFORMED

Stim Site

RIGHT

LAT

AMP

CV

nAUC

2.6

19.0

2.4

3.8

23.0

Elbow

3.7

22.8

Below SG

3.7

22.7

Above SG

3.2

23.4

6.9

54.8

Below SG

6.8

46.5

Above SG

6.2

44.6

LAT

AMP

CV

nAUC

2.6^

28.7^

2.2^

6.4^

32.9^

2.6^

7.1^

45.6^

SENSORY Superficial Radial

MOTOR Radial-EI

Mid-FA

Radial-ED

Elbow

2.8

^These values are the ones recorded during the previous EDX study

At this point, the demyelinating conduction block has essentially resolved (both studies show a drop across the spiral groove segment, but these drops are not diagnostic of demyelinating conduction block). There is a distal Radial-EI motor response asymmetry, indicative of axon loss – the amplitude and negative AUC values are slightly lower than they were on the previous study. Using the negative AUC values of the two sides, it involves approximately 30% of the motor axons to the EI muscle (23.0/32.9
100% = 70%). There is no motor NCS evidence of axon loss to the EDC muscle. The needle EMG study is more sensitive to this, but it cannot be used to estimate the severity. Follow-Up Needle EMG Study

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 12

Insertional Activity Normal IPSWs

SCP Other

Spontaneous Activity None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

RIGHT

358

EI

X

X

Normal

Mild CMAL

Brachioradialis

X

X

Normal

Mild CMAL

TC, LH

X

X

Normal

Normal

EDC

X

X

Normal

Mild CMAL

Case 1 through Case 50

(cont.)

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 12

Insertional Activity Normal

IPSWs

Spontaneous Activity

SCP Other

None

Fibs Fascs

MUAP Analysis

Other

MUAP Recruitment

MUAP Morphology

LEFT EI

X

X

Normal

Normal

EDC

X

X

Normal

Normal

Brachioradialis

X

X

Normal

Normal

The lack of fibrillation potentials indicates reinnervation of the denervated muscle fibers. There is also evidence of chronic motor axon loss in the radial nerve distribution, indicative of reinnervation via collateral sprouting, consistent with the resolution of the fibrillation potentials. The lack of fibrillation potentials also indicates that further functional motor recovery will not occur. Also, it is possible that some of the demyelinated axons innervating the extensor indicis muscles underwent axon loss given that the side-to-side distal motor response amplitude difference is greater on this study than the previous one. Finally, the previously noted neurogenic MUAP recruitment pattern has resolved, consistent with remyelination.

Exercise 13 A 32-year-old right hand–dominant male is referred for EDX assessment of bilateral hand numbness and tingling. The symptoms are episodic, have been present for 4 months, and are more pronounced on the right. In addition, he notes that the symptoms are present upon awakening, are precipitated by driving and relieved with limb lowering, and occur spontaneously while seated at rest. He denies neck pain. He works on an assembly line and uses both hands simultaneously. He denies symptoms prior to this time. Given that the right side is more symptomatic than the left side, and that the patient is right hand dominant, the screening sensory NCS are performed on the right side first. Depending on the Median-D2 response, palmar NCS may or may not be required. The findings on the right side will dictate the studies required on the left side. Nerve Conduction Studies

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 13 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Median-D2

NR

Ulnar-D5

2.9

21.3

Superficial Radial

2.4

24.0

The screening sensory NCS show an absent right Median-D2 response, indicating an axon loss process. Although this is consistent with right carpal tunnel syndrome, the lack of focal demyelination excludes focal localization. Thus, the lesion could be localized to the median nerve, lateral cord, upper or middle trunk, or C6 or C7 APR/DRG. At this point, the ipsilateral median palmar NCS is added (it may show a delay and allow more precise

359

Section 5: Case Studies in Electrodiagnostic Medicine

localization), and the contralateral Median-D2 response is indicated. A contralateral median palmar NCS may be necessary as well, depending on the Median-D2 response. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 13 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Median-D2

NR

NR

Ulnar-D5

2.9

21.3

Superficial Radial

2.4

24.0

Median Palmar

NR

NR

The right Median-D2 response is absent, as are the two median responses on the left side. Thus, focal localization is not yet possible. Because carpal tunnel syndrome affects the sensory responses earlier and to a greater extent than the motor responses, it may be that the median motor responses will allow focal localization. The initial motor NCS should include the left Median-APB. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 13 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Median-D2

NR

NR

Ulnar-D5

2.9

21.3

Superficial Radial

2.4

24.0

Median Palmar

NR

NR

MOTOR Median-APB

9.8

6.8

10.9

5.7 5.7

Ulnar-ADM

2.8

43.6

11.8 11.5

55.3

The median motor responses display very delayed onset latencies, indicating that the process has a demyelinating component and, thus, permitting focal localization (between the stimulation site and the E1 electrode [e.g., the carpal tunnel]). Also, the calculated right forearm motor NCV is mildly reduced. As discussed previously, this is observed when all of the fastest conducting fibers are delayed at the carpal tunnel, permitting slower fibers to reach the E1 electrode first (see Chapter 7). The amplitude of the median motor response is reduced, indicating some axon loss. The right median motor response is normal. However, this does not mean that there is no motor axon loss, because when reinnervation keeps pace with denervation, the motor response remains normal. Because carpal tunnel syndrome is a slowly progressive disorder, reinnervation typically keeps pace with denervation. Thus, motor response diminution is often seen with advanced disease. The presence of motor axon loss will be sought during

360

Case 1 through Case 50

the needle EMG study of the left APB muscle. The degree of prolongation of the duration of the MUAP is an indicator of the degree of collateral sprouting and, hence, of the degree of denervation. Because the patient presented with classic carpal tunnel syndrome and without features to suggest an alternate or concomitant diagnosis, and because the NCS showed carpal tunnel syndrome, in our EMG laboratories, the needle EMG study is abbreviated. Needle EMG Study

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 13

Insertional Activity Normal

IPSWs

SCP Other

Spontaneous Activity None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

RIGHT APB

X

X

Mild neurogenic

Severe CMAL

FDI

X

X

Normal

Normal

Pron teres

X

X

Normal

Normal

X

X

Normal

Moderate CMAL

LEFT APB

The needle EMG showed neurogenic recruitment in the right APB muscle and long-duration MUAPs in both APB muscles, right more pronounced than left, consistent with slowly progressive motor axon loss. Without the needle EMG study of the left APB muscle, the presence of motor axon loss would have been missed. Moreover, the severity of CMAL is much greater than expected based on the motor NCS. Thus, in our EMG laboratories, when the NCS identify carpal tunnel syndrome, we always perform a needle EMG study of the APB muscle to avoid significantly underestimating the severity of the lesion. EDX Study Conclusion 1. Bilateral Median Neuropathies (carpal tunnel syndrome) –

The above are demyelinating and axon loss in nature, involve the sensory and motor nerve fibers, and are located at or distal to the wrists. Both lesions are severe in degree, right worse than left. These EDX features are consistent with carpal tunnel syndrome.

Although the patient reports a 4-month history of symptoms, these findings indicate a much more chronic process. Following the study, the patient recalled remote (more than ten years ago) hand tingling when driving.

Exercise 14 A 38-year-old right hand–dominant male is referred for EDX assessment of a right-sided foot drop that began 39 days ago. According to the patient, he developed cellulitis of the right leg (related to an MRSA infection) that resulted in severe edema of the deep peroneal compartment of the right leg. As the swelling increased, he noted burning pain along the anterolateral aspect of the leg and the dorsal aspect of the foot. He then noted impaired ankle dorsiflexion and foot eversion, along with numbness involving the wedge of skin between the first and second toes that was much more pronounced than the new numbness across the top of the foot. Serum CK was

361

Section 5: Case Studies in Electrodiagnostic Medicine

elevated (14,000 units). He has a 20-year history of type 2 diabetes mellitus and has been insulin dependent for two years. Clinically, the deep peroneal compartment edema and the ankle dorsiflexion and foot eversion weakness suggest deep peroneal nerve fiber involvement, as does the severe numbness in the cutaneous distribution of this nerve. The lesser foot eversion weakness and the lesser numbness in the cutaneous distribution of the superficial peroneal nerve suggest lesser involvement of this nerve. The difference in clinical severity (deep peroneal nerve more severely affected than superficial peroneal nerve) raises the possibility of two separate lesions, one involving the deep peroneal nerve and one involving the superficial peroneal nerve. The elevated CK value indicates muscle fiber disruption. The contralateral superficial peroneal sensory NCS is added to the initial sensory NCS. LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 14 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

4.1

7.0

CV

nAUC

SENSORY Sural Supfcl Peroneal

3.0

8.3

NR

The initial sensory NCS show an absent right superficial peroneal response, indicating an axon loss process that is located at or distal to the DRG and at or proximal to the superficial peroneal nerve. Sparing of the sural response argues against a sciatic neuropathy. Overall, the most likely lesion localization is the superficial peroneal nerve or the common peroneal nerve. For this reason, the initial motor NCS should include bilateral peroneal motor NCS, recording EDB, and bilateral peroneal motor NCS, recording TA.

LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 14 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Sural Supfcl Peroneal

4.1 3.0

8.3

4.0

11.0

7.0 NR

MOTOR Tibial-AH

4.3

8.0 5.8

362

Peroneal-EDB

4.6

6.5

Peroneal-TA

3.2

6.0

21.8 46.2

NR

2.8

1.0

+dip

1.0

+dip

0.8

+dip

M wave

5.0

9.3

H wave

33.4

1.9

19.9

Case 1 through Case 50

The routine motor NCS are remarkable for an absent Peroneal-EDB response and a very-low-amplitude PeronealTA response (1 mV versus 6 mV on the contralateral side). The Peroneal-TA response had a positive dip that did not disappear with E1 electrode repositioning or with stimulation at different sites. Thus, the positive dip likely represents a CMAP from an adjacent muscle distant to the E1 electrode. With the Peroneal-TA motor NCS, the common peroneal nerve is stimulated and the tibialis anterior motor response is recorded. For this reason, the positive dip was presumed to be coming from the peroneus longus and peroneus brevis muscles through the superficial peroneal motor axons. When the E1 electrode was repositioned over the peroneus longus muscle, a motor response of similar size and without a positive dip was collected (a peroneal motor response, recording peroneus longus). Because the two motor responses were of similar size, it suggests that the peroneal motor response recording TA might actually be absent. Thus, at this point in the study, an axon loss process is present that is severe in degree for the motor axons of the deep peroneal nerve: axon loss involves at least 83% (1/6
100% = 17%) of the motor axons to the tibialis anterior muscle (and possibly 100% based on the above discussion). The needle EMG study is expanded to better assess the superficial and deep peroneal nerve muscle domains.

LOWER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 14

Insertional Activity Normal

IPSWs

SCP Other

Spontaneous Activity None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

RIGHT FHB

X

X

Normal

Normal

FDL

X

X

Normal

Normal

TA

None

X

No MUAPs

n/a

Gastroc, MH

X

X

Normal

Normal

Vast lateralis

X

X

Normal

Normal

BF, SH

X

X

Normal

Normal

Glut medius

X

X

Normal

Normal

Peron longus

X

X

Normal

Normal

EHL

None

X

No MUAPs

n/a

EDB

X

No MUAPs

n/a

Low L psp

X

X

–

–

High S psp

X

X

–

–

4+

The needle EMG study showed no insertional activity in the tibialis anterior and extensor hallucis longus muscles, indicating that these two muscles are not viable. In addition, although these two muscles were denervated, fibrillation potentials were not observed. Finally, neither of these two muscles showed volitional MUAPs. These findings indicate that the muscle tissue within the deep peroneal compartment is no longer viable. Notably, the EDB muscle, which also showed no volitional MUAPs, showed high-density fibrillation potentials. This is consistent with the fact that the EDB muscle is not located in the deep peroneal compartment and, thus, is viable. Despite the viability, it is nonetheless innervated by the deep peroneal nerve, which is completely affected by axon loss.

363

Section 5: Case Studies in Electrodiagnostic Medicine

EDX Conclusions

1. Deep Peroneal Neuropathy Likely due to Compartment Syndrome –

The above is axon loss in nature and involves the deep peroneal compartment. The relationship between the deep peroneal nerve innervated muscles located within the deep peroneal compartment and those located outside of this compartment is important to recognize because it indicates the etiology of the deep peroneal neuropathy.

–

The lack of a peroneal motor response recording from the EDB indicates that the deep peroneal motor axons to this muscle are completely involved, which is supported by the lack of voluntary MUAPs in this muscle on needle EMG study. As expected, this muscle shows a large number of fibrillation potentials. However, the deep peroneal nerve innervated muscles located within the deep peroneal compartment show no insertional activity, no fibrillation potentials, and no voluntary MUAPs. Thus, there is no EDX evidence that these two muscles are viable, indicating muscle ischemia. Because the EDB muscle is outside of the deep peroneal compartment, the denervated muscle fibers generate fibrillation potentials. Because the muscles within the deep peroneal compartment have suffered ischemia, there is essentially no chance of functional recovery. Regarding the EDB muscle, the completeness of the lesion indicates that reinnervation via collateral sprouting is not possible. Finally, it is unlikely that reinnervation via proximodistal axon regrowth will occur, given the severity of the deep peroneal nerve involvement.

2. Right Superficial Peroneal Neuropathy –

The above is axon loss in nature and complete. There is no EDX evidence of concomitant motor axon involvement (needle EMG of the peroneus longus muscle is normal).

–

At the request of the referring physician, an abbreviated EDX study was performed 3 months later. The NCS were unchanged. Needle EMG of the EHL muscle showed no insertional activity, no fibrillation potentials, and high resistance to needle advancement, consistent with the transformation of the ischemic muscle tissue into fibrotic tissue. Although the tibialis anterior muscle showed three extremely low-amplitude, polyphasic MUAPs firing at a slow rate, this degree of recovery would not improve the motor function of this muscle. The EDB muscle still showed 4+ fibrillation potentials without voluntary MUAPs. The needle EMG study of the peroneus longus muscle was again normal. As expected, there was no clinical improvement.

Exercise 15 A 67-year-old right hand–dominant male is referred for EDX assessment of neck pain. The pain began 9 months earlier, without a recognized precipitant, is right paracentral in location, and radiates to the right hand. He also reports episodic right hand tingling present upon awakening and precipitated by driving. He denies left-sided neck or limb symptoms. Clinically, the radiating nature of the neck pain suggests a radiculopathy, while the features of the associated hand tingling are more consistent with carpal tunnel syndrome. The initial sensory NCS are expanded to include palmar NCS. Nerve Conduction Studies

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 15 NCS PERFORMED

DRG

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

SENSORY

364

Median-D2

C6,7

3.9

6.0

Ulnar-D5

C8

2.9

8.7

CV

nAUC

Case 1 through Case 50

(cont.)

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 15 NCS PERFORMED

DRG

Superficial Radial

C6,7

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

2.5

11.4

Median Palmar

2.8

6.0

Ulnar Palmar

2.0

7.4

CV

nAUC

The initial sensory NCS show a mildly delayed Median-D2 response (0.1 msec for age), a more significantly delayed median palmar response (0.6 msec), and a significant palmar interpeak latency delay (0.8 msec). These delays indicate focal demyelination between the stimulation and recording sites (i.e., from just proximal to the wrist to 8 cm more distally), which is what is typically observed with carpal tunnel syndrome. In our EMG laboratories, when we identify carpal tunnel syndrome, we assess the contralateral side for both comparative purposes and diagnostic purposes. In general, carpal tunnel syndrome initially involves the dominant limb, unless the patient has a hobby or profession in which the nondominant limb performs sustained gripping (Ferrante, 2016).

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 15 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.3

15.2

Ulnar-D5 Superficial Radial

RIGHT CV

nAUC

LAT

AMP

3.9

6.0

C8

2.9

8.7

C6,7

2.5

11.4

2.8

6.0

2.0

7.4

CV

nAUC

SENSORY

Median Palmar

2.0

16.7

Ulnar Palmar

The contralateral sensory NCS are normal. By comparison, it is now clear that there is concomitant axon loss given the asymmetric amplitudes of the two sides. At this point, we have a right median neuropathy that is demyelinating and axon loss in nature and involves the sensory axons. It is located between the distal forearm and palm (i.e., the carpal tunnel). The initial motor NCS are expanded to include the contralateral median motor NCS. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 15 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.3

15.2

Ulnar-D5 Superficial Radial

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY

Median Palmar Ulnar Palmar

3.9

6.0

C8

2.9

8.7

C6,7

2.5

11.4

2.8

6.0

2.0

7.4

2.0

16.7

365

Section 5: Case Studies in Electrodiagnostic Medicine

(cont.)

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 15 NCS PERFORMED

DRG

LAT

AMP

3.8

8.7

RIGHT CV

nAUC

LAT

AMP

4.3

6.0

CV

nAUC

MOTOR Median-APB

5.8 Ulnar-ADM

3.3

46.4

7.3 7.1

55.3

The motor NCS demonstrate a delayed right median motor response. Its amplitude is normal. Although there is a side-to-side asymmetry, the degree of the asymmetry is less than 50% and, hence, does not meet our criteria for relative abnormal. The needle EMG study should include bilateral APB muscles to look for evidence of chronic motor axon loss, which, if present, would indicate concomitant motor axon involvement. The needle EMG study should also be extensive enough to screen for a radiculopathy. Because the pain radiates to the right hand, C6, C7, or C8 nerve root involvement is possible (C5 does not radiate below the elbow and T1 would not be expected to radiate to the hand).

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 15

Insertional Activity Normal

IPSWs

SCP Other

Spontaneous Activity None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

RIGHT APB

X

X

Normal

Mild CMAL

FDI

X

X

Normal

Mild CMAL

EI

X

Normal

Mild CMAL

FPL

X

Normal

Normal

Pron teres

X

Normal

Mild CMAL

BC, MH

X

Normal

Normal

TC, LH

X

Mild neurogenic

Mild CMAL

Deltoid, MH

X

X

Normal

Normal

Brachioradialis

X

X

Normal

Normal

ECR longus

X

X

Normal

Normal

3+

Mod neurogenic

Mild CMAL

1+

–

–

–

–

FCR

366

1+ X 1+ X 2+

1+

Low cerv psp

X

High thor psp

X

X

Case 1 through Case 50

(cont.)

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 15

Insertional Activity Normal IPSWs

SCP

Spontaneous Activity

Other None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

LEFT APB

X

X

Normal

Normal

EI

X

X

Normal

Normal

TC, LH

X

X

Normal

Normal

FCR

X

X

Normal

Normal

The needle EMG study shows fibrillation potentials in the right lower cervical paraspinal muscles, indicating an intraspinal canal lesion. The fibrillation potentials in the limb muscles are in the muscle domains of the right C7 and C8 nerve roots, as are the chronic changes. Despite an onset of right-sided neck pain 9 months ago, there are insertional positive sharp waves, indicating concomitant recent denervation (typically within the previous 14–21 days) and, hence, suggesting a progressive process. Many of the fibrillation potentials in the C7 myotome were high in amplitude (indicating recent denervation), whereas others were low in amplitude (indicating their generation by atrophied muscle fibers). This distribution of fibrillation potential amplitudes also indicates a progressive process because the varied amplitude sizes suggest varied degrees of muscle fiber atrophy. The chronic motor axon loss in the right APB muscle could be related to the median neuropathy or the C8 radiculopathy. EMG Conclusions

1. Right C7 and C8 Intraspinal Canal Disorder (probable radiculopathies) –

The above is axon loss in nature and localizes to the C7 and C8 segments of the intraspinal canal (e.g., radiculopathies). The presence of insertional positive sharp waves and the wide range of fibrillation potential amplitudes suggests a slowly progressive process, as do the presence of chronic changes noted on the needle EMG. Because the high-amplitude fibrillation potentials were confined to the C7 myotome and the insertional positive waves were also seen in a C7 muscle (FCR is C6,7), the C7 nerve root appears to be responsible for the current symptoms.

2. Right Median Neuropathy (carpal tunnel syndrome) –

The above is demyelinating and axon loss in nature, involves the sensory nerve fibers (and possibly the motor nerve fibers), and is located at or distal to the wrist (e.g., the carpal tunnel).

Exercise 16 A 62-year-old right hand–dominant male is referred for EDX assessment of lower back pain. The pain started about 4 years ago without a recognized precipitant. It is central and left paracentral in location and occasionally radiates to the left foot. He also reports numbness and tingling in both great toes, the adjacent toes, the medial aspects of both feet, and the lateral aspect of the left foot. He denies lower extremity weakness and has not noted any muscle wasting. He is able to walk on his heels and toes. Clinically, the radiating nature of the left-sided lower back pain suggests a radiculopathy, and the distribution of the sensory symptoms suggests left S1 and bilateral L5 nerve root involvement. The initial sensory NCS are performed on the left side.

367

Section 5: Case Studies in Electrodiagnostic Medicine

Nerve Conduction Studies

LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 16 NCS PERFORMED

Stim Site

LAT

AMP

Sural

3.5

4.6

Supfcl Peroneal

3.0

5.3

RIGHT CV

nAUC

LAT

AMP

2.9

5.2

CV

nAUC

SENSORY

The routine sensory NCS performed on the left side are normal, consistent with the clinical impression of radiculopathy. Because the symptoms are bilateral, a sensory NCS is also performed on the right side. The superficial peroneal sensory NCS was chosen because of involvement of the medial aspect of the right foot. As expected, this sensory NCS was also normal. Thus, the sensory NCS are normal and, hence, support an intraspinal canal lesion. The routine motor NCS are performed next. Because a left L5 and a left S1 are both suspected, the motor NCS are performed bilaterally, along with bilateral H reflexes. LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 16 NCS PERFORMED

Stim Site

LAT

AMP

Sural

3.5

4.6

Supfcl Peroneal

3.0

5.3

5.6

2.1

RIGHT CV

nAUC

LAT

AMP

3.8

5.2

5.5

1.4

4.2

1.3

3.7

4.2

CV

nAUC

SENSORY

MOTOR Tibial-AH

1.9 Peroneal-EDB

4.4

1.7 1.3

Peroneal-TA

3.6

38.1

38.9

3.4 3.3

52.9

M wave

5.5

10.0

5.8

14.6

H wave

34.0

4.6

33.7

6.8

The tibial and distal peroneal motor responses are low in amplitude on both sides, indicative of an axon loss process. For this reason, the peroneal motor responses, recording TA, are added bilaterally (surround abnormal with normal). The latter are normal. The combination of low-amplitude motor responses with normal sensory responses suggests a preganglionic (intraspinal canal) localization. The H waves were high in amplitude and raise the possibility of spinal cord compression (i.e., suspicious for concomitant cervical spondylosis). Thus, at this point, there appears to be bilateral L5 and S1 nerve root involvement, although, in theory, the reduced amplitude peroneal-EDB responses could be explained by bilateral S1 involvement (because the S1 nerve root is involved in the innervation of the AH and EDB muscles). The needle EMG is expanded to include contralateral muscles for both comparison and diagnostic purposes.

368

Case 1 through Case 50

LOWER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 16

Insertional activity Normal

IPSWs

SCP

Spontaneous Activity Other

None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

LEFT FHB

X

3+

Normal

Mild CMAL

FDL

X

1+

Normal

Moderate CMAL

TA

X

X

Normal

Severe CMAL

Gastroc, MH

X

X

Normal

Moderate CMAL

Vast lateralis

X

X

Normal

Mild CMAL

BF, SH

X

X

Normal

Mild CMAL

Glut medius

X

X

Normal

Moderate CMAL

Low L psp

X

–

–

High S psp

X

–

–

Normal

Mild CMAL

1+ X

RIGHT FHB

X

2+

FDL

X

X

Normal

Severe CMAL

TA

X

X

Mod neurogenic

Severe CMAL

Gastroc, MH

X

X

Mild neurogenic

Moderate CMAL

Vast lateralis

X

X

Normal

Normal

Needle EMG Examination

Fibrillation potentials were noted in the bilateral FHB muscles, the left FDL muscle, and the lower lumbar paraspinal muscles. Involvement of the paraspinal muscles indicates that the lesion is at or proximal to the posterior primary ramus level (i.e., it is at the root level). Except for the paraspinal muscles, the fibrillation potentials are limited to the distal lower extremity muscles. With subacute and chronic radiculopathies, the fibrillation potentials are often restricted to the distal muscles of the myotome, especially the foot intrinsic muscles. Notably, a small proportion of the fibrillation potentials were high in amplitude, indicating that they are being generated by denervated muscle fibers that have not yet undergone significant atrophy (i.e., the denervation is not chronic). As is often the case with lumbosacral spondylosis-related radiculopathies, the chronic abnormalities are much more pronounced than the acute abnormalities. In this case, EDX features of chronic motor axon loss (increased MUAP duration) are present in muscles belonging to the muscle domains of the left L4, left L5, left S1, right L5, and right S1 nerve roots. The chronic changes are most pronounced in the L5 distribution, right worse than left. The relationship between the acute and chronic needle EMG abnormalities suggests a slowly progressive process.

369

Section 5: Case Studies in Electrodiagnostic Medicine

EDX Study Conclusions

1. Bilateral, Multi-Level Intrapinal Canal Lesion (e.g., lumbosacral radiculopathies) –

The EDX study shows evidence of a slowly progressive axon loss process involving the right L5 > left L5 > bilateral S1  left L4 nerve roots (e.g., spondylosis).

2. Possible Myelopathy –

The high-amplitude H waves suggest possible spinal cord compression. Given that the findings of this EDX study are most consistent with bilateral lumbosacral spondylosis-related radiculopathies, it is likely that the high-amplitude H waves reflect concomitant cervical spondylosis. Thus, an MRI of the cervical spine may be of diagnostic utility in this regard. If normal, a thoracic spine MRI should be considered. The cervical spine MRI showed severe cervical spondylosis without apparent intrinsic spinal cord signal changes. Thus, the most likely explanation for the high-amplitude H waves is cervical myelopathy related to spondylosis. To be on the safe side, the referring physician ordered a thoracic spine MRI, which was read as normal.

Exercise 17 A 53-year-old left hand–dominant male is referred for EDX assessment of right upper extremity pain. According to the patient, about 4 years ago, he developed episodic right paracentral neck pain. The pain radiates to the right fifth digit. He has associated numbness and tingling along the medial aspect of the right hand and digits, as well as grip weakness. He denies left-sided symptoms. On examination, sustained neck extension precipitates the right-sided neck pain and, within 10 seconds, reproduces the radiating pain to the fifth digit. In addition, this maneuver produces tingling in the second through fifth digits. He has significant wasting in the ulnar nerve distribution, with muscle atrophy, and obvious weakness of distal thumb flexion (C8-median nerve innervated flexor pollicis longus) and index finger extension (C8-radial nerve innervated extensor indicis). He has a Tinel sign at the right ulnar groove but not the left one. These clinical features suggest a right C8 radiculopathy (sustained neck extension precipitates radiating neck pain extending to the fifth digit of the right hand; weakness in a C8 distribution), possibly with a concomitant right ulnar neuropathy (wasting is profound in the right ulnar distribution, and there is a Tinel sign restricted to the right side). Although an ulnar groove Tinel sign may be seen among normal individuals without complaints of ulnar nerve dysfunction, when it is ipsilateral or obviously asymmetric, it is more likely to be pathological. The initial sensory NCS are expanded to include the DUC sensory NCS. Whenever the Ulnar-D5 or DUC response is abnormal, the MABC NCS is added (to surround abnormal with normal). Nerve Conduction Study

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 17 NCS PERFORMED

DRG

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

SENSORY

370

Median-D2

C6,7

3.1

32.3

Ulnar-D5

C8

2.9

10.3

Superficial Radial

C6,7

2.1

38.3

DUC

C8

2.0

8.3

MABC

T1

2.3

16.3

CV

nAUC

Case 1 through Case 50

The initial sensory NCS show a low-amplitude right DUC response. In addition, although the Ulnar-D5 response is normal, it is at the lower limit of normal, whereas the Median-D2 and superficial radial responses are more than twice the lower limit of normal. This is suspicious for a relative abnormality of this response. Thus, the contralateral Ulnar-D5 NCS is required. The contralateral DUC is also required. Also, because the right DUC response is low in amplitude, the MABC sensory NCS is required. Depending on its amplitude (reduced or near the lower limit of normal), the contralateral response may also be required.

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 17 NCS PERFORMED

DRG

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

3.1

32.3

2.9

10.3

2.1

38.3

2.0

8.3

2.3

16.3

CV

nAUC

SENSORY Median-D2

C6,7

Ulnar-D5

C8

Superficial Radial

C6,7

DUC

C8

MABC

T1

2.7

1.9

29.3

33.3

The amplitude of the contralateral ulnar-D5 response is nearly three times larger than the ipsilateral response, indicating a relative abnormality. The amplitude of the MABC is normal and negates the need for a contralateral study. At this point, the sensory NCS indicate that the lesion is either ganglionic or postganglionic and is axon loss in nature. If postganglionic, sparing of the MABC response suggests an ulnar neuropathy that is located proximal to the DUC departure site. The initial motor NCS should include bilateral ulnar studies recording from the ADM and from the FDI. Regarding the clinical features, a concomitant C8 radiculopathy cannot yet be excluded (the needle EMG will better address this issue).

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 17 NCS PERFORMED

DRG

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

3.1

32.3

2.9

10.3

2.1

38.3

2.0

8.3

2.3

16.3

3.1

8.3

CV

nAUC

SENSORY Median-D2

C6,7

Ulnar-D5

C8

Superficial Radial

C6,7

DUC

C8

MABC

T1

2.7

1.9

29.3

33.3

MOTOR Median-APB

8.0 Ulnar-ADM

2.3

10.0

32.2

2.4

5.6

55.2 13.8

371

Section 5: Case Studies in Electrodiagnostic Medicine

(cont.)

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 17 NCS PERFORMED

DRG

LAT

Ulnar-FDI

AMP

3.8

RIGHT CV

nAUC

13.4

23.5

LAT

3.8

AMP

CV

nAUC

5.4

13.0

2.1

6.8

8.5

14.0

6.3

13.2

2.5

5.7

The screening motor NCS identify a low-amplitude distal right Ulnar-ADM response, consistent with an axon loss process. In addition, there is an amplitude drop across the elbow segment, indicating concomitant demyelinating conduction block and localizing the lesion. The amplitude of the left Ulnar-ADM response is much larger, indicating that the amount of axon loss is significant. Thus, at least 57% of the motor axons innervating the ADM muscle are affected by axon loss (13.8/32.2
100% = 43%; 100% – 43% = 57%). Of the remaining 43%, 48% are affected by demyelinating conduction block (6.8/13.0
100% = 52%; 100% – 52% = 48%). In other words, 48% of the 43% are involved, which equates to 21% (0.48
0.43 = 0.21 = 21%). In summary, regarding the motor axons to the ADM muscle (actually to the hypothenar eminence), 57% are affected by axon loss, 21% by demyelinating conduction block, and 22% are unaffected (normal). The Ulnar-FDI responses indicate axon loss and demyelinating conduction block. At least 40% of the motor axons innervating the FDI muscle are affected by axon loss (14.0/23.5
100% = 60%; 100% – 60% = 40%). Of the remaining axons, 34% are affected by demyelinating conduction block: 5:7=13:2 x 100% ¼ 43% ðthe percentage getting through the lesion at the elbowÞ 100%  43 % ¼ 57% ðthe percentage not getting through due to the blockÞ Thus, 57% of the 60% not affected by axon loss are affected by demyelinating block, which equates to 34% (0.57
0.60 = 0.34
100% = 34%). Thus, regarding the motor axons to the FDI muscle, 40% are affected by axon loss, 34% are affected by demyelinating conduction block, and 26% are normal. The needle EMG study is expanded to include additional ulnar nerve innervated muscles and contralateral comparison studies, as well as a thorough assessment of the right C8 nerve root. Importantly, if evidence of a C8 root lesion is identified, the motor axon loss component of the right ulnar neuropathy will not be determinable (it could reflect the C8 radiculopathy or the ulnar neuropathy). The Needle EMG Study

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 17

Insertional activity Normal IPSWs

SCP

Spontaneous Activity

Other None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

RIGHT

372

FDI

X

1+

Mod neurogenic

Normal

EI

X

1+

Severe neurogenic

Severe CMAL

FPL

X

X

Mod neurogenic

Moderate CMAL

Pron teres

X

X

Normal

Mild CMAL

Case 1 through Case 50

(cont.)

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 17

Insertional activity Normal

IPSWs

SCP Other

Spontaneous Activity None

Fibs

X

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

Normal

Normal

Mild neurogenic

Moderate CMAL

Normal

Normal

Mod neurogenic

Mod CMAL

BC, MH

X

TC, LH

X

Deltoid, MH

X

FDP-3,4

X

Low cerv psp

X

X

–

–

High thor psp

X

X

–

–

FDI

X

X

Normal

Normal

FDP-3,4

X

X

Normal

Normal

EIP

X

X

Normal

Normal

FPL

X

X

Normal

Normal

Pron teres

X

X

Normal

Normal

BC, MH

X

X

Normal

Normal

TC, LH

X

X

Normal

Normal

Deltoid, MH

X

X

Normal

Normal

1+ X 1+

LEFT

The needle EMG study shows fibrillation potentials in the muscle domain of the right C8 nerve root. Chronic motor changes are noted in the muscle domain of the right C8 nerve root and, to a much lesser degree, in the muscle domain of the right C7 nerve root (pronator teres shows chronic changes) but not the right C6 nerve root (the C5,6 muscles [biceps and deltoid] are spared). Importantly, the right ulnar neuropathy was initially characterized as axon loss and demyelinating conduction block on the motor NCS. However, the axon loss could be related to right C8 nerve root involvement or could be a combination of the two lesions. EDX Study Conclusion

1. Right C7 and C8 Intraspinal Canal Lesion (e.g., radiculopathies) –

The above are axon loss in nature and located in the intraspinal canal. Clinically, the radicular features (radiating neck pain to the fifth digit and associated tingling of the second through fifth digits) precipitated by sustained neck extension suggest right C7 and C8 radiculopathies.

373

Section 5: Case Studies in Electrodiagnostic Medicine

2. Right Ulnar Neuropathy –

The above is demyelinating conduction block and likely axon loss in nature, involves the sensory and motor axons, and is located between the above-elbow and below-elbow stimulation sites (e.g., along the elbow segment). The degree of associated motor axon loss cannot be determined due to #1 above.

Exercise 18 A 69-year-old right hand–dominant male was referred for EDX assessment of left foot drop. According to the patient, he awoke 26 days prior to the study and noted an inability to dorsiflex or evert his left foot. He also noted numbness, tingling, and hypersensitivity along the top of the foot and the lateral aspect of the leg, distally. Clinically, the motor abnormalities are in the distribution of the left common peroneal nerve (the major ankle dorsiflexor is the tibialis anterior, which is innervated by the deep peroneal nerve, and foot eversion is primarily through the peroneus longus, which is innervated by the superficial peroneal nerve). Therefore, the lesion must be at or proximal to this element. Thus, the initial group of sensory NCS includes a contralateral superficial peroneal sensory NCS. Nerve Conduction Studies

LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 18 NCS PERFORMED

Stim Site

LAT

AMP

3.5

5.7

RIGHT CV

nAUC

LAT

AMP

2.9

6.3

CV

nAUC

SENSORY Sural Supfcl Peroneal

NR

On sensory NCS, the screening sensory NCS are remarkable for an absent left superficial peroneal sensory response (normal on the right side), indicative of an axon loss process that is ganglionic or postganglionic in location. Possible localizations include the superficial peroneal nerve, the common peroneal nerve, the sciatic nerve, the lumbosacral plexus, and the DRG from which the sensory fibers destined to become the superficial peroneal nerve originate (usually the L5 DRG, but, unlike with the upper extremity sensory NCS, this is not known with certainty). Sparing of the sural response argues against a sciatic nerve localization, but a partial lesion can never be excluded with certainty. To further address this list of potential lesion sites and to quantify the degree of severity of the foot drop, the initial motor NCS are expanded to include bilateral Peroneal-TA motor NCS. Also, whenever an individual is studied for a foot drop, the common peroneal nerve is stimulated below and above the fibular head, as this is the most common compression site. When evidence of a demyelinating conduction block across the fibular head is not noted, we also stimulate the nerve in the popliteal fossa. When a demyelinating conduction block is noted across the fibular head, there is no need for stimulation at the popliteal fossa, because the degree of block and axon loss can be calculated without its inclusion. However, as in this case, we often include it as a technical double check of the above-fibular head value (the above-fibular head and popliteal fossa responses should be roughly the same).

374

Case 1 through Case 50

LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 18 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Sural

3.5

Supfcl Peroneal

5.7 NR

2.9

6.3

3.6

2.6

3.3

4.5

MOTOR Tibial-AH

4.5

6.9 6.1

Peroneal-EDB

Peroneal-TA

Ankle

3.8

1.4

Below FH

1.0

Above FH

0.5

Below FH

4.4

41

3.3

23.0

Above FH

1.3

9.0

Pop fossa

1.3

9.0

29.1

H reflex M wave

5.0

8.1

5.3

8.2

H wave

33.4

1.2

32.8

1.5

On motor NCS, the left Peroneal-EDB response is reduced in amplitude and the below-fibular head and abovefibular head stimulation sites show different response sizes. The left Peroneal-TA motor response shows the same pattern. Regarding the underlying pathophysiology, the Peroneal-EDB and Peroneal-TA motor responses show a mixed lesion. The distal motor responses of the two sides differ, indicating axon loss, and the above-fibular head and below-fibular head responses differ, indicating concomitant demyelination conduction block. Regarding the motor axons innervating the EDB muscle, the motor response amplitude of the symptomatic side is 54% as large as the asymptomatic side (1.4/2.6
100% = 54%). Therefore, 46% of the motor axons innervating the EDB muscle are affected by axon loss. The 54% value represents the motor axons contributing to the response with below-fibular head stimulation. With above-fibular head stimulation, the response amplitude value is 50% smaller, indicating that there is a 50% demyelinating conduction block across the fibular head (0.5/1.0
100% = 50%). Thus, because the below-fibular head response represents 54% of the motor axons, the above-fibular head represents 50% of 54%, which is 27%. Thus, for the motor axons innervating the EDB muscle, 46% are affected by axon loss, 27% are affected by demyelinating conduction block, and 27% are normal. Note that there is a small drop between the ankle (1.4 mV) and the below-fibular head (1.0 mV) amplitude values. This reflects temporal dispersion. Thus, it might have been slightly more accurate to use the negative AUC values for the calculation rather than the amplitude values. For the peroneal motor axons innervating the tibialis anterior muscle, 21% are affected by axon loss (23.0/ 29.1
100% = 79%; 100% – 79% = 21%). Of the 79% not affected by axon loss, 31% are affected by demyelinating conduction block (9/23
79% = 31%). Thus, 21% are affected by axon loss, 31% are affected by demyelinating conduction block, and the remaining 48% are normal.

375

Section 5: Case Studies in Electrodiagnostic Medicine

On the needle EMG study, additional muscles are added to better assess the muscle domain of the common peroneal nerve. In this case, the EHL (deep peroneal nerve innervated) and the peroneus longus (superficial peroneal nerve innervated) muscles are added. Unlike other nerves, the sciatic nerve is composed of two separate nerves throughout its length, the tibial nerve and the peroneal nerve. The sciatic nerve innervates the hamstring muscles. Of those four muscles, the biceps femoris short head (BFSH) is the only muscle that receives innervation via the peroneal fibers of the sciatic nerve. Thus, the BFSH muscle should be thoroughly studied for abnormalities. LOWER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 18

Insertional Activity Normal IPSWs

SCP

Spontaneous Activity Other None

Fibs

Fascs Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

LEFT FHB

X

X

Normal

Normal

FDL

X

X

Normal

Normal

Severe neurogenic

Normal

TA

1+

3+

Gastroc, MH

X

X

Normal

Normal

Vast lateralis

X

X

Normal

Normal

BF, SH

X

X

Normal

Normal

Glut medius

X

X

Normal

Normal

2+

Mild neurogenic

Normal

2+

Normal

Normal

Peroneus long

1+

EHL

X

Low L psp

X

X

–

–

High S psp

X

X

–

–

X

X

Normal

Normal

RIGHT TA

Needle EMG Study

The motor NCS showed a demyelinating conduction block across the fibular head, thereby localizing the lesion to the common peroneal nerve at this site. The needle EMG study confirmed this localization – all of the abnormal muscles are in the distribution of the left common peroneal nerve and the BFSH is spared. The motor NCS and the needle EMG study identified concomitant axon loss. The lack of CMAL is consistent with symptom onset 26 days prior.

EDX Examination Conclusion

1. Left Common Peroneal Neuropathy

376

Case 1 through Case 50

The above is axon loss and demyelinating conduction block in nature, involves the sensory and motor axons, and is located at the fibular head. It involves the majority of motor axons innervating the EDB and TA muscles and, thus, is severe in degree: approximately 73% of the motor axons to the EDB muscle are affected (46% by axon loss and 27% by demyelinating conduction block) and approximately 52% of the motor axons to the tibialis anterior muscle are affected (31% by demyelinating conduction block and 21% by axon loss). Although the lesion is severe in degree, assuming the underlying cause is no longer present, the motor axons affected by demyelinating conduction block should recover through remyelination within 3–4 months (often within 3 weeks), and because the motor axon loss is incomplete, reinnervation through collateral sprouting is expected. Thus, the prognosis for significant motor recovery is good. The patient should return for an abbreviated EDX reassessment in 4 months. The patient returned to the EMG laboratory 3 months later. At that time, he was only slightly improved. A focused repeat EDX study was performed. LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 18 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

3.8

4.9

14.7

3.1

4.9

33.9

SENSORY Supfcl Peroneal

NR

MOTOR Peroneal-EDB

Ankle

Peroneal-TA

2.0

6.5

Below FH

1.6

6.5

Above FH

1.6

6.5

4.2

26.4

3.8

26.2

Below FH

4.2

3.2

Above FH

The left superficial peroneal sensory response is absent, which is unchanged from the previous study. The peroneal motor response, recording EDB, is low in amplitude. The demyelinating conduction block previously present across the fibular head is no longer present. Thus, the lesion is now solely axon loss in nature and, using the negative AUC values, involves 56% of the motor axons to the EDB muscle (6.5/14.7
100% = 44%; 100% – 44% = 56%). Previously, this lesion was 46% axon loss, 27% demyelinating conduction block, and 27% normal axons. The peroneal motor response, recording TA, is normal for age. Comparing the negative AUC values of the two sides shows that 22% of the motor axons to the tibialis anterior muscle are affected by axon loss (26.4/33.9
100% = 78%; 100% – 78% = 22%). Thus, regarding the motor axons to TA in comparison to the previous study, the demyelinating conduction block component has resolved and the axon loss component is essentially unchanged. A focused needle EMG study is performed to assess the degree of CMAL (reinnervation via collateral sprouting results in lesion underestimation) and to assess the density of any fibrillation potentials (the denser the fibrillation potentials, the greater the chance of further improvement via reinnervation). LOWER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 18

Insertional Activity Normal

IPSWs

SCP

Spontaneous Activity Other

None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

Normal

Moderate CMAL

LEFT EHL

X

X

377

Section 5: Case Studies in Electrodiagnostic Medicine

(cont.)

LOWER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 18

Insertional Activity Normal

IPSWs

SCP

Spontaneous Activity Other

None

Fibs

Fascs

MUAP Analysis

Other

MUAP Recruitment

MUAP Morphology

Peron longus

X

X

Normal

Moderate CMAL

TA

X

X

Normal

Severe CMAL

The needle EMG study shows no fibrillation potentials. The lack of fibrillation potentials indicates that further functional improvement via reinnervation is unlikely (i.e., there are no denervated muscle fibers to reinnervate). Regarding the MUAPs, significant reinnervation via collateral sprouting has occurred, as evidenced by the longduration MUAPs (ranging from moderate to severe in duration prolongation). Thus, although the demyelinating conduction block component of the lesion resolved and the axon loss remained similar by motor NCS, the presence of the significant CMAL indicates that much of the demyelinating conduction block transformed into axon loss. The motor NCS underestimate lesion severity once enough time for reinnervation (via collateral sprouting) has elapsed, as in this case. Thus, little recovery occurred and, moreover, further recovery is not expected. Because the degree of recovery was unexpectedly poor and because he was now complaining of pain in the region of the fibular head, an MRI of this region, with and without contrast, was ordered. It showed a nondisplaced fracture of the proximal left fibula with callus formation around the fracture that was displacing the deep peroneal nerve. At this point, an orthopedic consultation was placed for further management.

Exercise 19 A 30-year-old right hand–dominant male is referred for EDX assessment of left foot numbness and burning. Six years ago, the patient was involved in a motor vehicle accident in which his left fibula and ankle were fractured and his left foot was partially torn at the ankle such that his foot and toes faced rearward. He underwent multiple surgeries. Since the accident, he has had continuous burning pain and numbness involving all aspects of the left foot. He also has centrally located, non-radiating lower back pain. The initial sensory NCS are expanded to include the plantar mixed NCS. Under the age of 45 years, these responses should be present, whereas after that age, they may be absent among normal individuals. Nerve Conduction Studies

LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 19 NCS PERFORMED

Stim Site

LAT

AMP

Sural

4.0

17.3

Supfcl Peroneal

3.1

8.3

Medial Plantar

3.4

5.3

Lateral Plantar

3.7

3.1

SENSORY

378

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

Case 1 through Case 50

The sensory NCS show the left sural response to be twice the size of the left superficial peroneal response. Thus, the contralateral superficial peroneal study is required. Also, the plantar responses may or may not be normal and need to be compared to the contralateral side for this purpose (i.e., we judge normal based on the contralateral response values). Given the high amplitude of the left sural response, a contralateral sural response was not required, but was added when the other three sensory responses were noted to be relatively abnormal. LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 19 NCS PERFORMED

Stim Site

LAT

AMP

Sural

4.0

Supfcl Peroneal

RIGHT CV

nAUC

LAT

AMP

17.3

4.2

18.1

3.1

8.3

3.3

20.7

Medial Plantar

3.4

5.3

3.3

13.2

Lateral Plantar

3.7

3.1

3.7

9.5

CV

nAUC

SENSORY

The contralateral sensory NCS are normal and indicate that the left superficial peroneal response is relatively abnormal, as are both left plantar responses. Thus, this is an axon loss process that is ganglionic or postganglionic. Together, the abnormal responses suggest a lesion at or proximal to the sciatic nerve. However, sparing of the sural response is atypical and suggests that multiple mononeuropathies may be responsible, which is also supported by the significant ankle injury incurred in the motor vehicle accident. The motor NCS are performed next. LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 19 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Sural

4.0

17.3

4.2

18.1

Supfcl Peroneal

3.1

8.3

3.3

20.7

Medial Plantar

3.4

5.3

3.3

13.2

Lateral Plantar

3.7

3.1

3.7

9.5

5.7

10.0

MOTOR Tibial-AH

10.0 Peroneal-EDB

4.2

59.4

5.2 4.7

17.9 55.6

16.4

H reflex: M wave

5.2

9.8

5.3

8.7

H wave

33.0

0.8

32.3

2.8

379

Section 5: Case Studies in Electrodiagnostic Medicine

The screening motor NCS are normal. The left H wave is significantly reduced in amplitude. The latter indicates involvement of the sensory or motor S1 axons. Given the normal sural response, this suggests an intraspinal canal process at the S1 segment. The needle EMG study is performed next. Comparison needle EMG studies are required in the S1, sciatic, tibial, common peroneal, and plantar nerve distributions. Needle EMG Study

LOWER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 19

Insertional activity Normal

Spontaneous Activity

IPSWs SCP Other None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

LEFT ADQP

X

X

Normal

Mild CMAL

FHB

X

X

Normal

Moderate CMAL

FDL

X

X

Normal

Normal

TA

X

X

Normal

Normal

Gastroc, MH

X

X

Normal

Mild CMAL

Vast lateralis

X

X

Normal

Normal

BF, SH

X

X

Normal

Mild CMAL

Glut medius

X

X

Normal

Normal

EDB

X

X

Normal

Mild CMAL

EHL

X

X

Normal

Normal

Peron longus

X

X

Normal

Normal

Low L psp

X

1+

–

–

High S psp

X

1+

–

–

RIGHT

380

ADQP

X

X

Normal

Normal

FHB

X

X

Normal

Normal

EDB

X

X

Normal

Normal

TA

X

X

Normal

Normal

Gastroc, MH

X

X

Normal

Normal

EHL

X

X

Normal

Normal

Peron longus

X

X

Normal

Normal

Case 1 through Case 50

The needle EMG study shows fibrillations potentials in the left lower lumbar and upper sacral paraspinal muscles, consistent with an intraspinal canal lesion. The distribution of the chronic changes is in the muscle domain of the left S1 nerve root. Except for the FHB changes, the chronic changes are mild in degree. It is likely that the FHB changes reflect medial plantar nerve involvement. Likewise, the EDB changes could reflect distal deep peroneal nerve involvement. EDX Study Conclusions

1. Left Superficial Peroneal Neuropathy –

The above is axon loss in nature, mild in degree, and likely related to the left foot trauma.

2. Left Medial and Lateral Plantar Neuropathies –

The above are axon loss in nature, mild-moderate in degree, and likely related to the left foot trauma.

3. Left S1 Radiculopathy –

The above is axon loss in nature. Because some of the fibrillations are high in amplitude (i.e., generated by muscle fibers recently denervated), this is likely a progressive process.

Exercise 20 A 59-year-old right hand–dominant male is referred for EDX assessment of right hand numbness and loss of grip strength. He first noted bilateral hand tingling about 5–10 years ago. The tingling is episodic in nature and involves the medial two digits of both hands. It is equally symptomatic on the two sides. He also reports loss of grip strength on the right side and non-radiating right-sided neck pain. On examination of the right upper extremity, he has significant wasting of the intrinsic hand muscles and less pronounced wasting of the proximal forearm muscles, ventromedially. He has weakness in the right ulnar nerve distribution (C8, T1-ulnar nerve), but also has significant weakness of the right extensor indicis (C8-radial nerve) and flexor pollicis longus (C8, T1-median nerve) muscles. The left upper extremity demonstrates normal strength. He has sensory loss in the ulnar distribution on both sides, including the dorsal ulnar cutaneous branch distributions, and he splits the fourth digit on both sides. Thus, clinically, the distribution of the right upper extremity weakness suggests right C8 nerve root involvement, whereas the distribution of the sensory features suggests bilateral ulnar nerve involvement and places the lesions above the dorsal ulnar cutaneous branch exit site. The initial sensory NCS are expanded to include bilateral Ulnar-D5 and DUC NCS. The MABC NCS is added whenever the Ulnar-D5 NCS or the DUC NCS is abnormal. Because the clinical features are worse on the right side, the initial sensory NCS are performed on the right side. Nerve Conduction Studies

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 20 NCS PERFORMED

DRG

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

3.3

16.0

2.9

6.9

2.3

39.0

CV

nAUC

SENSORY Median-D2

C6,7

Ulnar-D5

C8

Superficial Radial

C6,7

DUC

C8

1.8

15.7

1.8

5.9

MABC

T1

2.3

16.3

2.2

18.0

2.8

7.7

381

Section 5: Case Studies in Electrodiagnostic Medicine

The initial sensory NCS on the right side show a low-amplitude right Ulnar-D5 response and a low-amplitude right DUC response, indicating that the lesion is ganglionic or postganglionic and axon loss in nature. Thus, the possible lesion sites include ulnar nerve (at or proximal to the DUC branch exit point), medial cord, lower trunk, or C8 APR/DRG. For this reason, the MABC sensory NCS was added and was normal, arguing against a medial cord or lower trunk process. On the left side, because the Ulnar-D5 sensory response is low in amplitude, the MABC NCS is added. The DUC and MABC responses are normal. The isolated low-amplitude ulnar response indicates that the lesion involves the ulnar nerve, medial cord, lower trunk, or C8 APR/DRG, and that it is axon loss in nature. The initial motor NCS should include the Ulnar-ADM and Ulnar-FDI studies bilaterally.

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 20 NCS PERFORMED

DRG

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

3.3

16.0

2.9

6.9

2.3

39.0

CV

nAUC

SENSORY Median-D2

C6,7

Ulnar-D5

C8

Superficial Radial

C6,7

DUC

C8

1.8

15.7

1.8

5.9

MABC

T1

2.3

16.3

2.2

18.0

3.6

8.3

2.8

7.7

MOTOR Median-APB

8.1 Ulnar-ADM

Ulnar-FDI

2.7

4.5

10.0

27.8

9.9

54.7

9.7

57.1

9.6

4.4

27.4 27.1 18.5

6.0

64.2

1.2

3.6

1.0

52.0

0.8

44.7

0.1

9.1

52.2

17.7

NR

8.7

56.3

17.4

NR

0.3

The right ulnar motor responses are very low in amplitude, indicating an axon loss process. Regarding the severity of involvement, at least 87% of the motor axons to the right ADM muscle are involved (3.6/ 27.8
100% = 13%; 100% – 13% = 87%), and at least 98% of the motor axons to the right FDI muscle are involved (0.3/18.5
100% = 2%; 100% – 2% = 98%). It is important to recognize that when the motor responses are more affected than the sensory responses and both assess the same peripheral nervous system segments, then the responsible lesion is either preganglionic or it lies distal to the takeoff site of the sensory branch of a mixed nerve. Thus, it suggests a preganglionic lesion (C8 radiculopathy) or a distal one (hand lesion distal to the site at which the sensory branch leaves the ulnar nerve). With a mononeuropathy involving a mixed element (i.e., an element containing both sensory and motor axons), the sensory response is more affected than the motor response. The needle EMG study is expanded to include additional ulnar nerve innervated and C8,T1 muscles.

382

Case 1 through Case 50

Needle EMG Study

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 20

Insertional Activity Normal

IPSWs

SCP Other

Spontaneous Activity None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

Normal

Mild CMAL

RIGHT APB

X

X

FDI

X

2+

Severe neurogenic

Moderate CMAL

EI

X

1+

Severe neurogenic

Severe CMAL

FPL

X

X

Normal

Mild CMAL

Pron teres

X

X

Normal

Normal

BC, MH

X

X

Normal

Normal

TC, LH

X

X

Normal

Normal

FDP-3,4

X

X

Mild neurogenic

Moderate CMAL

Low cerv psp

X

X

–

–

High thor psp

X

–

–

1+

LEFT APB

X

X

Normal

Normal

FDI

X

X

Normal

Moderate CMAL

EI

X

Normal

Mild CMAL

Pron teres

X

X

Normal

Normal

TC, LH

X

X

Normal

Normal

FDP-3,4

X

X

Normal

Normal

1+

The needle EMG shows fibrillation potentials in the right high thoracic paraspinal muscles, consistent with an intraspinal canal process. The right FDI and EI fibrillation potentials are consistent with right C8 involvement. The chronic changes are also in a right C8 distribution. The APB muscle is less involved because it is more heavily T1 innervated than C8 innervated. This innervation pattern (T1 motor axon input > C8 motor axon input) may be true for the FPL muscle as well. A similar distribution of less intense changes is noted on the left side as well. EDX Study Conclusion

1. Bilateral C8 Radiculopathies The above are axon loss in nature. On the right side, the abnormalities are severe in degree, whereas on the left side, they are less pronounced.

383

Section 5: Case Studies in Electrodiagnostic Medicine

The relationship between the sensory and motor responses indicates that the right upper extremity weakness and atrophy is primarily due to a preganglionic lesion. The relationship between the acute and chronic changes suggests a slowly progressive process, such as spondylosis. Based on the EDX study findings, an MRI of the cervical spine was obtained. That study showed multilevel degenerative disk and joint disease with a large osteophyte compressing the right C8 nerve root. 2. Bilateral Ulnar Neuropathies –

The above are axon loss in nature and involve the sensory axons. On the right side, the lesion is at or above the departure site of the DUC branch (most likely at the elbow segment), and on the left side, the localization is unclear.

Whether or not the ulnar motor axons are also involved cannot be determined due to #1 above. However, if they are involved, the severity of the involvement is minimal in comparison to #1 above. In conclusion, the right hand and forearm muscle weakness and atrophy are solely or predominantly related to the C8 nerve root involvement and not the ulnar neuropathy.

Exercise 21 A 54-year-old right hand–dominant male is referred for EDX assessment of lower back pain. The lower back pain started about 2 years ago, is right paracentral in location, and radiates to the right ankle. He denies weakness and muscle atrophy, but notes occasional tingling of the right foot, although he cannot recall its precise distribution. Because the pain radiates to the foot, a radiculopathy is likely. Because it is right paracentral in location, the sensory NCS are first performed on the right lower extremity. LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 21 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

Sural

4.1

5.5

Supfcl Peroneal

3.1

5.3

CV

nAUC

SENSORY

The screening right lower extremity sensory NCS are normal. This is consistent with the clinical impression of a radiculopathy, because intraspinal canal lesions spare the sensory NCS. There is no reason to screen the left lower extremity at this point. Thus, the screening right lower extremity motor NCS are performed next. LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 21 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Sural

4.1

5.5

Supfcl Peroneal

3.1

5.3

4.8

13.4

MOTOR Tibial-AH

9.2

384

26.1 41.2

23.6

Case 1 through Case 50

(cont.)

LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 21 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

Peroneal-EDB

LAT

AMP

4.3

3.4 3.0

M wave

4.7

H wave

CV

nAUC

41.5

9.4 NR

The motor responses are also normal. Although the overwhelming majority of lumbosacral monoradiculopathies involve the L5 or S1 nerve root, the motor responses tend to be normal in the setting of slowly progressive disease (reinnervation via collateral sprouting keeps pace with denervation, thereby maintaining the clinical strength and the motor responses of the affected muscles). The absent right H wave indicates either focal involvement of the S1 loop, from the stimulation site to the recording site, including the sensory and motor axons of the tibial nerve, sciatic nerve, sacral plexus (only those regions containing S1-derived sensory and motor axons), the S1 root (sensory and motor axons; preganglionic and postganglionic, including the portions within the parenchyma of the spinal cord), the S1 sensory and motor neurons, and the motor branch to the gastrocnemius/soleus complex. The H reflex is also sensitive to sensory polyneuropathies and can be absent due to technical reasons. Thus, it should be assessed on the contralateral side.

LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 21 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Sural

4.1

5.5

Supfcl Peroneal

3.1

5.3

4.8

13.4

MOTOR Tibial-AH

9.2 Peroneal-EDB

4.3

4.9

9.7

H wave

34.3

1.4

4.7

41.2

23.6

3.4 3.0

M wave

26.1

41.5

9.4 NR

The left H wave is normal, indicating that this is not a sensory polyneuropathy. Based on the clinical presentation and the sparing of the sensory NCS, at this point, the most likely explanation is a right S1 radiculopathy. The needle EMG is performed next.

385

Section 5: Case Studies in Electrodiagnostic Medicine

LOWER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 21

Insertional Activity Normal

IPSWs

SCP

Spontaneous Activity Other None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

Normal

CMAL moderate

RIGHT FHB

X

3+

FDL

X

X

Normal

Normal

TA

X

X

Normal

Normal

Gastroc, MH

X

Normal

CMAL, mild

Vast lateralis

X

X

Normal

Normal

BF, SH

X

X

Normal

CMAL moderate

Glut medius

X

X

Normal

Normal

Low L psp

X

X

–

–

–

–

1+

High S psp

2+

LEFT FHB

X

X

Normal

Normal

TA

X

X

Normal

Normal

Gastroc, MH

X

X

Normal

Normal

Needle EMG evidence of acute motor axon loss (fibrillation potentials) were noted in the right S1 myotome. Both low-amplitude fibrillation potentials and high-amplitude fibrillation potentials were noted, suggesting a progressive process (the low-amplitude fibrillation potentials suggest muscle fiber atrophy and, thus, chronicity, whereas the high-amplitude fibrillation potentials suggest recent muscle fiber denervation). Chronic changes are also present in a right S1 nerve root distribution. The lower lumbar paraspinal muscles were studied first and were normal. Thus, the upper sacral paraspinal muscles were studied and were abnormal, indicating an intraspinal canal localization. Had the upper sacral paraspinal muscles been studied first, the lesion would have been localized to the intraspinal canal, and hence, the lower lumbar paraspinal muscles would not have been studied. This is because the paraspinal muscles receive their innervation from multiple nerve roots, including those superior to, at, and inferior to the paraspinal muscles under study (i.e., assessing the lower lumbar paraspinal muscles assesses the middle lumbar, lower lumbar, and upper sacral roots). Thus, the EDX study identifies a lesion that localizes to the intraspinal canal and that involves the right S1 nerve fibers, consistent with the clinical impression of a right S1 radiculopathy.

386

Case 1 through Case 50

Exercise 22 A 41-year-old right hand–dominant female was referred for EDX assessment of neck pain and episodic hand tingling. Approximately two years ago, she developed neck and scapular region pain, left worse than right, that radiates to the medial aspects of both hands and that is associated with tingling along the medial aspects of both hands, also left worse than right. She also awakens with hand tingling and notes hand tingling when driving and while seated at rest. The radiating neck pain and the distribution of the associated tingling suggests bilateral C8 radiculopathies, and the episodic hand tingling upon awakening and precipitated by driving suggests carpal tunnel syndrome. Because the patient reports that the left upper extremity is more symptomatic, screening sensory NCS are performed on that side first. Depending on the routine median sensory response findings, palmar mixed NCS may or may not be required. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 22 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.7

18.3

Ulnar-D5

C8

2.9

18.0

Superficial Radial

C6,7

2.3

29.9

Median Palmar

2.6

27.4

Ulnar Palmar

1.9

15.5

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY

The median sensory response and the median palmar mixed response both show delayed peak latencies, indicating a demyelinating process distal to the wrist stimulation site. The amplitude of the median sensory response is reduced, indicating concomitant axon loss. Because the patient has bilateral hand symptoms, the right side is also studied, the results of which are shown below.

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 22 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.7

Ulnar-D5

C8

Superficial Radial

C6,7

RIGHT CV

nAUC

LAT

AMP

18.3

4.2

14.7

2.9

18.0

3.0

16.4

2.3

29.9

Median Palmar

2.6

27.4

2.9

20.3

Ulnar Palmar

1.9

15.5

1.8

17.3

CV

nAUC

SENSORY

The right upper extremity shows the same features, but to a more pronounced degree. With bilateral carpal tunnel syndrome, the dominant side is usually affected to a greater degree than the nondominant side is. A major exception to this statement is observed among individuals with hobbies or professions that require them to perform sustained gripping with the nondominant side, in which case the nondominant limb often is more involved than the dominant one. Thus, this patient has bilateral median neuropathies distal to the distal forearm stimulation site. They are demyelinating and axon loss in nature and involve the sensory nerve fibers. Given the presence of axon loss,

387

Section 5: Case Studies in Electrodiagnostic Medicine

there will likely be motor axon involvement as well. The screening motor NCS are performed next; both upper extremities are required.

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 22 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.7

Ulnar-D5

C8

Superficial Radial

C6,7

RIGHT CV

nAUC

LAT

AMP

18.3

4.2

14.7

2.9

18.0

3.0

16.4

2.3

29.9

Median Palmar

2.6

27.4

2.9

20.3

Ulnar Palmar

1.9

15.5

1.8

17.3

3.8

5.6

4.3

7.6

CV

nAUC

SENSORY

MOTOR Median-APB

5.6 Ulnar-ADM

3.0

48.1

7.6

3.9

3.0

50.2

5.6

3.7

50.4

4.8

50.0

3.7

50.9

4.7

52.5

The right median motor response is delayed, consistent with right carpal tunnel syndrome. In addition, the left median motor response amplitude is reduced, as are the ulnar motor responses, left worse than right. Because both ulnar sensory responses are normal, this suggests an intraspinal canal process at the C8 level. To ensure there is no concomitant postganglionic process, the DUC sensory NCS are added and the left MABC sensory NCS. In addition, to better address the severity of involvement, the ulnar motor responses, recording FDI, are added. The radial motor responses, recording EI, could also be added for this purpose. In this case, they were not because the lesion is already localized to the intraspinal canal and because both sides are involved (side-to-side differences will likely underestimate the severity). UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 22 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.7

Ulnar-D5

C8

Superficial Radial

C6,7

RIGHT CV

nAUC

LAT

AMP

18.3

4.2

14.7

2.9

18.0

3.0

16.4

2.3

29.9

Median Palmar

2.6

27.4

2.9

20.3

Ulnar Palmar

1.9

15.5

1.8

17.3

DUC

2.0

16.3

1.8

17.3

MABC

2.6

9.3

SENSORY

388

CV

nAUC

Case 1 through Case 50

(cont.)

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 22 NCS PERFORMED

DRG

LAT

AMP

3.8

5.6

RIGHT CV

nAUC

LAT

AMP

4.3

7.6

CV

nAUC

MOTOR Median-APB

5.6 Ulnar-ADM

3.0

Ulnar-FDI

3.4

48.1

7.6

3.9

3.0

50.2

5.6

3.7

54.6

4.8

50.0

3.7

55.1

4.7

56.0

3.7

3.3

6.2

3.5

50.4

6.1

50.0

3.4

50.9

5.9

52.5

The added sensory and ulnar motor responses confirm an intraspinal canal process that is axon loss in nature and more pronounced on the left. Regarding the median motor responses, although carpal tunnel syndrome causes axon loss, it is worse on the right, and there is no evidence of median motor response amplitude decrement on that side. Thus, most likely, the low-amplitude left median motor response reflects the intraspinal canal process. This will be assessed on the needle EMG. UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 22

Insertional Activity Normal

Spontaneous Activity

IPSWs SCP Other None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

Normal

Normal

LEFT APB

X

X

FDI

X

1+

Normal

CMAL, moderate

EI

X

2+

Neurogenic, mild

CMAL, moderate

FPL

X

X

Normal

Normal

Pron teres

X

X

Normal

Normal

BC, MH

X

X

Normal

Normal

TC, LH

X

X

Normal

CMAL, moderate

Deltoid, MH

X

X

Normal

ECR, longus

X

1+

Neurogenic, mod

CMAL, moderate

389

Section 5: Case Studies in Electrodiagnostic Medicine

(cont.)

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 22

Insertional Activity Normal

Spontaneous Activity

IPSWs SCP Other None

Fibs 1+

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

–

–

Low cerv psp

X

High thor psp

X

X

–

–

APB

X

X

Normal

Normal

FDI

X

2+

Normal

CMAL, mild

EI

X

3+

Normal

CMAL, moderate

Pron teres

X

1+

Normal

CMAL, mild

BC, MH

X

Normal

Normal

TC, LH

X

Normal

Normal

RIGHT

X 1+

The needle EMG study shows acute motor axon loss in the right and left C7 and C8 myotomes. Some of the fibrillation potentials are high in amplitude. Chronic changes are noted in the same distribution and are more pronounced on the left. Thus, in this case, there is both a preganglionic process and a postganglionic process. EDX Study Conclusion

1. Bilateral Median Neuropathies (e.g., carpal tunnel syndrome) The above are demyelinating and axon loss in nature, affect the sensory and motor nerve fibers on the right and the sensory nerve fibers on the left, and are located at or distal to the wrist. 2. Bilateral C7 and C8 Radiculopathies –

The above are axon loss in nature and worse on the left. In the setting of chronic changes, the presence of high-amplitude fibrillation potentials suggests a slowly progressive process.

Exercise 23 A 71-year-old male is referred for a 1-year history of right calf atrophy. The patient had a long history of lower back pain and underwent an L3/4 discectomy in 2012, with good relief. He currently notes occasional left-sided lower back pain that is non-radiating in nature. He first noted the right calf muscle atrophy 1 year ago and states that it has not changed since its recognition. He was diagnosed with type 2 diabetes mellitus approximately 10 years ago. His examination is remarkable for right calf muscle atrophy, as well as thinning of the more distal musculature on that side, and has sensory loss to the mid-tibial level bilaterally.

390

Case 1 through Case 50

Given the indication for the EDX study, the right lower extremity will be studied. In addition, given the bilateral sensory symptoms, a partial study is performed the left side. The initial sensory NCS include routine sensory NCS on the right side and at least one sensory NCS on the left side. LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 23 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Sural

NR

Supfcl Peroneal

NR

NR

The routine sensory NCS on the right side and the left superficial peroneal sensory NCS on the left side showed absent responses, consistent with an axon loss process involving the two sides, such as an axon loss sensory polyneuropathy. At this point, the motor NCS are performed on the right side, along with the H reflex study. LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 23 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Sural

NR

Supfcl Peroneal

NR

NR

MOTOR Tibial-AH

5.2

6.7

14.4

5.1 Peroneal-EDB

4.3

38.7

1.8

13.7 5.1

1.2

37.7

3.9

Peroneal-TA

M wave

5.7

H wave

4.9 NR

The peroneal-EDB motor response is low in amplitude, the tibial motor response is normal, and the H wave is absent. For this reason, the right peroneal motor response, recording tibialis anterior, is required, along with contralateral motor responses and an H reflex study. LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 23 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Sural

NR

391

Section 5: Case Studies in Electrodiagnostic Medicine

(cont.)

LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 23 NCS PERFORMED

Stim Site

LAT

Supfcl Peroneal

AMP

RIGHT CV

nAUC

LAT

NR

AMP

CV

nAUC

NR

MOTOR Tibial-AH

4.6

14.4

29.5

12.8 Peroneal-EDB

3.8

42.4

1.6

4.0

26.6

38.2

3.7

6.7

14.4

5.1

3.1

1.4 Peroneal-TA

5.2

4.3

3.1

38.7

1.8

5.1

1.2

24.2

4.0

37.7

3.9

5.7

H wave

7.3

5.7

NR

3.9 26.1

3.8 M wave

13.7

48.0

4.9 NR

The additional motor responses reveal a low-amplitude left peroneal-EDB response, an absent left H wave, and that the amplitude of the right tibial motor response is relatively abnormal. The asymmetry of the tibial motor responses suggests a possible right S1 radiculopathy, but could be due to an axon loss lesion involving the S1 or S2 motor axons anywhere from the anterior horn of the spinal cord to the motor branch to the abductor hallucis muscle. The right S1 nerve root localization is consistent with the low-amplitude right M wave. The bilaterally low-amplitude peroneal-EDB responses could be due to an axon loss lesion involving the L5 or S1 motor axons anywhere from the anterior horn of the spinal cord to the motor branch to the EDB muscle, including a sensorimotor axon loss polyneuropathy. Regarding the latter consideration, the tibial motor responses would most likely also be involved to some degree. Thus, the left tibial motor response argues against this possibility. Trauma related to shoe wear also results in low-amplitude peroneal-EDB responses, but this can only be determined when the changes are isolated (i.e., normal superficial peroneal sensory responses and normal peroneal-TA motor responses), which is not the case here. The absent left H wave also suggests possible left S1 nerve root involvement, but could be related to the sensory polyneuropathy. Consequently, at this point, there are a number of possibilities, including bilateral S1 radiculopathies (right worse than left), axon loss sensory polyneuropathy, axon loss sensorimotor polyneuropathy, or a combination of these abnormalities. The needle EMG study is expanded to tease out these possibilities. Extensive contralateral studies are required to look for asymmetries indicative of lumbosacral radiculopathies. In our EMG laboratories, we find that it is best to study the contralateral muscle right after studying the ipsilateral one, to better appreciate side-to-side differences, rather than studying one limb and then the other. Thus, in this case, the first four muscles studied would be ipsilateral TA (it is one of the least painful muscles), contralateral TA, ipsilateral gastrocnemius, and contralateral gastrocnemius. LOWER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 23

Insertional Activity Normal

IPSWs

SCP

Spontaneous Activity Other None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

RIGHT

392

FHB

X

X

Normal

Severe CMAL

FDL

X

X

Normal

Moderate CMAL

Case 1 through Case 50

(cont.)

LOWER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 23

Insertional Activity Normal

IPSWs

SCP Other

Spontaneous Activity None

Fibs

Fascs

0

1+

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

Normal

Moderate CMAL

TA

X

Gastroc, MH

X

X

Normal

Severe CMAL

Vast lateralis

X

X

Normal

Normal

BF, SH

X

X

Normal

Moderate CMAL

Glut medius

X

X

Normal

Mild CMAL

FHB

X

X

Normal

Normal

FDL

X

X

Normal

Moderate CMAL

TA

X

Normal

Moderate CMAL

Gastroc, MH

X

X

Normal

Mild CMAL

Vast lateralis

X

X

Normal

Normal

BF, SH

X

X

Normal

Mild CMAL

LEFT

1+

In general, we do not study the paraspinal muscles in the setting of a previous lower back surgery. A small number of fasciculation potentials were noted in the right TA muscle. In our EMG laboratories, whenever we identify fasciculation potentials without fibrillation potentials, we record a zero in the fibrillation potential column so that it is clear that fibrillation potentials were not accidently entered in the incorrect box. The right TA muscle showed a small number of very-low-amplitude fibrillation potentials (indicates significant muscle fiber atrophy and, hence, chronicity). Moderate to severe chronic changes were noted in the L5 and S1 myotomes, right worse than left. On the right side, the S1 muscles were more affected than the L5 muscles, whereas on the left side, the L5 muscles showed more involvement than the S1 muscles. The relationship between the acute (minimal) and chronic (significant) changes indicates a slowly progressive process, such as spondylosis. The normal left FHB muscle argues against a polyneuropathy involving the motor axons, as the distal intrinsic foot muscles are the earliest involved. The asymmetry of involvement of the FHB muscles supports a right S1 radiculopathy. Bilateral sciatic neuropathies (extremely rare) would not account for the changes noted in the right gluteus medius muscle. Bilateral lumbosacral plexus lesions are seen with cancer and other entities, but are uncommon and present differently. Because of the lack of availability of a large number of reliable lower extremity sensory NCS, lesion localization is much more challenging with lower extremity EDX studies than with upper extremity EDX studies. For this reason, the needle EMG study often must be significantly expanded.

393

Section 5: Case Studies in Electrodiagnostic Medicine

EDX Study Conclusion

1. Bilateral Lumbosacral Radiculopathies The above are axon loss in nature and involve the L5 and S1 nerve roots. The abnormalities are most pronounced in the right S1 nerve root distribution and are worse on the right than on the left. 2. Polyneuropathy The above is axon loss in nature and confined to the sensory nerve fibers.

Exercise 24 A 62-year old female, over a 3-week period, developed generalized, symmetric weakness of all four extremities, easy fatigability, and shortness of breath just ambulating around the house. Although the weakness was generalized, on examination it was more pronounced in the proximal muscles (3 to 4-) than in the distal muscles (4- to 4). On the day of admission, she was unable to arise from her couch and was ambulanced to our institution. In addition, she reported a 3year history of carpal tunnel syndrome and a 1-year history of dysphagia. Her examination reveals obvious firmness of the tongue and the submental, submandibular, and sternocleidomastoid muscles, along with mild tongue enlargement and scalloping of its lateral borders. The muscular firmness, tongue enlargement, and bilateral carpal tunnel syndrome, suggests amyloidosis. Sensory loss is confined to the cutaneous distributions of the two median nerves. Given the four-extremity involvement, one upper and one lower extremity will require study, with some contralateral comparison studies performed to demonstrate contralateral involvement and to look for symmetry. In our EMG laboratories, when patients have symmetric disease, we begin the EDX study on the right side. The routine screening sensory NCS are performed on the right upper and lower extremities. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 24 NCS PERFORMED

DRG

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Median-D2

C6,7

NR

NR

Ulnar-D5

C8

2.7

28.3

Superficial Radial

C6,7

2.4

25.7

Median Palmar

NR LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

CASE # NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

3.7

7.7

2.7

13.3

CV

nAUC

SENSORY Sural Supfcl Peroneal

2.3

16.3

The upper extremity sensory responses are consistent with her history of carpal tunnel syndrome. At this point, there is evidence of severe sensory axon loss, but no EDX evidence of demyelination. Thus, the process cannot be localized to the carpal tunnel. Because the median motor responses are typically affected later and to a lesser degree than the median sensory responses in the setting of carpal tunnel syndrome, more accurate localization may be provided by the motor NCS. Importantly, except for the median nerve distributions, there is no evidence of sensory nerve fiber involvement. The lower extremity sensory responses are normal. The routine motor NCS are performed next.

394

Case 1 through Case 50

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 24 NCS PERFORMED

DRG

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Median-D2

C6,7

NR

NR

Ulnar-D5

C8

2.7

28.3

Superficial Radial

C6,7

2.4

25.7

Median Palmar

NR

MOTOR Median-APB

NR

11.0

1.4 1.3

Ulnar-ADM

3.2

43.0

5.6 5.3

65.0

LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 24 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

3.7

7.7

CV

nAUC

SENSORY Sural Supfcl Peroneal

2.3

16.3

2.7

13.3

4.0

5.4

4.9

8.3

MOTOR Tibial-AH

6.6 Peroneal-EDB

2.8

1.2

Peroneal-TA

2.8

4.0

H wave

4.5

6.1 NR

51.7

24.8

NR

3.0

3.1 2.7

M wave

27.6

4.3

26.7 52.0

22.7

8.1 NR

Regarding the upper extremity motor NCS, the left median motor response is absent and the right median motor response is severely reduced in amplitude and very severely prolonged. Thus, on the left, we have a median neuropathy that is axon loss in nature and that involves the sensory and motor nerve fibers. There is no evidence of demyelination by the NCS, and thus, the lesion can only be localized to the median nerve distal to the cord level of the brachial plexus. On the right there is evidence of both axon loss and demyelination. The demyelinating component indicates that the lesion lies distal to the distal stimulation site, consistent with carpal tunnel syndrome. Based on the presence of carpal tunnel syndrome on the right side, it is likely that there is carpal tunnel syndrome on the left side, given the clinical similarity of the hand symptoms. This can be addressed again during the needle

395

Section 5: Case Studies in Electrodiagnostic Medicine

EMG study, because, with carpal tunnel syndrome, the median nerve innervated muscles located proximal to the carpal tunnel (e.g., flexor pollicis longus and pronator teres) are spared with carpal tunnel syndrome. Regarding the lower extremity motor NCS, the right peroneal-EDB response is absent and the left one is severely reduced in amplitude. The H waves are absent bilaterally. Given that the superficial peroneal sensory responses and the peroneal-TA motor responses are normal, it is possible that the isolated peroneal-EDB motor responses are related to the trauma associated with shoe wear. However, this is typically not asymmetric, suggesting that there is at least some dysfunction on the left. Because the left superficial peroneal sensory response is normal, this suggests an intraspinal canal contribution. This should be determinable during the needle EMG study. At this point, the sensory and motor NCS demonstrate very severe right carpal tunnel syndrome and likely extremely severe left carpal tunnel syndrome, along with a possible lesion causing the left peroneal EDB motor response abnormality, such as an L5 or S1 radiculopathy given that the superficial peroneal sensory response is spared. There may also be right-sided involvement or the latter could be related to trauma related to shoe wear. UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 24

Insertional Activity Normal IPSWs

SCP

Spontaneous Activity Other None

Fibs

Fascs

MUAP Analysis

Other

MUAP Recruitment

MUAP Morphology

RIGHT APB

2+

1+

Normal

Many mild CMAL

FDI

2+

2+

Normal

Normal

EI

3+

1+

Normal

Normal

FPL

1+

3+

Normal

Normal

Pron teres

1+

1+

Early

All severe SDLA

Early

Most mod SDLA

1+

Normal

Normal

2+

Early

Most mod SDLA

3+

SMU rapid

4+ polyphasic

BC, MH

X

TC, LH

X

Deltoid, MH

X

2+

Low cerv psp High thor psp

LEFT APB

X

LOWER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 24

Insertional Activity Normal IPSWs

SCP

Spontaneous Activity Other None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

RIGHT TA

396

X

X

Some mild SDLA

Case 1 through Case 50

(cont.)

LOWER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 24

Gastroc, MH Vast lateralis

Insertional Activity Normal

IPSWs

SCP Other

X 1+

Spontaneous Activity None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

2+

Many mod SDLA

2+

Many mod SDLA

LEFT

Needle EMG of the upper extremities showed generalized insertional positive sharp waves and fibrillation potentials, consistent with the symptom onset 3 weeks before the EDX study. The more proximally located muscles showed short-duration, low-amplitude MUAPs with early recruitment. Similar features were noted in the right lower extremity. The lower extremity needle EMG study was incomplete, because the patient’s caregiver requested that the study be stopped. Thus, the EDB muscles were not studied, and therefore, the issue regarding the peroneal motor response asymmetry and possible underlying lumbosacral radiculopathies cannot be rectified. Nonetheless, the study is diagnostic with regard to the question prompting the EDX study. The weakness is related to a myopathy that is generalized in distribution and worse proximally. Based on these EDX findings, the patient underwent a left deltoid muscle biopsy. Prior to the muscle biopsy, because amyloidosis was so strongly suspected, the previous tongue biopsy was restudied using a congo red stain. The tissue was congo red stain negative. This was followed by a fat pad biopsy, which was also congo red stain negative. At this point, the left deltoid muscle was biopsied. There was no necrosis, phagocytosis, or inflammation. Routine stains were unremarkable. MHC staining showed patchy increased activity. It was congophilic red negative. Monoclonal Immunoglobulin Deposition Disease

When monoclonal immunoglobulin protein collections do not polymerize, the affected tissue is negative on congophilic red staining. In this setting, the deposited immunoglobulin protein forms an amorphous mass. This condition is referred to as monoclonal immunoglobulin deposition disease (MIDD). As its name implies, it is a plasma cell dyscrasia that deposits in the tissue. The monoclonal protein deposits may be composed of light chains, heavy chains, or both. Of these categories, light-chain deposition disease (LCDD) is the most common and is usually associated with multiple myeloma. These deposits are capable of infiltrating skeletal muscle and producing a myopathy. Again, because these light chain deposits do not polymerize, they do not have amyloidotic properties and, hence, are congophilic red negative. Thus, clinicians need to be aware of LCDD myopathy, especially when amyloidosis is suspected and tissue testing is congophilic red negative. A serum protein electrophoresis showed hypogammaglobulinemia and a very faint band in the gamma region, prompting further studies, including a UPEP with IFE (showed free kappa monoclonal light chains), serum kappa (4165), kappa/lambda ratio (737), and bone marrow biopsy and aspirate (Congo red negative; 20–25% IgG kappa restricted plasma cells). Based on these findings, kappa light chain deposition disease was suspected, and the muscle biopsy was restudied. It showed kappa light chain positivity (anti-kappa antibody testing), and a diagnosis of kappa light chain deposition disease with myopathy was rendered.

397

Section 5: Case Studies in Electrodiagnostic Medicine

Exercise 25 A 62-year-old right hand–dominant male is referred for EDX assessment of lower back pain that began approximately 25 years ago. It was precipitated by a railway accident in which, while he was moving a railroad car, he was suddenly rear-ended by another individual who was also moving a railroad car. The pain was severe at onset and persisted for about 1 year, at which point it lessened in intensity. It remained low intensity until 10 years later, when it began to slowly intensify. Approximately 15 years later, he was referred for EDX assessment. At that time, the lower back pain was centrally located and radiated to the right great toe on the top of the left foot. There was associated numbness in the right great toe and the top of the left foot as well. He recently tripped and fell while ascending stairs when the toes of his left foot caught the top of one of the stairs. Clinically, the lower back pain radiates to both feet, suggesting bilateral radiculopathies. The numbness is in the cutaneous distribution of the right and left L5 nerve roots. Catching the left foot while ascending stairs suggests ankle dorsiflexion weakness. Thus, the history is suggestive of bilateral L5 radiculopathies. For this reason, bilateral sensory NCS are initially performed. Nerve Conduction Studies

LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 25 NCS PERFORMED

Stim Site

LAT

AMP

3.6

RIGHT CV

nAUC

LAT

AMP

7.0

3.4

7.3

NR

2.9

8.5

CV

nAUC

SENSORY Sural Supfcl Peroneal

The initial sensory NCS are remarkable for an absent left superficial peroneal sensory response. This is consistent with a ganglionic or postganglionic localization, such as the superficial peroneal nerve, common peroneal nerve, sciatic nerve, lumbosacral plexus, or the DRG from which the superficial peroneal sensory axons emanate. The normal sural argues against a sciatic neuropathy but does not exclude this possibility, because, with sciatic neuropathies, the peroneal nerve fibers are more susceptible than the tibial fibers. Because intraspinal canal lesions do not typically affect the sensory responses, this finding is unexpected and suggests the presence of a second disorder. Because of the absent left superficial peroneal sensory response and the possible left foot drop while ascending stairs, the Peroneal-TA motor NCS are included in the initial motor NCS. LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 25 NCS PERFORMED

Stim Site

LAT

AMP

3.6

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

7.0

3.4

7.3

NR

2.9

8.5

7.4

4.0

8.5

21.2

7.1

18.3

7.3

15.3

SENSORY Sural Supfcl Peroneal

MOTOR Tibial-AH

4.0

6.1 Peroneal-EDB

3.6

1.9 1.7

398

42.1 4.0 40.8

3.9

5.4

44.3

13.0

Case 1 through Case 50

(cont.)

LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 25 NCS PERFORMED

Stim Site

Peroneal-TA

LAT

AMP

3.9

5.2

RIGHT CV

nAUC

LAT

AMP

3.8

5.2

48.7 M wave

5.5

H wave

5.0

8.8

5.5

CV

nAUC

51.3

8.9

NR

NR

The motor NCS are remarkable for a very-low-amplitude left peroneal-EDB response, indicative of an axon loss process. When considered with the absent superficial peroneal sensory response, the lesion must lie at or proximal to the common peroneal nerve. The absent H waves suggest possible bilateral S1 disease, early polyneuropathy (there are no clinical features to support this), normal aging (the patient is over the age of 60 years), or any lesion affecting the sensory or motor axons derived from the S1 spinal cord segment. The initial needle EMG studies should include muscles of the L5 and S1 myotomes bilaterally and additional common peroneal nerve innervated muscles (e.g., peroneus longus and EHL).

LOWER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 25

Insertional Activity Normal

IPSWs

SCP Other

Spontaneous Activity None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

RIGHT FHB

X

FDL

X

TA

1+

Normal

Moderate CMAL

X

Normal

Normal

X

X

Normal

Mild CMAL

Gastroc, MH

X

X

Normal

Mild CMAL

EHL

X

X

Normal

Mild CMAL

Peron longus

X

X

Normal

Normal

Vast lateralis

X

X

Normal

Normal

Low L psp

X

1+

–

–

High S psp

X

1+

–

–

Normal

Mod CMAL

LEFT FHB

X

X

399

Section 5: Case Studies in Electrodiagnostic Medicine

(cont.)

LOWER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 25

Insertional Activity Normal

FDL

X

TA

X

Gastroc, MH

IPSWs

SCP

Spontaneous Activity Other None

Fibs 3+

Fascs Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

Normal

Severe CMAL

X

Normal

Severe CMAL

X

X

Normal

Normal

Vast lateralis

X

X

Normal

Normal

BF, SH

X

X

Normal

Mild CMAL

Glut medius

X

X

Normal

Severe CMAL

Low L psp

X

–

–

High S psp

X

–

–

2+ X

The needle EMG is remarkable for fibrillation potentials in the right and left lumbosacral paraspinal muscles consistent with an intraspinal canal lesion, such as bilateral radiculopathies. There are also fibrillation potentials in the right FHB and left FDL muscles, but in none of the other studied skeletal muscles. The needle study also shows evidence of reinnervation via collateral sprouting (e.g., long duration MUAPs). These changes are most pronounced in the muscle domain of the left L5 nerve root. These changes are less pronounced in the left S1, right L5, and right S1 nerve root distributions. EMG Conclusion

1. Bilateral Lumbosacral Intraspinal Canal Lesion (e.g., bilateral multilevel radiculopathies) –

The above is axon loss in nature and involves the bilateral L5 and S1 nerve root domains. The abnormalities are most pronounced in the muscle domain of the left L5 nerve root. The relationship between the acute and chronic features suggests a slowly progressive process, such as spondylosis.

2. Absent Left Superficial Peroneal Sensory Response –

This is of unclear etiology and may represent a remote injury. Uncommonly, when the L5 DRG is located within the intraspinal canal rather than within the intervertebral foramen, the superficial peroneal sensory response may be reduced or absent (Levin, 1998).

Exercise 26 A 25-year-old male was referred for EDX assessment of left upper extremity numbness and a flail arm following a motorcycle accident that occurred 9 weeks earlier. The numbness involved all aspects of the upper extremity except for the axillary skin and a small area along the medial aspect of the arm just below the axilla. Given the diffuse distribution of the numbness, the initial sensory NCS are expanded.

400

Case 1 through Case 50

Nerve Conduction Studies

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 26 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

2.9

24.7

Ulnar-D5

C8

2.6

15.9

Superficial Radial

C6,7

2.4

22.1

LABC

C6

2.4

12.1

Median-D1

C6

3.1

22.7

MABC

T1

2.6

9.4

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY

The initial sensory NCS are normal. To address the possibility of relative abnormalities, a few sensory NCS were performed on the contralateral side. The contralateral studies strongly assess the sensory axons derived from the C6, C8, and T1 DRG. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 26 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

2.9

24.7

Ulnar-D5

C8

2.6

15.9

Superficial Radial

C6,7

2.4

22.1

LABC

C6

2.4

12.1

Median-D1

C6

3.1

22.7

MABC

T1

2.6

9.4

RIGHT CV

nAUC

LAT

AMP

2.7

14.7

2.6

13.9

2.5

8.5

CV

nAUC

SENSORY

The contralateral sensory NCS are symmetric with the ipsilateral ones, indicating that there are also no relative abnormalities. The normal sensory responses, when considered in the context of diffuse extremity numbness, suggest a preganglionic lesion (an intraspinal canal localization). The routine motor NCS will also require expansion, given the distribution of the limb weakness and the limited motor axons assessed by the routine motor NCS (i.e., the median and ulnar motor NCS only assess the motor axons derived from the C8 and T1 spinal cord segments). Thus, motor NCS assessing the C5, C6, and C7 segments are added. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 26 NCS PERFORMED

DRG

LAT

AMP

C6,7

2.9

24.7

RIGHT CV

nAUC

LAT

AMP

2.8

26.7

CV

nAUC

SENSORY Median-D2

401

Section 5: Case Studies in Electrodiagnostic Medicine

(cont.)

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 26 NCS PERFORMED

DRG

LAT

AMP

Ulnar-D5

C8

2.6

Superficial Radial

C6,7

LABC

RIGHT CV

nAUC

LAT

AMP

15.9

2.7

14.7

2.4

22.1

2.4

19.0

C6

2.4

12.1

2.6

13.9

Median-D1

C6

3.1

22.7

2.9

25.9

MABC

T1

2.6

9.4

2.5

8.5

Median-APB

NR

3.5

12.7

Ulnar-ADM

NR

Axillary-Deltoid

NR

3.4

13.8

Musculocutan-BC

NR

Radial-ED

NR

2.7

10.0

CV

nAUC

MOTOR

The routine motor NCS are diffusely absent, consistent with the significant muscle weakness noted on the clinical examination. Thus, the lesion localizes to the intraspinal canal and involves the C5 through T1 segments. It is axon loss in nature, but because the motor responses are absent, there is no NCS evidence of nerve fiber continuity. This will need to be addressed on the needle EMG study. UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 26

Insertional Activity Normal

Spontaneous Activity

IPSWs SCP Other None Fibs

Fascs Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

LEFT

402

APB

X

3+

None firing

n/a

FDI

X

3+

None firing

n/a

EI

X

4+

None firing

n/a

FPL

X

3+

None firing

n/a

Pron teres

X

3+

None firing

n/a

BC, MH

X

4+

None firing

n/a

TC, LH

X

4+

None firing

n/a

Deltoid, MH

X

4+

None firing

n/a

Infraspinatus

X

4+

None firing

n/a

Rhomb major

X

3+

None firing

n/a

Serr anterior

X

3+

None firing

n/a

Case 1 through Case 50

(cont.)

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 26

Trapezius

Insertional Activity Normal

Spontaneous Activity

IPSWs SCP Other None Fibs

X

X

Fascs Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

Normal

Normal

Middle cerv psp

3+

None firing

n/a

Low cerv psp

3+

None firing

n/a

3+

MUAPs noted

–

High thor psp

X

Needle EMG Study

A large number of fibrillation potentials (3+ to 4+) and an absence of voluntarily activated MUAPs were noted in muscles belonging to the C5 through T1 spinal cord segments, including those innervated via motor axons derived from the APR or proximal trunk level (serratus anterior, rhomboideus major, infraspinatus). The cervical paraspinal muscles did not show MUAPs with attempted neck extension. The upper thoracic paraspinal muscles showed some fibrillation potentials and MUAPs. The trapezius muscle is normal. Because it is innervated by the spinal accessory nerve and the C3 and C4 nerve roots, the C3 and C4 nerve roots are likely spared. Although the T1 nerve root is avulsed, the upper thoracic paraspinal muscles showed some MUAPs. Because the paraspinal muscles are innervated by more than one nerve root, these MUAPs likely reflect more inferior roots (e.g., T2 and T3). Thus, the EDX study could not demonstrate evidence of nerve root continuity in the C5 through T1 nerve roots, consistent with avulsion injuries. EDX Conclusion

1. C5 through T1 nerve root avulsions. Whenever the motor axons are affected to a greater extent than the sensory axons derived from the same spinal cord segment, the lesion is preganglionic, assuming that more than 10 days had elapsed from the time of symptom onset (i.e., enough time for Wallerian degeneration to have occurred along the affected sensory axons). In this case, the sensory responses were normal, which could also be seen with a distal process (e.g., terminal nerve, NMJ, and muscle fiber disorders). However, these disorders produce disintegration of the motor unit rather than complete loss. Moreover, these latter entities are not in the differential diagnosis, because the patient has profound sensory loss on examination. Although this pattern of EDX abnormalities – normal sensory responses, absent motor responses, significant numbers of fibrillation potentials, and absent MUAPs – can be seen with avulsion injuries, in our experience we often observe concomitant postganglionic damage, most commonly affected axons derived from the C5,6 spinal cord segments (Ferrante and Wilbourn, 1995).

Exercise 27 A 31-year-old male was referred for EDX assessment of diffuse left upper extremity numbness, total left hand paralysis, and minimal movement of the left shoulder and proximal upper extremity muscles, all of which followed a motorcycle accident 7 months prior. The diffuse numbness and weakness indicates that the lesion must involve multiple PNS elements (e.g., multiple roots, trunks, divisions, cords, or nerves). Absence of hand movements suggests extremely severe or complete

403

Section 5: Case Studies in Electrodiagnostic Medicine

involvement of the C8 and T1 motor axons, whereas slight movement of the shoulder and proximal upper extremity muscles suggests the presence of at least some C5 or C6 motor axon function. Given the diffuse distribution of the numbness, the initial sensory NCS are expanded. Nerve Conduction Studies

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 27 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.1

11.1

Ulnar-D5

C8

2.7

16.5

Superficial Radial

C6,7

2.4

22.8

LABC

C6

2.4

7.7

Median-D1

C6

3.2

9.7

MABC

T1

2.5

14.0

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY

The initial sensory NCS show low-amplitude LABC, Median-D1, and Median-D2 sensory responses. This indicates an axon loss process involving the lateral cord, upper plexus, or C6 APR/DRG as the smallest focus capable of explaining all three sensory response abnormalities. However, these possibilities are inconsistent with the clinical examination showing diffuse sensory abnormalities, raising the possibility of concomitant demyelinating conduction block lesion or an intraspinal canal lesion, both of which would spare the sensory NCS. The low sensory responses are compared to those on the contralateral side. The contralateral superficial radial nerve was also studied because it is abnormal in at least 60% of upper plexopathies (Ferrante and Wilbourn, 1995). UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 27 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.1

11.1

Ulnar-D5

C8

2.7

16.5

Superficial Radial

C6,7

2.4

LABC

C6

Median-D1 MABC

RIGHT CV

nAUC

LAT

AMP

3.0

26.3

22.8

2.5

20.7

2.4

7.7

2.5

16.3

C6

3.2

9.7

3.2

24.8

T1

2.5

14.0

2.4

14.5

CV

nAUC

SENSORY

The contralateral sensory NCS verify the initial suspicion and also suggest that the sensory axons composing the superficial radial nerve predominantly derive from the C7 DRG in this patient. The initial motor NCS are expanded to better assess the distribution of the clinical weakness.

404

Case 1 through Case 50

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 27 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.1

18.1

Ulnar-D5

C8

2.7

16.5

Superficial Radial

C6,7

2.4

22.8

RIGHT CV

nAUC

LAT

AMP

3.0

26.3

2.5

20.7

CV

nAUC

SENSORY

LABC

C6

2.4

7.7

2.5

16.3

Median-D1

C6

3.2

9.7

3.2

24.8

MABC

T1

2.5

14.0

MOTOR Median-APB

NR

Ulnar-ADM

NR

Ulnar-FDI

NR

Radial-ED

NR

X

14.9

Axillary-Deltoid

X

0.5

X

12.4

Musculocutan-BC

X

1.1

X

8.7

The expanded motor NCS are indicative of a more diffuse axon loss process that involves the C5 through T1 segments and that supports an intraspinal canal lesion at the C5 through T1 levels. Regarding the latter, the absent median and ulnar motor responses indicate no evidence of C8 and T1 nerve root continuity. Because the Radial-ED motor response is also absent, this indicates the C6 and C7 nerve roots are also involved (the ED is innervated by motor axons derived from the C6, C7, and C8 spinal cord segments). The axillary and musculocutaneous motor responses are very low in amplitude, consistent with absent C6 input and partial C5 input. Because the C6 motor involvement is more pronounced than the C6 sensory involvement, at this level, the process must be both preganglionic (motor worse than sensory) and postganglionic (based on the initial sensory NCS). The degree of involvement of the musculocutaneous and axillary motor responses indicates that more than one root is involved. Because C6 shows no continuity, it is likely that the C5 root is only partially disrupted. Thus, at this point, the NCS indicate an axon loss process involving the C5 through T1 roots, that the C6 through T1 nerve roots are likely avulsed, that the C5 root is likely partially avulsed, and that in addition to the root avulsions, there is concomitant postganglionic axon loss involving the C6 sensory axons. Whether the C5 sensory axons are also affected at the postganglionic level cannot be determined by the sensory NCS, because a reliable sensory NCS to assess them is not available. The needle EMG will be expanded to include enough muscles to determine which roots are avulsed, which roots are partially avulsed, and which roots are spared. UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 27

Insertional Activity Normal

Spontaneous Activity

IPSWs SCP Other None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

LEFT APB

X

4+

None firing

n/a

FDI

X

3+

None firing

n/a

405

Section 5: Case Studies in Electrodiagnostic Medicine

(cont.)

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 27

Insertional Activity Normal

Spontaneous Activity

IPSWs SCP Other None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

EI

X

4+

None firing

n/a

FPL

X

3+

None firing

n/a

Pron teres

X

3+

None firing

n/a

BC, MH

X

3+

Severe neurogenic

Mild CMAL

TC, LH

X

3+

None firing

n/a

Deltoid, MH

X

3+

Severe neurogenic

Moderate CMAL

ECR

X

3+

None firing

n/a

Trapezius

X

X

Normal

Normal

Middle cerv psp

X

X

–

–

Low cerv psp

X

X

–

–

High thor psp

X

X

–

–

Needle EMG Examination

The needle EMG study showed a large number of fibrillation potentials and an absence of MUAPs in all of the muscles belonging to the C6 through T1 myotomes. The absence of MUAPs in the C6,7 muscles (pronator teres and ECR) indicates both roots are avulsed, whereas the presence of MUAPs in the C5,6 muscles indicates that some motor axons are functioning. Consequently, the C5 root is only partial disrupted. Also, the involvement of the rhomboideus major indicates that the lesion is at or proximal to the APR level, as previously indicated by the relationship between the sensory and motor NCS. Sparing of the trapezius argues against C3 and C4 involvement, because the trapezius receives motor axons from the C3 and C4 roots in addition to its innervation via the spinal accessory nerve. There is evidence of some reinnervation via collateral sprouting occurring through the functioning C5-derived motor axons. The trapezius muscle and the middle and lower cervical and upper thoracic paraspinal muscles are normal. In the setting of avulsion injuries, it is not infrequent to see a lack of fibrillation potentials in the paraspinal muscles, likely due to the multi-root innervation of these muscles. EDX Conclusion

The EDX abnormalities indicate an intraspinal canal lesion involving the C5 through T1 nerve roots. There is no evidence of continuity for the C6 through T1 nerve roots, suggesting avulsion. There is neurogenic MUAP recruitment in muscles receiving C5 input, suggesting partial avulsion. In addition, at the C6 level, there is concomitant postganglionic involvement (indicated by the sensory response abnormalities). When avulsion injuries (preganglionic) are combined with postganglionic injuries, it is referred to as a 2-level process. The textbook EDX features of nerve root avulsions – normal SNAPs, absent CMAPs, absent MUAPs, and abundant fibrillation potentials in the extremity and paraspinal muscles – in our opinion, are less frequent than the pattern reported in this exercise, which includes sensory response involvement and paraspinal muscle sparing.

406

Case 1 through Case 50

In this exercise, the sensory responses indicate postganglionic involvement at the C6 level, indicating a 2-level process for the C6 nerve fibers. There are no sensory NCS to assess the sensory axons of the C5 DRG. Thus, it may also be involved. Anatomically, there is connective tissue anchoring the C5 and C6 mixed spinal nerves to the transverse processes just after their exit from the spinal column. The C8 and T1 mixed spinal nerves are not bound down in this manner. Thus, with traction injuries, the C5 and C6 axons are often ruptured at their connective tissue anchorage sites (i.e., a postganglionic level), and the C8 and T1 roots are avulsed from the spinal cord. With more pronounced degrees of traction, the C5 and C6 roots are often ruptured and avulsed, as in this case. The degree of connective tissue anchoring at the C7 mixed spinal nerve varies. Thus, it may be ruptured, avulsed, or both. In this case, the normal superficial radial sensory response argues against a 2-level process at the C7 level.

Exercise 28 A 42-year-old right hand–dominant male is referred for EDX assessment of right hand numbness and tingling. According to the patient, this started 3–4 months ago and involves the medial two digits and the medial aspect of the hand. He also complains of loss of grip strength. He denies neck pain, and there are no left-sided symptoms. His examination shows that he splits the fourth digit, has sensory loss in the DUC distribution but not the MABC distribution, and has significant weakness of finger abduction and distal flexion of the fourth and fifth digits (i.e., FDP muscle), suggesting an elbow segment lesion. Based on this presentation, the initial sensory NCS include routine right-sided NCS plus bilateral Ulnar-D5, DUC, and MABC sensory NCS, unless the ipsilateral responses are unquestionably normal. Nerve Conduction Studies

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 28 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

3.3

45.0

3.0

7.7

2.0

26.0

CV

nAUC

SENSORY Median-D2 Ulnar-D5

2.6

20.7

Superficial Radial DUC

1.8

11.6

1.9

4.9

MABC

2.1

8.6

2.0

9.3

The initial sensory NCS identify low-amplitude right Ulnar-D5 and DUC responses (by absolute criteria) and normal MABC responses. The contralateral ulnar sensory responses are normal. These studies indicate an axon loss process involving the ulnar nerve (at or proximal to the DUC branch point), medial cord, lower trunk, C8 APR/DRG. The normal MABC argues against a medial cord or lower trunk lesion (but cannot exclude a partial lesion involving one of these two sites). The degree of clinical weakness seems too great to be associated with such a mild degree of sensory axon loss, raising the possibility of concomitant demyelinating conduction block. Although the EDX study is an extension of the clinical examination, it is an independent study. Thus, throughout the EDX study, the clinical information is considered but is not substituted for the EDX information. The initial motor NCS include the ipsilateral routine motor NCS, the ipsilateral Ulnar-FDI, and the contralateral ulnar motor NCS. Because the localization differential includes the C8 APR, the Radial-EI motor NCS may be required. Because the clinical features strongly suggest an ulnar neuropathy at the elbow segment and because the sensory NCS suggest a demyelinating conduction block, if the ulnar motor NCS identify the lesion, the radial motor NCS is not needed.

407

Section 5: Case Studies in Electrodiagnostic Medicine

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 28 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

3.3

45.0

3.0

7.7

2.0

26.0

CV

nAUC

SENSORY Median-D2 Ulnar-D5

2.6

20.7

Superficial Radial DUC

1.8

11.6

1.9

4.9

MABC

2.1

8.6

2.0

9.3

3.3

7.5

MOTOR Median-APB

7.3 Ulnar-ADM

2.5

Ulnar-FDI

3.9

7.6

23.3

14.7

2.7

25.3

4.0

56.3

4.3

11.8

4.0

54.7

10.6

0.9

28.1

2.9

4.0

7.0

4.0

50.5

6.9

1.5

15

2.9

On motor NCS, ulnar motor responses are reduced in amplitude, indicating an axon loss process. In addition, both identify a demyelinating conduction block between the above-elbow and below-elbow stimulation sites, thereby localizing the lesion to the elbow segment. Regarding the motor axons to the ADM (actually to the hypothenar eminence), the degree of axon loss is 49% (11.8/ 23.3
100% = 51%; 100% – 51% = 49%). Of the 51% of motor axons not affected by axon loss, 73% are blocked (2.9/ 10.6
100% = 27%; 100% – 27% = 73%). Thus, 73% of 51% are blocked (0.73
0.51 = 0.37
100% = 37%). Thus, of the motor axons innervating the hypothenar eminence, axon loss involves 49%, demyelinating conduction block involves 37%, and 14% are unaffected. Regarding the motor axons to the FDI muscle, axon loss involves 72% (7.0/25.3
100% = 28%; 100% – 28% = 72%). Demyelinating conduction block involves 58% of the remaining motor axons (2.9/6.9 = 42%; 100% – 42% = 58%). Thus, 16% of the motor axons are affected by demyelinating conduction block (0.58
0.28 = 0.16
100% = 16%). At this point, the lesion is localized to the elbow segment of the ulnar nerve, involves the sensory and motor axons, and is both demyelinating conduction block and axon loss in nature. Needle EMG Study

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 28

RIGHT FDI

408

Insertional Activity Normal IPSWs X

SCP Other

Spontaneous Activity None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

V severe neurogenic

Severe CMAL

X 1+

3+

Case 1 through Case 50

(cont.)

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 28

Insertional Activity Normal IPSWs

SCP Other

Spontaneous Activity None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

EI

X

X

Normal

Normal

FPL

X

X

Normal

Normal

Pron teres

X

X

Normal

Normal

BC, MH

X

X

Normal

Normal

TC, LH

X

X

Normal

Normal

Severe neurogenic

Moderate CMAL

FDP

3+

3+

Low cerv psp

X

X

–

–

High thor psp

X

X

–

–

FDI

X

X

Normal

Normal

FDP

X

X

Normal

Normal

LEFT

The needle EMG study shows EDX evidence of both acute (insertional positive sharp waves and fibrillation potentials) and chronic (long-duration MUAPs) motor axon loss in the muscle domain of the right ulnar nerve. In addition, there is neurogenic MUAP recruitment in this distribution, severe in degree, consistent with concomitant demyelinating conduction block. The degree of chronic motor axon loss is greater than would be expected for a lesion of 3–4 months’ duration, and the presence of insertional positive sharp waves indicates denervation after the onset of symptoms. Thus, this is a progressive lesion. EDX Study Conclusion

1. Right Ulnar Neuropathy –

The above is axon loss and demyelinating conduction block in nature, involves the sensory and motor nerve fibers, and is located at the elbow segment. The lesion is severe in degree. For the motor axons to the hypothenar eminence, axon loss involves at least 49% and demyelinating conduction block involves 37%. For the motor axons to the FDI muscle, axon loss involves at least 72% and demyelinating conduction block involves at most 16%.

Because of the timing of the study (performed more than 3 months after symptom onset) and the chronic motor axon loss noted on the needle EMG study, the term “at least” is used when defining the degree of axon loss (reinnervation by collateral sprouting causes lesion underestimation). For this reason, the calculated percentage of motor axons affected by demyelinating conduction block is an overestimate, hence the term “at most” when defining its percentage.

409

Section 5: Case Studies in Electrodiagnostic Medicine

Exercise 29 A 55-year-old right hand–dominant male is referred for EDX assessment of bilateral upper extremity weakness. According to the patient, about 18 years ago, he had a bout of transverse myelitis with associated bilateral upper extremity weakness. He also has decreased sensation in both hands. Over the past 2 years, his hands have stopped moving and are more painful. Based on this, the initial sensory NCS include ipsilateral routine and median palmar NCS and contralateral Median-D2, Ulnar-D5, and median palmar NCS. Nerve Conduction Studies

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET EXERCISE 29 NCS PERFORMED

LEFT Stim Site

LAT

AMP

Median-D2

3.0

Ulnar-D5

RIGHT CV

nAUC

LAT

AMP

37.0

3.1

40.0

2.8

19.0

3.1

21.7

Superficial Radial

2.4

30.0

Median palmar

1.7

41.3

1.8

39.6

CV

nAUC

SENSORY

The initial sensory NCS are normal. Given the peak latencies of the median palmar responses (1.7 and 1.8 msec), ulnar palmar responses to look for an interpeak latency difference exceeding 3 msec were not pursued. Because the hand weakness is bilateral, the initial motor NCS include the screening studies on both sides.

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET EXERCISE 29 NCS PERFORMED

LEFT Stim Site

LAT

AMP

Median-D2

3.0

Ulnar-D5

RIGHT CV

nAUC

LAT

AMP

37.0

3.1

40.0

2.8

19.0

3.1

21.7

Superficial Radial

2.4

30.0

MABC

1.9

15.3

Median palmar

1.7

41.3

1.8

39.6

CV

nAUC

SENSORY

MOTOR Median-APB

NR

NR

Ulnar-ADM

NR

NR

The screening motor NCS are performed and the responses are absent bilaterally. For this reason, the MABC sensory NCS was added on the more symptomatic side (it assesses sensory axons derived from the T1 DRG). It was also normal. When an absent motor response is coupled with a normal sensory response and both assess the same spinal cord segment, an intraspinal canal lesion is indicated. Such pairings include the Ulnar-D5 (C8 DRG) sensory

410

Case 1 through Case 50

NCS with the Ulnar-ADM (C8 > T1) motor NCS and the MABC (T1 DRG) sensory NCS with the Median-APB (T1 > C8) motor NCS. Thus, at this point, an axon loss process involving at least the C8 and T1 intraspinal canal segments that is severe in degree is identified. Additional motor NCS can be added to assess the C5 through C7 segments, or the needle EMG study can be employed to look at the C5 through T1 segments. In this case, we opted to expand the needle EMG study. Needle EMG Study

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 29

Insertional Activity Normal IPSWs SCP Other

Spontaneous Activity None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

LEFT APB

Decr

1+

None fire

n/a

FDI

Decr

1+

None fire

n/a

EI

Decr

X

None fire

n/a

FPL

Decr

X

None fire

n/a

Pron teres

X

X

Mild Decreased

Moderate CMAL

BC, MH

X

X

Normal

Mild CMAL

TC, LH

Decr

X

None fire

n/a

Deltoid, MH

X

X

Normal

Mild CMAL

Low cerv psp

Patient refused

High thor psp

Patient refused

RIGHT FDI

X

Pron teres

Decr

BC, MH

2+

None fire

n/a

X

None fire

n/a

X

X

Normal

Normal

TC, LH

Decr

X

None fire

n/a

Deltoid, MH

X

X

Normal

Mild CMAL

The needle EMG study shows decreased insertional activity in muscles of the C7, C8, and T1 spinal cord segments, consistent with loss of viable muscle tissue. The bilateral C8 and T1 spinal cord segments innervate muscles from which MUAPs cannot be voluntarily generated. This is less pronounced for the left C7 segment (pronator teres shows some firing). On the right, there are no MUAPs noted in the pronator teres (C6,7) and triceps

411

Section 5: Case Studies in Electrodiagnostic Medicine

(C6,7,8), yet the right biceps (C5,6) is normal. Chronic changes in the left C6,7 distribution (pronator teres) are more pronounced than those noted in muscles of the bilateral C5,6 segments. Some very-low-amplitude fibrillation potentials (indicative of chronicity) are present in the C8 and T1 myotomes. Their etiology is unclear. Like post-poliomyelitis syndrome, they may represent loss of anterior horn cells with normal aging or excessive metabolic demand. The needle EMG study indicates involvement of the C7, C8, and T1 segments to a greater extent than the C6 segment and suggests that the triceps is likely C7,8 innervated (it is most frequently C7 > C6  C8 innervated but, less frequently, receives significant innervation via C8. EDX Study Conclusion

1. Intraspinal Canal Lesion The above involves the C8 and T1 spinal cord segments to a greater extent than the C7 spinal cord segment; the C6 spinal cord segment is least involved. Due to myotomal overlap, it cannot be determined whether the C5 segment is also involved (a C4,5 muscle, such as the levator scapulae, could have been studied, but the patient did not wish further needle EMG studies). These EDX features are consistent with his history of transverse myelitis. There is no EDX evidence of a postganglionic lesion.

Exercise 30 A 48-year-old right hand–dominant male was referred for EDX assessment of bilateral hand muscle wasting and weakness suspected to be amyotrophic lateral sclerosis. According to the patient, he first noted trouble buttoning his shirt about one year ago. The hand weakness progressed, and he eventually became unable to squeeze the trigger of his weapon. He denied bulbar, proximal upper extremity, and lower extremity weakness. He also denied muscle cramping and sensory loss. He had bilateral hand intrinsic muscle weakness and wasting, severe in degree; forearm muscle weakness with distal wasting, mild in degree; and ankle dorsiflexion weakness, mild in degree. His sensory examination was normal. Muscle stretch reflexes were absent in the upper extremities. Based on the symmetric distribution of the weakness and wasting, amyotrophic lateral sclerosis seemed unlikely. The study was started with screening sensory NCS of the right upper extremity.

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 30 NCS PERFORMED

DRG

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Median-D2

C6,7

2.8

27.0

Ulnar-D5

C8

2.6

17.6

Superficial Radial

C6,7

2.2

23.9

The screening sensory NCS are normal. Given that the patient denied sensory symptoms, this is not unexpected. The screening motor NCS are performed next.

412

Case 1 through Case 50

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 30 NCS PERFORMED

DRG

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Median-D2

C6,7

2.8

27.0

Ulnar-D5

C8

2.6

17.6

Superficial Radial

C6,7

2.2

23.9

2.8

4.5

MOTOR Median-APB

2.8

5.1 4.8

Ulnar-ADM

2.8

58.9

4.4

5.2

2.8

5.0

52.3

6.0

55.2

5.5

55.3

The motor responses are mildly reduced in amplitude. Motor response involvement with sensory response sparing is seen with proximal lesions (e.g., bilateral C8 radiculopathies), distal lesions (i.e., distal to the departure sites of the sensory branches, including the neuromuscular junction and muscle fiber level), and lesions of roughly 7 days of age. The latter is excluded by the history of weakness for at least one year. Routine upper extremity muscles are studied next. UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 30

Insertional Activity Normal IPSWs

Spontaneous Activity

SCP Other None

Fibs

Fascs Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

RIGHT APB

1+

2+

Early

All SDLA, severe

FDI

1+

2+

Early

Most SDLA, severe

Normal

Most SDLA, mod

Normal

Most SDLA, mod

Normal

Many SDLA, mod

Normal

Many SDLA, mod

EI

X

FPL

X

Pron teres

X

BC, MH

X

TC, LH

X

X

Normal

Few SDLA, mild

Deltoid, MH

X

X

Normal

Many SDLA, mild

–

–

Mid cerv psp

X 1+ X 1+

1+

1+

413

Section 5: Case Studies in Electrodiagnostic Medicine

(cont.)

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 30

Low cerv psp

Insertional Activity Normal IPSWs

Spontaneous Activity

SCP Other None

X

Fibs

MUAP Analysis

Fascs Other

1+

MUAP Recruitment

MUAP Morphology

–

–

LEFT FDI

X

X

Early

Most SDLA, mod

BC, MH

X

X

Normal

Many SDLA, mod

The needle EMG showed short-duration, low-amplitude motor unit APs, consistent with a distal myopathy. Early recruitment was present in the most severely affected muscles. Fibrillation potentials were also present. Based on the needle EMG findings, the lower extremities were also studied. LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 30 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

Sural

3.2

11.6

Supfcl Peroneal

2.7

9.7

4.3

4.5

CV

nAUC

SENSORY

MOTOR Tibial-AH

4.3 Peroneal-EDB

4.3

2.8 2.1

Peroneal-TA

2.8

44.2

44.2

5.4

LOWER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 30

Insertional Activity

Spontaneous Activity

Normal IPSWs SCP Other

None

FHB

X

FDL

X

Fibs Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

X

Normal

Many SDLA, mild

X

Normal

Normal

RIGHT

414

Case 1 through Case 50

(cont.)

LOWER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 30

Insertional Activity Normal IPSWs SCP Other

Spontaneous Activity None

Fibs Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

TA

X

2+

Normal

Many SDLA, mild

Gastroc, MH

X

2+

Normal

Most SDLA, mod

Vast lateralis

X

X

Normal

Many SDLA, mild

Rectus femoris

X

X

Normal

Many SDLA, mod

Glut medius

X

X

Normal

Few SDLA, mild

Iliacus

X

X

Normal

Many SDLA, mod

–

–

Low L psp

1+

2+

LEFT TA

X

X

Normal

Many SDLA, mild

Gastroc, MH

X

X

Normal

Many SDLA, mild

Vast lateralis

X

1+

Normal

Some SDLA, mild

Rectus femoris

X

1+

Normal

Some SDLA, mild

The lower extremity study showed similar features, less pronounced in degree, and clinically, the patient had ankle dorsiflexion weakness. Muscle biopsy showed desmin staining, consistent with a myofibrillary myopathy.

Exercise 31 A 44-year-old right hand–dominant male was referred for EDX assessment of generalized weakness and distal 4-extremity numbness. According to the patient, he was in his usual state of health (running several miles per day and performing weight training exercises three times per week) until 8 months ago, when he developed lower extremity edema, anasarca, orthopnea, weakness, and fatigue. The weakness worsened and he could not arise from a chair, and with ambulation, he was scuffing his toes. He noted profound thigh muscle atrophy that was worse on the right. Over the next two weeks, severe hand weakness and bilateral foot drop developed, at which point he noted numbness and tingling in all toes and fingers. The latter progressed and maintained a stocking-glove

415

Section 5: Case Studies in Electrodiagnostic Medicine

distribution. In addition, he developed a severe nephropathy and a severe retinopathy. He was diagnosed with type 2 diabetes mellitus and started on insulin. General examination showed generalized cachexia, severe lower extremity edema, and anasarca. Neurological examination was remarkable for severe weakness involving the C5 and C6 myotomes, the distal upper extremity muscles, and the proximal and distal lower extremity muscles (distal lower extremity weakness more pronounced than proximal lower extremity weakness). There was diminished pinprick perception to the mid-metatarsal level and mildly decreased vibratory perception at both great toes. Muscle stretch reflexes were absent at the ankles and knees and reduced elsewhere. The patient was referred for EDX testing 18 months after the onset of the disease. Given the distribution of the sensory, motor, and reflex abnormalities, a 4-extremity study was planned. At this point, his clinical phenotype demonstrated a stocking-glove distribution of sensory and motor loss (consistent with an axon loss polyneuropathy) and superimposed proximal muscle wasting involving at least the C5, C6, and L4 nerve roots, bilaterally. The lower extremities were studied first, beginning with the screening sensory NCS. LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 31 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Sural

NR

NR

Supfcl Peroneal

NR

NR

The sensory responses are absent on both sides, consistent with an axon loss process involving the sensory nerve fibers. The screening motor NCS are performed next. LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 31 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Sural

NR

NR

Supfcl Peroneal

NR

NR

Tibial-AH

NR

NR

Peroneal-EDB

NR

NR

MOTOR

M wave H wave

5.5

3.4 NR

5.3

3.9 NR

The distal motor responses are absent, consistent with an axon loss process involving the motor nerve fibers. The H waves are absent bilaterally, whereas the M waves are reduced in amplitude bilaterally. This indicates a length-dependent distribution (distal worse than proximal). The peroneal motor responses, recording tibialis anterior, are added.

416

Case 1 through Case 50

LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 31 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Sural

NR

NR

Supfcl Peroneal

NR

NR

Tibial-AH

NR

NR

Peroneal-EDB

NR

NR

MOTOR

Peroneal-TA

4.9

1.4

5.4

1.2

M wave

5.5

H wave

32.8

1.9 1.8

3.4

5.3

NR

31.3

3.9 NR

The peroneal motor responses, recording tibialis anterior, are also significantly reduced in amplitude, similar to the M waves, and consistent with a length dependent process. Femoral motor responses, recording rectus femoris, might have been helpful but were not performed. At this point, an axon loss polyneuropathy, involving the sensory and motor nerve fibers and severe in degree, is apparent. The upper extremity screening sensory NCS are performed next, beginning with the right upper extremity. Because length-dependent polyneuropathies are symmetric, some NCS will be required on the contralateral side. Also, additional NCS may be required to identify the proximal extent of the process. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 31 NCS PERFORMED

DRG

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Median-D2

C6,7

NR

NR

Ulnar-D5

C8

NR

NR

Superficial Radial

C6,7

3.2

Dorsal Ulnar Cutan MABC

5.7

3.0

NR 4.3

8.9

8.8 NR

4.6

7.8

The digital sensory responses and the dorsal ulnar cutaneous responses are absent, consistent with an axon loss process. The superficial radial sensory responses are low in amplitude bilaterally, consistent with a lesser degree of axon loss more proximally (i.e., a length-dependent pattern). Given that the superficial radial sensory responses are present and that the superficial radial and dorsal ulnar cutaneous NCS are recorded from approximately the same level of the upper extremity, the possibility of superimposed ulnar neuropathies is raised. This will need to be addressed on the motor NCS. The medial antebrachial cutaneous responses are normal, again consistent with a length-dependent process. The motor NCS are performed next.

417

Section 5: Case Studies in Electrodiagnostic Medicine

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 31 NCS PERFORMED

DRG

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Median-D2

C6,7

NR

NR

Ulnar-D5

C8

NR

NR

Superficial Radial

C6,7

3.2

Dorsal Ulnar Cutan

5.7

3.0

NR

8.8 NR

MABC

4.1

4.2

4.6

3.8

LABC

3.5

6.2

3.6

5.5

7.5

3.2

8.6

1.1

MOTOR Median-APB

2.6 Ulnar-ADM

3.3

37.0

0.8

5.9

3.0

5.5

39.9

4.9

3.5 3.4

5.4 Ulnar-FDI

38.3

1.9

1.5

5.4

1.2

1.5 1.3

The median and ulnar motor responses are very low in amplitude, and the distal latencies of the median motor responses are delayed out of proportion to the degree of amplitude decrement. This indicates bilateral median neuropathies (e.g., carpal tunnel syndrome), very severe in degree (focal demyelination and axon loss), as well as bilateral ulnar neuropathies. The latter are axon loss in nature on the left and mixed axon loss and demyelinating conduction block on the right. The demyelinating conduction block is present between the above- and belowelbow stimulation sites, consistent with an elbow segment lesion. The motor NCS explain why the sensory responses did not follow a length-dependent process – there are superimposed bilateral median and ulnar neuropathies. The low superficial radial sensory responses are consistent with a sensory polyneuropathy, but whether or not the motor axons are involved by the polyneuropathy cannot be determined and will need to be determined by the needle EMG findings of distal muscles not innervated by the ulnar or median nerve (e.g., the extensor indicis). The needle EMG is performed next. LOWER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 31

Insertional Activity Normal IPSWs

SCP

Spontaneous Activity Other None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

RIGHT FHB

1+

2+

Neurogenic, severe

CMAL, severe

FDL

1+

2+

Neurogenic, mod

CMAL, moderate

1+

Neurogenic, mod

CMAL, severe

TA

418

X

Case 1 through Case 50

(cont.)

LOWER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 31

Insertional Activity Normal

Gastroc, MH

IPSWs

SCP Other

Spontaneous Activity None

1+

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

2+

Neurogenic, mild

CMAL, moderate

2+

Normal

Normal

Neurogenic, mild

CMAL, mild

Normal

CMAL, moderate

Vast lateralis

X

BF, SH

X

Glut medius

X

Rectus Femoris

X

X

Neurogenic, mod

CMAL, moderate

Mid L psp

X

X

–

Obvious CMAL

High S psp

X

X

–

Obvious CMAL

X 1+

LEFT FDL

X

2+

Neurogenic, mod

CMAL, moderate

TA

X

1+

Neurogenic, mod

CMAL, moderate

BF, SH

X

1+

Normal

CMAL, mild

Vast lateralis

X

2+

Normal

CMAL, mild

The needle EMG of the lower extremities demonstrates both a length-dependent distribution (distal muscles worse than proximal) and a polyradiculopathy distribution (bilateral L4, L5, and S1 involvement). The presence of pronounced chronic changes in the paraspinal muscles also indicates an intraspinal canal localization. In our EMG laboratories, like most others, we do not grade the MUAPs of the paraspinal muscles. However, when there are extreme chronic changes (i.e., MUAPs that would be large for an extremity muscle), we include this in the report.

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 31

Insertional Activity Normal

IPSWs

SCP Other

Spontaneous Activity None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

Neurogenic, mod

Mod CMAL

RIGHT APB

X

X

419

Section 5: Case Studies in Electrodiagnostic Medicine

(cont.)

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 31

Insertional Activity Normal

FDI

IPSWs

SCP

Spontaneous Activity Other None

2+

Fibs 2+

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

Neurogenic, mod

Mod CMAL

EI

X

X

Normal

Mild CMAL

FPL

X

X

Normal

Normal

Pron teres

X

1+

Normal

Mild CMAL

BC, MH

X

1+

Neurogenic, mild

Mod CMAL

TC, LH

X

Neurogenic, mild

Mild CMAL

Deltoid, MH

X

Neurogenic, mild

Mod CMAL

Low cerv psp

X

X

–

Obvious CMAL

High thor psp

X

X

–

–

X 1+

LEFT APB

1+

1+

Neurogenic, mild

Mild CMAL

FDI

2+

2+

Neurogenic, mod

Mild CMAL

EI

X

1+

Normal

Mild CMAL

Pron teres

X

1+

Normal

Mild CMAL

BC, MH

X

1+

Neurogenic, mild

Mod CMAL

TC, LH

X

Normal

Mild CMAL

Deltoid, MH

X

Neurogenic, mild

Mod CMAL

X 1+

The presence of fibrillation potentials and mild chronic changes in the extensor indicis muscle allows the polyneuropathy to be characterized as axon loss in nature and involving the sensory and motor nerve fibers. In addition, there is median and ulnar nerve involvement, as well as bilateral C6 involvement. The C5 nerve roots may also be involved, given that the C5,6 muscles (biceps and deltoid) are affected out of proportion to the C6,7 muscles (pronator teres and triceps). The patient was diagnosed with diabetic neuropathic cachexia. Among the various subgroups of diabetic neuropathy, by far the most disabling are those forms involving the nerve roots. When the lumbosacral nerve roots are involved, the terms Bruns-Garland syndrome, diabetic amyotrophy, and, most recently, diabetic

420

Case 1 through Case 50

radiculoplexus neuropathy have been applied. Frequently, there is an associated diabetic polyneuropathy. Rarely (the author has only come across this entity on a single occasion), this entity becomes diffuse, involving the cervical, thoracic, and lumbosacral roots, as well as the somatic plexuses and nerves. This has been termed diabetic neuropathic cachexia and it shows the most marked involvement for the C5 and C6 myotomes of the upper extremities and the L2 through S2 nerve roots of the lower extremities. The lower cervical nerve roots and thoracic nerve roots are also involved, as is the autonomic nervous system. Weight loss typically is profound. In this case, there were also superimposed ulnar neuropathies and carpal tunnel syndrome. With this disorder, the nerve involvement (and associated organ involvement) begins with the complications of end-stage diabetes (i.e., retinal and renal involvement; distal symmetric polyneuropathy; autonomic neuropathy; radiculoplexus neuropathy). This occurs over a subacute period and then shows variable recovery (Riley and Shields, 1984; Wilbourn, 1993).

Exercise 32 A 70-year-old left hand–dominant male was referred for EDX assessment of bilateral upper extremity weakness. According to the patient, about 7 months ago he noted bilateral hand weakness, left more pronounced than right. The weakness progressed proximally to the shoulder girdle muscles on both sides and muscle atrophy followed, severe in degree. He denies bulbar and lower extremity weakness, and there is no sensory loss. On examination, he has severe weakness and muscle wasting distally and proximally, including the shoulder girdle muscles. Because the left upper extremity is slightly weaker than the right upper extremity, the sensory responses are collected from the left upper extremity first. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 32 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.0

18.0

Ulnar-D5

C8

2.5

14.0

Superficial Radial

C6,7

2.3

19.7

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY

The left upper extremity screening sensory responses are normal, consistent with the clinical history of no sensory loss. The motor responses are performed next.

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 32 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.0

18.0

Ulnar-D5

C8

2.5

14.0

Superficial Radial

C6,7

2.3

19.7

4.3

2.9

RIGHT CV

nAUC

LAT

AMP

3.9

4.0

CV

nAUC

SENSORY

MOTOR Median-APB

2.4

50.5

4.0

51.6

421

Section 5: Case Studies in Electrodiagnostic Medicine

(cont.)

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 32 NCS PERFORMED

DRG

Ulnar-ADM

LAT

AMP

2.7

4.3

CV

3.9 Ulnar-FDI

4.3

RIGHT nAUC

LAT

AMP

3.0

5.7

52.8

5.9

4.7

4.1

4.6

CV

nAUC

51.7

5.1

52.4

4.9

55.3

The motor responses are reduced in amplitude, consistent with motor axon loss, and involving the median motor axons to the greatest degree. The pattern of motor response involvement with sensory response sparing is consistent with an intraspinal canal lesion (anterior horn cells; nerve roots) or a lesion involving the peripheral neuromuscular system distal to the sensory branches (e.g., distal motor axons; neuromuscular junction; muscle). Of these, motor neuron disease, as suggested by the history, is most likely. Because both upper extremities are involved, the left lower extremity is added. LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 32 NCS PERFORMED

Stim Site

LAT

AMP

Sural

3.5

5.2

Supfcl Peroneal

3.0

5.6

5.3

6.5

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY

MOTOR Tibial-AH

6.5 Peroneal-EDB

4.5

41.3

3.8 3.6

M wave

5.8

11.5

H wave

34.3

1.2

43.2

The sensory and motor responses collected from the left lower extremity are normal. The needle EMG study is performed next. UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 32

Insertional Activity Normal IPSWs SCP

Spontaneous Activity

Other None

Fibs

Fascs

3+

2+

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

Neurogenic, severe

CMAL, severe

LEFT APB

422

X

Case 1 through Case 50

(cont.)

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 32

Insertional Activity Normal IPSWs SCP

Spontaneous Activity

Other None

Fibs

Fascs

MUAP Analysis

Other

MUAP Recruitment

MUAP Morphology

FDI

X

2+

1+

Neurogenic, mod

CMAL, mod

EI

X

3+

1+

Neurogenic, severe

CMAL, severe

FPL

X

3+

1+

Neurogenic, mod

CMAL, mod

Pron teres

X

3+

3+

Neurogenic, mod

CMAL, mod

BC, MH

X

3+

3+

Neurogenic, severe

CMAL, severe

TC, LH

X

3+

2+

Neurogenic, severe

CMAL, severe

3+

1+

Neurogenic, severe

CMAL, severe

1+

2+

–

–

–

–

Deltoid, MH

1+

Low cerv psp

X

High thor psp

X

X

RIGHT FDI

X

2+

2+

Normal

CMAL, moderate

BC, MH

X

3+

2+

Neurogenic, severe

CMAL, severe

Deltoid, MH

X

2+

3+

Neurogenic, severe

CMAL, severe

The needle EMG findings show generalized fibrillations (ranging in size from high amplitude to low amplitude), the presence of insertional positive sharp waves (left deltoid), generalized fasciculations, and features of advanced chronic motor axon loss with long duration MUAPs and a neurogenic MUAP recruitment pattern. At this point, midthoracic paraspinal muscles and a lower extremity are added, along with tongue assessment. LOWER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 32

Insertional Activity Normal IPSWs SCP

Spontaneous Activity

Other None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

Normal

Normal

LEFT Tongue

X

X

423

Section 5: Case Studies in Electrodiagnostic Medicine

(cont.)

LOWER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 32

Insertional Activity Normal IPSWs

Mid-thor psp X

SCP Other

Spontaneous Activity None

Fibs

Fascs

X

Low thor psp X

2+

2+

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

–

–

–

–

Normal

Normal

FHB

X

0

2+

Normal

Normal

TA

X

0

2+

Normal

Normal

Gastroc, MH

X

Normal

Normal

Vast lateralis

X

Normal

Normal

BF, SH

X

Normal

Normal

Glut medius

X

X 0

1+

X 0

1+

Normal

Normal

1+

0

–

–

X

2+

1+

–

–

FHB

X

0

3+

Normal

Normal

TA

X

0

1+

Normal

Normal

Gastroc, MH

X

Normal

Normal

Vast lateralis

X

Normal

Normal

Low L psp High S psp

2+

RIGHT

X 0

3+

The needle EMG of the lower extremities shows that fasciculation potentials have entered the L4, L5, and S1 myotomes. Thus, by this study, there is involvement of the cervical, thoracic, and lumbosacral levels. In general, with amyotrophic lateral sclerosis, there is a focal onset from which the disease progresses. In this case, the onset site is unclear given that the patient noted bilateral hand weakness at onset. Frequently, fasciculation potentials precede fibrillation potentials, which, in turn, precede chronic changes. As more and more anterior horn cells are lost, the number of fasciculation potentials decreases and the degree of chronic change increases. Because this is a rapidly progressive process, fibrillation potentials of varying size are typically observed (recall that their amplitudes decrease as the muscle fibers atrophy, with high-amplitude fibrillation potentials representing recently denervated muscle fibers and low-amplitude fibrillation potentials representing more prolonged denervation). Thus, the onset site typically shows the most pronounced chronic MUAP changes, and the most recently involved sites show isolated fasciculation potentials. However, this is not invariable, though it is well demonstrated by this case.

Exercise 33 A 74-year-old right hand–dominant male was referred for EDX assessment of an 18-year history of slowly progressive 4-extremity weakness. He was a former triathlon competitor and was in his usual state of excellent health until the age of 56 years. At that time, while still running about 120 miles per week, he developed left lower extremity

424

Case 1 through Case 50

cramping, followed by weakness in that limb. Shortly thereafter, the right lower extremity developed similar symptoms, and then he developed upper extremity cramping with effort (e.g., painting). He noted muscle twitching in the arms and tongue. Due to bilateral foot drop, he started using a walker, and then a wheelchair when his knees started buckling. At the 3-year mark, he noted upper extremity weakness and fasciculations, as well as tongue fasciculations. He had to chew his food much longer in order to swallow it. He denies sensory symptoms. He was referred with a diagnosis of amyotrophic lateral sclerosis. However, based on the slow progression and symmetry of the process, this seemed unlikely. Perioral quivering movements were intermittently noted. The presentation, along with the perioral fasciculations, suggests Kennedy disease (XLR bulbospinal sensorimotor neuronopathy). The sensory NCS are performed first. When the clinical features are symmetric, we typically start on the right upper extremity. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 33 NCS PERFORMED

DRG

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Median-D2

C6,7

3.6

5.3

Ulnar-D5

C8

3.3

4.7

Superficial Radial

C6,7

2.6

6.3

The screening sensory responses are mildly reduced in amplitude. Thus, to determine the distribution and symmetry of the process, the contralateral sensory NCS are performed.

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 33 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.6

Ulnar-D5

C8

Superficial Radial

C6,7

RIGHT CV

nAUC

LAT

AMP

4.3

3.6

5.3

3.2

4.0

3.3

4.7

2.7

5.0

2.6

6.3

CV

nAUC

SENSORY

The same findings are present, indicating sensory axon loss, related to either a sensory polyneuropathy or a sensory neuronopathy. This will be further assessed during the lower extremity NCS. At this point, the upper extremity motor NCS are performed, beginning with the right upper extremity. The left upper extremity will also require study.

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 33 NCS PERFORMED

DRG

LAT

AMP

C6,7

3.6

4.3

RIGHT CV

nAUC

LAT

AMP

3.6

5.3

CV

nAUC

SENSORY Median-D2

425

Section 5: Case Studies in Electrodiagnostic Medicine

(cont.)

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 33 NCS PERFORMED

DRG

LAT

AMP

Ulnar-D5

C8

3.2

Superficial Radial

C6,7

RIGHT CV

nAUC

LAT

AMP

4.0

3.3

4.7

2.7

5.0

2.6

6.3

4.9

2.1

4.9

1.6

CV

nAUC

MOTOR Median-APB

2.0 Ulnar-ADM

3.7

45.7

1.5

2.9

4.3

2.4

50.2

48.7

2.1 2.1

49.2

The motor responses are reduced in amplitude, consistent with generalized motor axon or anterior horn cell loss. The abnormalities are symmetric in severity. The lower extremities are studied next.

LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 33 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Sural

NR

NR

Supfcl Peroneal

NR

MOTOR Tibial-AH

7.1

0.8 0.6

Peroneal-EDB

4.7 33.7

NR

0.6 0.5

4.9

0.4 0.2

Peroneal-TA

6.0

0.7 0.7

M wave H wave

7.0

1.3 NR

6.3 36.5

31.6

0.9 0.9

7.0

35.3

35.8

1.2 NR

The lower extremity sensory responses are absent and the motor responses are severely reduced in amplitude, including the M wave, consistent with generalized sensory and motor axon or cell body loss. The needle EMG study will be helpful to assess the rate of progression of the disorder and to better define its boundaries. At this point, the initial impression of Kennedy disease seems accurate. Many of these patients deny sensory symptoms but demonstrate them on sensory NCS.

426

Case 1 through Case 50

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 33

Insertional Activity Normal

IPSWs

SCP

Spontaneous Activity Other

None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

RIGHT FDI

X

X

Neurogenic, severe

CMAL, severe

EI

X

X

Neurogenic, severe

CMAL, severe

FPL

X

X

Neurogenic, severe

CMAL, severe

Pron teres

X

X

Neurogenic, severe

CMAL, severe

BC, MH

X

X

Neurogenic, severe

CMAL, severe

TC, LH

X

X

Neurogenic, severe

CMAL, severe

Low cerv psp

X

–

–

1+

1+

LEFT FDI

X

X

Neurogenic, severe

CMAL, severe

Pron teres

X

X

Neurogenic, severe

CMAL, severe

Right tongue

X

X

Neurogenic, mod

CMAL, moderate

Left tongue

X

X

Neurogenic, mod

CMAL, moderate

LOWER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 33

Insertional Activity Normal

IPSWs

SCP

Spontaneous Activity Other

None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

RIGHT FHB

X

X

Neurogenic, severe

CMAL, severe

FDL

X

X

Neurogenic, severe

CMAL, severe

TA

X

Neurogenic, severe

CMAL, severe

1+

427

Section 5: Case Studies in Electrodiagnostic Medicine

(cont.)

LOWER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 33

Insertional Activity Normal

IPSWs

SCP

Spontaneous Activity Other

None

Fibs

Fascs

X

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

Neurogenic, severe

Rare MUAP

Neurogenic, severe

CMAL, severe

Gastroc, MH

Decr

Vast lateralis

X

BF, SH

X

X

Neurogenic, severe

CMAL, severe

Glut medius

X

X

Neurogenic, severe

CMAL, severe

Low L psp

X

2+

–

Obvious CMAL

Mid T psp

X

1+

–

–

Rare MUAP

Normal

1+

1+

LEFT FDL

X

X

TA

X

1+

Neurogenic, severe

CMAL, severe

Gastroc, MH

X

1+

No MUAPs

–

Vast lateralis

X

Neurogenic, severe

CMAL, severe

X

The needle EMG showed sparse, low-amplitude fibrillation potentials in some of the studied muscles and neurogenic MUAP recruitment with severe chronic motor axon loss. A small number of fasciculation potentials were noted the paraspinal muscles. Although we do not typically assess the paraspinal muscle MUAPs, when there are obvious chronic changes, we report it. In this case, the relationship between the acute changes (sparse) and the generalized chronic changes (pronounced) is consistent with a slowly progressive process, whereas the involvement of the paraspinal muscles is consistent with an intraspinal canal process, such as a motor neuronopathy. The sensory NCS are consistent with a sensory neuronopathy. Thus, the study supports the clinical impression of Kennedy disease. The patient underwent genetic testing, and this diagnosis was confirmed (41 repeats, consistent with his late onset age).

Exercise 34 A 20-year-old right hand–dominant male is referred for EDX assessment of bilateral lower extremity weakness. According to the patient, he has slowly progressive calf muscle wasting that started about one year ago. It interferes with his ability to climb a ladder and he is having difficulty with jump shots during basketball. He has more recently

428

Case 1 through Case 50

noted that his thighs are thinner (not filling his pants legs as much) and he has lower extremity fatigue from walking just 0.5 miles. His examination was remarkable for bilateral lower extremity weakness and wasting, distal muscles weaker than proximal muscles, and severe gastrocnemius atrophy. The ankle muscle stretch reflexes are absent. The two sides are symmetrically affected. Thus, we will start with sensory NCS of the right lower extremity.

UPPER and LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 34 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Sural

3.9

25.9

Supfcl Peroneal

2.8

13.7

The lower extremity sensory NCS are normal. The screening lower extremity motor responses are collected next.

UPPER and LOWER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 34 NCS PERFORMED

Stim Site

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Sural

3.9

25.9

Supfcl Peroneal

2.8

13.7

Median-D2

2.8

30.4

5.3

24.8

MOTOR Tibial-AH

21.1 Peroneal-EDB

4.3

44.0

12.0 10.7

49.2

H reflex M wave

3.8

6.1

H wave

31.1

1.3

The motor responses are also normal. The M wave is mildly reduced in amplitude, although, when considered in relation to the amplitudes of the tibial and peroneal motor responses, it is probably more than mildly reduced in amplitude. The needle EMG is performed next.

429

Section 5: Case Studies in Electrodiagnostic Medicine

LOWER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 34

Insertional Activity Normal

Spontaneous Activity

IPSWs SCP Other None Fibs

Fascs Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

RIGHT FHB

X

1+

Normal

Normal

FDL

X

2+

Normal

Normal

TA

X

1+

Normal

Normal

Gastroc, MH

X

2+

Early

SDLA, moderate

Vast lateralis

X

1+

Normal

Normal

Rectus femoris

X

1+

Normal

SDLA, mild

Add longus

X

2+

Early

SDLA, moderate

Glut medius

X

Normal

Normal

Iliacus

X

1+

Normal

SDLA, mild

3+

–

–

Low L psp

X

2+

FDI

X

X

Normal

Normal

Brachioradialis

X

X

Normal

SDLA, mild

Biceps

X

1+

Normal

SDLA, mild

Triceps

X

1+

Normal

Normal

Deltoid

X

1+

Normal

Normal

Lower cerv psp

X

1+

–

–

The needle EMG study shows low-density fibrillation potentials in most of the lower extremity muscles and the proximal upper extremity muscles. The primary finding was short-duration, low-amplitude, polyphasic MUAPs that were most pronounced in the lower extremities. Thus, the EDX study is indicative of a myopathy. Clinically, the phenotype (severe plantar flexion weakness and mild knee flexion weakness) suggested Miyoshi myopathy and genetic testing verified that diagnosis.

Exercise 35 The patient is a 59-year-old right hand–dominant male referred for EDX assessment of right-sided neck pain and right upper extremity weakness. According to the patient, about 5 months ago he noted right paracentral neck pain and pain along the posterior aspect of his shoulder blade. At night, when his head is turned rightward on the pillow, he develops painful tingling that runs down his arm to his right hand. He has noted weakness pushing doors open with his right upper extremity. He has noted numbness of the tips of his fingers, especially digit three.

430

Case 1 through Case 50

The radiating neck pain suggests a radiculopathy, and the weakness noted with pushing open doors and the numbness most pronounced at digit three suggest right C7 nerve root involvement. The initial sensory NCS are limited to the screening studies. Nerve Conduction Studies

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 35 NCS PERFORMED

DRG

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Median-D2

C6,7

3.5

29.4

Ulnar-D5

C8

2.8

19.0

Superficial Radial

C6,7

2.4

27.0

The screening sensory NCS are normal, and because the amplitude values are roughly 3 times the lower limit of normal, contralateral comparison sensory NCS are not performed. Routine motor NCS are performed next. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 35 NCS PERFORMED

DRG

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Median-D2

C6,7

3.5

29.4

Ulnar-D5

C8

2.8

19.0

Superficial Radial

C6,7

2.4

27.0

3.5

6.8

MOTOR Median-APB

6.7 Ulnar-ADM

3.1

54.3

7.8 7.5

55.4

The motor responses are normal. Thus, the needle EMG study is performed next, with special attention focused on the right C7 nerve root distribution. Needle EMG Study

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 35

Insertional Activity

Spontaneous Activity

Normal IPSWs SCP Other None

Fibs Fascs Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

RIGHT FDI

X

X

Normal

Normal

431

Section 5: Case Studies in Electrodiagnostic Medicine

(cont.)

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 35

Insertional Activity

Spontaneous Activity

Normal IPSWs SCP Other None

Fibs Fascs Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

EI

X

X

Normal

Normal

FPL

X

X

Normal

Normal

Pron teres

X

Normal

CMAL, moderate

BC, MH

X

Normal

Normal

Neurogenic, mild

CMAL, moderate

Normal

Normal

–

–

TC, LH

2+ X 1+

Deltoid, MH

X

Low cerv psp

X

3+ X

2+

LEFT APB FDI EI Pron teres BC, MH Brachioradialis The needle EMG study shows insertional positive sharp waves and fibrillation potentials in the right C7 myotome (involvement of C6,7 muscles with sparing of C5,6 muscles), as well as chronic changes in the right C7 nerve root distribution. The fibrillation potentials ranged from low amplitude to high amplitude, indicating various degrees of associated muscle fiber atrophy. The presence of fibrillation potentials in the right lower cervical paraspinal muscles localizes the lesion to the intraspinal canal. Thus, overall, the needle EMG identifies a slowly progressive right C7 radiculopathy that has had a recent worsening. EDX Study Conclusion

1. Right C7 Radiculopathy –

The above is axon loss in nature. The particular abnormalities identified indicate that the underlying process is slowly progressive in nature (e.g., spondylosis), and that there has been a recent worsening. The neurogenic MUAP recruitment pattern noted in the lateral head of the right triceps muscle indicates that the process is severe in degree, at least for those nerve fibers.

Exercise 36 A 78-year-old right hand–dominant male is referred for EDX assessment of neck pain. According to the patient, he has a 1-year history of right-sided neck pain. It is non-radiating in nature. He also has bilateral hand numbness,

432

Case 1 through Case 50

especially the right fifth digit, hand tingling upon awakening and when driving, and diminished grip strength. He has a long history of type 2 diabetes mellitus, and his lower extremities are numb to the calf muscle level. The presentation suggests a possible right-sided radiculopathy, right ulnar neuropathy, and bilateral carpal tunnel syndrome. The level of lower extremity sensory involvement (i.e., the calf level) is consistent with possible sensory polyneuropathy in the hands. Thus, this study will need to be fairly extensive to tease out all of these possibilities. Because the neck pain is right-sided and the fifth digit numbness is on the right, the sensory NCS are performed on that limb first. Because of the fifth digit involvement, the DUC and MABC sensory NCS are added to the screening studies. Palmar mixed NCS may also be required, depending on the median sensory response findings. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 36 NCS PERFORMED

DRG

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Median-D2

C6,7

NR

NR

Ulnar-D5

C8

NR

NR

Superficial Radial

C6,7

NR

NR

Median Palmar

NR

DUC

1.9

3.4

1.8

3.0

MABC

2.6

5.1

2.7

5.3

The three screening sensory NCS are absent, as is the median palmar mixed response. Because the latter is absent, there is no reason to perform an ulnar palmar mixed NCS. Involvement of all three sensory responses indicates an axon loss process and suggests a sensory polyneuropathy. At this point, it is unclear whether there are superimposed mononeuropathies (e.g., median or ulnar neuropathies or if it is all polyneuropathy). The DUC response is reduced in amplitude and the MABC response is normal. This still does not clarify whether this is a sensory polyneuropathy (i.e., with a glove distribution to the distal forearm level) or a sensory polyneuropathy with superimposed mononeuropathies. The left upper extremity shows the three screening sensory responses to be absent. The DUC and MABC responses are similar to those recorded on the right side. This is consistent with a symmetric sensory polyneuropathy, axon loss in nature. It will be important to look for asymmetries on the motor NCS to better determine if there are superimposed focal mononeuropathies. Again, we begin on the right side because it is the more symptomatic side. Because of the right fifth digit symptoms, the ulnar motor response, recording FDI, is added bilaterally. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET EXERCISE 36 NCS PERFORMED

LEFT DRG

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Median-D2

C6,7

NR

NR

Ulnar-D5

C8

NR

NR

Superficial Radial

C6,7

NR

NR

Median Palmar

NR

DUC

1.9

3.4

1.8

3.0

MABC

2.6

5.1

2.7

5.3

433

Section 5: Case Studies in Electrodiagnostic Medicine

(cont.)

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 36 NCS PERFORMED

DRG

LAT

AMP

4.7

4.6

RIGHT CV

nAUC

LAT

AMP

5.1

4.1

CV

nAUC

MOTOR Median-APB

4.4 Ulnar-ADM

3.5

45.6

4.8

3.7

4.6

Ulnar-FDI

4.6

3.2

41.3

11.5

1.6

46.4

5.9

11.8

5.3

1.0

47.3

0.9

47.4

1.2

5.7

45.5

0.9

37.9

5.6

46.4

0.8

37.1

The right median motor response is delayed in onset and is reduced in amplitude. The delay is more than expected for the degree of amplitude reduction and, thus, suggests carpal tunnel syndrome on that side. This is also true for the left side. Thus, focal delays indicate superimposed bilateral carpal tunnel syndrome, right worse than left. The four ulnar motor responses are all reduced in amplitude. Although this can be seen with sensorimotor axon loss, the asymmetry could not be explained by a polyneuropathy. Thus, if the polyneuropathy involves the motor axons, there is a superimposed process on the right. The latter could represent a right ulnar neuropathy or a more proximal lesion, such as a C8 radiculopathy (a medial cord or lower trunk lesion makes less sense given the normal MABC). Thus, there are three possibilities to account for the asymmetric low-amplitude ulnar motor responses: (1) a right ulnar neuropathy superimposed on a sensory polyneuropathy; (2) bilateral ulnar neuropathies, right greater than left, superimposed on a sensory polyneuropathy; or (3) bilateral ulnar neuropathies, right greater than left, superimposed on a sensorimotor polyneuropathy. Needle EMG of the C8-radial nerve innervated muscles will be important to tease out these possibilities and additional muscles will need to be sampled on both sides. UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 36

Insertional Activity Normal

Spontaneous Activity

IPSWs SCP Other None

Fibs Fascs Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

RIGHT

434

APB

X

FDI

X

EI

X

FPL

X

Normal

CMAL, severe

Neurogenic, severe

CMAL, severe

X

Normal

CMAL, mild

X

X

Normal

CMAL, mild

Pron teres

X

X

Normal

CMAL, moderate

BC, MH

X

X

Normal

Normal

TC, LH

X

X

Normal

CMAL, mild

Deltoid, MH

X

X

Normal

Normal

2+

Case 1 through Case 50

(cont.)

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 36

Insertional Activity Normal

FDP-3,4

X

Low cerv psp

X

High thor psp

X

APB

X

FDI

X

FDP-3,4

X

EI

X

Pron teres

X

BC, MH TC, MH

Spontaneous Activity

IPSWs SCP Other None

Fibs Fascs Other

X

MUAP Analysis MUAP Recruitment

MUAP Morphology

Neurogenic, mod

CMAL, severe

–

–

X

–

–

X

Neurogenic, severe

CMAL, moderate

Neurogenic, mod

CMAL, severe

Normal

CMAL, mild

Normal

CMAL, moderate

X

Normal

Normal

X

X

Normal

Normal

X

X

Normal

Normal

2+

LEFT

2+ X 1+

Based on the needle EMG study, the following conclusions can be made. 1. Polyneuropathy –

Based on bilateral involvement of the superficial radial sensory responses, it can be concluded that the sensory NCS show a symmetric polyneuropathy that is axon loss in nature. It has a glove distribution to approximately the wrist level (it involves the three screening sensory NCS, it partially involves the DUC sensory responses, and it spares the MABC sensory responses). Whether there is concomitant motor axon involvement is unclear due to #2 below.

2. Bilateral Cervical Radiculopathies –

The distribution of the acute and chronic changes on the needle EMG study indicates bilateral involvement of the right C7 (involvement of C6,7 muscles [pronator teres and triceps] with sparing of the C5,6 muscles [biceps and deltoid]) and bilateral C8 nerve roots (because the extensor indicis muscles are involved on both sides).

3. Bilateral Median Neuropathies –

The prolonged onset latencies of the distal median motor responses indicate superimposed bilateral carpal tunnel syndrome, right worse than left.

4. Possible Ulnar Neuropathies –

Because of the symmetry of the absent ulnar sensory responses and the low-amplitude DUC responses, these changes more likely reflect the previously identified sensory polyneuropathy. On the right side, it is

435

Section 5: Case Studies in Electrodiagnostic Medicine

unclear whether the low-amplitude ulnar motor responses are related to an ulnar neuropathy or the C8 radiculopathy described above. The degree of involvement suggests at least concomitant ulnar nerve involvement (the response amplitudes are unexpectedly low for an isolated C8 radiculopathy and would be expected to involve C8 and T1 to generate this degree of amplitude decrement). Because T1 radiculopathies are uncommon, a concomitant right ulnar neuropathy is suggested. A concomitant left ulnar neuropathy may also be present. Because of the axon loss nature of the ulnar motor responses and #2 above, localization is imprecise for these changes.

Exercise 37 A 25-year-old female was referred for EDX assessment of left shoulder pain followed by upper extremity weakness and wasting. The wasting and weakness was noted about 21 days ago. The shoulder pain was sudden in onset and severe in degree. It awoke her from sleep and persisted for 8 days. As it subsided, she became aware of significant weakness with lifting and overhead activities. Her examination showed left biceps and deltoid muscle wasting, winging of the left scapula, and sensory loss in the cutaneous distributions of the axillary (superior lateral brachial cutaneous nerve) and lateral antebrachial cutaneous nerves. This history is suggestive of neuralgic amyotrophy, a presumable autoimmune disorder that typically presents with severe shoulder pain followed shortly thereafter by weakness and wasting, typically in the distribution of a single nerve (mononeuropathy) or of multiple nerves (multiple mononeuropathy). The weakened muscles suggest involvement of the long thoracic (scapular winging worse with upper extremity elevation anteriorly), axillary (deltoid wasting and overhead weakness), and musculocutaneous (biceps wasting and trouble carrying groceries) nerves. If the lesion is more proximal, it must lie at or proximal to the APR level, because the long thoracic nerve emanates from the C5 through C7 APR elements. Because of this distribution of potential lesion sites, additional sensory NCS to assess the C6 segment of the PNS are included in the initial group of sensory NCS. Also, the contralateral LABC sensory NCS is added, given that the patient has sensory loss in the cutaneous distribution of the ipsilateral LABC nerve. Nerve Conduction Studies

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 37 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

2.8

37.8

Ulnar-D5

C8

2.5

19.9

Superficial Radial

C6,7

2.4

28.1

LABC

C6

2.4

3.5

Median-D1

C6

3.3

26.1

RIGHT CV

nAUC

LAT

AMP

2.5

18.4

CV

nAUC

SENSORY

The three screening sensory NCS are normal, but the added sensory NCS show a low-amplitude LABC response. The latter indicates an axon loss process that could reflect a lesion involving the LABC nerve, musculocutaneous nerve, lateral cord, upper trunk, or C6 APR/DRG. However, in the presence of a normal Median-D1 response, it is very unlikely that the lesion could involve the lateral cord, upper trunk, or C6 APR/DRG (Ferrante and Wilbourn, 1995). This pattern of sensory NCS involvement strongly suggests that the lesion involves either the LABC nerve or the musculocutaneous nerve. As previously stated, there is no sensory NCS to assess the sensory branch of the axillary nerve (the superior lateral brachial cutaneous nerve), and the long thoracic nerve does not have a cutaneous

436

Case 1 through Case 50

branch. Of the two possibilities listed above, an LABC lesion would not be associated with biceps weakness and wasting, suggesting that the lesion involves the musculocutaneous nerve. Based on the sensory NCS and the clinical examination, the Musculocutan-BC and Axillary-Deltoid motor NCS are included in the initial motor NCS.

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 37 NCS PERFORMED

Stim Site

LAT

AMP

Median-D2

2.8

37.8

Ulnar-D5

2.5

19.9

Superficial Radial

2.4

28.1

LABC

2.4

3.5

Median-D1

3.3

26.1

3.5

8.3

RIGHT CV

nAUC

LAT

AMP

2.5

18.4

3.1

6.9

3.4

17.6

CV

nAUC

SENSORY

MOTOR Median-APB

8.3 Ulnar-ADM

2.5

54

10.5 10.4

Musculocutan-BC

3.0

55

3.1 3.0

Axillary-Deltoid

3.6

58

3.8

The low-amplitude musculocutaneous motor response is consistent with an axon loss process involving this nerve, as suspected from the sensory NCS. The very-low-amplitude axillary motor response indicates an axon loss process involving this nerve, as suspected by the obvious weakness and wasting observed on the clinical examination. Because the serratus anterior muscles are actually composed of a number of individual muscle bellies, the long thoracic nerve is a less reliable motor NCS. However, it is easily studied by needle EMG. Thus, in addition to the routine muscles, the serratus anterior is added. Due to the predilection of neuralgic amyotrophy to involve purely or predominantly motor nerves (Ferrante and Wilbourn, 2017), one of the spinati muscles should be studied. Also, in the setting of an axillary neuropathy, it is our practice to include the posterior head of the deltoid. Needle EMG Examination

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 37

Insertional Activity Normal

Spontaneous Activity

IPSWs SCP Other None

Fibs Fascs Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

LEFT FDI

X

X

Normal

Normal

EI

X

X

Normal

Normal

437

Section 5: Case Studies in Electrodiagnostic Medicine

(cont.)

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 37

Insertional Activity Normal

Spontaneous Activity

IPSWs SCP Other None

Fibs Fascs Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

FPL

X

X

Normal

Normal

Pron teres

X

X

Normal

Normal

3+

Mod neurogenic

Normal

3+

Mod neurogenic

Normal

Normal

Normal

3+

Severe neurogenic

Normal

3+

Severe neurogenic

Normal

2+

Normal

Normal

BC, MH

1+

Brachialis

X

TC, LH

X

Deltoid, MH Deltoid, PH

X 1+

X

Serratus ant

1+

Infraspinatus

X

X

Normal

Normal

Low cerv psp

X

X

–

–

High thor psp

X

X

–

–

Deltoid, MH

X

X

Normal

Normal

BC, LH

X

X

Normal

Normal

RIGHT

The needle EMG study shows acute changes in the lateral head of the biceps, brachialis, the middle and posterior heads of the deltoid, and the serratus anterior muscles. Given the degree of severity of involvement of the middle and posterior deltoid heads, the anterior head was not studied. For the deltoid and biceps muscles, the changes are very severe (neurogenic MUAP recruitment). The infraspinatus muscle was normal (the supraspinatus was not studied). The presence of insertional positive sharp waves, which are typically seen in the 14–21-day window, are consistent with the report of the patient of weakness and wasting beginning 21 days earlier. When calculating the age of the lesion, although the pain started 30 days before the study, the weakness and wasting were not apparent until 9 days later. Of course, in this case, the delay may reflect the fact that the patient was favoring the left upper extremity due to the severe pain. Because the abnormalities were most pronounced in the axillary and musculocutaneous nerve distributions, the deltoid and biceps were studied contralaterally so that the durations of the MUAPs could be compared (to look for relative abnormalities). Consistent with the age of the lesion, no changes were noted. Notice that the musculocutaneous motor response is 55% smaller than the contralateral side (i.e., about 50%) and that the LABC sensory response is 81% smaller. In general, when the motor response is 50% lower that the contralateral, the sensory response from the same element is approximately 90% lower or absent.

438

Case 1 through Case 50

EDX Conclusion

The EDX study shows multiple mononeuropathies, axon loss in nature, involving the musculocutaneous (severe in degree), axillary (very severe in degree), and long thoracic nerves. The severity of the long thoracic neuropathy is unclear because this is best determined through motor NCS. The clinical features and the pattern of nerve involvement are indicative of neuralgic amyotrophy. Although the LABC nerve is the most commonly affected sensory nerve (Sumner and England, 1997; Ferrante and Wilbourn, 2017), the relationship between the musculocutaneous motor response amplitude (about 50% lower than the contralateral side) and the LABC sensory response amplitude (about 80% lower than the contralateral side) suggests it reflects the musculocutaneous nerve lesion rather than being a fourth lesion site. Neuralgic Amyotrophy

Although a large number of terms have been applied to this entity, we prefer neuralgic amyotrophy (NA) because it conveys the two quintessential features of this disorder: nerve pain and muscle atrophy. Because NA has a predilection for purely motor or predominantly motor nerves, it frequently involves the suprascapular, long thoracic, axillary, and musculocutaneous nerves (proximally) and motor nerve branches to individual muscles (distally) (Ferrante and Wilbourn, 2017). It also frequently involves the anterior interosseous nerve and extraplexal nerves, such as the spinal accessory nerve. About 50% of patients present with a mononeuropathy and the other 50% present with a multiple mononeuropathy (Ferrante and Wilbourn, 2017). Clinically, abrupt and excruciating shoulder pain, typically with a nocturnal onset, brings the patient to medical attention. Most commonly, the shoulder pain is located laterally or dorsally. Other sites of pain depend on the particular nerves involved – lateral shoulder (axillary nerve), dorsal shoulder (suprascapular nerve), lateral chest wall (long thoracic), antecubital fossa (anterior interosseous nerve), and lateral arm and forearm (musculocutaneous nerve). Shoulder movement intensifies the pain, but neck movement does not (i.e., it is not radicular pain). The shoulder pain usually lasts 1–2 weeks and then resolves or is replaced by a dull ache. Patients usually do not recognize the limb weakness until the pain lessens and they begin to use the limb again. Typically, the weakness is accompanied by profound muscle wasting. In our study of 281 neuralgic amyotrophy patients, a trigger was noted in 75%, pain was present in 88%, and weakness with wasting was noted in 99% (Ferrante and Wilbourn, 2017).

Exercise 38 A 58-year-old female with a 6-week history of left upper extremity pain, tingling, and weakness is referred for EDX testing. These symptoms started 6 weeks ago. The tingling involves the dorsolateral aspect of the left hand. Strength assessment is limited by pain. Her past medical history is remarkable for breast cancer 11 years earlier. She did not receive radiation therapy. Initial sensory NCS are performed. The Nerve Conduction Studies

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 38 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.2

28.0

Ulnar-D5

C8

2.8

18.3

Superficial Radial

C6,7

2.4

4.1

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY

The three screening sensory NCS are remarkable for a very-low-amplitude superficial radial response. Because the superficial radial response assesses axons originating from the C6 and C7 dorsal root ganglia (DRG), the possible lesion localization sites include the superficial radial nerve, the radial nerve, the posterior cord, the upper and middle trunks, and the C6 and C7 APR/DRG (see Appendix 2). Because of sensory response involvement, the lesion cannot

439

Section 5: Case Studies in Electrodiagnostic Medicine

be preganglionic (preganglionic lesions spare the sensory NCS). The other two sensory responses are normal and do not affect this list of possibilities. Whenever either the Superficial Radial or Median-D2 response is abnormal, both of which assess sensory axons derived from the C6 and C7 DRG, the lateral antebrachial cutaneous (LABC) and median sensory, recording first digit (Median-D1) sensory NCS, both of which assess sensory axons derived from the C6 DRG, are added to further narrow the list of possible localization sites (see Appendix 2). UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 38 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.2

28.0

Ulnar-D5

C8

2.8

18.3

Superficial Radial

C6,7

2.4

4.1

LABC

C6

NR

Median-D1

C6

NR

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY

The LABC and Median-D1 sensory responses are both absent. The LABC response assesses the lateral antebrachial cutaneous nerve, the musculocutaneous nerve, the lateral cord, the upper plexus, and the C6 DRG. The Median-D1 response assesses the median nerve, lateral cord, upper plexus, and C6 DRG. When the three abnormal responses are considered simultaneously, the original list can be shortened to the C6 fibers of the upper plexus (i.e., the upper trunk and the C6 APR/DRG). Thus, at this point, the lesion can be characterized as an axon loss upper plexopathy. We can now assess the severity of the lesion using the motor NCS. Because the lesion involves the upper plexus, the initial motor NCS should include the routine motor NCS (median, recording thenar eminence; ulnar, recording hypothenar eminence) plus the axillary, recording deltoid, and the musculocutaneous, recording biceps, motor NCS. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 38 NCS PERFORMED

Stim Site

LAT

AMP

3.6

8.1

RIGHT CV

nAUC

LAT

AMP

3.4

6.0

3.8

9.0

CV

nAUC

MOTOR Median-APB

8.0 Ulnar-ADM

2.9

9.1 8.7

Musculocutan-BC

3.4

53.4

2.5

55.7

2.4 Axillary-Deltoid

3.9

4.0

The routine motor NCS, as expected with an upper plexopathy, are normal, whereas the two added motor NCS are abnormally low in amplitude. The low-amplitude motor responses are consistent with an axon loss upper plexopathy. Because there are motor response abnormalities, the upper plexopathy cannot be restricted to the C6 DRG. Thus, at this point, we have an upper plexopathy that involves the upper trunk or the C6 APR. Because there

440

Case 1 through Case 50

are no sensory NCS to assess the C5 APR, its concomitant involvement cannot be excluded. As stated earlier, the motor NCS are especially helpful for determining the severity of the lesion. Musculocutan-BC

X

Axillary-Deltoid

X

2.5

X

6.0

2.4

6.0

4.0

9.0

The lesion involves approximately 56% of the axillary motor fibers innervating the deltoid muscle: 4=9
100% ¼ 44% ðthe percentage of normal motor axonsÞ 100%  44% ¼ 56% ðthe percentage of abnormal motor axonsÞ The lesion involves approximately 58% of the musculocutaneous motor fibers innervating the biceps muscle: 2:5=6:0
100% ¼ 42% ðthe percentage of normal motor axonsÞ 100%  42% ¼ 58% ðthe percentage of abnormal motor axonsÞ Thus, at this point, we have an axon loss upper plexopathy that is severe in degree. In addition to the routine LUE screening muscles, additional upper plexus muscles are incorporated into the needle EMG. When desired, it is possible to differentiate an upper trunk localization from a C6 APR localization by adding muscles that are innervated via motor axons leaving the C6 APR (serratus anterior) or exiting the upper trunk just after its formation (spinati). The Needle EMG Study

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 38

Insertional Activity Normal

Spontaneous Activity

IPSWs SCP Other None

Fibs Fascs Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

LEFT FDI

X

X

Normal

Normal

Ext indicis

X

X

Normal

Normal

FPL

X

X

Normal

Normal

Pron teres

X

3+

Mild neurogenic

Normal

BC, MH

X

1+

Mild neurogenic

Normal

TC, LH

X

1+

Normal

Normal

2+

Mod neurogenic

Normal

2+

Normal

Normal

Deltoid, MH

1+

Brachioradialis

X

Serratus ant

X

X

Normal

Normal

Infraspinatus

X

X

Normal

Normal

Low cerv psp

X

X

–

–

High thor psp

X

X

–

–

441

Section 5: Case Studies in Electrodiagnostic Medicine

(cont.)

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 38

Insertional Activity Normal

Spontaneous Activity

IPSWs SCP Other None

Fibs Fascs Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

RIGHT BC, MH

X

X

Normal

Normal

Pron teres

X

X

Normal

Normal

Deltoid

X

X

Normal

Normal

The abnormal muscles are all within the muscle domain of the upper plexus (Appendix 4). Sparing of the suprascapular and long thoracic nerve innervated muscles argues against a C6 APR lesion (it cannot exclude fascicular sparing). Thus, the lesion best localizes to the upper trunk. The lack of chronic changes (e.g., longduration MUAPs) indicates that appreciable collateral sprouting has not yet occurred, consistent with the onset of symptoms 6 weeks prior to the study. The EDX Examination Conclusions

The EDX findings indicate a subacute, axon loss, upper plexopathy. The pattern of needle EMG abnormalities favors an upper trunk localization over a C6 APR localization. Attempting to differentiate an upper trunk localization from a C6 APR localization is not mandatory. The infraspinatus and serratus anterior muscles are helpful in this regard.

Exercise 39 A 41-year-old female is referred for EDX assessment of distal left upper extremity sensory and motor abnormalities that followed a fall onto her outstretched arm approximately 6 months ago. Initial sensory NCS are performed. The Nerve Conduction Studies

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 39 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.0

30.8

Ulnar-D5

C8

Superficial Radial

C6,7

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY

NR 2.2

24.6

The screening sensory NCS are remarkable for an absent Ulnar-D5 SNAP. Thus, this is an axon loss lesion involving the ulnar nerve, medial cord, lower trunk, C8 APR, or C8 DRG. The lesion cannot be preganglionic. The other responses are normal and have no effect on this list of possible lesion localizations. Whenever the Ulnar-D5 sensory response, which assesses sensory axons derived from the C8 DRG, is abnormal, we add the MABC sensory

442

Case 1 through Case 50

NCS, which assesses sensory axons derived from the T1 DRG, to better define the localization of the lesion. An abnormal MABC response would exclude the ulnar nerve, C8 APR, and C8 DRG as possible localization sites. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 39 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.0

30.8

Ulnar-D5

C8

NR

Superficial Radial

C6,7

24.6

MABC

T1

NR

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY

The MABC response is absent. Thus, as previously stated, the list of possible lesion localizations is reduced to the medial cord or lower trunk. If the lesion is at the APR or DRG level, it must involve the C8 and T1 fibers. In other words, it cannot be restricted to the C8 APR or DRG (because the MABC is affected and does not traverse these two elements) and it cannot be restricted to the T1 APR or DRG (because the Ulnar-D5 is affected and does not traverse these two elements). The sensory NCS cannot differentiate a medial cord lesion from a lower plexus lesion. For this distinction, the Radial-EI motor NCS is added to the routine motor NCS. The extensor indicis (and the extensor pollicis brevis) is a radial nerve innervated muscle with strong C8 input. The C8-derived motor axons traverse the C8 root and lower trunk (supraclavicular elements) before entering the posterior cord via the posterior division of the lower trunk. Thus, this muscle is spared with medial cord lesions and potentially affected with lower trunk lesions. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 39 NCS PERFORMED

DRG

LAT

AMP

3.0

30.8

RIGHT CV

nAUC

LAT

AMP

3.3

13.6

2.8

13.1

3.9

9.1

3.0

3.4

CV

nAUC

SENSORY Median-D2

C6,7

Ulnar-D5

C8

NR

Superficial Radial

C6,7

24.6

MABC

T1

NR

MOTOR Median-APB

Stim Site 3.5

3.7 3.5

Ulnar-ADM

2.9

4.0 4.0

Ulnar-FDI

4.1

4.5 4.4

Radial-EI

3.3

1.3 1.3

443

Section 5: Case Studies in Electrodiagnostic Medicine

The routine motor responses are both low in amplitude, indicating an axon loss process. These two abnormal responses do not further localize the lesion because they could be affected by a medial cord lesion or a lower plexus lesion. However, the involvement of the Radial-EI motor response excludes a medial cord localization. Thus, the lesion involves the lower plexus, either the lower trunk or both the C8 and T1 APR. The lesion cannot be restricted to the C8 and T1 DRG because DRG lesions do not produce motor axon involvement. To grade the severity of the lesion, the Ulnar-FDI motor NCS is added along with the contralateral distal motor responses. It can be concluded that the process is axon loss in nature and that it involves approximately 63% of the motor axons to the thenar eminence (3.7/13.6 = 0.37), 70% of the motor axons to the hypothenar eminence (4.0/13.1 = 0.30), 50% of the motor axons to the FDI muscle (4.5/9.1 = 0.50), and 62% of the motor axons to the EI muscle (1.3/3.4 = 0.38). The initial calculations indicate the percentage of functional motor axons. When this value is subtracted from 100%, the percentage of nonfunctioning axons is defined. To better define the lesion, the needle EMG study is expanded to include more muscles from the lower plexus muscle domain. UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 39

Insertional Activity Normal

IPSWs

Spontaneous Activity

SCP Other None Fibs

Fascs Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

LEFT APB

Few

3+

Severe neurogenic

FDI

X

3+

Mod neurogenic

Ext indicis

X

2+

Mod neurogenic

Normal

FPL

X

2+

Normal

Normal

Pron teres

X

X

Normal

Normal

BC, MH

X

X

Normal

Normal

TC, LH

X

Normal

Normal

Deltoid

X

Normal

Normal

FDP-3,4

X

Mild neurogenic

Normal

Low cerv psp

X

X

–

–

High thor psp

X

X

–

–

APB

X

X

Normal

FDI

X

X

Normal

Ext indicis

X

X

Normal

Few X 2+

RIGHT

444

Case 1 through Case 50

The abnormal muscles are in the muscle domain of the lower plexus. Involvement of the extensor indicis is helpful in differentiating a medial cord lesion from a lower plexus lesion. When this muscle is involved, it excludes a medial cord lesion because the motor axons innervating the extensor indicis muscle traverse the posterior cord, not the medial cord. Importantly, however, the converse of this statement – when the extensor indicis is spared, the lesion localizes to the medial cord – is not true, because the lower plexus lesion could be incomplete (fascicular sparing). Thus, although the extensor indicis muscle has localizing utility when it is involved, it has no localizing potential when it is spared. EDX Study Conclusion

The EDX study localizes the lesion to the lower plexus and characterizes it as axon loss in nature. It involves the sensory and motor nerve fibers and is severe to very severe in degree. The presence of chronic changes (longduration MUAPs) is consistent with the 6-month time frame, and the presence of a large number of fibrillation potentials indicates that further reinnervation via collateral sprouting is possible.

Exercise 40 A 26-year-old female is referred for EDX assessment of intermittent numbness along the medial aspect of her left forearm and hand whenever she assumes the supine position. She also reports a 10-year history of aching along the medial aspect of her left arm and forearm. Clinically, her symptoms involve the C8 and T1 cutaneous distributions, suggesting that the lesion lies between the C8/T1 DRG, proximally, and the medial antebrachial cutaneous and ulnar nerves, distally. The screening sensory NCS are performed. Thus, regarding the sensory NCS, the Ulnar-D5 response is particularly important. Nerve Conduction Studies

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 40 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

2.9

57.0

Ulnar-D5

C8

2.5

16.1

Superficial Radial

C6,7

2.3

48.4

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY

The screening sensory NCS are normal by absolute criteria (see Appendix 6). However, the Ulnar-D5 response seems low in comparison to the other responses. It is helpful to assess the amplitude values in relation to the lower limit of normal for the response. Using this approach, the Ulnar-D5 response is about 1.25 times the lower limit of normal (12.0), whereas the other two responses are nearly three times the value of their lower limits of normal. For this reason, the contralateral Ulnar-D5 response is added to look for a relative abnormality. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 40 NCS PERFORMED

DRG

LAT

AMP

C6,7

2.9

57.0

RIGHT CV

nAUC

LAT

AMP

2.6

41.3

CV

nAUC

SENSORY Median-D2 Ulnar-D5

C8

2.5

16.1

Superficial Radial

C6,7

2.3

48.4

445

Section 5: Case Studies in Electrodiagnostic Medicine

The value of the Ulnar-D5 response is less than 50% of that recorded from the contralateral side, indicating a relative abnormality. Although not as severe as an absolute abnormality, it is nonetheless abnormal. Because the Ulnar-D5 response is abnormal, the lesion must involve the ulnar nerve, medial cord, lower trunk, C8 APR, or C8 DRG. It cannot be preganglionic. Again, to better localize the lesion, whenever the Ulnar-D5 response is abnormal (assesses sensory axons emanating from the C8 DRG and traversing the ulnar nerve), we add the MABC sensory NCS (assesses sensory axons emanating from the T1 DRG and traversing the medial antebrachial cutaneous nerve). UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 40 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

2.9

57.0

Ulnar-D5

C8

2.5

16.1

Superficial Radial

C6,7

2.3

48.4

MABC

T1

RIGHT CV

nAUC

LAT

AMP

2.6

41.3

2.4

15.1

CV

nAUC

SENSORY

NR

The MABC response is absent. To exclude a technical reason, the contralateral MABC sensory NCS is also performed. Because the MABC is abnormal, the possibility of an ulnar nerve localization is eliminated. Thus, at this point, we can conclude that the lesion is axon loss in nature and involves the medial cord, the lower trunk, or the C8 and T1 APR/DRG. Because the MABC response is much more affected than the Ulnar-D5 response is (absent versus relatively abnormal), it indicates that the T1 fibers are more involved than the C8 fibers are. This suggests that the lesion might be affecting the C8 and T1 fibers from below (medial cord or lower trunk) or, if it is at the APR/DRG level, that it affects the T1 APR/DRG more than it does the C8 APR/DRG. Again, with this localization differential, the Radial-EI motor NCS is added to the routine motor NCS. In addition, because the Ulnar-D5 response is abnormal, the Ulnar-FDI motor NCS is added (to better define lesion severity). The contralateral distal motor responses of the affected motor NCS are added to semiquantify the degree of axon loss. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 40 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

2.9

57.0

Ulnar-D5

C8

2.5

16.1

Superficial Radial

C6,7

2.3

48.4

MABC

T1

RIGHT CV

nAUC

LAT

AMP

NR

2.4

15.1

1.5

3.1

12.1

2.6

14.2

3.6

15.1

SENSORY

MOTOR Median-APB

3.2

1.4 Ulnar-ADM

2.5

12.3 12.1

Ulnar-FDI

3.7

54

10.7 10.6

446

47

53

CV

nAUC

Case 1 through Case 50

(cont.)

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 40 NCS PERFORMED

DRG

Radial-EI

LAT

AMP

2.9

2.0

RIGHT CV

2.0

nAUC

LAT

AMP

2.8

4.2

CV

nAUC

52

The Median-APB response is extremely reduced in amplitude, and the Radial-EI response is mild to moderately reduced in amplitude. The ulnar motor responses are asymmetric but do not meet the criteria for absolute or relative abnormal. The presence of motor response abnormalities excludes a DRG localization. The abnormal RadialEI response excludes a medial cord localization. Because the distal motor responses are reduced in amplitude, this is an axon loss process. Consequently, at this point in the EDX study, we know that the lesion is an axon loss process, that it is localized to the lower trunk or to both the C8 and T1 APR, and that it is very severe in degree. It is important to notice the pattern of motor NCS involvement. The Median-APB (T1 > C8 innervation) is very severely affected, the Radial-EI response (C8 innervation) is mildly to moderately affected, and the Ulnar-ADM and Ulnar-FDI responses (more balanced C8 and T1 innervation) are spared. This indicates greater T1 involvement, which was also indicated by the sensory NCS. This pattern of T1 > C8 fiber involvement raises the possibility of true neurogenic thoracic outlet syndrome. Based on the NCS findings, the needle EMG is expanded to better evaluate the lower plexus. UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE C

Insertional Activity Normal

Spontaneous Activity

IPSWs SCP Other None

Fibs

Fascs Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

LEFT APB

X

2+

Severe neurogenic

Severe CMAL

FDI

X

1+

Mod neurogenic

Mod CMAL

EI

X

1+

Mod neurogenic

Mod CMAL

FPL

X

X

Mod neurogenic

Mod-severe CMAL

Pron teres

X

X

Normal

Normal

BC, MH

X

X

Normal

Normal

TC, LH

X

X

Normal

Normal

Deltoid, MH

X

X

Normal

Normal

Low cerv paraspinal

X

X

–

–

High thor paraspinal

X

X

–

–

447

Section 5: Case Studies in Electrodiagnostic Medicine

(cont.)

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE C

Insertional Activity Normal

Spontaneous Activity

IPSWs SCP Other None

Fibs

Fascs Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

RIGHT APB

X

X

Normal

Normal

FDI

X

X

Normal

Normal

The needle EMG abnormalities are confined to muscles in the lower plexus muscle domain. Involvement of the extensor indicis muscle serves to exclude a medial cord lesion (which was already excluded by the motor NCS). The majority of the visualized fibrillation potentials were very low in amplitude, indicating chronicity. A chronic lesion is confirmed by the presence of the chronic changes, which also are in a lower plexus distribution. Similar to the sensory and motor NCS, the needle EMG shows that the T1 motor axons are more affected than the C8 motor axons (the APB muscle is heavily T1 innervated and is the most severely affected muscle). The relationship between the acute changes (not too pronounced) and the chronic changes (quite pronounced) suggests a slowly progressive process. EDX Conclusion

The EDX study localizes the lesion to the lower plexus and characterizes it as axon loss in nature and involving the sensory and motor nerve fibers. It further indicates that the process is severe in degree, is slowly progressive, and involves the T1 fibers to a greater extent than the C8 fibers. These EDX features, when combined with the clinical features, strongly support true neurogenic thoracic outlet syndrome as the underlying etiology.

Exercise 41 A 71-year-old male is referred for EDX assessment of weakness and numbness of the left hand. The weakness involves gripping, and the numbness involves the medial aspect of the left hand and digits. These symptoms were immediately apparent following open heart surgery 10 weeks ago. The surgical procedure utilized a median sternotomy to gain access. The referral diagnosis noted on the EDX consult is postoperative ulnar neuropathy. The lack of sensory involvement along the medial aspect of the forearm and the isolated grip weakness are consistent with that consideration. In addition, there is sensory involvement in the cutaneous distribution of the dorsal ulnar cutaneous nerve. In addition to the screening sensory NCS, because the clinical features suggest an ulnar neuropathy, the DUC is added to the initial studies. If our pre-study assessment had generated a different history and examination, we would still include the DUC sensory NCS because the referring physician suggested an ulnar neuropathy. The clinical considerations of the referring provider should always be addressed by the EDX study. In this case, an ulnar neuropathy seems likely. Nerve Conduction Studies

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 41 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.3

14.3

Ulnar-D5

C8

2.9

NR

Superficial Radial

C6,7

2.3

18.7

SENSORY

448

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

Case 1 through Case 50

(cont.)

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 41 NCS PERFORMED

DRG

DUC

C8

LAT

AMP

RIGHT CV

nAUC

NR

LAT

AMP

X

6.1

CV

nAUC

The initial sensory NCS are remarkable for absent Ulnar-D5 and DUC responses. Thus, this is an axon loss process that involves the ulnar nerve, medial cord, lower trunk, C8 APR, or C8 DRG. Because of the abnormal DUC response, if the ulnar nerve is involved, it must be located proximal to the takeoff site of the dorsal ulnar cutaneous nerve branch (i.e., the lesion is proximal to the wrist). Again, whenever the Ulnar-D5 response is absent, we add the MABC sensory NCS to assist with lesion localization.

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 41 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.3

14.3

Ulnar-D5

C8

2.9

NR

Superficial Radial

C6,7

2.3

18.7

DUC

C8

MABC

T1

RIGHT CV

nAUC

LAT

AMP

NR

3.0

6.1

11.2

2.6

12.1

CV

nAUC

SENSORY

2.5

The MABC response is normal. Because the DUC response was absent on the symptomatic side, it was performed on the asymptomatic side. The MABC response was normal. It was assessed on the contralateral side to make sure that it was not relatively abnormal (unlikely given its value of 11.2 microV). The normal MABC response lessens the likelihood of a medial cord or lower trunk lesion, given that the ulnar sensory responses are absent and the MABC sensory response is normal (a partial lesion at one of these two sites cannot be excluded with certainty due to possible fascicular sparing). Thus, this most likely represents an ulnar neuropathy proximal to the wrist or a C8 APR or DRG lesion and less likely a partial medial cord or lower trunk lesion. Regardless of localization, it is an axon loss lesion. Because the medial cord and lower trunk are potential lesion sites, the Radial-EI is added to the routine motor NCS. The Ulnar-FDI is also added to better assess severity. The distal motor responses of any low-amplitude motor responses will also be collected for lesion severity assessment.

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 41 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.3

14.3

Ulnar-D5

C8

2.9

18.7

Superficial Radial

C6,7

2.3

NR

DUC

C8

RIGHT CV

nAUC

LAT

AMP

3.0

6.1

CV

nAUC

SENSORY

NR

449

Section 5: Case Studies in Electrodiagnostic Medicine

(cont.)

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 41 NCS PERFORMED

DRG

LAT

AMP

MABC

T1

2.5

3.6

RIGHT CV

nAUC

LAT

AMP

11.2

2.6

12.1

7.7

3.6

8.9

2.7

10.1

3.7

9.1

3.1

2.3

CV

nAUC

MOTOR Median-APB

7.7 Ulnar-ADM

2.8

Ulnar-FDI

4.2

Radial-EI

3.3

52

3.5 3.4

54

3.4

56

2.0 1.9

54

1.8

54

0.8 0.8

56

Both ulnar motor responses and the radial motor response are reduced in amplitude, consistent with axon loss. The amplitude values of the abnormal responses indicate that the lesion is severe in degree (the conduction velocity values were normal and marked with an X). The abnormal Radial-EI response excludes an ulnar neuropathy, a medial cord lesion, and a ganglionic process. Thus, at this point in the EDX study, the lesion localizes to the lower plexus and is either a partial lower trunk lesion or a C8 APR lesion. Given the severity of the motor response involvement, a C8 APR localization is favored, because when the motor responses are more than 50% down, the sensory responses are typically 90% down or absent. Thus, MABC sparing favors the C8 APR location. This pattern of NCS abnormalities suggests a median sternotomy brachial plexopathy, which is associated with surgical procedures requiring a median sternotomy. The NEE is expanded to better assess the lower plexus. UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 41

Insertional Activity

Spontaneous Activity

Normal IPSWs SCP Other None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

LEFT

450

APB

X

FDI

X

EI

X

Normal

Normal

2+

Normal

Mild CMAL

X

1+

Normal

Mild CMAL

FPL

X

1+

Normal

Normal

Pron teres

X

X

Normal

Normal

BC, MH

X

X

Normal

Normal

TC, LH

X

X

Normal

Normal

Case 1 through Case 50

(cont.)

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 41

Insertional Activity

Spontaneous Activity

Normal IPSWs SCP Other None

Deltoid, MH

X

FDP-3,4

X

Low cerv psp

X

X

High thor psp

X

X

APB

X

FDI

Fibs

X

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

Normal

Normal

Normal

Mild CMAL

X

Normal

Normal

X

X

Normal

Normal

EIP

X

X

Normal

Normal

FDP-3,4

X

X

Normal

Normal

1+

RIGHT

The needle EMG abnormalities are in the muscle domain of the lower plexus. The pattern of abnormalities is most consistent with a C8 APR localization. EDX Conclusion

The EDX study localizes the lesion to the lower plexus and indicates that it is axon loss in nature, that it involves the sensory and motor nerve fibers, and that it is severe in degree. The pattern of abnormalities is more consistent with a lesion involving the C8 APR. The explanation of C8 APR involvement following median sternotomy is C8 APR trauma related to first rib movement during the median sternotomy. Typically, the pathophysiology of median sternotomy brachial plexopathy is demyelinating conduction block more pronounced than axon loss, and hence, the prognosis for recovery is good. Had the EDX provider assumed that the patient had a postoperative ulnar neuropathy and performed a limited study to confirm the clinical impression, the lesion might have been misdiagnosed as an ulnar neuropathy at the elbow segment. This exercise illustrates the importance of surrounding the abnormal findings with normal ones. When one of the ulnar sensory responses is abnormal, add the MABC sensory NCS. When the differential diagnosis includes the ulnar nerve and the C8 APR, always include C8 radial motor axon assessments (i.e., Radial-EI on NCS and EI and extensor pollicis brevis on needle EMG) and C8 median motor axon assessment (flexor pollicis longus on needle EMG). The author has seen one case in which the EDX provider (like the referring health care provider), suspecting a postoperative ulnar neuropathy at the elbow segment, performed an abbreviated EDX assessment (only median and ulnar sensory and motor NCS were performed) and a limited needle EMG study. The latter led to the erroneous conclusion that the patient had an ulnar neuropathy at the elbow segment, which ultimately led to an inappropriate ulnar transposition. Because the conclusion of the EDX study was erroneous and because the error resulted in a negative outcome (claw hand due to operative insult), the patient sought legal advice and litigation was initiated. Even if the EDX provider had not been familiar with postmedian sternotomy brachial plexopathy, by following the basic rule of surrounding the abnormal responses with normal ones, the correct localization would have been identified and a postoperative ulnar neuropathy excluded. This approach is analogous to the clinical examination.

451

Section 5: Case Studies in Electrodiagnostic Medicine

When there is sensory loss involving the fifth digit, the medial aspect of the fourth digit (superficial terminal branch), the dorsomedial aspect of the hand (DUC branch), and the medial aspect of the forearm (MABC nerve) are assessed for sensory loss. And when there is ulnar nerve distribution weakness, the flexor pollicis longus (C8-median) and the extensor indicis (C8-radial) muscles are assessed. These same assessments must also be performed during the EDX study. The EDX examination is the electrical version of the clinical examination.

Exercise 42 A 56-year-old female is referred for EDX assessment of left-sided biceps weakness and pain and numbness along the lateral aspect of the forearm following pacemaker placement 11 weeks prior to the EDX study. The pacemaker was placed using an infraclavicular portal of entry. Thus, the history suggests an infraclavicular process. The sensory symptoms are in an LABC distribution, and the isolated biceps weakness suggests musculocutaneous nerve involvement, which would also account for the sensory abnormalities. Because the screening sensory NCS poorly assess the C6 DRG-derived sensory axons and because the screening motor NCS do not assess the C5 or C6 motor axons, additional sensory and motor NCS are required. In addition to the three screening sensory NCS, the LABC and Median-D1 sensory NCS should be included in the first group of sensory NCS. Nerve Conduction Studies

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 42 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.2

14.7

Ulnar-D5

C8

2.8

22.7

Superficial Radial

C6,7

2.5

32.0

LABC

C6

2.6

5.7

Median-D1

C6

3.2

12.1

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY

The screening sensory NCS show a low-amplitude Median-D2 sensory response. This indicates that the lesion involves the median nerve, lateral cord, upper or middle trunk, or the C6 or C7 APR/DRG. Of these possibilities, because the symptoms followed the infraclavicular placement of a pacemaker, an infraclavicular localization is more likely (from the list of potential localizations, only the lateral cord is infraclavicular). When the Median-D2 response is abnormal, the LABC and Median-D1 sensory NCS are added. Thus, had the need for the LABC and Median-D1 sensory NCS not been recognized from the history, the screening sensory NCS results would have indicated their addition. The LABC response appears to be low in amplitude (some individuals with large-girth forearms have lower than expected LABC response amplitudes). Thus, this sensory NCS should be performed bilaterally when it is done. The Median-D1 amplitude is usually at least two-thirds the size of the Median-D2 amplitude, but because the Median-D2 response is abnormal, this comparison cannot be used. The Median-D1 response is lower in amplitude than the Ulnar-D5 response, suggesting that it is indeed involved. For diagnostic clarification, the LABC and MedianD1 sensory NCS should be performed contralaterally. To better assess the degree of Median-D2 decrement, it should also be performed contralaterally.

452

Case 1 through Case 50

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 42 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.2

14.7

Ulnar-D5

C8

2.8

22.7

Superficial Radial

C6,7

2.5

32.0

RIGHT CV

nAUC

LAT

AMP

3.1

32.3

CV

nAUC

SENSORY

LABC

C6

2.6

5.7

2.5

16.3

Median-D1

C6

3.2

12.1

3.2

28.4

The contralateral LABC, Median-D1, and Median-D2 sensory responses are normal in amplitude and more than twice the amplitude of the ipsilateral responses. Thus, this is an axon loss process. Of the original list of localization possibilities, the median nerve, musculocutaneous nerve, and the C7 APR/DRG can be eliminated, leaving the following possibilities: the lateral cord, the upper trunk, or the C6 APR/DRG. A subtler clue is that the severity of Median-D1 and Median-D2 response involvement is similar. With supraclavicular lesions involving the C6 fibers, the Median-D1 response is frequently affected out of proportion to the Median-D2 response (80% of the time, this response is spared; see Appendix 2). To distinguish among these possibilities, the Axillary-Deltoid and Musculocutaneous-BC motor NCS are added to the routine motor NCS. Although an upper trunk or C6 APR/DRG lesion can affect both of these motor NCS, a lateral cord lesion can only affect the Musculocutan-BC response (because the Axillary-Deltoid assesses the posterior cord). UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 42 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.2

14.7

Ulnar-D5

C8

2.8

22.7

Superficial Radial

C6,7

2.5

32.0

LABC

C6

2.6

Median-D1

C6

RIGHT CV

nAUC

LAT

AMP

3.1

32.3

5.7

2.5

16.3

3.2

12.1

3.2

28.4

3.6

6.1

CV

nAUC

SENSORY

MOTOR Median-APB

6.1 Ulnar-ADM

3.0

54

7.4 7.2

56

Axillary-Deltoid

4.5

9.2

4.5

8.6

Musculocutan-BC

3.3

2.0

3.1

4.2

2.0

56

453

Section 5: Case Studies in Electrodiagnostic Medicine

Involvement of the Musculocutan-BC response with sparing of the Axillary-Deltoid response supports a lateral cord localization. The needle EMG examination is expanded to confirm this localization. It should include suprascapular, axillary, and radial nerve innerved C5,6 muscles, which would be spared with a lateral cord process. UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 42

Insertional Activity

Spontaneous Activity

Normal IPSWs SCP Other None

Fibs Fascs Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

LEFT APB

X

X

Normal

Normal

FDI

X

X

Normal

Normal

EI

X

X

Normal

Normal

FPL

X

X

Normal

Normal

Mild Neurogenic

Normal

Mod Neurogenic

Normal

Pron teres

2+

3+

BC, MH TC, LH

X

X

Normal

Normal

Deltoid, MH

X

X

Normal

Normal

FCR

X

Normal

Normal

Infraspinatus

X

X

Normal

Normal

Brachioradialis

X

X

Normal

Normal

Low cerv psp

X

X

–

–

High thor psp

X

X

–

–

Pron teres

X

X

Normal

Normal

BC, MH

X

X

Normal

Normal

2+

RIGHT

The needle EMG abnormalities are confined to the muscle domain of the lateral cord. The C5,6 suprascapular (infraspinatus), axillary (deltoid), and radial (brachioradialis) muscles are unaffected, further confirming the lateral cord localization. EDX Conclusion

The EDX study localizes the lesion to the lateral cord and characterizes it as axon loss in nature, involving the sensory and motor nerve fibers, and moderate in degree (based on the musculocutaneous motor response comparison). The density of fibrillation potentials is not used to assess severity (see Chapter 14), and the degree of increment of MUAP duration tends to underestimate the lesion.

454

Case 1 through Case 50

Exercise 43 A 31-year-old male with a 12-year history of slowly worsening left wrist drop is referred for EDX assessment. Further history and examination findings are intentionally not included in this exercise. The screening sensory NCS are performed. Nerve Conduction Studies

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 43 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

2.7

32.7

Ulnar-D5

C8

2.3

22.1

Superficial Radial

C6,7

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY

NR

The screening sensory NCS are remarkable for an absent Superficial Radial response. Thus, this is an axon loss process that is localized to the superficial radial nerve, radial nerve, posterior cord, upper or middle trunk, or the C6 or C7 APR/DRG. Whenever one of the C6,7 sensory responses is abnormal, we add the LABC and Median-D1 sensory NCS bilaterally.

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 43 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

2.7

32.7

Ulnar-D5

C8

2.3

22.1

Superficial Radial

C6,7

LABC

C6

2.5

26.2

Median-D1

C6

3.0

25.7

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY

NR

The LABC and Median-D1 responses are normal. Given their high amplitude, the possibility of these values representing a relative abnormality is minimal. Thus, their performance on the contralateral side was not indicated. Of the initial list or potential lesion localizations, these findings exclude the possibility of an upper trunk or C6 APR/DRG localization because the sensory axons subserving the LABC and Median-D1 sensory NCS emanate from the C6 DRG in 100% of cases (Ferrante and Wilbourn, 1995). The normal Median-D2 response argues against a middle trunk or C7 APR/DRG localization but does not exclude these sites. Thus, the lesion best localizes to the superficial radial nerve, radial nerve, or posterior cord and, less likely, to the middle trunk or the C7 APR/DRG. To further localize the lesion, the Axillary-Deltoid, Radial-ED, and Radial-EI motor NCS are added to the routine motor NCS.

455

Section 5: Case Studies in Electrodiagnostic Medicine

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 43 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

2.7

32.7

Ulnar-D5

C8

2.3

22.1

Superficial Radial

C6,7

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY

NR

LABC

C6

2.5

26.2

Median-D1

C6

3.0

25.7

3.4

18.4

MOTOR Median-APB

18.4 Ulnar-ADM

2.9

X

10.1 10.0

X

Radial-ED

X

1.6

X

10.3

Axillary-Deltoid

X

0.3

X

15.7

The routine motor NCS are normal, as expected. The axillary and radial motor responses are severely reduced in amplitude. To better assess severity, these studies were performed on the contralateral side. Involvement of the axillary motor response excludes radial nerve, middle trunk, and C7 APR/DRG from the original list of potential lesion localizations. Thus, at this point in the EDX study, the lesion localizes to the posterior cord, although a lesion involving both the axillary and radial nerves near the bifurcation site cannot be excluded. It is characterized as axon loss and very severe in degree. The needle EMG study is expanded to include more muscles of the posterior cord muscle domain. Needle EMG Examination

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 43

Insertional Activity Normal

Spontaneous Activity

IPSWs SCP Other None

Fibs Fascs Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

Normal

Normal

Severe Neurogenic

Severe CMAL

LEFT

456

FDI

X

X

EI

Decr

FPL

X

X

Normal

Normal

Pron teres

X

X

Normal

Normal

BC, MH

X

X

Normal

Normal

TC, LH

Decr

1+

Severe Neurogenic

Severe CMAL

Brachioradialis

Decr

1+

None fire

2+

Case 1 through Case 50

(cont.)

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 43

Insertional Activity Normal

Spontaneous Activity

IPSWs SCP Other None

MUAP Analysis

Fibs Fascs Other

MUAP Recruitment

MUAP Morphology

1+

Severe Neurogenic

Severe CMAL

Deltoid, MH

Decr

FCR

X

X

Normal

Normal

Low cerv psp

X

X

–

–

High thor psp

X

X

–

–

Deltoid

X

X

Normal

Normal

EI

X

X

Normal

Normal

RIGHT

The contralateral deltoid and EI muscles were added to compare the MUAP durations on the two sides to better characterize them. The FCR muscle was added because it is a C6,7 median nerve innervated muscle. Thus, it receives motor axons via the middle trunk. Sparing of the FCR and pronator teres muscles therefore argues against middle trunk involvement, as was suggested by the spared Median-D2 sensory response (traverses the middle trunk 80% of the time; see Appendix 2). EDX Conclusion

The EDX study localizes the lesion to the posterior cord and characterizes it as axon loss in nature. It involves the sensory and motor axons and is extremely severe in degree.

Exercise 44 A 32-year-old female with a 10-year history of slowly progressive left-hand atrophy and recent onset axillary pain radiating to the hand was referred for EDX assessment. Further clinical details are intentionally not included. The distribution of the wasting indicates involvement of C8,T1-derived elements (i.e., C8/T1 AHCs, APR, and roots; lower trunk; medial cord; ulnar and median nerves). Axillary pain suggests an infraclavicular process (the distribution of the pain has lesser localizing value than the distribution of weakness). Routine screening studies are performed first. Nerve Conduction Studies

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 44 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.2

38.3

Ulnar-D5

C8

2.7

4.7

Superficial Radial

C6,7

2.4

40.1

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY

457

Section 5: Case Studies in Electrodiagnostic Medicine

On screening sensory NCS, the Ulnar-D5 response is low in amplitude, consistent with an axon loss process involving the ulnar nerve, medial cord, lower plexus, or C8 DRG. Because the Ulnar-D5 response is reduced in amplitude, the MABC NCS is added.

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 44 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.2

38.3

Ulnar-D5

C8

2.7

4.7

Superficial Radial

C6,7

2.4

40.1

MABC

T1

RIGHT CV

nAUC

LAT

AMP

2.8

28.7

2.6

20.6

CV

nAUC

SENSORY

NR

The MABC response is absent. Thus, the ulnar nerve, C8 APR, and C8 DRG are eliminated as potential sites of lesion localization. Thus, potential lesion localization sites include the medial cord or lower trunk, although a fascicular lesion involving the C8 and T1 APR or the C8 and T1 DRG could also account for these NCS findings. For comparison purposes, the contralateral Ulnar-D5 NCS and MABC are also performed. To better localize the lesion (infraclavicular versus supraclavicular), the Radial-EI motor NCS is added to the routine motor NCS. Because the Ulnar-D5 sensory response is low in amplitude, the Ulnar-FDI motor NCS is also added (to better define the severity of the lesion). UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 44 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.2

38.3

Ulnar-D5

C8

2.7

4.7

Superficial Radial

C6,7

2.4

40.1

MABC

T1

RIGHT CV

nAUC

LAT

AMP

2.8

28.7

NR

2.6

20.6

0.6

3.5

16.7

SENSORY

MOTOR Median-APB

3.8

0.5 Ulnar-ADM

2.9

3.5 3.5

Ulnar-FDI

3.8

47.8

8.1

16.7 3.0

14.2

3.9

14.4

2.8

7.9

51

8.1 Radial-EI

2.8

8.1 8.0

458

CV

nAUC

Case 1 through Case 50

On motor NCS, the Median-APB response is extremely low in amplitude, the Ulnar-ADM response is severely reduced in amplitude, the Ulnar-FDI response is normal (but 44% lower than the contralateral side), and the RadialEI response is normal. At this point, it cannot be determined whether the lesion is infraclavicular or supraclavicular, because a normal Radial-EI motor response cannot be used to “rule in” a medial cord lesion or to exclude a lower trunk lesion because it may represent a partial lower trunk lesion that is sparing the C8-radial nerve fibers. Because this is a slowly progressive process, the needle EMG study is more sensitive (reinnervation produces long-duration MUAPs) than motor NCS (reinnervation normalizes the CMAPs). Thus, the C8-radial muscles should be carefully assessed. If the extensor indicis is normal, the extensor pollicis brevis should be added and at least one of these muscles should be studied on the contralateral side. Needle EMG Examination

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 44

Insertional Activity Normal

Spontaneous Activity

IPSWs SCP Other None

Fibs

MUAP Analysis

Fascs Other

MUAP Recruitment

MUAP Morphology

LEFT APB

X

3+

Severe neurogenic

Severe CMAL

FDI

X

2+

Mod neurogenic

Severe CMAL

EI

X

Normal

Normal

FPL

X

Mod neurogenic

Severe CMAL

Pron teres

X

X

Normal

Normal

BC, MH

X

X

Normal

Normal

TC, LH

X

X

Normal

Normal

Deltoid, MH

X

X

Normal

Normal

FDP-3,4

X

Mod neurogenic

Moderate CMAL

EPB

X

X

Normal

Normal

Low cerv psp

X

X

–

–

High thor psp

X

X

–

–

APB

X

X

Normal

Normal

FDI

X

X

Normal

Normal

EI

X

X

Normal

Normal

X 2+

2+

RIGHT

459

Section 5: Case Studies in Electrodiagnostic Medicine

The needle EMG abnormalities are distributed within the muscle domain of the medial cord, although a partial lower plexus lesion cannot be excluded. The extensor pollicis brevis was added to further assess C8-radial motor axons; it was normal. Although the presence of C8-radial motor axon involvement can be used to identify a supraclavicular process, C8-radial motor axon sparing cannot be used to exclude one. However, considering the severity of involvement of the ulnar and median motor responses, at least some radial motor involvement would be expected with a supraclavicular process, and thus, a medial cord lesion is favored. EDX Conclusion

The EDX study localizes the responsible lesion to the lower trunk or medial cord and characterizes it as an axon loss process involving sensory and motor axons. It is severe in degree. The needle EMG findings also suggest a progressive process given the density of fibrillation potentials and the severity of the chronic MUAP changes. The pattern of motor NCS and needle EMG study findings favors a medial cord localization over a lower plexus localization. Clinically, the history of axillary pain also supports an infraclavicular process.

Exercise 45 A 22-year-old male with diffuse upper extremity weakness and sensory loss following a loading dock accident 4 weeks earlier is referred for EDX assessment. The distribution of the weakness and sensory loss exceeds any one element of the peripheral nervous system. Routine sensory NCS are performed first. Nerve Conduction Studies

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 45 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.3

3.7

Ulnar-D5

C8

NR

Superficial Radial

C6,7

NR

RIGHT CV

nAUC

LAT

AMP

3.2

34.1

CV

nAUC

SENSORY

The ulnar and superficial radial responses are absent, and the median response is severely reduced in amplitude, consistent with an axon loss process involving the ganglionic or postganglionic elements of the peripheral nervous system. For comparison purposes, the right Median-D2 NCS was added. To better define the extent of the lesion, additional sensory NCS are added. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 45 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.3

8.7

Ulnar-D5

C8

NR

Superficial Radial

C6,7

NR

LABC

C6

RIGHT CV

nAUC

LAT

AMP

3.2

34.1

2.6

17.8

SENSORY

460

2.8

2.0

CV

nAUC

Case 1 through Case 50

(cont.)

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 45 NCS PERFORMED

DRG

LAT

AMP

Median-D1

C6

3.3

MABC

T1

2.7

RIGHT CV

nAUC

LAT

AMP

6.1

3.3

27.7

3.9

2.7

14.2

CV

nAUC

The added sensory NCS are severely reduced in amplitude. Thus, the lesion involves sensory axons derived from at least the C6 through T1 DRG. Based on the distribution of the sensory NCS, the motor NCS require expansion. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 45 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.3

8.7

Ulnar-D5

C8

NR

Superficial Radial

C6,7

NR

LABC

C6

2.8

Median-D1

C6

MABC

T1

RIGHT CV

nAUC

LAT

AMP

3.2

34.1

2.0

2.6

17.8

3.3

6.1

3.3

27.7

2.7

3.9

2.7

14.2

3.4

8.1

3.3

12.3

2.8

13.6

X

14.5

X

13.2

X

7.2

X

11.3

CV

nAUC

SENSORY

MOTOR Median-APB

8.0 Ulnar-ADM

2.8

7.3 7.1

Ulnar-FDI

X

X

X

X

X

3.7 3.6

Axillary-Deltoid

X

5.7 5.7

Musculocutan-BC

X

6.1 6.1

Radial-ED

X

5.4

X

The motor response abnormalities also indicate a diffuse lesion of the brachial plexus. The Ulnar-FDI, Radial-ED, Axillary-Deltoid, and Musculocutaneous-BC responses are low in amplitude. The Median-APB and Ulnar-ADM responses do not meet our criteria for absolute or relative abnormal, although there is an obvious asymmetry. The asymmetry will likely be associated with needle EMG abnormalities.

461

Section 5: Case Studies in Electrodiagnostic Medicine

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 45

Insertional Activity Normal

Spontaneous Activity

IPSWs SCP Other None

Fibs

Fascs Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

LEFT APB

X

3+

Normal

Normal

FDI

X

2+

Normal

Normal

EI

X

3+

Normal

Normal

FPL

X

2+

Normal

Normal

Pron teres

X

2+

Mild neurogenic

Normal

BC, MH

X

2+

Mod neurogenic

Normal

TC, LH

X

2+

Mild neurogenic

Normal

Deltoid, MH

X

3+

Mod neurogenic

Normal

Serr anterior

X

2+

Mild neurogenic

Normal

Infraspinatus

X

3+

Mod neurogenic

Normal

Trapezius

X

X

Normal

Normal

Low cerv psp

X

X

–

–

High thor psp

X

X

–

–

Needle EMG Study

The needle EMG examination shows fibrillation potentials in all of the limb and shoulder girdle muscles studied, including the routine screening muscles and two preterminal nerve innervated muscles, the serratus anterior (long thoracic nerve) and the infraspinatus (suprascapular nerve). The trapezius and paraspinal muscles are normal. The trapezius was added to attempt to identify a superior boundary (in addition to the spinal accessory nerve, it receives innervation from the C3 and C4 spinal cord segments). Because the long thoracic nerve is derived from the APR level of the brachial plexus and because the suprascapular nerve departs from the initial portion of the upper trunk, involvement of the serratus anterior and the infraspinatus indicates that the lesion reflects a diffuse supraclavicular plexopathy. The other important observation is that the severity of the neurogenic recruitment follows a root distribution (i.e., C5,6,7 muscles demonstrate neurogenic recruitment, but the C8,T1 muscles do not) rather than a nerve distribution (i.e., pronator teres involvement, without APB or FPL involvement; triceps involvement, without EI involvement). This also supports a supraclavicular process. Supraclavicular processes mimic lesions involving multiple roots, whereas infraclavicular processes mimic lesions involving multiple nerves.

462

Case 1 through Case 50

EDX Conclusion

The EDX study identifies a diffuse supraclavicular plexopathy, axon loss in nature, involving the sensory and motor nerve fibers, and severe in degree. The lack of chronic changes is consistent with an onset 4 weeks prior to this study. Final Comment

Because of the large size of the brachial plexus, most brachial plexopathies with C5 through T1 axon involvement are either diffuse supraclavicular processes (termed pan-supraclavicular brachial plexopathies) or diffuse infraclavicular processes (termed pan-supraclavicular brachial plexopathies). In determining whether the lesion is supraclavicular or infraclavicular in location, muscles innervated by preterminal brachial plexus nerves (e.g., dorsal scapular, long thoracic, and suprascapular) are helpful because their involvement indicates a supraclavicular process. Useful muscles in this regard include the levator scapulae and the rhomboids (dorsal scapular nerve), the serratus anterior (long thoracic nerve), and the spinati (suprascapular nerve). In addition, because the pectoral nerves exit the brachial plexus from the proximal aspects of the cords, these muscles are also helpful in differentiating diffuse supraclavicular processes (involved) from diffuse infraclavicular processes (spared).

Exercise 46 A 45-year-old male developed numbness of his entire arm and severe weakness of all upper extremity muscles following a gunshot wound to the left shoulder region 31 days ago. The proximal shoulder girdle muscles were unaffected. (The history of a shoulder level injury suggests an infraclavicular process.) Because the numbness involves all aspects of the extremity, if the screening sensory NCS are abnormal, the study will need to be expanded. Nerve Conduction Studies

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 46 NCS PERFORMED

DRG

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Median-D2

C6,7

NR

Ulnar-D5

C8

NR

Superficial Radial

C6,7

NR

LABC

C6

NR

Median-D1

C6

NR

MABC

T1

NR

The screening sensory NCS are absent, indicating a diffuse axon loss process. Additional sensory NCS are added to “surround the abnormal with the normal” but are also absent. This indicates a diffuse, axon loss process involving at least the C6 through T1 DRG-derived sensory axons. The motor NCS must also be expanded.

463

Section 5: Case Studies in Electrodiagnostic Medicine

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 46 NCS PERFORMED

DRG

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Median-D2

C6,7

NR

Ulnar-D5

C8

NR

Superficial Radial

C6,7

NR

LABC

C6

NR

Median-D1

C6

NR

MABC

T1

NR

MOTOR Median-APB

NR

Ulnar-ADM

NR

Radial-ED

NR

Musculocutan-BC

NR

Axillary-Deltoid

NR

The absent motor responses also indicate a diffuse axon loss process, extremely severe in degree, and involving motor axons derived from the C5 through T1 spinal cord segments. Because the sensory responses are affected, the lesion is postganglionic (it cannot be a ganglionic disorder because the motor responses are also affected). The NCS findings are consistent with a diffuse supraclavicular plexopathy or a diffuse infraclavicular plexopathy. To better localize the process, muscles innervated by preterminal nerves of the brachial plexus, as discussed in the last exercise, are added to the routine needle EMG study.

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 46

Insertional Activity

Spontaneous Activity

Normal IPSWs SCP Other None Fibs

Fascs Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

LEFT

464

APB

X

3+

Severe neurogenic

Normal

FDI

X

3+

Mod neurogenic

Normal

EI

X

3+

Mod neurogenic

Normal

FPL

X

3+

Severe neurogenic

Normal

Pron teres

X

4+

Severe neurogenic

Normal

Case 1 through Case 50

(cont.)

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 46

Insertional Activity

Spontaneous Activity

Normal IPSWs SCP Other None Fibs

Fascs Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

BC, MH

X

3+

Mod neurogenic

Normal

TC, LH

X

2+

Mod neurogenic

Normal

Deltoid, MH

X

3+

Mod neurogenic

Normal

Serr anterior

X

X

Normal

Normal

Infraspinatus

X

X

Normal

Normal

Low cerv psp

X

X

–

–

High thor psp

X

X

–

–

The needle EMG study shows dense fibrillation potentials and neurogenic MUAP recruitment in all muscles innervated via the cords or terminal nerves of the brachial plexus. The muscles innervated by preterminal nerves of the brachial plexus are spared (serratus anterior via long thoracic nerve; infraspinatus via suprascapular nerve). The lack of chronic MUAP changes is consistent with the symptom onset time provided by the patient (i.e., 31 days). EDX Study Conclusion

The EDX study identifies a diffuse infraclavicular plexopathy that is axon loss in nature, involves sensory and motor axons, and is extremely severe in degree. There was no EDX evidence of an avulsion injury. As pointed out in the previous exercise, muscles innervated by preterminal nerves (long thoracic, dorsal scapular, suprascapular, and pectoral nerves) are helpful in differentiating diffuse supraclavicular plexopathies from diffuse infraclavicular plexopathies. Again, because the lateral and medial pectoral nerves exit the brachial plexus from the proximal aspects of the lateral and medial cords, respectively, they are of value in differentiating diffuse supraclavicular lesions (involved) from diffuse infraclavicular ones (spared). In this exercise, they were not necessary.

Exercise 47 A 17-year-old male is referred for EDX assessment of abductor pollicis brevis and biceps muscle wasting and weakness, as well as numbness along the lateral aspect of the left forearm and thumb. These clinical features started approximately 1 year ago and have been slowly progressive in nature. Because of the thumb numbness and the brachioradialis weakness, the referring neurosurgeon suspects a C6 radiculopathy. However, this would not account for the thenar eminence wasting unless it is unrelated (e.g., concomitant recurrent thenar neuropathy) or congenital (e.g., congenital absence of the thenar eminence). Given the distribution of the numbness, the LABC and Median-D1 sensory NCS are added to the routine sensory NCS.

465

Section 5: Case Studies in Electrodiagnostic Medicine

Nerve Conduction Studies

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 47 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.0

6.5

Ulnar-D5

C8

2.7

15.0

Superficial Radial

C6,7

2.4

LABC

C6

Median-D1 Median-D3

RIGHT CV

nAUC

LAT

AMP

2.9

24.6

21.3

2.5

21.8

2.4

3.6

2.5

16.5

C6

3.2

6.9

3.1

18.4

C6,7,8

2.9

6.1

2.9

21.7

CV

nAUC

SENSORY

The initial sensory NCS demonstrate low-amplitude LABC, Median-D1, and Median-D2 responses consistent with an axon loss process involving lateral cord, upper trunk, or the C6 APR/DRG. For comparison purposes, contralateral LABC and median responses are also collected. The degree of decrement is similar for the two median responses, favoring a lateral cord localization. For this reason, bilateral Median-D3 sensory NCS are added. These responses are also low in amplitude. The degree of decrement of all three median responses is similar, favoring a lateral cord process. The contralateral Superficial Radial NCS was added to ensure that the ipsilateral response was not relatively abnormal. Whenever the three median sensory responses are similarly affected, it favors a lateral cord lesion over an upper trunk or C6 APR/DRG process, because the percentage of sensory axons emanating from the C6 DRG varies among these three responses (100% C6 for Median-D1, 20% C6 for Median-D2, and 10% C6 for Median-D3). Thus, with upper trunk and C6 APR/DRG lesions, the Median-D1 response tends to be affected to a greater extent than are the Median-D2 and Median-D3 responses. To better localize the lesion, the Axillary-Deltoid and Musculocutan-BC motor NCS are added to the routine screening studies. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 47 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.0

6.5

Ulnar-D5

C8

2.7

15.0

Superficial Radial

C6,7

2.4

LABC

C6

Median-D1 Median-D3

RIGHT CV

nAUC

LAT

AMP

2.9

24.6

21.3

2.5

21.8

2.4

3.6

2.5

16.5

C6

3.2

6.9

3.1

18.4

C6,7,8

2.9

6.1

2.9

21.7

3.7

2.1

3.4

8.3

SENSORY

MOTOR Median-APB

2.0

466

55

CV

nAUC

Case 1 through Case 50

(cont.)

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 47 NCS PERFORMED Ulnar-ADM

DRG

LAT 2.7

AMP

RIGHT CV

nAUC

8.0 8.0

LAT

AMP

2.7

8.5

CV

nAUC

56

Musculocutan-BC

3.1

3.5

3.2

7.6

Axillary-Deltoid

4.2

15.7

4.3

16.3

The median motor response is severely reduced in amplitude, and the musculocutaneous motor response is moderate to severely reduced in amplitude. The axillary motor response is normal, arguing against an upper plexus localization. To ensure that a relative abnormality was not being missed, the axillary motor response was collected on the contralateral side as well. Because the median motor response was affected in the setting of normal median sensory responses, the ulnar motor response was collected from the contralateral side; a significant side-to-side difference was not present. The pattern of NCS abnormalities suggests a multiple mononeuropathy involving the median and musculocutaneous nerves or a lateral cord lesion extending distally into the proximal portion of the terminal median nerve of the brachial plexus. To differentiate between these two possibilities, the NEE is lightly expanded to include additional lateral cordmedian (i.e., FCR) and musculocutaneous (brachialis) nerve innervated muscles. For more accurate MUAP grading, two contralateral muscles were also studied. The Needle EMG Study

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 47

Insertional Activity

Spontaneous Activity

Normal IPSWs SCP Other None

Fibs

APB

X

2+

FDI

X

EI

Fascs Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

LEFT Severe neurogenic

Moderate CMAL

X

Normal

Normal

X

X

Normal

Normal

FPL

X

X

Mod neurogenic

Moderate CMAL

Pron teres

X

Mod neurogenic

Mild CMAL

BC, MH

X

X

Mod neurogenic

Severe CMAL

TC, LH

X

X

Normal

Normal

Deltoid, MH

X

X

Normal

Normal

FCR

X

X

Mod neurogenic

Moderate CMAL

1+

467

Section 5: Case Studies in Electrodiagnostic Medicine

(cont.)

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 47

Insertional Activity

Spontaneous Activity

Normal IPSWs SCP Other None

Fibs

Brachialis

X

1+

Low cerv psp

X

High thor psp

MUAP Analysis

Fascs Other

MUAP Recruitment

MUAP Morphology

Mod neurogenic

Moderate CMAL

X

–

–

X

X

–

–

FPL

X

X

Normal

Normal

BC

X

X

Normal

Normal

RIGHT

The needle EMG study is consistent with either a multiple mononeuropathy involving the musculocutaneous and median nerves or a brachial plexus lesion involving the lateral cord with extension into the proximal portion of the terminal median nerve. The relationship between the acute and chronic changes is consistent with a slowly progressive process. EDX Conclusion

The EMG study findings are consistent with two possibilities: (1) a brachial plexopathy involving the lateral cord and the proximal aspect of the terminal median nerve that is axon loss in nature, involves the sensory and motor axons, and is severe in degree; or (2) a multiple mononeuropathy involving the musculocutaneous and median nerves that is axon loss in nature and severe in degree. It was suggested in the EDX report that an MRI of the brachial plexus might be of further diagnostic utility. An MRI of the brachial plexus was obtained and showed a mass involving the lateral cord and proximal portion of the terminal median nerve, as suggested by the EDX study findings. Histological assessment showed that the mass was a perineuroma.

Exercise 48 An 18-year-old male is referred for EDX assessment of distal left upper extremity weakness and whole hand numbness secondary to a shoulder fracture-dislocation that followed a motor vehicle accident 5 months prior to this study. He had minimal ability to extend his fingers and wrist and minimal ability to flex his fingers and wrist. He could not pronate his forearm. (The history of a shoulder fracture-dislocation suggests an infraclavicular process and the clinical features indicate involvement at more than one PNS element.) Clinically, the distribution of the sensory and motor abnormalities suggests involvement of the median, ulnar, and radial nerves (the absence of numbness in the forearm argues against a lesion involving all three cords, as dose the lack of elbow flexor weakness). Routine screening sensory NCS are performed first.

468

Case 1 through Case 50

Nerve Conduction Studies

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 48 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.1

Ulnar-D5

C8

Superficial Radial

C6,7

RIGHT CV

nAUC

LAT

AMP

16.7

3.0

35.1

2.7

25.9

2.6

29.9

2.4

28.3

2.5

41.8

CV

nAUC

SENSORY

The screening sensory NCS are remarkable for a low-amplitude Median-D2 response. The Superficial Radial response asymmetry does not meet our EMG laboratory criteria for relative abnormal. The Ulnar-D5 responses are symmetric. Thus, the low-amplitude Median-D2 response indicates an axon loss process involving the median nerve, lateral cord, upper or middle trunk, or the C6 or C7 APR/DRG. Importantly, this isolated median sensory abnormality does not account for the clinical features (median, ulnar, and radial nerve distribution), suggesting that the lesion might not represent a predominantly axon loss process (possible demyelinating conduction block), might be an intraspinal canal disorder, or might be less than 10 days old (the latter possibility is excluded by the fact that the symptoms started 5 months ago with the motor vehicle accident). Additional sensory NCS are added to better define the distribution of the lesion.

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 48 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.1

Ulnar-D5

C8

Superficial Radial

RIGHT CV

nAUC

LAT

AMP

16.7

3.0

35.1

2.7

25.9

2.6

29.9

C6,7

2.4

28.3

2.5

41.8

LABC

C6

2.5

18.7

Median-D1

C6

3.3

16.1

3.1

36.7

MABC

T1

2.6

18.9

CV

nAUC

SENSORY

The amplitude of the Median-D1 response is also reduced, indicating axon loss. Again, the sensory NCS do not account for the clinical findings, suggesting a predominantly demyelinating conduction block lesion (i.e., the lesion is proximal to the stimulus sites for the sensory NCS). Thus, if true, it lies proximal to the sensory NCS stimulation sites. For this reason, more proximal stimulation sites must be incorporated into the motor NCS. In addition to the routine motor NCS, a Radial-ED motor NCS is indicated.

469

Section 5: Case Studies in Electrodiagnostic Medicine

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 48 NCS PERFORMED

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

SENSORY

DRG

Median-D2

C6,7

3.1

16.7

3.0

35.1

Ulnar-D5

C8

2.7

25.9

2.6

29.9

Superficial Radial

C6,7

2.4

28.3

2.5

41.8

3.1

36.7

3.4

9.0

LABC

C6

2.5

18.7

Median-D1

C6

3.3

16.1

MABC

T1

2.6

18.9

MOTOR

STIM SITE

Median-APB

Wrist

3.5

7.5

Ulnar-ADM

Radial-ED

Musculocutan-BC

Axilla

6.5

52

8.6

54

SCF

0.5

44

6.5

57

Wrist

2.9

10.6

2.9

9.5

52

10.5

51

SCF

0.4

48

9.5

56

Below SG

3.0

6.5

3.1

9.1

Above SG

6.5

53

9.0

54

SCF

0.6

51

8.1

57

Axilla

3.1

SCF

5.2 5.1

4.3

12.4

3.1

nAUC

12.3

Axilla

SCF Axillary-Deltoid

CV

6.1

59

58 4.1

13.5

SCF = supraclavicular fossa; SG = spiral groove

The screening motor responses show a focal demyelinating conduction block involving the median, ulnar, and radial nerves that is located between the axillary and supraclavicular fossa stimulation sites. To ensure that the lesion has been fully demarcated, the Musculocutan-BC and the Axillary-deltoid motor NCS are added bilaterally; the latter studies were normal. Thus, at this point, the lesion: (1) is located between the axillary and supraclavicular fossa stimulation sites; (2) involves the median, ulnar, and radial nerves; and (3) is predominantly demyelinating conduction block in nature. There is concomitant axon loss involving the median nerve. Regarding the ulnar and radial nerves, because the side-to-side differences are below 50%, there is no motor NCS evidence of concomitant axon loss. This will be further addressed during the needle EMG study of ulnar and radial nerve innervated muscles.

470

Case 1 through Case 50

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 48

Insertional Activity

Spontaneous Activity

Normal IPSWs SCP Other None

Fibs

Fascs

Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

Severe neurogenic

Normal

LEFT APB

X

X

FDI

X

1+

Severe neurogenic

Normal

EI

X

1+

Severe neurogenic

Normal

FPL

X

2+

Severe neurogenic

Normal

Pron teres

X

1+

Severe neurogenic

Normal

BC, MH

X

X

Normal

Normal

TC, LH

X

X

Severe neurogenic

Normal

Deltoid, MH

X

X

Normal

Normal

Infraspinatus

X

X

Normal

Normal

Low cerv psp

X

X

–

–

High thor psp

X

X

–

–

On needle EMG testing, the abnormalities were confined to the muscle domains of the radial, ulnar, and median nerves. Fibrillation potentials were seen, consistent with axon loss. All muscles showed a neurogenic MUAP recruitment pattern, severe in degree, consistent with the severe demyelinating conduction block pathophysiology affecting these three nerves. Muscles outside of the median, ulnar, and radial nerve distributions were normal. EDX Conclusion

The EDX study identifies a brachial plexopathy involving the median, ulnar, and radial terminal nerves. The lesion is predominantly demyelinating conduction block in nature. There is concomitant axon loss, mild in degree, in all three nerves; it is most pronounced in the median nerve. Based on these findings, significant functional improvement is expected. The density of fibrillation potentials is more reflective of the timing of the EDX study rather than of the degree of severity. In general, given the degree of demyelinating conduction block, it would be unusual for there to be no fibrillation potentials. Again, for lesion severity estimation, the motor response amplitude differences across the block sites are utilized. In this case, the discordance between the clinical examination and the sensory NCS prompted consideration of underlying demyelination conduction block. Had this not been considered and the motor NCS performed without adding more proximal stimulation sites, the initial motor responses would have been normal. However, once the severe neurogenic MUAP recruitment pattern was observed on the needle EMG study of median, ulnar, and radial nerve innervated muscles, the true nature of the lesion would have become apparent, prompting repeat motor NCS with more proximal sites of stimulation.

471

Section 5: Case Studies in Electrodiagnostic Medicine

The important concept is that whenever a neurogenic MUAP recruitment is recorded from a muscle demonstrating a normal or near-normal amplitude on motor NCS, an underlying demyelinating conduction block is present and must be sought.

Exercise 49 A 70-year-old female was referred for EDX assessment of diffuse left upper extremity weakness and patchy numbness following a shoulder procedure that was performed using an interscalene block 5 weeks prior. The localization question is whether the new deficits are related to the interscalene block (supraclavicular plexus involvement) or the shoulder procedure (infraclavicular plexus involvement). Given that the sensory abnormalities are diffuse, the initial sensory NCS are expanded to include all DRG levels. Nerve Conduction Studies

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET EXERCISE 49 NCS PERFORMED

LEFT DRG

LAT

AMP

Median-D2

C6,7

3.3

21.7

Ulnar-D5

C8

2.8

16.5

Superficial Radial

C6,7

2.5

20.1

LABC

C6

2.5

13.1

Median-D1

C6

3.4

17.1

MABC

T1

2.6

13.9

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY

The initial sensory NCS are normal by absolute criteria. Thus, there is no evidence of axon loss, raising the possibility of an intraspinal canal lesion or a demyelinating lesion. Importantly, when judging the amplitudes in relation to each other, the two median responses seem low with respect to the ulnar response, as does the radial response, because, in general, the median and radial response amplitudes are 1.5 to 2.0 times the ulnar response amplitude. For this reason, contralateral sensory NCS were added to look for relative abnormalities. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET EXERCISE 49 NCS PERFORMED

LEFT DRG

LAT

AMP

Median-D2

C6,7

3.3

Ulnar-D5

C8

Superficial Radial

RIGHT CV

nAUC

LAT

AMP

21.7

3.2

34.1

2.8

16.5

2.8

27.3

C6,7

2.5

20.1

2.4

31.1

LABC

C6

2.5

13.1

2.5

28.0

Median-D1

C6

3.4

17.1

3.3

27.2

MABC

T1

2.6

13.9

2.7

16.8

SENSORY

472

CV

nAUC

Case 1 through Case 50

Based on side-to-side comparisons, the LABC response is relatively abnormal. The other sensory responses are normal (by absolute and relative criteria), but an obvious asymmetry is present. This suggests possible axon loss not meeting the criteria for absolute or relative abnormal. The low-amplitude LABC response generates a lesion localization differential of lateral antebrachial cutaneous nerve, musculocutaneous nerve, lateral cord, upper trunk, C6 APR/DRG. At this point, the sensory NCS findings are too focal to account for the clinical features of the patient. Thus, there is the possibility that the lesion is demyelinating or that it is a two-level process, with elements of both preganglionic and ganglionic/postganglionic involvement. Given that the weakness is generalized, the initial motor NCS are expanded. Given that there is a possibility of underlying demyelinating conduction block, proximal stimulation sites are included. UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 49 NCS PERFORMED

DRG

LAT

AMP

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY Median-D2

C6,7

3.3

21.7

3.2

34.1

Ulnar-D5

C8

2.8

16.5

2.8

27.3

Superficial Radial

C6,7

2.5

20.1

2.4

31.1

LABC

C6

2.5

13.1

2.5

28.0

Median-D1

C6

3.4

17.1

3.3

27.2

MABC

T1

2.6

13.9

2.7

16.8

MOTOR Median-APB

Ulnar-ADM

Musculocutan-BC

Radial-ED

Axillary-Deltoid

Wrist

12.4

12.0

Elbow

10.9

11.5

Axilla

9.0

SCF

8.8

Wrist

9.0

11.0

Elbow

8.5

10.0

Axilla

7.3

SCF

7.2

Axilla

4.3

4.3

SCF

3.6

3.7

Elbow

7.7

12.3

Below SG

7.4

11.5

Axilla

6.5

SCF

6.5

SCF

5.5

12.4

473

Section 5: Case Studies in Electrodiagnostic Medicine

The routine median and ulnar motor responses, as well as the musculocutaneous response, are normal, and there is no evidence of a demyelinating conduction block. The axillary motor response is severely reduced in amplitude, indicating axon loss, and the radial motor response is lower in amplitude than the contralateral side, but does not meet our criteria for relative abnormal. Thus, like the sensory NCS, the motor NCS abnormalities are too limited to account for the distribution of the clinical abnormalities. Thus, at this point, the sensory NCS indicate axon loss involving the C6 segment, and the motor NCS indicate axon loss involving the C5 and C6 segments (given the degree of axillary motor response decrement, both innervating roots are likely involved). If there is underlying demyelinating conduction block, the lesion must lie proximal to the supraclavicular fossa stimulation site (i.e., proximal to the mid-trunk level), and given the severity of the weakness on clinical examination, it should be associated with a neurogenic MUAP firing pattern (i.e., a decreased number of MUAPs firing at a rapid rate is expected). Needle EMG Examination

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 49

Insertional Activity

Spontaneous Activity

Normal IPSWs SCP Other None

Fibs Fascs Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

LEFT

474

APB

X

FDI

X

EI

X

Normal

Normal

1+

Severe neurogenic

Normal

X

1+

Severe neurogenic

Normal

FPL

X

1+

Severe neurogenic

Normal

Pron teres

X

2+

Severe neurogenic

Normal

BC, MH

X

2+

Severe neurogenic

Normal

TC, LH

X

2+

Severe neurogenic

Normal

Deltoid, MH

X

2+

Severe neurogenic

Normal

Infraspinatus

X

2+

Severe neurogenic

Normal

Brachioradialis

X

2+

Severe neurogenic

Normal

Rhomb major

X

X

Normal

Normal

Serr anterior

X

X

Normal

Normal

Low cerv psp

X

X

High thor psp

X

X

Case 1 through Case 50

The muscle distribution of the observed neurogenic MUAP recruitment pattern indicates involvement of the C5 through C8 myotomes. From the motor NCS, we know that the lesion lies proximal to the mid-trunk level. Because the rhomboideus major and serratus anterior muscles are both normal, we can conclude that the lesion is at the trunk level because, based on the sensory and motor NCS, the lesion is most pronounced at the C5 and C6 level. Therefore, these two muscles should be involved. The exception would be if the lesion were distal to the APR (these two muscles are innervated via motor axons exiting the APR elements of the brachial plexus, which lie proximal to the trunk elements). Despite the fibrillation potentials, the severity of the axon loss is dictated by the NCS response amplitudes (the LABC sensory response and the axillary motor response). Therefore, the lesion is very predominantly demyelinating conduction block in nature, and the axon loss is most pronounced at C5 and C6. The C5 axons of the upper trunk appear to be more involved than the C6 axons given that the LABC sensory response amplitude is only relatively abnormal. Again, in general, when the motor response amplitude is 50% or more decreased (the axillary motor response), the sensory response amplitude (the LABC sensory response) is typically absent of nearly so. Finally, sparing of the APB suggests that the T1 fibers may not be involved. EDX Study Conclusion

The study indicates a brachial plexopathy with elements of both demyelinating conduction block and axon loss with the former predominating. The lesion involves the C5 through C8 sensory and motor nerve axons and is located distal to the APR level (because of the rhomboideus major and serratus anterior sparing) and proximal to the midtrunk level (because the lesion was not noticeable on motor NCS but was present based on the neurogenic MUAP recruitment pattern noted on the needle EMG study). Also, given the large demyelination conduction block component, significant recovery is expected. The axon loss is most pronounced at the C5 and C6 levels. Based on its severity and the short distance between the lesion and the target muscles, reasonably good recovery is expected. This case demonstrates how proximally located demyelinating conduction block lesions can be identified whenever a neurogenic MUAP recruitment pattern is observed during the needle EMG study of a muscle from which a normal or near-normal distal motor response was recorded.

Exercise 50 A 45-year-old male was referred for EDX assessment of left hand pain associated with numbness and weakness in a median nerve distribution. The clinical features started about 10 weeks ago and followed an axillary arteriogram 5 days prior. Given the distribution of the weakness, the screening sensory NCS are performed first. Nerve Conduction Studies

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 50 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.2

5.0

Ulnar-D5

C8

2.8

16.1

Superficial Radial

C6,7

2.5

22.4

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY

The screening sensory NCS reveal a low-amplitude Median-D2 response, indicating an axon loss process located in the median nerve, lateral cord, upper or middle trunk, or the C6 or C7 APR/DRG. To better refine the localization of the underlying lesion, the LABC and Median-D1 sensory NCS are added.

475

Section 5: Case Studies in Electrodiagnostic Medicine

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 50 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.2

5.0

Ulnar-D5

C8

2.8

16.1

Superficial Radial

C6,7

2.5

22.4

RIGHT CV

nAUC

LAT

AMP

CV

nAUC

SENSORY

LABC

C6

2.6

9.1

Median-D1

C6

3.3

3.1

The additional sensory NCS show a very-low-amplitude Median-D1 response and a normal LABC response. This pattern argues against a lateral cord, upper trunk, or C6 APR/DRG localization because the LABC and Median-D1 are typically both affected or both spared. In one study of the 26 upper trunk lesions, both responses were affected in 25 of 26 and both responses were spared in 1 of 26 (Ferrante and Wilbourn, 1995). Thus, they were affected identically in 100% (i.e., both abnormal or both normal). Therefore, the combination of Median-D1 involvement with LABC sparing strongly argues for a median nerve localization, as suggested by the clinical features. For this reason, the Median-D3 sensory NCS (to determine the degree of median nerve involvement) and the contralateral three median sensory NCS (to assess the side-to-side differences and to determine whether the LABC sensory response is relatively abnormal) were added.

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET EXERCISE 50 NCS PERFORMED

LEFT DRG

LAT

AMP

Median-D2

C6,7

3.2

5.0

Ulnar-D5

C8

2.8

16.1

Superficial Radial

C6,7

2.5

22.4

LABC

C6

2.6

Median-D1

C6

Median-D3

C6,7,8

RIGHT CV

nAUC

LAT

AMP

3.1

26.4

9.1

2.5

8.7

3.3

3.1

3.3

18.6

3.2

4.7

3.2

27.9

CV

nAUC

SENSORY

The added studies show that the three median sensory responses are roughly affected to the same degree and that the LABC sensory response is definitely spared (it is not relatively abnormal). Therefore, at this point, the sensory NCS indicate that this is an axon loss process involving the median nerve. The uniformity of involvement of the three median sensory responses could be seen with a distal lateral cord lesion but would be very uncommon with a supraclavicular process (upper trunk or C6 APR/DRG). Based on these findings, the motor NCS are expanded to include the Median-L2 motor NCS.

476

Case 1 through Case 50

UPPER EXTREMITY NERVE CONDUCTION STUDY WORKSHEET LEFT

EXERCISE 50 NCS PERFORMED

DRG

LAT

AMP

Median-D2

C6,7

3.2

5.0

Ulnar-D5

C8

2.8

16.1

Superficial Radial

C6,7

2.5

22.4

RIGHT CV

nAUC

LAT

AMP

3.1

26.4

CV

nAUC

SENSORY

LABC

C6

2.6

9.1

2.5

8.7

Median-D1

C6

3.3

3.1

3.3

18.6

Median-D3

C6,7,8

3.2

4.7

3.2

27.9

3.7

5.9

3.5

8.9

3.1

8.2

2.5

2.3

MOTOR Median-APB

5.9 Ulnar-ADM

2.9

57

7.8 7.7

Median-L2

2.7

57

0.9

The routine motor NCS show the median motor responses to be low in amplitude, consistent with the impression that the lesion represents an axon loss median neuropathy. Also, involvement of these two motor responses eliminates the possibility of a lateral cord, upper trunk, or C8 APR/DRG localization. For severity assessment, the median motor responses were recorded on the contralateral side. The FCR muscle is added to the routine needle EMG study to better assess the median nerve and two contralateral median nerve innervated muscles are added to better define the degree of severity of any associated CMAL. Needle EMG Study

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 50

Insertional Activity Normal

Spontaneous Activity

IPSWs SCP Other None

Fibs

Fascs Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

Mild neurogenic

Mild CMAL

LEFT APB

X

3+

FDI

X

X

Normal

Normal

EI

X

X

Normal

Normal

FPL

X

2+

Normal

Normal

Pron teres

X

2+

Normal

Mild CMAL

BC, MH

X

X

Normal

Normal

TC, LH

X

X

Normal

Normal

477

Section 5: Case Studies in Electrodiagnostic Medicine

(cont.)

UPPER EXTREMITY NEEDLE EMG WORKSHEET EXERCISE 50

Insertional Activity Normal

Spontaneous Activity

IPSWs SCP Other None

Fibs

X

Fascs Other

MUAP Analysis MUAP Recruitment

MUAP Morphology

Normal

Normal

Mild neurogenic

Moderate CMAL

Deltoid, MH

X

FCR

X

Low cerv psp

X

X

–

–

High thor psp

X

X

–

–

FPL

X

X

Normal

Normal

Pron teres

X

X

Normal

Normal

3+

RIGHT

The needle EMG abnormalities are restricted to the muscle domain of the median nerve, consistent with the suspected axon loss median neuropathy. Involvement of the pronator teres and FCR muscles indicates that the lesion is situated proximal to the elbow. The presence of neurogenic recruitment indicates that the lesion is at least moderate-severe in degree. The presence of CMAL indicates reinnervation via collateral sprouting, and the presence of fibrillation potentials indicates the potential for further reinnervation. EDX Conclusion

The EDX study shows a median neuropathy, axon loss in nature, involving the sensory and motor nerve fibers, and located proximal to the elbow level. The side-to-side median motor response asymmetries and the presence of neurogenic MUAP recruitment in two of the needle EMG studied muscles suggest that the severity of the lesion is at least moderate-severe. Because the symptoms started within 14 days of an axillary arteriogram, the most likely explanation is that the median neuropathy is related to medial brachial fascial compartment syndrome. This disease has a predilection for the median nerve and typically affects it in isolation or out of proportion to any concomitantly affected nerves (Ferrante, 2014). Medial Brachial Fascial Compartment Syndrome

Medial brachial fascial compartment (MBFC) syndrome, which was first described in 1966 by Staal and colleagues, is a neurologic emergency that must be recognized by EDX providers. Anatomically, the brachial fascia surrounds the arm along its length, forming a cylindrical enclosure. The medial and lateral intermuscular septums extend from the medial and lateral aspects of the humerus, respectively, to the brachial fascia, generating anterior and posterior arm compartments. The medial edge of the medial intermuscular septum divides into two fascial flaps as it approaches the brachial fascia, forming a separate, triangular-shaped compartment that runs from the clavicle to the elbow. This compartment, termed the MBFC, contains the five terminal nerves of the brachial plexus (median, ulnar, radial, axillary, and musculocutaneous) and the axillary vessels. The five nerves exit from this compartment in the following order, from proximal to distal: musculocutaneous, axillary, radial, ulnar, and median. As stated earlier, with MBFC syndrome, the median nerve is typically affected first and, in the setting of multiple nerve involvement, most severely. About 50% of the time, it is the only nerve affected and when two nerves are involved, the median and ulnar nerves are the most common combination (Tsao and Wilbourn, 2003). The radial, axillary, and musculocutaneous nerves are affected less frequently. Lesions within the MBFC capable of producing

478

Case 1 through Case 50

mass effect (e.g., hematomas, aneurysms, pseudoaneurysms) cause the intracompartmental pressure to rise, thereby impeding or eliminating the microcirculation of the nerve fibers. Clinically, patients develop pain or paresthesias in the distribution of the affected nerves, followed shortly thereafter by weakness in the same distribution. Without prompt surgical intervention, recovery is unlikely. For this reason, it is imperative that these lesions be recognized immediately so that emergent decompression can be performed (Staal, 1966; Tsao and Wilbourn, 2003). In one study, complete recovery was 8.3 times more likely when surgical intervention occurred within the first 4 hours of symptom onset (Chitwood et al., 1996). Electrodiagnostic testing has no role in the acute setting. It is of value postoperatively to assess the prognosis for recovery. It is important to realize that the intracompartmental pressure blocks the microcirculation but not the major arteries of the limb. Consequently, the presence of a normal radial artery pulse does not exclude this entity. Additionally, the lack of a palpable cord along the medial aspect of the arm does not exclude this entity. It must be presumed to be present whenever hand pain follows a procedure performed within the MBFC, and it must be treated emergently at the first sign of median nerve abnormalities.

References Chitwood RW, Shepard AD, Shetty PC, et al. Surgical complications of transaxillary arteriography: a casecontrol study. J Vasc Surg 1996;23:44–85.

peripheral nerve injuries. In Mackinnon SE, editor, Nerve surgery. New York: Thieme Medical Publishers, 2015:59–74.

Ferrante MA. Brachial plexopathies. Continuum 2014;20:1323–1342.

Ferrante MA, Wilbourn AJ. The lesion distribution among 281 patients with sporadic neuralgic amyotrophy. Muscle Nerve 2017;55:858–861.

Ferrante MA. The relationship between sustained gripping and the development of carpal tunnel syndrome. Federal Practitioner 2016;33:10–15.

Levin KH. L5 radiculopathy with reduced superficial peroneal sensory responses: intraspinal and extraspinal causes. Muscle Nerve 1998;21:3–7.

Ferrante MA, Wilbourn AJ. The utility of various sensory nerve conduction responses in assessing brachial plexopathies. Muscle and Nerve 1995;18:1–11.

Riley DE, Shields RW. Diabetic amyotrophy with upper extremity involvement. Neurology 1984;34:216.

Ferrante MA, Wilbourn AJ. The electrodiagnostic examination of

following arterial catheterization by the axillary approach. Br J Radiol 1966;39:115–116. Sumner AJ, England JD. Neuralgic amyotrophy: an increasingly diverse entity. Muscle Nerve 1987;10:60–68. Tsao BE, Wilbourn AJ. The medial brachial fascial compartment syndrome following axillary arteriography. Neurology 2003;61:1037–1041. Wilbourn AJ. Diabetic neuropathies. In Brown WF, Bolton CF, editors, Clinical electromyography, 2nd ed. Boston: Butterworth-Heinemann, 1993:477–515.

Staal A, Voorthuisen AE, Dijk LM. Neurological complications

479

Section

6

Appendices

Appendix 1: Plexus Anatomy

The Brachial Plexus

C5

The Lumbosacral Plexus

L4

Posterior spinal roots Anterior spinal roots Dorsal root ganglia Mixed spinal nerves Post. primary rami Ant. primary rami (roots of plexus)

Roots

L5 Lumbosacral trunk S1

C6

Superior gluteal n. S2

C7 Trunks

C8

S4

Up

r

M

Post.cutancous nerve of the thigh Pudendal n.

pe

T1

Inferior gluteal n.

S3

Divisions

idd

le

Lo we r

Sciatic n. La

Cords

l ra te r te

s Po r

io

l

ia

ed

M

The formation of the superior gluteal, inferior gluteal, posterior femoral cutaneous, and sciatic nerves from the lumbosacral trunk and sacral plexus nerve fibers is depicted. (Illustration courtesy of Asa J. Wilbourn, MD)

Terminal nerves

Although anatomists define the roots of the brachial plexus as being equivalent to the anterior primary rami, clinicians specializing in brachial plexopathies define the roots as that portion of the brachial plexus located proximal to the trunks. Using this definition, lesions involving the dorsal or ventral roots, the mixed spinal nerves, or the anterior or posterior primary rami are brachial plexopathies. This approach has considerable clinical utility (see text). (Illustration courtesy of Asa J. Wilbourn, MD)

483

Appendix 2: Nerve Anatomy

The Median Nerve LATERAL CORD

MEDIAL CORD MEDIAL BRACHIAL CUTANEOUS NERVE

MUSCULOCUTANEOUS NERVE

MEDIAL ANTEBRACHIAL CUTANEOUS NERVE LATERAL HEAD OF MEDIAN NERVE

ULNAR NERVE MEDIAL HEAD OF MEDIAN NERVE

MEDIAN NERVE Pronator teres Palmaris longus Flexor carpi radialis Flexor digitorum superficialis

ANTERIOR INTEROSSEOUS NERVE Flexor digitorum profundus Flexor pollicis longus Pronator quadratus

PALMAR CUTANEOUS NERVE CARPAL TUNNEL RECURRENT THENAR NERVE Abductor pollicis brevis Opponens pollicis Flexor pollicis brevis, superficial head PALMAR DIGITAL NERVES

The first common palmar digital nerve supplies two branches to the volar aspect of the thumb and a third branch to the radial aspect of the index finger. It also gives off a branch to the first lumbrical. The second common palmar digital branch supplies the adjacent sides of the index and middle fingers and gives off a branch to the second lumbrical muscle. The third

484

common palmar digital branch supplies the adjacent sides of the middle and ring fingers. Thus, in general, the median nerve innervates the lateral 3.5 digits and the ulnar nerve innervates the medial 1.5 digits, although, in some individuals, the ring finger may be supplied solely by the median nerve or the ulnar nerve, or the split may occur at the third digit rather than the fourth digit.

Appendix 2: Nerve Anatomy

The Ulnar Nerve MEDIAL CORD

MEDIAL BRACHIAL CUTANEOUS NERVE MEDIAL ANTEBRACHIAL CUTANEOUS NERVE

ULNAR NERVE

ULNAR GROOVE ELBOW CUBITAL TUNNEL

Flexor carpi ulnaris Flexor digitorum profundus

PALMAR CUTANEOUS NERVE

DORSAL ULNAR CUTANEOUS NERVE WRIST GUYON CANAL SUPERFICIAL TERMINAL NERVE Palmaris brevis Hypothenar muscles Abductor digiti minimi Opponens digiti minimi Flexor digiti minimi Interosseous muscles (4 dorsal; 3 palmar) Adductor pollicis Deep head of the flexor pollicis brevis

485

Section 6: Appendices

The Radial Nerve POSTERIOR CORD Inferior lateral cutaneous nerve of arm Posterior cutaneous nerve of the arm Triceps Anconeus Posterior cutaneous nerve of the forearm

AXILLARY NERVE RADIAL NERVE

SPIRAL GROOVE Brachialis Brachioradialis Extensor carpi radialis longus ELBOW

POSTERIOR INTEROSSEOUS NERVE ARCADE OF FROHSE

Supinator Extensor carpi radialis brevis Extensor digitorum Extensor digiti minimi Extensor carpi ulnaris Abductor pollicic longus Extensor pollicis longus Extensor pollicis brevis Extensor indicis

486

SUPERFICIAL RADIAL NERVE

Appendix 2: Nerve Anatomy

The Obturator Nerve L2 L3 L4 OBTURATOR NERVE OBTURATOR CANAL Obturator externus

POSTERIOR DIVISION Obturator externus

ANTERIOR DIVISION

Pectineus Adductor longus

Adductor magnus Adductor brevis Adductor brevis

Gracilis SENSORY TERMINAL BRANCH

487

Section 6: Appendices

The Femoral Nerve L2

L3

L4

INGUINAL LIGAMENT

ANTERIOR DIVISION POSTERIOR DIVISION Pectineus

Quadriceps femoris

Sartorius MEDIAL CUTANEOUS NERVE OF THIGH

SAPHENOUS NERVE

488

Appendix 2: Nerve Anatomy

The Sciatic Nerve Proper SCIATIC NERVE

Semitendinosus

Semimembranosus

Biceps femoris, long head

Branch to adductor magnus

TIBIAL NERVE

Biceps femoris, short head

COMMON PERONEAL NERVE

489

Section 6: Appendices

The Tibial Nerve

TIBIAL NERVE

COMMON PERONEAL NERVE

MEDIAL SURAL CUTANEOUS NERVE Popliteus Plantaris Soleus Gastrocnemius

Tibialis posterior Flexor hallucis longus Flexor digitorum longus TARSAL TUNNEL

MEDIAL CALCANEAL NERVE MEDIAL PLANTAR NERVE Abductor hallucis Flexor digitorum brevis Flexor hallucis brevis Lumbricals 1 and 2 Cutaneous branch

490

LATERAL PLANTAR NERVE Abductor digiti minimi Flexor digiti minimi Adductor hallucis Interossei Lumbricals 3 and 4 Cutaneous branch

Appendix 2: Nerve Anatomy

The Peroneal Nerve

TIBIAL NERVE

COMMON PERONEAL NERVE LATERAL SURAL CUTANEOUS NERVE

FIBULAR TUNNEL SUPERFICIAL PERONEAL NERVE

DEEP PERONEAL NERVE Tibialis anterior

Peroneus longus Peroneus brevis

Extensor hallucis longus Extensor digitorum longus

Extensor digitorum brevis TERMINAL SENSORY BRANCH

TERMINAL SENSORY BRANCH

491

Appendix 3: Myotome Tables for the Upper and Lower Extremities C5 C6 C7 C8 T1

ANTERIOR PRIMARY RAMI

ANTERIOR PRIMARY RAMI

PROXIMAL NERVES

PROXIMAL NERVES

Rhomboid major/minor (dorsal scapular)

Iliacus (lumbar plexus)

Supraspinatus (suprascapular)

Adductor longus (obturator)

L2 L3 L4 L5 S1 S2

Vastus lateralis/medialis (fermoral)

Infraspinatus (suprascapular)

Rectus femoris (femoral)

Deltoid (axillary)

Tensor tascia lata (superior gluteal)

Biceps brachii (musculocutaneous)

Gluteus medius (superior gluteal)

Brachialis (musculocutaneous)

Gluteus maximus (inferior gluteal) RADIAL NERVE

SCIATIC NERVE

Triceps

Semitendinosus/membranosus (tibial)

Anconeus

Biceps femoris (sht. hd.)(peroneal)

Brachioradialis

Biceps femoris (long. hd.)(tibial)

Extensor carpi radialis

PERONEAL NERVE

Extensor pollicis brevis

Tibialis anterior

Extensor indicis proprius

Extensor hallucis

MEDIAN NERVE

Peroneus longus

Pronator teres

Extensor digitorum brevis

Flexor carpi radialis

TIBIAL NERVE

Flexor pollicis longus

Tibialis posterior

Pronator quadratus

Flexor digitorum longus

Abductor pollicis brevis

Gastrocnemius lateral

ULNAR NERVE

Gastrocnemius medial Soleus

Flexor carpi ulnaris

Abductor hallucis

Flexor digitorum profundus (med)

Abductor digiti quinti pedis

Abductor digiti minimi

POSTERIOR PRIMARY RAMI

Adductor pollicis

Lumbar paraspinals

First dorsal interosseous C5 C6 C7 C8 T1

POSTERIOR PRIMARY RAMI Cervical paraspinals High thoracic paraspinals Often predominant contribution

Sometimes significant contribution

Minor/equivocal contribution

492

L2 L3 L4 L5 S1 S2

Often predominent Sometimes significant contribution contribution Minor/equivocal contribution

Appendix 4: The SNAP, CMAP, and Needle EMG Domains of the Brachial Plexus Elements^ The Upper Plexus

SNAP Domain

CMAP Domain

Needle EMG Domain

LABC (100%)

Musculocutaneous (biceps)

Levator scapulae

Median-D1 (100%)

Axillary (deltoid)

Rhomboids

Superficial radial (60%)

Radial (EDC)

Serratus anterior

Median-D2 (20%)

Supraspinatus, infraspinatus

Median-D3 (10%)

Biceps, brachialis Deltoid, teres minor Brachioradialis Triceps Extensor carpi radialis Pronator teres Flexor carpi radialis

The Middle Plexus

SNAP Domain

CMAP Domain

Needle EMG Domain

Median-D2 (80%)

Radial (anconeus)

Pronator teres

Median-D3 (70%)

Flexor carpi radialis

Superficial radial (40%)

Triceps Anconeus Extensor carpi radialis Extensor digitorum communis Serratus anterior

The Lower Plexus

SNAP Domain

CMAP Domain

Needle EMG Domain

Ulnar-D5 (100%)

Ulnar (ADM)

Abductor pollicis brevis

MABC (100%)

Ulnar-FDI

Flexor pollicis longus

493

Section 6: Appendices

(cont.)

SNAP Domain

CMAP Domain

Needle EMG Domain

Median-D3 (20%)

Median-APB

Extensor indicis proprius

Radial-EIP

Extensor pollicis brevis Extensor carpi ulnaris First dorsal interosseous Abductor digiti minimi Adductor pollicis Flexor digitorum profundus-4,5 Flexor carpi ulnaris

The Lateral Cord

SNAP Domain

CMAP Domain

Needle EMG Domain

LABC (100%)

Musculocutaneous (biceps)

Biceps

Median-D1 (100%)

Brachialis

Median-D2 (100%)

Pronator teres

Median-D3 (80%)

Flexor carpi radialis

The Posterior Cord

SNAP Domain

CMAP Domain

Needle EMG Domain

Superficial radial (100%)

Axillary (deltoid)

Latissimus dorsi

Radial (EDC)

Deltoid; teres minor

Radial (EIP)

Triceps; anconeus Brachioradialis Extensor carpi radialis Extensor digitorum communis Extensor pollicis brevis Extensor carpi ulnaris Extensor indicis proprius

The Medial Cord

SNAP Domain

CMAP Domain

Needle EMG Domain

Ulnar-D5 (100%)

Ulnar (ADM)

Abductor pollicis brevis

MABC (100%)

Ulnar-FDI

Opponens pollicis

Median-D3 (20%)

Median-APB

Flexor pollicis longus First dorsal interosseous

494

Appendix 4: The SNAP, CMAP, and needle EMG Domains of the brachial plexus Elements

(cont.)

SNAP Domain

CMAP Domain

Needle EMG Domain Abductor digiti minimi Adductor pollicis Flexor digitorum profundus-4,5 Flexor carpi ulnaris

^The complete needle EMG domains of each brachial plexus element are not provided, but rather only those muscles considered most helpful by the author. Regarding the percentages shown in parentheses in the SNAP domain column, these values represent the frequency with which the sensory nerve fibers studied by the specific sensory NCS traverse each element.

This reflects their DRG derivation frequency (Ferrante and Wilbourn, 1995).

Reference Ferrante MA, Wilbourn AJ. The utility of various sensory nerve conduction responses in assessing brachial plexopathies. Muscle Nerve 1995;18:1–11.

495

Appendix 5: The Sensory and Motor NCS Techniques Used in Our EMG Laboratories The sensory NCS assess the integrity of the sensory neurons, including their cell bodies within the dorsal root ganglia and their axons. For a number of reasons, the sensory NCS are indispensable, and, as stated throughout this textbook, EDX studies that do not include them are incomplete (see Chapter 8). Unfortunately, these studies are much more technically demanding, less standardized among EMG laboratories, can be performed in an orthodromic or an antidromic manner, have more than one latency available for measurement (onset versus peak), and have more than one amplitude for measurement. The motor NCS are much easier to perform and vary less among various EMG laboratories. The sensory and motor NCS techniques we use in our EMG laboratories are shown here. These techniques were introduced to me in 1993, during my EMG fellowship training at The Cleveland Clinic Foundation. I have performed them in the identical manner since that time and consider them reliable. A description of each technique, including the position of the limb and the stimulating and recording electrode placement sites, is provided. The age-related normal control values for each technique are provided in Appendix 6. For each technique described, the distance between the stimulating electrodes (two prong electrodes, the cathode and the anode, extending from the handheld stimulator) and the recording electrodes (the E1 and E2 electrodes) is measured between the cathode prong of the stimulator and the E1 electrode. The anode prong of the stimulator is maintained proximal to the cathode and is ideally rotated about the cathode so that the baseline disruption related to the shock artifact is minimized. The onset of the recorded response should originate from flat baseline, not from the shock artifact as it is returning to the baseline. The E2 electrode is located 3 cm distal to the E1 electrode for sensory NCS and mixed NCS, and is positioned over the tendon, distal to the muscle belly, for motor NCS. The ground electrode (E0) is ideally placed

496

between the stimulating and recording electrodes. Depending on body habitus, many individuals require longer distances between the stimulating cathode and the E1 recording electrode. Regarding normal latency, we add 0.2 msec for each 1-cm addition to the distance (this equates to a nerve conduction velocity of 50 m/sec, as described in the textbook).

The Median Sensory NCS, Recording Second Digit

R S

Limb position

 Forearm supinated Recording electrode positions

 The E1 electrode is placed over the center of the proximal phalange of the second digit  The E2 electrode is placed 3 cm distal to the E1 electrode Stimulating electrode positions

 The cathode is positioned 13 cm proximal to the E1 electrode, over the median nerve in the distal forearm, between the palmaris longus tendon and the flexor carpi radialis tendon Comments

 We use digital clip electrodes with the contact sites along the lateral and medial sides of the second digit (i.e., superficial to the digital nerves under study)  Avoid flexing or extending the wrist, as this changes the distance between the stimulating cathode and the E1 recording electrode

Appendix 5: The Sensory and Motor NCS Techniques Used in Our EMG Laboratories

The Median Sensory NCS, Recording Third Digit

R

S

Limb position

 Forearm supinated

Stimulating electrode positions

 The cathode is positioned 13 cm proximal to the E1 electrode, over the median nerve in the distal forearm, between the palmaris longus tendon and the flexor carpi radialis tendon. The 13 cm distance in nonlinear and consists of a 10-cm segment (10 cm proximal to the E1 electrode, over the median nerve in the distal forearm, between the flexor carpi radialis and the flexor pollicis longus tendons) and a 3-cm segment (3 cm proximal to the 10 cm point). Comments

Recording electrode positions

 The E1 electrode is placed over the center of the proximal phalange of the third digit  The E2 electrode is placed 3 cm distal to the E1 electrode Stimulating electrode positions

 The cathode is positioned 13 cm proximal to the E1 electrode, over the median nerve in the distal forearm, between the palmaris longus tendon and the flexor carpi radialis tendon

 We use digital clip electrodes with the contact sites along the lateral and medial sides of the second digit (i.e., superficial to the digital nerves under study)  Avoid flexing or extending the wrist, as this changes the distance between the stimulating cathode and the E1 recording electrode

The Ulnar Sensory NCS, Recording Fifth Digit

Comments

 We use digital clip electrodes with the contact sites along the lateral and medial sides of the second digit (i.e., superficial to the digital nerves under study)  Avoid flexing or extending the wrist, as this changes the distance between the stimulating cathode and the E1 recording electrode

S R

Limb position

The Median Sensory NCS, Recording First Digit

 Forearm supinated Recording electrode positions

R

S

Limb position

 Forearm supinated Recording electrode positions

 The E1 electrode is placed over the center of the proximal phalange of the first digit  The E2 electrode is placed 3 cm distal to the E1 electrode

 The E1 electrode is placed over the center of the proximal phalange of the fifth digit  The E2 electrode is placed 3 cm distal to the E1 electrode Stimulating electrode positions

 The cathode is positioned 11 cm proximal to the E1 electrode, over the ulnar nerve in the distal forearm, between the flexor carpi ulnaris and the flexor digitorum profundus tendons (i.e., just radial to the flexor carpi ulnaris tendon) Comments

 We use digital clip electrodes with the contact sites along the lateral and medial sides of the fifth digit (i.e., superficial to the digital nerves under study)

497

Section 6: Appendices

The Lateral Antebrachial Cutaneous Sensory NCS

Comments

 We use disposable surface recording electrodes  I typically place the recording electrodes adjacent to the ulna, stimulate as described earlier, moving the cathode medially and laterally until the maximum response is collected. This identifies the ideal stimulation site. I then establish the ideal recording site by moving the recording electrodes medially (away from the ulna) until the maximal response is collected.

S

Limb position

 Forearm supinated; elbow extended Recording electrode positions

Dorsal Ulnar Cutaneous Sensory NCS

 The E1 electrode is placed 12 cm distal to the elbow crease, along a line connecting the stimulating cathode with the radial pulse  The E2 electrode is placed 3 cm distal to the E1 electrode Stimulating electrode positions

 The cathode is positioned at the elbow crease, just lateral to the biceps tendon Comments

 We use disposable surface recording electrodes

The Medial Antebrachial Cutaneous Sensory NCS

S

 Forearm pronated Recording electrode positions

 The E1 electrode is positioned in the proximal aspect of the interspace at the apex of the fourth and fifth metacarpals  The E2 electrode is placed 3 cm distal to the E1 electrode

X R

Limb position

 Forearm supinated; elbow extended Recording electrode positions

 The E1 electrode is placed 8 cm distal to the point (marked as “X” in the figure) at which the elbow crease intersects the medial aspect of the bicep tendon  The E2 electrode is placed 3 cm distal to the E1 electrode Stimulating electrode positions

 The cathode is positioned 4 cm proximal to the previously defined point, between the biceps and the triceps muscles

498

Limb position

Stimulating electrode positions

 The cathode is positioned 8 cm proximal to the E1 electrode, between the ulna and the flexor carpi ulnaris tendon Comments

 We use disposable surface recording electrodes

Superficial Radial Sensory NCS

Appendix 5: The Sensory and Motor NCS Techniques Used in Our EMG Laboratories

Limb position

 Forearm pronated (or midway between pronation and supination)

Ulnar Palmar Mixed NCS

Recording electrode positions

 The E1 electrode is placed on the dorsolateral aspect of the hand, overlying the superficial radial nerve as it crosses the extensor pollicis longus tendon at the base of the thumb (with the thumb extended, this nerve is easily palpated as it crosses this tendon)  The E2 electrode is placed 3 cm distal to the E1 electrode Stimulating electrode positions

 The cathode is positioned overlying the radius, 10 cm proximal to the E1 electrode Comments

 We use disposable surface recording electrodes

R

S

Limb position

 Forearm supinated Recording electrode positions

 The E1 electrode is placed along the proximal wrist crease, between the flexor carpi ulnaris tendon and the flexor digitorum profundus tendon  The E2 electrode is placed 3 cm proximal to the E1 electrode Stimulating electrode positions

Median Palmar Mixed NCS

 The cathode is positioned 8 cm distal to E1 electrode, along a line dividing the fourth and fifth digits (i.e., within the fourth metacarpal interspace, along a line dividing the fourth web space) R

Comments S

Limb position

 We use disposable surface recording electrodes

Median Motor NCS, Recording Abductor Pollicis Brevis (Thenar Eminence)

 Forearm supinated Recording electrode positions

 The E1 electrode is placed along the proximal wrist crease, between the flexor carpi radialis tendon and the flexor pollicis longus tendon  The E2 electrode is placed 3 cm proximal to the E1 electrode

R S2

Stimulating electrode positions

Limb position

 The cathode is positioned 8 cm distal to E1 electrode, along a line dividing the second and third digits (i.e., within the second metacarpal interspace, along a line dividing the second web space)

 Forearm supinated

Comments

 We use disposable surface recording electrodes

S1

Recording electrode positions

 The E1 electrode is over the thenar eminence, just proximal to its midpoint  The E2 electrode is placed over the tendon, distally, away from the belly of the muscle

499

Section 6: Appendices

Stimulating electrode positions

Comments

 Distal stimulation

 We use disposable surface recording electrodes  When the median sensory, median palmar, and median motor APB responses are absent, an axon loss process involving the median nerve is indicated. However, where along the median nerve the lesion is located cannot be determined by the NCS findings. When these responses are absent due to advanced carpal tunnel syndrome, this study is helpful when it is present, because the delayed distal latency permits localization (i.e., the lesion lies distal to the distal stimulation site). If it is also absent, then localization will need to be determined by the needle EMG study, which is less definitive.

:

The cathode is positioned 5 cm proximal to the E1 electrode, between the palmaris longus tendon and the flexor carpi radialis tendon

 Proximal stimulation

:

The cathode is positioned along the elbow crease, medial to the biceps tendon (just medial to the brachial pulse)

Comments

 We use disposable surface recording electrodes  In the setting of concomitant ulnar nerve stimulation, the anode of the stimulator is rotated away from the ulnar nerve  If the median motor response shows an initial positive phase (positive dip), the E1 electrode should be relocated

Median Motor NCS, Recording Second Lumbrical

Ulnar Motor NCS, Recording Abductor Digiti Minimi (Hypothenar Eminence)

E2 S2

S1

S3

E1

S2

S1 R

Limb position

 Forearm supinated Recording electrode positions

 The E1 electrode is situated in the palm, at the intersection of a line bisecting the second and third digits and another line bisecting the two distal palmar creases  The E2 electrode is placed over the middle phalange of the second digit Stimulating electrode positions

 Distal stimulation

:

The cathode is positioned between the flexor carpi radialis tendon and the palmaris longus tendon in the same spot used to stimulate the median nerve for the median motor NCS, recording thenar eminence

 Proximal stimulation

:

500

The cathode is positioned along the elbow crease, medial to the biceps tendon

Limb position

 Forearm supinated Recording electrode positions

 The E1 electrode is positioned over the center of the hypothenar eminence  The E2 electrode is placed distal to the hypothenar eminence, over the medial aspect of the middle phalange of the fifth digit Stimulating electrode positions

 Distal stimulation

:

:

The cathode is positioned 5 cm proximal to the E1 electrode, between the flexor carpi ulnaris and flexor digitorum profundus tendons (i.e., just radial to the flexor carpi ulnaris tendon) In the setting of concomitant median nerve stimulation, the anode of the stimulator is rotated away from the median nerve

Appendix 5: The Sensory and Motor NCS Techniques Used in Our EMG Laboratories

 Below-elbow stimulation

:

The cathode is positioned 5 cm distal to the medial epicondyle, along the course of the nerve

 Above-elbow stimulation

:

The cathode is positioned 5 cm proximal to the medial epicondyle, along the course of the nerve

Comments

 We use disposable surface recording electrodes

Ulnar Motor NCS, Recording First Dorsal Interosseous

 Above-elbow stimulation

:

The cathode is positioned 5 cm proximal to the medial epicondyle, along the course of the nerve

Comments

 The stimulation site is identical to that used for the Ulnar-ADM motor NCS  We use disposable surface recording electrodes  With this technique, an initial positive dip is typical  Although some place the E2 electrode over the extensor pollicis longus tendon adjacent to the snuff box to avoid the initial positive dip, we do not because we did not collect our normal values in that manner

Radial Motor NCS, Recording Extensor Digitorum Communis S2

E2

S1

Limb position

 Forearm pronated for recording electrode placement and supinated for stimulation

E1

Recording electrode positions

 The E1 electrode is positioned over the center of the first dorsal interosseous muscle  The E2 electrode is placed distal to this muscle, along the radial (lateral) aspect of the middle phalange of the second digit Stimulating electrode positions

 Distal stimulation

: :

The cathode is positioned 5 cm proximal to the E1 electrode, between the flexor carpi ulnaris and flexor digitorum profundus tendons (i.e., just radial to the flexor carpi ulnaris tendon) In the setting of concomitant median nerve stimulation, the anode of the stimulator is rotated away from the median nerve

 Below-elbow stimulation

:

The cathode is positioned 5 cm distal to the medial epicondyle, along the course of the nerve

Limb position

 Forearm pronated Recording electrode positions

 The E1 electrode is positioned over the extensor digitorum communis muscle, between the radius and ulna bones, at about the junction where the proximal one-third of the forearm and the middle one-third of the forearm meet  The E2 electrode is placed just distal to the ulnar styloid process Stimulating electrode positions

 Distal stimulation

:

The cathode is positioned over the radial nerve, just lateral to the biceps tendon

501

Section 6: Appendices

 Above–spiral groove stimulation

:

The cathode is positioned between the posterior aspect of the middle head of the deltoid and the triceps, proximal to the level of the deltoid tendon insertion

Comments

 We use disposable surface recording electrodes  When there is a drop in the amplitude or negative area under the curve value between the elbow and above–spiral groove stimulation sites (i.e., evidence of a demyelinating conduction block), stimulate below the spiral groove to better localize the lesion  The nerve may also be stimulated at the medial arm/axillary level, just medial to the triceps muscle

Radial Motor NCS, Recording Extensor Indicis

deltoid and the triceps, proximal to the level of the deltoid tendon insertion Comments

 We use disposable surface recording electrodes  If there is a drop in the amplitude or negative area under the curve value between the elbow and above–spiral groove stimulation sites (i.e., evidence of a demyelinating conduction block), then we stimulate below the spiral groove (i.e., in the biceps–triceps groove) to better localize the lesion  The nerve may also be stimulated at the medial aspect of the proximal arm/axilla, just medial to the triceps muscle

Musculocutaneous Motor NCS, Recording Biceps

S2

E2 S1

E1

Limb position

 Forearm pronated Recording electrode positions

 The E1 electrode is positioned over the extensor indicis muscle, approximately three fingerbreadths (5 cm) proximal to the ulnar styloid and about one-third of the distance from the ulnar aspect of the forearm to the radial aspect  The E2 electrode is placed just distal to the ulnar styloid process Stimulating electrode positions

Limb position

 Forearm and arm supinated Recording electrode positions

 The E1 electrode is positioned over the midportion of the biceps muscle  The E2 electrode is placed over the biceps tendon in the antecubital fossa Stimulating electrode positions

 Distal stimulation

:

 Distal stimulation

:

The cathode is positioned over the radial nerve, just lateral to the biceps tendon

 Above–spiral groove stimulation

:

502

The cathode is positioned between the posterior aspect of the middle head of the

The cathode is positioned in the proximal arm/distal axilla, immediately inferior to the tendon of the short head of the biceps

 Supraclavicular fossa stimulation

:

The cathode is positioned posterior to the clavicle and lateral to the sternocleidomastoid muscle (at the junction of the medial one-third and the middle one-third of the clavicle)

Appendix 5: The Sensory and Motor NCS Techniques Used in Our EMG Laboratories

Comments

Limb position

 We use disposable surface recording electrodes  When the patient rotates the head contralaterally, the sternocleidomastoid muscle borders are better appreciated. Once identified, the head can be rotated ipsilaterally so that the stimulator advances slightly deeper into the fossa, lessening the amount of current required  The stimulator is relocated within the supraclavicular fossa as necessary to obtain the maximal response

 Patient supine with arm at side

Axillary Motor NCS, Recording Deltoid

Recording electrode positions

 The E1 electrode is positioned over the center of the deltoid muscle, midway between the acromion process and the deltoid insertion site  The E2 electrode is placed over the lateral epicondyle Stimulating electrode positions

 Supraclavicular fossa stimulation

:

The cathode is positioned posterior to the clavicle and lateral to the sternocleidomastoid muscle (at the junction of the medial one-third and the middle one-third of the clavicle)

Comments

 We use disposable surface recording electrodes  The most important comment is that E1 be placed in the center of the deltoid muscle and that E2 be positioned off of the muscle  When the patient rotates the head contralaterally, the sternocleidomastoid muscle borders are better appreciated. Once identified, the head can be rotated ipsilaterally so that the stimulator advances slightly deeper into the supraclavicular fossa, lessening the amount of current required.  The stimulator is relocated within the supraclavicular fossa as necessary to obtain the maximal response

Suprascapular Motor NCS, Recording Infraspinatus Scapular spine

Acromion process E2

E1

503

Section 6: Appendices

Limb position

 Patient seated with arm at side

 The E2 electrode is placed on the dorsolateral aspect of the foot, 3 cm distal to the E1 electrode

Recording electrode positions

Stimulating electrode positions

 The E1 electrode is positioned in the infraspinous fossa, 2 cm below the scapular spine and 2 cm lateral to its medial border  The E2 electrode is placed over the acromion process

 The cathode is positioned 14 cm proximal to the E1 electrode, on the posterior aspect of the leg, just lateral to the Achilles’ tendon Comments

 We use disposable surface recording electrodes

Stimulating electrode positions

 Supraclavicular fossa stimulation

:

Superficial Peroneal Sensory NCS

The cathode is positioned posterior to the clavicle and lateral to the sternocleidomastoid muscle (at the junction of the medial one-third and the middle onethird of the clavicle)

Comments

 We use disposable surface recording electrodes  The stimulator is relocated within the supraclavicular fossa as necessary to obtain the maximal response

Sural Sensory NCS Limb position

 The subject is positioned so that the studied limb is superior to the non-studied limb, with the medial aspect of the knee of the studied limb on the contralateral knee and the medial aspect of the foot on the examination table Recording electrode positions

Limb position

 With the subject lying on their side with the lower extremity under study superior to the lower extremity not under study, the knee is slightly flexed and resting on the contralateral lower extremity with the medial aspect of the foot under study contacting the examination table Recording electrode positions

 The E1 electrode is positioned inferoposterior to lateral malleolus

504

 The E1 electrode is positioned 3 cm proximal to the E2 electrode  The E2 electrode is placed on the dorsal aspect of the ankle, halfway between the edge of the lateral malleolus and the extensor digitorum longus tendon Stimulating electrode positions

 The cathode is positioned 10 cm proximal to the E1 electrode, on the anterolateral aspect of the leg, just in front of the fibula Comments

 We use disposable surface recording electrodes  It is important to realize that the E2 electrode is placed first

Appendix 5: The Sensory and Motor NCS Techniques Used in Our EMG Laboratories

Saphenous Sensory NCS

Limb position

 Supine Recording electrode positions

 The bar electrode is positioned between the tibia and the Achilles’ tendon with the E1 electrode distal to the E2 electrode Stimulating electrode positions

 Stimulation (medial plantar NCS)

:

The cathode is positioned 11 cm distal to the E1 electrode, along a nonlinear 11 cm line, extending from the E1 electrode to the line bisecting the first web space, and then along that bisecting line to the 11 cm mark

 Stimulation (lateral plantar NCS) Limb position

 Supine Recording electrode positions

 The E1 electrode is positioned 3 cm proximal to the E2 electrode  The E2 electrode is placed on the dorsal aspect of the ankle, halfway between the edge of the medial malleolus and the tibialis anterior tendon Stimulating electrode positions

 The cathode is positioned 10 cm proximal to the E1 electrode, on the medial aspect of the leg, just medial to the tibia Comments

 We use disposable surface recording electrodes  It is important to realize that the E2 electrode is placed first

:

The cathode is positioned 13 cm distal to the E1 electrode, along a nonlinear 13 cm line, extending from the E1 electrode to the line bisecting the fourth web space, and then along that bisecting line to the 13 cm mark

Comments

 We use disposable surface recording electrodes  Many individuals require longer distances, such as 15 cm or more for the medial plantar NCS and 17 cm or more for the lateral plantar NCS (ideally, a 2-cm difference is maintained)

Peroneal Motor NCS, Recording Extensor Digitorum Brevis

Medial and Lateral Plantar Mixed NCS

505

Section 6: Appendices

Limb position

 Supine

 The E2 electrode is placed on the dorsal aspect of the ankle

Recording electrode positions

Stimulating electrode positions

 The E1 electrode is positioned on the dorsolateral aspect of the foot, overlying the center of the extensor digitorum brevis muscle  The E2 electrode is placed on the dorsal aspect of the fifth metatarsal phalangeal joint

 Distal stimulation

Stimulating electrode positions

 Distal stimulation

:

The cathode is positioned over the deep peroneal nerve, 7 cm proximal to the E1 electrode and about 2 cm lateral to the tibialis anterior tendon

 Proximal stimulation

: :

The cathode is positioned over the common peroneal nerve in the lateral aspect of the popliteal fossa When evidence of a conduction block is noted, additional stimulation above and below the fibular head permits more accurate localization

Comments

 We use disposable surface recording electrodes  When the ankle and popliteal fossa stimulation sites yield identical motor responses, then there is no conduction block between these two sites and, consequently, no reason to stimulate above or below the fibular head

:

The cathode is positioned just inferior and distal to the fibular head

 Proximal stimulation

: :

The cathode is positioned over the common peroneal nerve in the lateral aspect of the popliteal fossa When evidence of a conduction block is present, subsequent stimulation above the fibular head permits more accurate localization

Comments

 We use disposable surface recording electrodes  When the below–fibular head and popliteal fossa stimulation sites yield identical motor responses, then there is no conduction block between these two sites and, consequently,no reason to stimulate above the fibular head

Tibial Motor NCS, Recording Abductor Hallucis

Peroneal Motor NCS, Recording Tibialis Anterior

Limb position

 Supine Recording electrode positions

 The E1 electrode is positioned on the anterior aspect of the leg, 4 cm below the tibial tuberosity and 1 cm lateral to the tibial crest

506

Limb position

 Supine Recording electrode positions

 The E1 electrode is positioned on the medial aspect of the foot, 1 cm inferior to the navicular process and at its proximal edge

Appendix 5: The Sensory and Motor NCS Techniques Used in Our EMG Laboratories

 The E2 electrode is placed on the medial aspect of the foot at the metatarsophalangeal joint of the great toe Stimulating electrode positions

 The E2 electrode is placed on the lateral aspect of the foot overlying the fifth metatarsophalangeal joint of the little toe

 Distal stimulation

Stimulating electrode positions

:

The cathode is positioned just inferior to the tibia, along a nonlinear line 10 cm proximal to the E1 electrode, as shown in the illustration

 Proximal stimulation

:

The cathode is positioned over the tibial nerve in the center of the popliteal fossa

Comments

 We use disposable surface recording electrodes  The 10-cm segment is measured from the E1 electrode, proximally to the inferoposterior aspect of the medial malleolus, and then proximally along the medial aspect of the tibia, which roughly approximates the course of the nerve  Distally, this technique assesses the motor axons of the medial plantar nerve from their anterior horn cell origins in the S1 and S2 spinal cord segments to their terminations in the abductor hallucis muscle

 Distal stimulation

:

The cathode is positioned at the same site used for the tibial motor NCS, recording abductor hallucis

 Proximal stimulation

:

The cathode is positioned over the tibial nerve in the center of the popliteal fossa

Comments

 We use disposable surface recording electrodes  Distally, this technique assesses the motor axons of the lateral plantar nerve from their anterior horn cell origins in the S1 and S2 spinal cord segments to their terminations in the abductor digiti quinti muscle

Femoral Motor NCS, Recording Rectus Femoris

Tibial Motor NCS, Recording Abductor Digiti Quinti

Limb position

 Supine Recording electrode positions

 The E1 electrode is positioned on the lateral aspect of the foot, at the center of the abductor digit quinti muscle, about halfway between the palpable bony prominence (the tuberosity of the fifth metatarsal bone) and the fifth metatarsophalangeal joint

507

Section 6: Appendices

Limb position

Comments

 Supine

 We use disposable surface recording electrodes  Like all motor NCS, it is important to observe the specific muscles contracting in response to the stimulation. When the medial thigh muscles are primarily contracting, the stimulator should be repositioned (usually slightly more laterally) until the rectus femoris is observed to contract.

Recording electrode positions

 The E1 electrode is positioned over the rectus femoris muscle halfway along a line joining the anterior superior iliac spine (marked with an “X” in the illustration) and the patellar tendon  The E2 electrode is placed on the patellar tendon Stimulating electrode positions

 Distal stimulation

:

508

The cathode is positioned inferior to the inguinal ligament, just lateral to the femoral artery pulse.

Appendix 6: The Age-Related, Normal Control Values Used in Our EMG Laboratories Age-Related Normal Control Values for Particular Nerve Conduction Studies AGE NERVE

Distance 5–9 years old (cm) DL

AMP

CV

10–29 years old

30–49 years old

50–59 years old

DL

DL

DL

AMP

CV

AMP

CV

AMP

CV

60+ years old

DL

AMP

CV

Median (s)

13

< 3.2

> 20

> 51

< 3.3

> 20

> 51

< 3.4

> 20

> 50

< 3.6

> 15

> 50

< 3.8

> 10

> 50

Ulnar (s)

11

< 2.9

> 18

> 51

< 3.0

> 18

> 51

< 3.1

> 12

> 50

< 3.1

> 10

> 50

< 3.2

>5

> 50

Radial (s)

10

< 2.6

> 18

> 51

< 2.7

> 18

> 51

< 2.7

> 18

> 50

< 2.7

> 14

> 50

< 2.8

> 10

> 50

LABC (s)

12

< 2.8

> 18

> 51

< 2.9

> 16

> 51

< 2.9

> 14

> 50

< 2.9

> 12

> 50

< 2.9

> 10

> 50

< 2.9

> 18

> 51

< 2.9

> 18

> 51

< 3.0

> 18

> 50

< 3.1

> 10

> 50

< 3.2

>5

> 50

< 2.9

> 5.0 > 50

DUC (s) MABC (s)

12 (8–4)

Median palmar (mx)

8

< 2.2

> 10

> 51

< 2.2

> 10

> 51

< 2.2

> 10

> 50

< 2.2

> 10

> 50

< 2.2

> 10

> 50

Ulnar palmar (mx)

8

< 2.2

>5

> 51

< 2.2

>5

> 51

< 2.2

>5

> 50

< 2.2

>5

> 50

< 2.2

>5

> 50

Median-APB

5

< 3.6

>6

> 51

< 3.9

>6

> 51

< 3.9

>6

> 50

< 4.0

>6

> 50

< 4.0

>5

> 50

Ulnar-ADM

5

< 2.9

>8

> 51

< 3.0

>8

> 51

< 3.1

>7

> 50

< 3.1

>7

> 50

< 3.1

>6

> 50

Ulnar-FDI

< 3.8

>8

> 51

< 3.8

>8

> 51

< 4.3

>7

> 50

< 4.5

>7

> 50

< 4.5

>6

> 50

Radial-EDC

< 3.0

>6

> 51

< 3.0

>6

> 51

< 3.1

>6

> 50

< 3.1

>5

> 50

< 3.1

>5

> 50

MusculoBiceps

< 3.5

>4

> 51

< 3.5

>4

> 51

< 3.5

>4

> 50

< 3.5

>4

> 50

< 3.8

>3

> 50

AxillaryDeltoid

< 4.8

>4

> 51

< 4.8

>4

> 51

< 4.8

>4

> 50

< 4.8

>4

> 50

< 5.0

>3

> 50

Sural (s)

14

< 4.3

>6

> 41

< 4.4

>6

> 41

< 4.5

>5

> 40

< 4.6

>4

> 40

< 4.6

>3

> 40

Spfcl peron (s)

10

< 4.3

>6

> 41

< 4.4

>6

> 41

< 4.5

>5

> 40

< 4.6

>4

> 40

< 4.6

>3

> 40

Saphenous (s)

10

< 4.3

>6

> 41

< 4.4

>6

> 41

< 4.5

>4

> 40

< 4.6

>4

> 40

< 4.6

>3

> 40

Post tibial (AH)

10

< 5.8

>8

> 41

< 5.8

>8

> 41

< 6.0

>8

> 40

< 6.0

>4

> 40

< 6.0

>4

> 40

< 6.0

>4

> 41

< 6.0

>4

> 41

< 6.5

>4

> 40

< 6.5

>3

> 40

< 6.5

>3

> 40

Post tibial (ADQP) Peroneal-EDB 7

< 5.5

>3

> 41

< 5.5

>3

> 41

< 5.5

>3

> 40

< 6.0

> 2.5 > 40

< 6.0

> 2.5

> 40

Peroneal-TA

< 4.0

>4

> 41

< 4.0

>4

> 41

< 4.0

>4

> 40

< 4.5

>3

> 40

< 4.5

>3

> 40

Femoral-RF

< 6.0

>4

> 41

< 6.0

>4

> 41

< 6.5

>4

> 40

< 6.5

>3

> 40

< 6.5

>3

> 40

H-reflex, M-wave

< 7.0

>8

> 41

< 7.0

>8

> 41

< 7.0

>8

> 40

< 7.5

>6

> 40

< 7.5

>6

> 40

H-reflex, H-wave

< 35.0

>1

> 41

< 35.0

>1

> 41

< 35.0

>1

> 40

< 35.0

>1

> 40

< 35.0

>1

> 40

509

Appendix 7: Our Screening Sensory and Motor NCS and Needle EMG Muscles

Upper Extremity

 Sensory NCS: median, recording second digit; ulnar, recording fifth digit; superficial radial, recording dorsum hand  Motor NCS: median, recording thenar eminence; ulnar, recording hypothenar eminence  Needle EMG: first dorsal interosseous, extensor indicis, flexor pollicis longus, pronator teres, biceps (lateral head), triceps (lateral head), deltoid (middle head), paraspinal muscles

510

Lower Extremity

 Sensory NCS: sural, recording adjacent to lateral malleolus; superficial peroneal, recording dorsum of ankle  Motor NCS: tibial, recording abductor hallucis; peroneal, recording extensor digitorum brevis  H reflex  Needle EMG: flexor hallucis brevis, flexor digitorum longus, tibialis anterior, gastrocnemius (medial head), biceps femoris (short head), vastus lateralis, gluteus medium, paraspinal muscles

Appendix 8: The Advantages and Disadvantages of the Individual EDX Studies Sensory NCS Advantages

The only component that assesses the sensory nerve fibers Identifies remote lesions Ideal for localizing axon loss plexopathies More sensitive than motor NCS for identifying mixed mononeuropathies Differentiates between preganglionic and ganglionic/postganglionic Disadvantages

Useful for severity assessment prior to collateral sprouting Can diagnose disorders of the distal aspects of the nerve fibers, as well as the NMJ and muscle Disadvantages

Do not evaluate the sensory nerve fibers

Needle EMG Advantages

The most sensitive component for motor axon loss

Technically demanding

Capable of widespread assessment (most skeletal muscles)

A minority of normal elderly may not have elicitable lower extremity sensory responses

Able to identify subclinical motor axon disruption

Do not assess the nerve fiber segments distal to the most distal electrode pair

Able to assess muscles not assessable clinically Best component to identify radiculopathies (when motor axons involved)

Do not evaluate the motor nerve fibers Overestimates axon loss

Motor NCS (and RNSS) Advantages

Capable of assessing long nerve fiber segments Useful for localizing focal demyelination Useful for differentiating acquired demyelination from hereditary dysmyelination

Best component to identify myopathies Disadvantages

Requires patient cooperation More uncomfortable Does not evaluate the sensory nerve fibers Insensitive for detecting focal demyelination

511

Appendix 9: Needle EMG Findings with Lesions at Various Levels of the Neuraxis Anterior Horn Cell

 Neurogenic recruitment when severity is at least moderate-severe; MUAP duration increases following collateral sprouting; MUAP amplitude typically normal unless the disorder is extremely severe in degree and very slowly progressive (e.g., SMA-3) or remote (e.g., previous poliomyelitis)

Nerve Root

 Sensory NCS: normal  Motor NCS: typically normal unless the root involves small muscles routinely studied by motor NCS (e.g., C8 radiculopathy with hand intrinsic muscle involvement) or it involves adjacent roots (e.g., combined C5 and C6 radiculopathies and the musculocutaneous motor response)  Needle EMG: fibrillation potentials; chronic changes appear following collateral sprouting (e.g., MUAP duration increases); with increased severity, neurogenic recruitment appears

Polyneuropathy, Length-Dependent, Axonal

 Symmetric, low-amplitude or absent sensory responses; low-amplitude or absent motor responses (typically, the sensory responses are affected to a greater extent than the motor responses); MUAP duration increases following collateral sprouting

512

Polyneuropathy, Acquired, Generalized Demyelinating

 Increased latencies, decreased CVs, pathologic temporal dispersion, conduction block, reduced recruitment when at least moderate-severe in degree

POLYNEUROPATHY, HEREDITARY DEMYELINATING (actually dysmyelination)  Like acquired demyelinating polyneuropathy, except that the increases in latency and the decreases in CV are uniform and focal DMCB does not occur

Presynaptic NMJ (e.g., LEMS)

 Low-amplitude, short-duration, polyphasic MUAPs; possible early recruitment; increment on high-frequency RNSS MMV on needle EMG

Postsynaptic NMJ (e.g., MG)

 Possible low-amplitude, short-duration, polyphasic MUAPs; possible early recruitment; decrement on low-frequency RNS; MMV on needle EMG

Myopathy

 Low-amplitude, short-duration, polyphasic MUAPs; possible early recruitment; may be normal; possible fibrillation potentials or myotonic potentials, depending on underlying etiology

Index

A waves, 148–149 AANEM. See American Association of Neuromuscular and Electrodiagnostic Medicine abductor pollicis brevis (APB), 318 aberrant reinnervation, 223 abnormal RNSS, 156 absolute abnormal, 284 absolute refractory period, 58–59 AC. See alternating current accessory deep peroneal nerve, 269–270 accommodation, 59 acetate, 69 acetylcholine (ACh), 68 binding, 70–71 exocytosis and, 69 release, 71–72 in synaptic space, 69 synthesized, 69 vesicles, 69, 71–72, 152–153 acetylcholine receptors (AChR), 152, 200–201 in LEMS, 253 in myasthenia gravis, 252 acetylcholinesterase (AChE), 69 ACh. See acetylcholine ACh receptors (AChRs), 68–70 AChE. See acetylcholinesterase AChRs. See acetylcholine receptors acquired axon loss polyneuropathies, 247–248 acquired demyelinating polyneuropathies, 248–250 actin, 74 action at a distance, 7 action potential (AP) generation, 59 action potential (AP) propagation, 61 antidromic, orthodromic conduction in, 92–93 block, 124 conduction velocity of, 62–63 toward E1 electrode, 102 in excitation-contraction coupling, 74 length constant in, 125 of myelin, 61–63 physiologic, 93 safety factor of, 125–126 saltatory conduction in, 64 speed, 62–65, 123–124, 132

action potentials (AP), 34, 49, 59. See also absolute; compound muscle action potential; motor unit action potential; muscle fiber APs; relative refractory period; sensory nerve action potential duration of, 63–64 of F wave, 146 focal axon loss and, 124 focal demyelination and, 123–124 of nerve fibers, 99–100, 111 of recording electrode, 90–91 activation gate, of VGNC, 258 activation phase of MUAP assessment, 193–194 of needle EMG, 173, 235 activation time, of muscle, 107 active electrode, 87–88. See also bipolar recordings; E1 electrode; referential recordings active release zones, 68 acute compression, 229–230 acute inflammatory demyelinating polyradiculoneuropathy (AIDP), 249 acute poliomyelitis, 237 ADC. See analog-to-digital converter adductor pollicis muscle, 264–265 adenosine monophosphate (AMP), 78 adenylate kinase, 78 adhesive surface recording electrodes, 89 ADP, 75–76, 78 age-related issues in NCS, 260–261 in needle EMG, 261 as pitfalls, 260–261 agrin, 70 AHCs. See anterior horn cells AIDP. See acute inflammatory demyelinating polyradiculoneuropathy aliasing, 42 alkaline batteries, 11 alpha gamma coactivation, 77 ALS. See amyotrophic lateral sclerosis alternating current (AC), 14, 23 cycle of, 24–25, 27

in magnetic field, 26–28 Ohm’s law on, 23 reversals in, 23, 36 voltage in, 23–24 alternating current (AC) signal of 60 Hz power artifact, 276–278 in capacitors, 36–37 dangers of, 290–291 in high pass filter, 38 in low pass filter, 38 quantification of, 24–26 aluminum, 27–28 American Association of Neuromuscular and Electrodiagnostic Medicine (AANEM), 290, 298, 306–307 AMP. See adenosine monophosphate amphiphilic molecules, 49 amplifiers ADC, DAC for, 45 differential, 45–47 for EMG machines, 45, 47 gain on, 44 inverting, 46 monopolar, 44–45 non-inverting, 46 resistance, 44 signal-to-noise ratio for, 47 amplitude, 25 absolutely, relatively abnormal, 102 baseline-to-peak, 113 body tissue and, 103 CMAP, 103–104 decrement, 100 discrepancy, in DMCB, 130 of fibrillation potentials, 185 of H wave abnormal, 145 in EMG laboratory, 144–145 measurements of, 144–145 low, peroneal-EDB, 285 measurement of, 102–104 of motor axons, 103 of motor NCS, 100 of motor response, 104, 135, 216 of MUAP, 165, 173–176, 195 negative AUC and, 104–105, 132 peak-to-peak, 113 response, very low, 284–285

513

Index

amplitude (cont.) in sensory NCS, 113–114 of sensory response, 135, 216 synchrony and, 103 of tibial response, 100 amplitude values, 85 amyotrophic lateral sclerosis (ALS), 132, 238 analog filters, 40–41 analog-to-digital converter (ADC), 41–43, 45 anastomosis. See also Martin-Gruber anastomosis Berretini, 269 in forearm, 268 median-to-ulnar, 262 Riche–Cannieu, 268–269 ulnar-to-median, 268 animal electricity (Galvani), 76 anisotropic band, 75 anodal block, 279 anode, 11, 21–22, 43 anode rotation, 273–275 anomalous innervations, 262 anterior horn cells (AHCs), 50, 68, 73, 79–80 acute poliomyelitis and, 237 disorders, 236–237 in motor unit, 79–80, 162–163 post-poliomyelitis syndrome and, 237–238 SMA-3 and, 238–239 antidromic conduction, 92–93 antidromic sensory NCS, 111, 114–116 AP. See action potentials APB. See abductor pollicis brevis apparent single fiber APs (ASFAPs), 206 arborization point, 183 area under the curve (AUC), 101 DMCB and, 129 negative, 104–105, 132 ASFAPs. See apparent single fiber APs atomic number, 5 atomic structure, in charge, 4–5 atoms, 4–5, 8 ATP, 75–76 ADP to, 78 of muscle fibers, 78 atypically proximal MGA, 267 AUC. See area under the curve axolemma, 50, 53 axon, 50–51. See also distal; initial; myelin; proximodistal diameter, 62–63 disruption in NCS, 137–138 Wallerian degeneration in, 137–138 motor, 103, 201 terminals, 68

514

axon loss, 123. See also acquired axon loss polyneuropathies; chronic motor axon loss approach to, in surgical intervention, 225–226 conduction failure and, 131–132 DMCB and, 214 EDX features of, 248 focal, 124, 202 lesions, 215–216 needle EMG study of, 215–216 proximal lesion and, 147 sensory response and, 135–136 axonotmesis, 222 axoplasm, 50 B. See bel backfire of motor neurons, 146 time, 146–147 bandpass filter, 40 bandwidth, of frequencies, 35 basal firing rate, 165 baseline train, 154 baseline-to-peak amplitude, 113 batteries alkaline, 11 capacitors and, 21–22 electrochemical cells and, 11 lead-acid automotive, 11 transistor, 11 voltage in, 10–11, 18 zinc-carbon, 11 bel (B), 26 Bell, Alexander Graham, 26 belly-tendon method, 95–96 Berretini anastomosis, 269 bidirectional propagation, 74 bilateral hand numbness and tingling, exercise, 332, 359 EDX study conclusion in, 361 EMG study conclusion in, 335 NCS in, 332–334, 359–361 needle EMG in, 334, 361 bilateral lower extremity weakness, exercise, 428–429 NCS in, 429 needle EMG in, 429–430 bilateral upper extremity weakness, exercise, 410, 421 EDX study conclusion in, 412 NCS in, 410–411, 421–422 needle EMG in, 411–412, 422–424 biological signals, unwanted, 46 bipennate muscle, 80 biphasic morphology, 95–96 bipolar recordings, 88 bits, 41

bleeding in needle EMG, 295–296 safety issues of, 295–296 blink reflexes, 149–150 blocking, 200, 205 blood-nerve barrier, 66 BMI. See body mass index body habitus-related issues gender in, 262 as pitfalls, 261–262 weight as, 262 body mass index (BMI), 262 body tissue, 103 brachial plexus composition of, 242–243 plexopathies of, 242–245 preterminal nerves of, 242–243 cable properties, 60 calcinosis cutis, 299 can’t let-go currents, 291 capacitance, 22 DC circuit and, 33–34 of membrane, 61–62 total membrane, 60 capacitive coupling, 45 capacitive current, 21 capacitive reactance, 36 capacitors, 8. See also RC circuit AC signal in, 36–37 batteries and, 21–22 charge in, 21–23 current and, 20–23 DC signal in, 38–39 defining, 21 in EDX medicine, 20–21 electrons and, 31 filters and, 34, 36 as frequency-dependent resistor, 38 impedance and, 33–34 RC circuit in, 37–39 steady state response of, 33 time constant in, 22, 60 voltage in, 37 capillaries, 66 cardiac injury, 291 carpal tunnel syndrome, 108, 126–127. See also bilateral hand numbness and tingling concomitant, MGA with, 265–266 exercise, 324, 347, 351 EDX impression in, 353–354 EDX study conclusion in, 327, 349, 354 NCS in, 324–326, 347–348, 351–353 needle EMG in, 325–327, 348, 353 cathode, 11, 21–22, 42–43, 273–274 cauda equina, 241 CDC. See Centers for Disease Control

Index

cell membrane, 49 in EDX medicine, 49 electrophysiological properties of, 49–50 ions of, 49–50 TMP and, 60 cells, 49, 52. See also anterior horn cells electrochemical, 11 excitatory, 57 Schwann, 52, 123, 131 Centers for Disease Control (CDC), 296 central nervous system (CNS), 50–52, 76–77 cervical plexus, 242 change in distance, 107, 277 change in time, 34, 107, 277 charge atomic structure in, 4–5 in capacitors, 21–23 conductors, in flow of, 10 conservation of, 36 current, as flow of, 8–10, 21, 43 current, as movement of, 8 in electricity, 4–8 electrostatic, 6–7 negative, of electrons, 6 separation, 22 separation, voltage and, 22 triboelectric effect in, 6 chassis current, 291–292 chemical denervation, 253 chemical force, 54 chemical synapses. See synapses choline, 69. See also acetylcholine choline acetyltransferase, 69 chronic compression, 230–231 chronic inflammatory demyelinating polyradiculoneuropathy (CIDP), 249–250 chronic motor axon loss (CMAL), 320 chronodispersion, 147 CIDP. See chronic inflammatory demyelinating polyradiculoneuropathy circuit breaker, main, 29 closed, 14 DC in, 14, 33–34 EMF in, 13 loop, 18 open, 14–15 parallel DC, 18–20 resistance of, 15–16, 19–20 series, 13–14 series DC, 14–18 short, 15 switches in, 15 voltage in, 10, 17 clean signal, 46

clinical bias, as pitfall, 285 features of fasciculation potentials, 189 of lesions, 214 grading, of lesion severity, 215 information, collected by EDX provider, 309 closed circuit, 14 CMAL. See chronic motor axon loss CMAP. See compound muscle action potential CMRR. See common mode rejection ratio CNS. See central nervous system coaxial cable, 31 collateral sprouting, 183 motor response normalization through, 132–133 reinnervation via, 195–196, 201, 218–220 common mode rejection ratio (CMRR), 47 common signal, 47 compartment syndrome, 232 complete interference pattern, 196 complex repetitive discharges (CRDs), 192–193 compound muscle action potential (CMAP), 95–96, 103–106 H wave as, 144 median, 201 SNAP and, 135 compression injuries, 228–229 acute, 229–230 chronic, 230–231 concentric needle electrodes, 206–207, 277 concentric needle recording electrodes, 179–180 conceptual pitfalls, 282–286 conductance, 19 conduction. See also demyelinating conduction block; demyelinating conduction slowing; nerve; proximal; saltatory conduction; transient axonal conduction block antidromic, orthodromic, 92–93 continuous, 63 ephaptic, 192–193 failure, axon loss and, 131–132 proximal, 147 slowing, focal demyelinating, 126 time, of muscle fibers, 107 values, 118 volume, 90–92 conduction block, 216 pattern, of MGA, 267

transient axonal, 131–132, 216 conduction velocity, 62–64 of forearm, 108, 126–127 latency and, 132 measurement of, 107–108 nerve, 107 in proximal stimulation site, 108 in sensory NCS, 114, 118–119 time of, 107 conductors, 8 copper as, 10 in flow, of charge, 10 metals as, 9 cone, lateral surface area of, 180 connective tissue, 65–66, 224 conservation, of charge, 36 constant current stimulators, 44, 272 constant voltage stimulators, 44, 272 consult. See also the encounter reviewing, 307–308 scheduling, 308 continuous conduction, 63 continuous propagation, 61 contractures, 189 conventional current, 9 cool limbs, studying MUAP of, 260 NCS of, 258–259 needle EMG for, 260 pitfalls of, 257–260 RNSS of, 259–260 cooperative binding, 70 copper, 9–10 correction factors, 259 corticobulbar tract, 50 corticospinal tract, 50 Coulomb’s inverse-square law, 7 CPS. See cycles per second cramp potentials, 188–190 CRDs. See complex repetitive discharges creatine kinase, 78 current. See also alternating current; direct current; Kirchhoff’s current law available, required, 125 capacitive, 21 capacitors and, 20–23 chassis, 291–292 conventional, 9 displacement, 21 dividers, 19 in electricity, 8–10 electron, 9 as flow, of charge, 8–10, 21, 43 intensity of, 293–294 leading source, 91 let-go, can’t let-go, 291 limits, safe, 293

515

Index

current. (cont.) loop, 293 as movement, of charge, 8 Ohm’s law and, 15 pathway, 291 sources, 91–92 stimulators and, 43–44 trailing source, 91 transmembrane, 60 transverse, 62 cutaneous domain, 134 cutoff frequency, 37, 39–40 cycles per second (CPS), 23 cytoplasm, 50 DAC. See digital-to-analog converter dark matter, 4 dB. See decibel DC. See direct current decibel (dB), 26 decreased insertional activity, 184 degeneration. See also Wallerian degeneration distal axon, 131 distal stump, 124 of NMJ, 137 delayed repair, 231 demyelinating conduction block (DMCB), 123–125 acute compression and, 229–230 amplitude discrepancy in, 130 AUC and, 129 axon loss and, 214 chronic compression and, 230 focal, 128–131, 202 lesion, 202 MGA and, 263–264 neurapraxia and, 221–222 proximal lesion and, 147 proximally-located, 201 spiral groove and, 128 stimulation sites, 130 ulnar motor response and, 129 demyelinating conduction slowing (DMCS), 123–126, 214 acute compression and, 229–230 chronic compression and, 230 focal, 126–127, 202 nonuniform, 127–128 proximal lesion and, 147–148 uniform, 126–127 demyelination, 123. See also acquired demyelinating polyneuropathies; focal demyelination focal, 123–126 identifying, 284–285 depolarization. See also endplate potential

516

of membrane, 63 phase, 57 postsynaptic membrane, 71–72 repolarization and, 57–58 by stimulator, 43–44 VGNCs in, 61 depolarization threshold, 54 desired signal, 33, 35, 45–46 diabetic neuropathic cachexia, 420–421 difference, in signal, 46 differential amplifiers, 45–47 differential signal, 87 diffuse numbness and weakness, exercise, 403–404 EDX conclusion in, 406–407 NCS in, 404–405 needle EMG in, 405–406 diffuse upper extremity weakness and sensory loss, exercise, 460 EDX conclusion in, 463 NCS in, 460–461 needle EMG in, 461–462 digital filters, 41 digital sensory NCS, 114 digital signal, 35 digital-to-analog converter (DAC), 41–43, 45 direct current (DC), 14. See also series DC circuits in circuit, 14, 33–34 circuits, parallel, 18–20 signal in capacitors, 38–39 danger of, 290–291 discharge time, 125 discrete interference pattern, 196 disintegration, of motor unit, 169 displacement current, 21 disposable gloves, 296–297 distal axon degeneration, 131 distal axon sprouting. See collateral sprouting distal latency CMAP and, 105–106 measurement of, 105–106 time, 106–107 distal left upper extremity sensory and motor abnormalities, exercise, 442 EDX study conclusion in, 445 NCS in, 442–444 needle EMG in, 444–445 distal muscles, 154 distal sprouting, 220 distal stump, 134 distal stump degeneration, 124 distance, measurement of, 278–279 DMCB. See demyelinating conduction block

DMCS. See demyelinating conduction slowing doping, 42 dorsal root ganglia (DRG), 51, 133–134, 138, 140, 241 drift velocity, 9 duration, of MUAP. See motor unit action potential dysmyelinating disorders, 250–251 E1 electrode. See also interelectrode distance APs propagating toward, 102 misplacement, in sensory NCS, 112 in motor NCS, 126 positive dip of, 97–98 in sensory NCS, 113 E2 electrode improper placement of, 97 in motor NCS, 126 needle for, 98 placement of, 96–97 in sensory NCS, 113 early recruitment, 169, 199–200 early repair, 231 ECF. See extracellular fluid EDB. See extensor digitorum brevis EDX. See electrodiagnostic efflux, 54 Einthoven galvanometer, 85 electric fields, 31 electric power single-phase, 28 three-phase, 28–29 electrical force, 54 electrical injury general guidelines, to avoid, 293 ground in, 292–293 safety issues and, 290–293 electricity charge in, 4–8 current in, 8–10 in EDX, 3 history of, 4 household, 26–30 Ohm’s law in, 12–13 resistance in, 11–12 static, 4 as term, 4 voltage in, 10–11 electrochemical cells, 11 electrochemical force, gradient, 54 electrodes. See also E1 electrode; E2 electrode; motor NCS; needle electrode; needle recording electrodes; surface recording electrodes active, 87–88 adhesive surface recording, 89

Index

ground, 46 history of, 86–87 impedance and, 89, 270 inactive, 87–88 nonadhesive, 89 recording APs and, 90–91 in sensory NCS, 115 reference, 87 stimulating, 93–94 surface stimulating, 88 electrodiagnostic (EDX) abnormalities, with nonorganic lesions, 318–319 electrodiagnostic (EDX) assessment issues, connective tissue and, 224 electrodiagnostic (EDX) examination late responses in, 141 of plexopathies, 241–242 sensory NCS in, 111 timing of, 303–304 transmission, of infection in, 296 electrodiagnostic (EDX) features of axon loss, 248 of fasciculation potentials, 189–190 of lesions, 214 of myotonic potentials, 191 electrodiagnostic (EDX) grading, of lesion severity, 215–216 electrodiagnostic (EDX) laboratory negative AUC in, 105 pathophysiology in, 105 safety issues in, 290 electrodiagnostic (EDX) manifestations, of myopathies, 254 electrodiagnostic (EDX) medicine APs in, 49 capacitors in, 20–21 cell membrane in, 49 electricity in, 3 focal demyelination in, 124 formulas in, 4 voltage in, 10 electrodiagnostic (EDX) practitioner, skin preparation by, 44–45 electrodiagnostic (EDX) provider, 236 clinical information collected by, 309 filters and, 34–35 in peripheral nerve injuries, 213 pitfalls of, 257, 270 RC circuits and, 38 electrodiagnostic (EDX) report, 312–313 electrodiagnostic (EDX) studies. See also consult; the encounter components, order of performance and, 305–306

dependent versus independent, 303 explaining, to patient, 308–309 findings of, 319 of focal neuropathies, 246–247 initial, 319 needle EMG in, 161–162 nerve fiber in, 52 preoperative versus postoperative, 354 on radiculopathies, 240 reduced temperature in, 258–260 signals in, 34–35 supervision of, 306–307 timing of, 216 variation, among EMG laboratories, 305 electrodiagnostic (EDX) technique, RNS as, 152–153 electrodiagnostic (EDX) testing, 33 as extension, of neurological examination, 302 filters in, 34–35 lesion localization in, 317–318 limitations of, 304–305 for peripheral neuropathies, 246 for polyneuropathies, 248 of radiculopathies, 241 for sensory neuronopathies, 251 utility of, 304 electrolyte, 11 electromagnetic field, 31 electromagnetic induction, 23, 45 electromagnetic radiation, 31 electromagnetic waves, 9 electromotive force (EMF), 10 in circuit, 13 potential difference in, 10 voltage and, 12 electromyography (EMG) abnormalities APB and, 318 of FDI, 317 disease, 170–171 electromyography (EMG) laboratory. See also needle EMG antidromic sensory NCS in, 116 EDX study variation among, 305 envelope pattern in, 154 F wave studies in, 148 facilitation in, 145 filters in, 41 H wave amplitude in, 144–145 high frequency RNSS in, 157 latency in, 116 NCS in, 85, 93 normal control values in, 282 sensory NCS in, 113 standard, nonstandard motor NCS in, 101

20:20 rule in, 100 voltage in, 46 electromyography (EMG) machine ADC, DAC in, 42 amplifier for, 45, 47 desired, undesired signal in, 33 isolated power systems and, 292 limitations of, 144 maximum stimulation in, 276 onset, peak latency in, 117 pitfalls of, 276 sensitivity of, 283 stimulator in, 44 time cursors on, 285–286 on time difference, 277–278 vertical resolution of, 283 electron current, 9 electrons, 4–5 capacitors and, 31 movement of, 8–9 negatively charged, 6 Pauli exclusion principle and, 5–6 in static electricity, 7–8 electrostatic charge, 6–7 electrostatic force, 31 protons and, 7 strength of, 7 electrotonic properties, 60 elements, 4 EMF. See electromotive force EMG. See electromyography the encounter activities occurring after, 312 activities occurring during limbs studied in, 309 NCS performed first, 309 needle EMG studies in, 309–312 activities occurring immediately before, 309 consult before reviewing, 307–308 scheduling, 308 verbal informed consent in, 308–309 endoneurial tube, 65 endoneurium, 65 endplate, 68, 70. See also miniature endplate potential activity, in needle EMG, 171 region, 95 spikes, 172 endplate potential (EPP), 71–72. See also miniature endplate potential end-to-end repair, 226 end-to-end suturing, 226–227 envelope pattern, 154 ephaptic conduction, 192–193 epineurial repair, 227

517

Index

epineurium, 65–66 EPP. See endplate potential EPSPs. See excitatory postsynaptic potentials equilibrium potential, 55–56. See also Nernst equation excessive stimulation, 275–276 excitation-contraction coupling AP propagation in, 74 sarcomere in, 74–76 sarcoplasmic reticulum in, 74 excitatory cells, 57 excitatory postsynaptic potentials (EPSPs), 61, 165 exhaustion, postexercise, 155–156 exocytosis, 69 extensor digitorum brevis (EDB), 285 extracellular fluid (ECF), 53, 56–57 extrafusal muscle fiber innervation, 76 extrafusal muscle fibers, 53 F wave, 146 APs of, 146 latency, 146–148 minimal latency of, 147 physiology, technique of, 146–147 repeater, 147 studies, in EMG laboratories, 148 utility of, 147–148 facilitation in EMG laboratory, 145 postexercise, 155–156 farad, 22 fascicles, 65–66 fascicular repair, 227 fascicular sparing, 318 fasciculation potentials, 188–189 clinical features of, 189 EDX features of, 189–190 grading of, 190 myogenic, 189 sites of origins of, 189 waveform morphology of, 189–190 FDA. See Food and Drug Administration FDI. See first dorsal interosseous muscle FDP. See flexor digitorum profundus fibrillation potentials, 185. See also needle EMG amplitude of, 185 duration, morphology of, 185–187 firing frequency of, 186–187 insertional positive sharp waves in, 188 of myopathies, 254–255 positive sharp wave form of, 186, 188 quantification of, 187 specificity, utility of, 187–188

518

spike form of, 185–186 time to appearance of, 187 filtering high frequency, 280–281 issues, 279–281 low frequency, 280 in sensory NCS, 119–120 filters, 33 analog, 40–41 arrangements of, 40–41 bandpass, 40 capacitors and, 34, 36 cutoff frequency for, 39–40 digital, 41 EDX providers and, 34–35 in EMG laboratory, 41 high pass, 38, 40 low pass, 38, 40 notch, 40–41 RC circuits in, 35–36 stopband, 40 transition band in, 39 final common path, 76 firing frequency of fibrillation potentials, 186–187 of motor units, 165 of MUAP, 165–168 first dorsal interosseous (FDI) muscle EMG abnormalities of, 317 MGA to, 264–267 fixed distances, landmarks, 116–117 flashover, 28 flexor digitorum profundus (FDP), 317 flow of charge, 8–10, 21, 43 hole, 9 ionic, 53–54, 56 of water, 12–14, 18–19 FM. See frequency modulation focal axon loss, 124, 202 focal demyelination, 123–124. See also demyelinating conduction slowing conduction, 126 in NCS, 137–138 pathology of, 125–126 in sensory NCS, 134 focal DMCB, 128–131, 202 focal DMCS, 126–127, 202 focal neuropathies, 246–247 Food and Drug Administration (FDA), 290 force generation, motor unit, 79 forearm anastomosis in, 268 conduction velocity, 108, 126–127 formulas, in EDX medicine, 4 frequencies bandwidth of, 35

cutoff, 37, 39–40 ideal sampling, 283 onset, 165–166 recruitment of, 165–166 frequency modulation (FM), 40 frequency-dependent resistors, 38 frequency-independent resistors, 36 full interference pattern, 168 fusimotor neurons, 76 gain on amplifiers, 44 sensitivity and, 282 Galvani, Luigi, 9–10, 76 galvanism, 9 gamma gain, 77 gases, 9 Gasser, Herbert, 62 gated ion channels, 50, 54, 57–58 gender, 262 generalized low motor normal sensory (GLMNS), 138 generalized weakness. See also diabetic neuropathic cachexia exercise, 415–416 diagnosis in, 420–421 NCS in, 416–418 needle EMG in, 418–420 GLMNS. See generalized low motor normal sensory gloves, disposable, 296–297 Goldman-Hodgkin-Katz voltage equation, 56 GRDs. See grouped repetitive discharges grid, 27, 42 ground. See also iso-grounds in electrical injury, 292–293 electrode, 46 loop current, 293 grouped repetitive discharges (GRDs), 192–193 H band, 75 H reflex study, 142 supine patient in, 142 technique of, 142–143 tibial nerve stimulation in, 142–144 H wave, 142 amplitude abnormal, 145 in EMG laboratory, 144–145 latency measurements of, 144–145 as CMAP, 144 testing technical errors in, 143–144 utility of, 145–146 hand muscle innervation patterns, 269

Index

accessory deep peroneal nerve in, 269–270 Berretini anastomosis in, 269 head, 75 Helmholtz, Herman von, 85 Henneman size principle, 164 hereditary dysmyelination, 250–251 hereditary polyneuropathies, 250–251 hertz (Hz), 23 high frequency filtering, 280–281 high frequency RNSS. See repetitive nerve stimulation studies high pass filter, 38, 40 hinge, 75 hole flow, 9 horizontal resolution, 283 household electricity, 26–30 hyperpolarization, 57–58 hypothenar eminence, MGA to, 263–264 Hz. See hertz iatrogenic pneumothorax, 298–299 ICF. See intracellular fluid ideal sampling frequencies, 283 immediate release pool, 69 impedance, 23 in capacitors, 33–34 electrodes and, 89, 270 mismatch, 47, 270, 272 between signal source, surface recording electrode, 286 skin, 273 inactivation gate, 258 inactive electrode, 87–88. See also E2 electrode incomplete interference pattern, 196–197 increased insertional activity, 184 inductive coupling, 30 inductors, 23–24 indwelling medical devices NCS with, 294–295 RNSS with, 295 safety issues of, 293–295 inert atoms, 8 infection, transmission of disposable gloves, for preventing, 296–297 in EDX examination, 296 needle electrodes in, 297–298 safety issues of, 296–298 skin in, 298 influx, 54 infraclavicular plexus, 243–244 inhibitory postsynaptic potentials (IPSPs), 61 initial axon segment, 183 initial positive phase, 91

innervation. See also hand muscle innervation patterns anomalous, 262 extrafusal muscle fiber, 76 ratio, 51, 73, 195, 219 zones, 103 insertional activity decreased, 184 increased, 184 in needle EMG, 162, 170–171, 184, 235 insertional positive sharp waves, 188 insulators, 8–9 intensity, of currents, 293–294 interconnections, 27 interelectrode distance, in sensory NCS, 120–121 internal longitudinal current, 63 internal neurolysis, 231 internodal distance, 63 internodal membrane segments, 63–64 internodes, 52 intervertebral foramina, 51 intracellular fluid (ICF), 53, 56–57 intrafusal muscle fibers, 53, 76 intraspinal canal disorders, 51 acute poliomyelitis as, 237 AHC disorders and, 236–237 ALS as, 238 background anatomy of, 236 Kennedy disease as, 239 NCS in, 236 post-poliomyelitis syndrome as, 237–238 radiculopathies as, 239–241 SMA-3 as, 238–239 intussusception, 214 inverting amplifier, 46 ion channels, 53–55 ionic current, 53–54 ionic flow, 53–54 ionic flux, 53–54 ionotropic receptors, 68 ions, 8. See also Nernst equation of cell membrane, 49–50 flow of, 54, 56 movement of, 56–57 IPSPs. See inhibitory postsynaptic potentials ischemic injuries, 231 iso-grounds, 46 isolated electrical myotonia, 191 isolated power systems, 292 isotopes, 5 isotropic bands, 75 jiggle, 201, 206 jitter, 200, 205

concentric needle electrode for, 206–207 potential, of muscle fiber APs, 205 value, of muscle fiber APs, 205–206 juxtaparanode, 52 Kennedy disease, 239 Keraunoparalysis, 290 Kirchhoff’s current law, 14–15, 36 Kirchhoff’s voltage law, 14 Lambert test, 157–158 Lambert-Eaton myasthenic syndrome (LEMS) background of, 253 needle EMG study of, 253–254 as NMJ disorder, 253–254 RNSS for, 253 routine NCS in, 253 late fibbers, 187 late responses, in EDX examination, 141 latency. See also onset latency conduction velocity and, 132 in EMG laboratories, 116 of F wave, 146–148 measurements, of H wave, 144–145 minimal, 147 peak, 112, 117–118 proximal, 105 in sensory NCS, 116 values, 118 lateral surface area, of cone, 180 LCDD. See light-chain deposition disease lead-acid automotive battery, 11 leading source current, 91 leading-off surface, 179 leading-off surface area, of needle electrodes, 204 leadwire, 45–46 least resistance, 18 left common peroneal neuropathy, 376–378 left foot numbness and burning, exercise, 378 EDX study conclusion in, 381 NCS in, 378–380 needle EMG in, 380–381 left hand atrophy, exercise, 457 EDX conclusion in, 460 NCS in, 457–459 needle EMG in, 459–460 left hip pain, exercise, 335 EMG conclusion in, 338 NCS in, 335–337 needle EMG in, 337–338

519

Index

left upper extremity pain, tingling, and weakness, exercise, 439 EDX examination conclusion in, 442 NCS in, 439–441 needle EMG in, 441–442 left wrist drop, exercise, 455 EDX conclusion in, 457 NCS in, 455–456 needle EMG in, 456–457 LEMS. See Lambert-Eaton myasthenic syndrome length constant, 62, 125 length-dependent polyneuropathies, 247 Lenz’s law, 23–24 lesion acuteness, 214–215 axon loss and, 215–216 clinical, EDX features of, 214 completeness, 225–226 DMCB, 202 localization in EDX testing, 317–318 in NCS, 137, 140 MUAP and, 202 nonorganic, 318–319 postganglionic, 140 proximal, 147–148 in sensory NCS, 134 severity clinical grading of, 215 EDX grading of, 215–216 motor response value and, 217–218 in needle EMG studies, 218 sensory response value and, 218 site, Wallerian degeneration at, 137–138 stimulation above, 131 stimulation below, 131 let-go currents, 291 Leyden jar, 4 ligand-gated channels, 54, 68 light-chain deposition disease (LCDD), 397 lightning, 8 limbs, studied, in NCS, 309. See also cool limbs, studying linear phase characteristic, 40–41 liquids, 9 LMN. See lower motor neurons load, 15, 18 local potentials, 61 local responses, 59–61 localization. See also lesion point, 317 of postganglionic lesions, 140 regional, 317 long duration MUAPs, 248

520

longitudinal tubules, 74 loop circuit, 18 current, 293 recorder, 294 low frequency filtering, 280 low frequency RNSS. See repetitive nerve stimulation studies low pass filter, 38, 40 low voltage, 29 lower back pain, exercise, 322, 367, 398 EDX study conclusion in, 323, 370 EMG conclusion in, 400 NCS in, 322–323, 368, 398–399 needle EMG in, 323, 368–369, 399–400 lower motor neurons (LMN), 50, 76 lower plexus, 243–244 lumbosacral plexus NCS of, 245 needle EMG study of, 245–246 plexopathies of, 245–246 M wave, 142, 276 machine maximum stimulation, 276 macro EMG, 204, 207–208 macro MUAP, 208 macroshock, 291 magnetic field, 26–28, 31 main circuit breaker, 29 main reserve, 69 maintenance of position, 235–236 major surgical interventions, 226–227 Martin-Gruber anastomosis (MGA), 262–263. See also anastomosis to adductor pollicis muscle, 264–265 atypically proximal, 267 with concomitant carpal tunnel syndrome, 265–266 with concomitant ulnar neuropathy, 266–267 conduction block pattern of, 267 DMCB and, 263–264 to FDI, 264–267 to hypothenar eminence, 263–264 proximal median motor response and, 266–267 unrecognizable, 268 mass number, 5 maximal temporal recruitment, 168–169 maximum response, to stimulus, 93 MBFC. See medial brachial fascial compartment syndrome MCD. See mean consecutive difference mean consecutive difference (MCD), 205–207 measurement of amplitude, 102–104

of conduction velocity, 107–108 of distal latency, 105–106 of distance, 278–279 errors, 277–279, 285–286 of motor response, 101–102 MUAP, 164 of muscle fibers, 101–102 needle EMG, 169–170 of negative AUC, 104–105 of negative phase duration, 109 in sensory NCS, 112–113 medial brachial fascial compartment (MBFC) syndrome, 478–479 median CMAP, 201 median-to-ulnar anastomosis, 262 Medical Research Council (MRC), 215 Medicare, 306 membrane, 54–55. See also ion channels; postsynaptic; presynaptic; total membrane capacitance; transmembrane current; transmembrane potential capacitance of, 61–62 depolarization of, 63 MEPP. See miniature endplate potential metals, as conductors, 9 MGA. See Martin-Gruber anastomosis microshock, 291 MIDD. See monoclonal immunoglobulin deposition disease middle plexus, 243–244 miniature endplate potential (MEPP), 71, 152, 171–172 minimal latency, of F waves, 147 missing relative abnormalities, 284 mixed NCS, 93, 121–122 mixed polyneuropathies, 250 mixed signal, 37–38 M-line, 75 MMV. See moment-to-moment variability mobilization store, 69 molecules, amphiphilic, 49 moment-to-moment variability (MMV), 178, 200–201 monoclonal immunoglobulin deposition disease (MIDD), 397 mononeuropathies. See focal monopolar amplifiers, 44–45 monopolar needle electrodes, 180 motor axon amplitude of, 103 loss, 201 motor NCS, 85, 92, 95 abbreviations for, 321

Index

advantages, disadvantages of, 132–133 amplitude of, 100 E1, E2 electrodes in, 126 exercise, for sensorimotor dysfunction, 354–355 focal demyelination and, 125–126 needle EMG and, 138–140, 161–162, 201 nerve stimulation in, 98–99 PNS elements assessed by, 138–140 proximal tibial motor response to, 100 radiculopathies and, 239–240 recording electrodes for, 95 belly-tendon method and, 95–96 biphasic morphology of, 95–96 standard, nonstandard, 101 temporal dispersion and, 131 motor nerve branching, 51 motor nerves, 85 motor neurons, 50–51, 146 motor point, of muscle, 95, 270–271 motor response amplitude, 104, 135, 216 in blink reflexes, 149 low, nonpathological causes of, 202 measurement of, 101–102 normalization, through collateral sprouting, 132–133 from spinal cord level, 138 ulnar, 96–97, 129 value of, lesion severity and, 217–218 motor unit, 50, 124–125. See also nonmyopathic motor unit disintegration disorders AHC and, 79–80, 162–163 anatomy, physiology, 162–163 disintegration disorders, 183 disintegration of, 169 firing frequency of, 165 force generation, 79 of muscle fibers, 73, 78–79 recruitment, 164–169 size, distribution of, 79–80 types of, 78 motor unit action potential (MUAP), 183 amplitude, 165, 173–176, 195 assessment activation phase of, 193–194 in needle EMG, 162–163, 173, 183–184 auditory characteristics of, 161 of cool limbs, 260 duration, 163–164, 175–177, 194–195, 240, 261, 281–282, 284 firing frequency of, 165–168 full interference pattern in, 168 GRD of, 192

lesions and, 202 long duration, 248 macro, 208 maximal temporal recruitment in, 168–169 measurements, 164 MMV of, 178, 200–201 morphology, grading, 320 muscle fiber APs and, 132 of myopathies, 254 neurogenic, 130–131, 176, 218, 243–245 neurogenic recruitment in, 168–169, 197–198, 243–245 phases of, 177–178, 195 recording, 179–180 recruitment, 196–197 in ALS, 238 grading, 320 in needle EMG, 165–166 pattern, 162, 169 ratio of, 166–167 in UMN disorders, 235–236 reinnervation via collateral sprouting in, 195–196 reinnervation via proximodistal axon advancement, 196 satellite potential and, 196 stability, 178, 200–201 of trapezius muscle, 299 trigger line of, 176 turns of, 177–178, 195 voluntary activity in, 193–194 movement of charge, 8 of electrons, 8–9 of ions, 56–57 MRC. See Medical Research Council MUAP. See motor unit action potential multipennate muscle, 80 muscle. See also first dorsal interosseous muscle; hand muscle innervation patterns; neural control, of muscle activation time, 107 adductor pollicis, 264–265 bipennate, 80 contraction, 78–79 distal, 154 innervation ratios, 51, 73, 219 lengthening, 77 motor point of, 95, 270–271 multipennate, 80 in needle EMG, 162 shortening, 77–78 spindles anatomy of, 77 physiology of, 77–78 structural organization of, 73–74

studied, in needle EMG, 310 tissue transit time, 107 tone, 77 trapezius, 299 unipennate, 80 muscle fiber APs, 74, 99, 102–103, 132, 173–174 concentric needle electrodes for, 206–207 jitter value of, 205–206 jittering potential of, 205 SFEMG for, 204–205 triggering potential of, 205 muscle fibers, 76 ATP of, 78 conduction time, 107 density of, 175–176 extrafusal, 53 intermingling, 51 intrafusal, 53, 76 measurement of, 101–102 motor units of, 73, 78–79 NMJ of, 80 primary, secondary endings in, 77 proximity of, 175–176 types of, 79 muscle-specific kinase (MuSK), 70 myasthenia gravis, 152 AChR in, 252 background of, 252 needle EMG study of, 253 as NMJ disorder, 252–253 RNSS in, 252–253 routine NCS in, 252 myelin, 52–53. See also demyelination; remyelination advantage of, 61–62 in AP propagation, 61–63 sheath, 123, 125 myofibrils, 74 myogenic fasciculation potentials, 189 myokymia, 192–193 myopathies, 132, 199–200 EDX manifestations of, 254 fibrillation potentials of, 254–255 MUAP of, 254 needle EMG and, 254 myoplasm, 73 myosin, 74–75 myotome, spinal cord and, 138 myotonia, 190–191. See also neuromyotonia myotonic potentials, 190–191 EDX features of, 191 in needle EMG studies, 191 train of, 191 NA. See neuralgic amyotrophy National Electrical Code, 29

521

Index

NCS. See nerve conduction studies neck pain, exercise, 349, 364, 370, 432–433 EDX study conclusion in, 373–374, 390, 435–436 EMG conclusion in, 351, 367 NCS in, 349–350, 364–366, 370–372, 387–389, 433–434 needle EMG in, 350–351, 366–367, 372–373, 389–390, 434–435 needle electrode, 98 concentric, 206–207, 277 jitter and, 206–207 leading-off surface dimensions of, 204 macro EMG, 207–208 monopolar, 180 recording surface, 277 single-fiber, 204–205 in transmission, of infection, 297–298 needle EMG. See also electromyography abbreviations for, 321 activation phase of, 173, 235 age-related issues in, 261 bleeding in, 295–296 for cool limbs, 260 in EDX studies, 161–162 endplate activity in, 171 spikes in, 172 exercise for bilateral hand numbness, tingling, 334, 361 for bilateral lower extremity weakness, 429–430 for bilateral upper extremity weakness, 411–412, 422–424 for carpal tunnel syndrome, 325–327, 348, 353 for diffuse numbness, weakness, 405–406 for diffuse upper extremity weakness, sensory loss, 461–462 for distal left upper extremity sensory, motor abnormalities, 444–445 for generalized weakness, 418–420 for left foot numbness, burning, 380–381 for left hand atrophy, 459–460 for left hip pain, 337–338 for left upper extremity pain, tingling, weakness, 441–442 for left wrist drop, 456–457 for lower back pain, 323, 368–369, 399–400 for neck pain, 350–351, 366–367, 372–373, 389–390, 434–435

522

for numbness, 402–403, 447–448 for pacemaker placement, 454 for peroneal neuropathy, 376 for radiculopathies, 385–386 for right calf atrophy, 392–393 for right foot drop, 331–332 for right hand numbness, tingling, 408–409 for right upper extremity weakness, 431–432 for sciatic neuropathy, 343–347 for sensorimotor dysfunction, 356–359 for shoulder fracture-dislocation, 328–329 for symmetric weakness, 395–397, 413–415, 426–428 for ulnar neuropathy, 341, 383 for upper extremity weakness, wasting, 437–438 for weakness and numbness, 450–451, 464–465, 467–468, 470–471, 474–475, 477–478 findings, range of, 320 history of, 161 insertional activity in, 162, 170–171, 184, 235 on Kennedy disease, 239 manifestations, of various pathophysiologies, 202 measurements, 169–170 MEPP in, 171–172 motor NCS and, 138–140, 161–162, 201 MUAP assessment in, 162–163, 173 183–184 duration in, 175–177, 284 recruitment in, 165–166 stability in, 178 muscles in, 162, 310 myopathies and, 254 myotonic potentials in, 191 pitfalls of, 281–282 provoked activity in, 170, 184 in radiculopathies, 240 resting phase of, 171, 184–185 spontaneous activity in, 162, 170–171, 184–185, 202 study of axon loss, 215–216 of calcinosis cutis, 299 in encounter, 309–312 on hereditary polyneuropathies, 251 of iatrogenic pneumothorax, pneumoperitoneum, 298 of LEMS, 253–254 in lesion severity assessment, 218

lessening discomfort associated with, 310–312 of lumbosacral plexus, 245–246 of myasthenia gravis, 253 of plexopathies, 242–245 in UMN disorders, 235 voluntary activity in, 162, 170, 173 needle recording electrodes, 178–179 concentric, 179–180 dimensions, areas of, 179 monopolar, 180 negative AUC amplitude and, 104–105, 132 in EDX laboratory, 105 measurement of, 104–105 negative charge, of electrons, 6 negative peak voltage, 25 negative phase duration, 109 negative sink, 91–92 Nernst equation, 55–56 nerve. See also sensory nerve action potential accessory deep peroneal, 269–270 activation time, 106 conduction time, 106–107 velocity, 107 motor, 51, 85 preterminal, 242–243 transfer, 227 trunk, 65–66, 228 nerve conduction studies (NCS). See also motor NCS; sensory NCS; surface recording electrodes abnormalities, 318 age-related issues in, 260–261 axon disruption in, 137–138 bipolar, referential recordings of, 88 concepts, principles of, 85–86 of cool limbs, 258–259 in EMG laboratory, 85, 93 exercise for bilateral hand numbness, tingling, 332–334, 359–361 for bilateral lower extremity weakness, 429 for bilateral upper extremity weakness, 410–411, 421–422 for carpal tunnel syndrome, 324–326, 347–348, 351–353 for diffuse numbness, weakness, 404–405 for diffuse upper extremity weakness, sensory loss, 460–461 for distal left upper extremity sensory, motor abnormalities, 442–444 for generalized weakness, 416–418

Index

for left foot numbness, burning, 378–380 for left hand atrophy, 457–459 for left hip pain, 335–337 for left upper extremity pain, tingling, weakness, 439–441 for left wrist drop, 455–456 for lower back pain, 322–323, 368, 398–399 for neck pain, 349–350, 364–366, 370–372, 387–389, 433–434 for numbness, 401–402, 445–447 or bilateral lower extremity weakness, 429 for pacemaker placement, 452–454 for peroneal neuropathy, 374–375 for radiculopathies, 384–385 for right calf atrophy, 391–392 for right foot drop, 330–331 for right hand numbness, tingling, 407–408 for right upper extremity weakness, 431 for sciatic neuropathy, 342–345 for sensorimotor dysfunction, 355–356, 358 for shoulder fracture-dislocation, 327–328 for symmetric weakness, 394–395, 412–413, 425–426 for ulnar neuropathy, 339–340, 381–382 for upper extremity weakness, wasting, 436–437 for weakness and numbness, 448–450, 463–464, 466–467, 469–470, 472–477 explaining, to patient, 308–309 findings of, 319 focal demyelination in, 137–138 focal DMCB and, 129 focal DMCS, 126 with indwelling medical devices, 294–295 in intraspinal canal disorders, 236 in lesion localization, 137, 140 limbs studied in, 309 of lumbosacral plexus, 245 manifestations, timing of, 135–136 mixed, 93, 121–122 performed first, during encounter, 309 pitfalls of, 281 routine, in LEMS, 253 routine, in myasthenia gravis, 252 surface recording electrodes in, 87 technician, 306 in UMN, 235

nerve fiber, 52–53, 99–100, 111. See also fascicles; myelin nerve injuries connective tissue in, 224 Seddon classification system for axonotmesis in, 222 neurapraxia in, 221–222 neurotmesis in, 222 Sunderland classification system for, 222 grade 1, 222 grade 2, 222–223 grade 3, 223 grade 4, 223 grade 5, 223–224 grade 6, 224 surgical intervention in, 224–225 types of, 227–228 compartment syndrome, 232 compression injuries, 228–231 ischemic injuries, 231 stretch injuries, 228 traction injuries, 228 transection injuries, 231 nerve stimulation in motor NCS, 98–99 nonuniform DMCS and, 128 proximal median motor response to, 99 tibial, 142–144 neural control, of muscle, 76 CNS influence in, 76–77 final common path in, 76 muscle spindles in, 77 neuralgic amyotrophy (NA), 439 neurapraxia, 221–222 neurogenic MUAP, 130–131, 176, 218, 243–245 neurogenic recruitment grades of, 320 grading, 197–198 MUAP in, 168–169, 197–198 243–245 as pathological, 319 neurological examination, 302 neuromuscular junctions (NMJ), 68–71 degeneration, 137 disorders, 152, 251–252 LEMS as, 253–254 myasthenia gravis as, 252–253 in high frequency RNSS, 157 of muscle fibers, 80 transmission, 200, 205 disorders, SFEMG for, 206 safety factor of, 153 neuromyotonia, 191–192 neuronopathies, sensory, 251 neurons, 50 fusimotor, 76

motor, 50–51, 146 resting, 57 sensory, 51–52 skeletofusimotor, 76 skeletomotor, 76 neurotization, 227 neurotmesis, 222 neurotonic discharge, 192 neurotransmitters, 68 neutrons, 4–5 NMJ transmission time, 107 NMJs. See neuromuscular junctions nodal intussusception, 230 nodal membrane segments, 63–64 nodes, of Ranvier, 52 noise, 35 nonadhesive surface recording electrodes, 89 non-dissipative opposition, 23 nongated ion channels, 54, 57–58 non-inverting amplifiers, 46 nonlinear phase characteristic, 40–41 non-myopathic motor unit disintegration disorders, 199–200 nonorganic lesions, 318–319 nonpathological causes, of low motor responses, 202 nonuniform DMCS, 127–128 normal control values, in EMG laboratory, 282 notch filter, 40–41 numbness. See also diffuse numbness and weakness; left foot numbness and burning; right hand numbness and tingling; weakness and numbness exercise, 400, 445 EDX conclusion in, 403, 448 NCS in, 401–402, 445–447 needle EMG in, 402–403, 447–448 Occupational Safety and Health Administration (OSHA), 290, 296 Ohm, Georg Simon, 12 Ohm’s law on AC, 23 current and, 15 in electricity, 12–13 reactance, resistance and, 33 voltage and, 15 onset frequency, 165–166 onset latency, 105–106, 112 changing sensitivity, to better identify, 282–283 peak latency and, 117–118 in SNAP, 116 open circuit, 14–15

523

Index

orthodromic conduction, 92–93 orthodromic sensory NCS, 111, 114–116 OSHA. See Occupational Safety and Health Administration output voltage, 17 paced fiber, 193 pacemaker placement, exercise, 452 EDX conclusion in, 454 NCS in, 452–454 needle EMG in, 454 pacing fiber, 193 parallel circuits, 13–14 parallel DC circuits, 18–20 parallel resistors, 19–20 pathophysiology, 105, 133–134, 202, 214–215 Pauli exclusion principle, 5–6 peak latency, 112, 117–118 peak voltage, 25, 37 peak-to-peak amplitude, 113, 173–175 voltage, 25 pennate muscle, 80 perineurium, 65 period, of wave, 24–25 peripheral nerve injuries demographics of, 213 proper initial management of, 213 types of, 213 peripheral nervous system (PNS), 50–52, 138–140 peripheral neuropathies acquired axon loss polyneuropathies as, 247–248 EDX testing for, 246 focal, 246–247 types of, 246 peroneal neuropathy. See also accessory deep peroneal nerve deep, 364 exercise, 361–363, 374 EDX conclusions of, 364, 376–378 NCS in, 374–375 needle EMG in, 376 left common, 376–378 right superficial, 364 peroneal-EDB, 285 phases, of MUAP, 177–178, 195 phosphocreatine, 78 phospholipid layers, 49 photons, 7, 9 physiologic temporal dispersion, 113–114, 131 Piper, H., 85 pitfalls age-related issues, 260–261 body habitus-related issues, 261–262

524

clinical biases as, 285 conceptual, 282–286 of EDX providers, 257, 270 of EMG machine, 276 impedance, between signal source, surface recording electrode as, 286 of NCS, 281 of needle EMG, 281–282 of recording electrodes, 270–271 of stimulation, 271–281 studying cool limbs, 257–260 technical, 270 time constraints as, 284 plexopathies of brachial plexus, 242–245 of cervical plexus, 242 EDX examination of, 241–242 of lumbosacral plexus, 245–246 needle EMG studies of, 242–245 neurogenic MUAP in, 243–245 sensory NCS in, 241–242 pneumoperitoneum, 298–299 PNS. See peripheral nervous system point localization, 317 polyneuropathies acquired axon loss, 247–248 acquired demyelinating, 248–249 AIDP as, 249 CIDP as, 249–250 EDX testing for, 248 hereditary, 250–251 length-dependent, 247 mixed, 250 sensory NCS in, 248 positive dip, of E1 electrode, 97–98, 270–271 positive sharp wave form of fibrillation potential, 186, 188 insertional, 188 postexercise exhaustion, 155–156 postexercise facilitation, 155–156 postganglionic lesions, 140 post-poliomyelitis syndrome, 237–238 postsynaptic folds, 70 postsynaptic membrane, 68 postsynaptic membrane depolarizations, 71–72 postsynaptic region, 70–71 potential difference, 10 power loss equation, 29–30 power stroke, 75 preamplifier, 87 precautions, of OSHA, 296 prepatterning, 70 presynaptic membrane, 68 presynaptic region, 69 preterminal nerves, of brachial plexus, 242–243

primary endings, in muscle fibers, 77 primary synaptic cleft, 70 progressive proximal spinal and bulbar muscular atrophy. See Kennedy disease propagation. See also action potential propagation bidirectional, 74 continuous, 61 protons, 4–5, 7 provoked activity, in needle EMG, 170, 184 proximal conduction time, 147 proximal latency, 105 proximal lesion, 147–148 proximal median motor response MGA and, 266–267 to nerve stimulation, 99 proximal stimulation site, 108, 111–112 proximal tibial motor response, to motor NCS, 100 proximally-located DMCB, 201 proximodistal axon advancement, 183, 196, 220 proximodistal axon regeneration, 219–220 pseudofacilitation, 155–157 quantal content, 69, 71 quantification, of fibrillation potentials, 187 quantum, 69 quarks, 5 radiculopathies, 134. See also neck pain EDX study of, 240 EDX testing of, 241 exercise NCS in, 384–385 needle EMG in, 385–386 as intraspinal canal disorders, 239–241 motor NCS and, 239–240 MUAP duration in, 240 needle EMG in, 240 raindrops striking a tin roof, 186–187 rapsyn, 70 RC circuits capacitors in, 37–39 EDX providers and, 38 filters in, 35–36 mixed signal in, 37–38 RCA. See Riche-Cannieu anastomosis reactance, 23 capacitive, 36 Ohm’s law and, 33 reciprocals, 19

Index

recording electrodes. See also electrodes; motor NCS; needle recording electrodes; surface recording electrodes APs of, 90–91 for MUAP, 179–180 pitfalls of, 270–271 in sensory NCS, 115 recruitment. See also motor unit action potential; neurogenic recruitment early, 169, 199–200 frequency, 165–166 maximal temporal, 168–169 motor unit, 164–169 ratio, 166, 168 full interference pattern in, 168 rule of fives in, 166–167 spatial, temporal, 169, 198–199 UMN, 198–199 red flags, 248 reduced interference pattern, 197 reduced temperature, in EDX study, 258–260 reference electrode, 87 referential recordings, 88 regenerative distance, 225–226 regional localization, 317 reinnervation, 183, 194–195 aberrant, 223 via collateral sprouting, 195–196, 201, 218–220 determining potential for, 220–221 mechanisms of, 218, 220 via proximodistal axon advancement, 196, 220 via proximodistal axon regeneration, 219–220 sensory receptor, 220 relative refractory period, 58–59 relatively abnormal, 284 Remak bundle, 52 remyelination, 221 repeater F waves, 147 repetitive nerve stimulation (RNS), 152–153 repetitive nerve stimulation studies (RNSS), 152–153 of cool limbs, 259–260 high frequency, 157 in EMG laboratories, 157 Lambert test in, 157–158 in NMJ transmission disorders, 157 pseudofacilitation in, 157 technique, 157–158 with indwelling medical devices, 295 for LEMS, 253 low frequency, 153

abnormal, 156 baseline train in, 154 distal muscles in, 154 postexercise facilitation, exhaustion in, 155–156 pseudofacilitation in, 155–156 technical errors in, 156–157 technique of, 154–155 for myasthenia gravis, 252–253 repolarization absolute, relative refractory period in, 58–59 depolarization and, 57–58 phase, 57 resistance, 23 amplifier, 44 of circuit, 15–16, 19–20 in electricity, 11–12 least, 18 Ohm’s law and, 33 of skin, 44, 293 of tissue, 291 total membrane, 60 voltage and, 15–16, 18 resistors, 15–17. See also RC circuit frequency-dependent, 38 frequency-independent, 36 parallel, 19–20 response dispersion, 109 resting membrane potential (RMP), 43, 55–56 resting neuron, 57 resting phase, of needle EMG, 171, 184–185 reversals in AC, 23, 36 stimulator, 279 ribbon pair leadwire, 46 Riche-Cannieu anastomosis (RCA), 268–269 right calf atrophy, exercise, 390 EDX study conclusion in, 394 NCS in, 391–392 needle EMG for, 392–393 right foot drop, exercise, 329–330 EDX study conclusion in, 332 NCS in, 330–331 needle EMG in, 331–332 right hand numbness and tingling, exercise, 407 EDX study conclusion in, 409 NCS in, 407–408 needle EMG in, 408–409 right superficial peroneal neuropathy, 364 right upper extremity weakness, exercise, 430–431 EDX study conclusion in, 432

NCS in, 431 needle EMG in, 431–432 rigor mortis, 76 RMP. See resting membrane potential RMS. See root-mean-square voltage value RNS. See repetitive nerve stimulation RNSS. See repetitive nerve stimulation studies roll-off slope, 39 root-mean-square (RMS) voltage value, 25–26 Rule of 20, 225 rule of fives, in recruitment ratio, 166–167 safe current limits, 293 safety factor, 71 of AP propagation, 125–126 of NMJ transmission, 153 safety issues of bleeding, 295–296 of calcinosis cutis, 299 in EDX laboratory, 290 electrical injury and, 290–293 of iatrogenic pneumothorax, pneumoperitoneum, 298–299 of indwelling medical devices, 293–295 of transmission, of infection, 296–298 saltatory conduction, 61, 63–64, 123 sarcolemma, 73–74 sarcomeres, 74 sarcoplasm, 73 sarcoplasmic reticulum, 74 satellite potential, MUAP and, 196 Schwann cells, 52, 123–124, 131 sciatic neuropathy. See also right foot drop exercise, 342 EDX study conclusion in, 344 NCS in, 342–345 needle EMG in, 343–347 secondary clefts, 70 secondary endings, in muscle fibers, 77 Seddon classification system axonotmesis in, 222 neurapraxia in, 221–222 neurotmesis in, 222 selective ionic permeability, 53, 56 semiconductors, 8–9 sensitivity changing, to better identify onset latency, 282–283 of EMG machines, 283 gain and, 282 sweep speed and, 283–284

525

Index

sensorimotor dysfunction, exercise, 354 EDX study conclusion in, 357 motor NCS in, 354–355 NCS in, 355–356, 358 needle EMG in, 356–359 sensory domains, 134 sensory NCS, 92–93 abbreviations for, 320–321 advantages, disadvantages of, 119, 135 amplitude in, 113–114 antidromic, orthodromic, 111, 114–116 conduction velocity in, 114, 118–119 digital, 114 in DRG, 138 E1, E2 electrodes in, 113 E1 electrode misplacement in, 112 in EDX examination, 111 in EMG laboratories, 113 filtering in, 119–120 focal demyelination in, 134 interelectrode distance in, 120–121 latency in, 116 lesions in, 134 in localization, of postganglionic lesions, 140 manifestations, of pathology, pathophysiology, 133–134 measurement in, 112–113 in plexopathies, 241–242 on PNS, 139–140 in polyneuropathies, 248 proximal stimulation site in, 111–112 recording electrodes in, 115 technique, 111–112 sensory nerve action potential (SNAP), 111, 116, 135 sensory nerve fibers, 52–53, 111 sensory neuronopathies, 251 sensory neurons, 51–52 sensory receptor reinnervation, 220 sensory response amplitude, 135, 216 axon loss and, 135–136 value of, lesion severity and, 218 series circuits, 13–14 series DC circuits, 14–18 SFEMG. See single fiber EMG shock artifact. See stimulus short circuit, 15 shoulder fracture-dislocation, exercise, 327 EDX study conclusion in, 329 NCS in, 327–328 needle EMG in, 328–329 signal clean, 46

526

common, 47 desired, undesired, in EMG machine, 33 difference, 46 in EDX study, 34–35 mixed, 37–38 source, 286 signal source generator, 46 signal-to-noise ratio (SNR), 35, 47, 277 silent contractures, 184 single fiber EMG (SFEMG), 204, 207–208 criteria, for assessment, 206 for muscle fiber APs, 204–205 needle electrodes, 204–205 for NMJ transmission disorders, 206 utility of, 206 single-phase power, 28 sink, 91–92 sixth-degree injury, 224 60 Hz power artifact, AC signal of, 276–278 skeletofusimotor neurons, 76 skeletomotor neurons, 76 skin impedance, 273 preparation, 44–45, 88–89 resistance, 44, 293 in transmission, of infection, 298 SMA-3. See spinal muscular atrophy type 3 SNAP. See sensory nerve action potential snap-crackle-pop, 170 SNR. See signal-to-noise ratio solid angle analysis, 92 spatial recruitment, 169, 198–199 spike form, of fibrillation potential, 185–186 spinal cord level, motor response from, 138 myotome and, 138 spinal muscular atrophy type 3 (SMA-3), 238–239 spiral groove, DMCB and, 128 spontaneous activity, in needle EMG, 162, 170–171, 184–185, 202 standard precautions, 296 staphylococcus aureus, 298 static, 6 static electricity, 4, 7–8 steady state, 33 stimulating electrodes, 93–94 stimulation. See also nerve stimulation; repetitive nerve stimulation studies above lesion, 131 below lesion, 131 excessive, 275–276 machine maximum, 276

pitfalls of, 271–281 sites of DMCB, 130 proximal, 108, 111–112 submaximal, 274–275, 279 stimulators constant current, 44, 272 constant voltage, 44, 272 current and, 43–44 depolarization by, 43–44 on EMG machines, 44 reversal of, 279 skin resistance to, 44 stimulus. See also nerve stimulation lead, 89, 93, 276 maximum response to, 93 spread, 276 strength, 89–90, 93 supramaximal, 93 stimulus artifact, 271–272 reduction, 272–273 anode rotation, about cathode in, 273–275 final options in, 274 future of, 274 stopband filter, 40. See also notch filter stretch injuries, 228 submaximal stimulation, 274–275, 279 Sunderland classification system, 222–223, 225–226 grade 1, 222 grade 2, 222–223 grade 3, 223 grade 4, 223 grade 5, 223–224 grade 6, 224 supervision, of EDX study, 306–307 supine patient, in H reflex study, 142 supraclavicular plexus, 243 supramaximal stimulus, 93 surface recording electrodes adhesive, nonadhesive, 89 basic technique of, 88–90 impedance, between signal source and, 286 in NCS, 87 proper placement of, 87–88 response to, 89–90 surface stimulating electrodes, 88 surgical intervention approach, to axon loss in, 225–226 in nerve injuries, 224–225 surgical interventions, major, 226–227 suturing, end-to-end, 226–227 sweep speed, 283–284 switches, 15 switchyards, 27

Index

symmetric weakness, exercise, 394, 412, 424–425 NCS in, 394–395, 412–413, 425–426 needle EMG in, 395–397, 413–415, 426–428 synapses, 68 synaptic space, 68–69 synchronized slowing. See uniform DMCS synchrony, 103 synthesized ACh, 69 tail, 75 technical errors in H wave testing, 143–144 in low frequency RNSS, 156–157 temporal dispersion motor NCS and, 131 physiologic, 99–100, 113–114, 131 temporal recruitment, 169, 198–199 terminal boutons, 68–69 terminal branch conduction time, 107 terminal cisternae, 74 terminal positive phase, 91 territorial distribution, 51 thick filaments, 75–76 thin filaments, 75–76 three-phase power, 28–29 threshold time, 106 tibial nerve stimulation, in H reflex study, 142–144 tibial response, amplitude, 100 time backfire, 146–147 of conduction velocity, 107 constant in capacitors, 22, 60 myelin sheath and, 125 constraints, as pitfalls, 284 cursors, on EMG machines, 285–286 difference, 277–278 discharge, 125 distal latency and, 106–107 of muscle activation, 107 of nerve activation, 106 of NMJ transmission, 107 of proximal conduction, 147 of terminal branch conduction, 107 threshold, 106 tissue transit, 106 tissue resistance of, 291 transit time, 106 TMP. See transmembrane potential total membrane capacitance, 60 total membrane resistance, 60 tourniquet paralysis, 230 traction injuries, 228

trailing source current, 91 transection injuries, 231 transformers, 27–31 transient axonal conduction block, 131–132, 216 transistor batteries, 11 transition band, 39 transmembrane current, 60 transmembrane potential (TMP), 43, 53, 185. See also resting membrane potential cell membrane and, 60 equilibrium potential and, 55–56 ion movement and, 56 of resting neuron, 57 transmission feeders, 27 transverse current, 62 transverse tubules, 74 trapezius muscle, 299 triad, 74 triboelectric effect, 6 trigger line, of MUAP, 176 triggering potential, of muscle fiber APs, 205 tropomyosin, 74–75 troponin, 74–75 turns, of MUAP, 177–178, 195 20:20 rule, 100 twisted pair leadwire, 46 ulnar motor response, 96–97, 129 ulnar neuropathy, 263, 266–267. See also neck pain exercise, 338, 381 EDX study conclusion in, 341–342, 383–384 NCS in, 339–340, 381–382 needle EMG in, 341, 383 ulnar-to-median anastomosis, 268 ultrahigh voltage, 29 UMN. See upper motor neuron undesired signal, 33, 35, 45–46. See also noise uniform DMCS, 126–127 unipennate muscle, 80 universal precautions, 296 upper extremity weakness and wasting, exercise, 436 EDX study conclusion in, 439 NCS in, 436–437 needle EMG in, 437–438 upper motor neuron (UMN), 50, 76. See also intraspinal canal disorders ALS and, 238 disorders, 235–236 MUAP recruitment in, 235–236 NCS in, 235 needle EMG in, 235

recruitment, 198–199 upper plexus, 243–244 vacuum tube, 42 velocity. See conduction velocity; drift velocity verbal informed consent, 308–309 vertical resolution, 42–43, 283 vesicles, of ACh, 69, 71–72, 152–153 VGCCs. See voltage-gated calcium channels VGKCs. See voltage-gated potassium channels VGNCs. See voltage-gated sodium channels voltage. See also constant voltage stimulators; direct current; Goldman-Hodgkin-Katz voltage equation; Kirchhoff’s voltage law; root-mean-square voltage value in AC, 23–24 in batteries, 10–11, 18 in capacitors, 37 charge separation and, 22 in circuit, 10, 17 dividers, as resistors, 16–17 in EDX medicine, 10 in electricity, 10–11 EMF and, 12 in EMG laboratory, 46 low, ultrahigh, 29 negative peak, 25 Ohm’s law and, 15 output, 17 peak, 25, 37 peak-to-peak, 25 resistance and, 15–16, 18 ultrahigh, 29 voltage-gated calcium channels (VGCCs), 68, 253 voltage-gated channels, 54 voltage-gated potassium channels (VGKCs), 54, 58 voltage-gated sodium channels (VGNCs), 54, 58–59 activation gate of, 258 in bidirectional propagation, 74 in depolarization, 61 volume conduction, 90–92 voluntary activity in MUAP assessment, 193–194 in needle EMG, 162, 170, 173 Wallerian degeneration, 123–124, 131, 134 in axon disruption, 137–138 conduction block in, 216 at lesion site, 137–138

527

Index

water, flow of, 12–14, 18–19 wave, period of, 24–25 weakness. See bilateral lower extremity weakness; bilateral upper extremity weakness; generalized weakness; symmetric weakness; upper extremity weakness and wasting

528

weakness and numbness, exercise, 448, 463, 465, 468, 472, 475 EDX conclusion in, 451–452, 465, 468, 471–472, 475, 478 NCS in, 448–450, 463–464, 466–467, 469–470, 472–477 needle EMG in, 450–451, 464–465, 467–468, 470–471, 474–475, 477–478

Weichers-Johnson syndrome, 170–171 weight, 262 X-linked recessive bulbospinal neuronopathy, 239 zinc-carbon batteries, 11 Z-line, 75